- Email: [email protected]
Involvement of SPARC in in Vitro Differentiation of Skeletal Myoblasts
Biochemical and Biophysical Research Communications 271, 630 – 634 (2000) doi:10.1006/bbrc.2000.2682, available online at http://www.idealibrary.com on
Involvement of SPARC in in Vitro Differentiation of Skeletal Myoblasts Won Jin Cho,* Eun Ju Kim,* Soo Jung Lee,* Han Do Kim,* Hae Ja Shin,† and Woon Ki Lim* ,1 *Department of Molecular Biology, College of Natural Sciences, Pusan National University, Pusan 609-735, Korea; and †Department of Environmental Engineering, Dongseo University, Pusan 616-716, Korea
Received April 8, 2000
SPARC (secreted protein acidic and rich in cysteine) is an extracellular Ca 2ⴙ-binding glycoprotein associated with the morphogenesis and remodeling of various tissues. Here, involvement of SPARC in the myogenesis of skeletal myoblasts was investigated in vitro. First, the differential expression of SPARC mRNA during the myogenesis was initially identified by a differential display reverse transcription (DDRT)-PCR method. The expression of the SPARC gene was significantly up-regulated during the differentiation of C2C12 mouse myoblasts. Second, the treatment with anti-SPARC antibody almost completely prevented the differentiation of myoblasts. Third, the treatment with EGTA, a Ca 2ⴙ chelator that is known to inhibit the fusion of C2C12 myoblasts, reversibly inhibited the up-regulation of SPARC gene expression. On the other hand, the treatment with A23187, a Ca 2ⴙ ionophore, rapidly and dramatically increased the level of SPARC transcript. Taken together, these results suggest that SPARC may play a critical role(s) in the morphological change of myoblasts, and that the expression of SPARC gene may be controlled by Ca 2ⴙ-dependent pathway in myogenesis. © 2000 Academic Press Key Words: SPARC; myogenesis; Ca 2ⴙ; morphogenesis; C2C12 myoblast; DDRT-PCR.
The differentiation of myoblasts into skeletal muscle is accompanied by the dramatic changes of cellular morphology such as the formation of bipolar myoblast and multinucleated myotube (1). Although our understanding of the molecular basis of myoblast differentiation has appreciably increased in recent years compared to that for other tissues, its many aspects still remain to be elucidated; for example, morphological remodeling. SPARC (secreted protein acidic and rich in cysteine, also known as osteonectin or BM 40) is a 1 To whom correspondence should be addressed at Department of Molecular Biology, College of Natural Sciences, Pusan National University, Jang Jeon-dong, Keum Jeong District, Pusan 609-735, Korea. Fax: 82-51-513-9258. E-mail: [email protected].
0006-291X/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
specific class of extracellular matrix (ECM)-associated components that regulate interactions between cells and ECM (2). It is composed of three structural domains; acidic, follistatin-like, and C-terminal extracellular Ca 2⫹-binding domain. One or more of these are implicated in the regulation of cell adhesion, proliferation and matrix synthesis/turnover. Therefore, SPARC is important in tissue remodeling and repair for developing embryos and adults as well as in invasiveness for angiogenesis (3). On the other hand, ECM proteins such as laminin and fibronectin have been reported to play a crucial role in skeletal muscle differentiation (4 –7). Recently, several lines of evidence have demonstrated that extracellular factors essential for myogenesis influence the initial specification and patterning events as well as the terminal differentiation, and that ECM is important as a part of the signaling mechanism in myogenesis (8, 9). In addition, it has previously been reported that the expression of SPARC is significantly high in some muscle tissues from fetuses and adults of mice and humans (10, 11). There is also a report that appropriate levels of SPARC are essential for normal muscle function in C. elegance (12). However, involvement of SPARC in myogenesis has not been clearly addressed yet. In the present paper, we demonstrate that SPARC plays critical roles in myoblast differentiation. The expression of SPARC gene was markedly up-regulated during in vitro myogenesis, and SPARC protein was required for myogenesis. The regulation of SPARC gene expression was influenced by both intra- and extra-cellular Ca 2⫹ concentrations in the same manner that the fusion of cells was affected by Ca 2⫹ level. Thus, it appears likely that one or more functions of SPARC may be involved in the various steps of skeletal myogenesis. MATERIALS AND METHODS Materials. Eagle’s minimum essential medium (MEM), Dulbecco’s MEM (DMEM), fetal bovine serum (FBS), horse serum, antibiotics,
630
Vol. 271, No. 3, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
adapter primer, 5⬘ RACE system (Version 2.0), and TRIZOL reagent were purchased from GIBCO BRL (Gaithersburg, MD). Zeta-probe membrane was from Bio-Rad (Hercules, CA). Monoclonal antibody anti-osteonectin (SPARC) was purchased from Haematologic Technologies Inc. (Burlington, VT). Rediprime DNA labeling system, 35 S-␣-dATP, and 32P-␣-dCTP were obtained from Amersham Co. (Piscataway, NJ). All primers were obtained from Korea Basic Science Center (Daejeon, Republic of Korea). MMLT reverse transcriptase and pGEM-T vector were from Promega (Madison, WI). Taq polymerase was purchased from Takara (Japan). All other reagents including A23187 were purchased from Sigma (St. Louis, MO). C2C12 myoblast cell line was kindly provided by Prof. M. S. Kang (Seoul National University, Korea). Myoblast cultures. Primary culture of myoblasts from the breast muscles of 12 day-old chick embryos was performed as described elsewhere (13). The myoblasts were plated at a concentration of 5 ⫻ 10 5 cells/ml on collagen-coated dishes in MEM containing 10% horse serum, 10% chick embryo extract, and 1% penicillin/streptomycin at 37°C in an atmosphere of 5% CO 2. The culture medium was changed 24 h after plating with the same medium but containing 2% chick embryo extract. Mouse myoblast C2C12 cells were grown in DMEM containing 10% FBS (growth medium, GM) for the proliferation, and induced to differentiate at 80 –90% confluence by being transferred into DMEM with 2% horse serum (differentiation medium, DM) (14). Thereafter medium was changed every other day. Treatment with antibody, ionophore or EGTA. Monoclonal antibody anti-osteonectin (SPARC) (10 g/ml) was directly added into DM for C2C12. Because anti-SPARC antibody was kept stored in 50% glycerol, control cells were treated with the same amounts of 50% glycerol. The media containing anti-SPARC antibody or glycerol was replaced every 12 h during the differentiation period (up to 4 days). A23187 (1 mM) was dissolved in dimethyl sulfoxide (DMSO), stored at ⫺20°C, and added at 0.5 or 1 M into media 12 h after the cells were switched to DM. To examine the effect of lowered [Ca2⫹] on the expression of SPARC, EGTA at 1.4 mM was added into medium. EGTA below 1.4 mM was not fully effective in blocking fusion, whereas EGTA at 1.8 mM had a tendency to make cells detached from the substratum (data not shown). The culture medium was recovered to normal Ca2⫹ level by substituting with the medium containing no EGTA. Measurement of cell fusion. Cultured cells were rinsed twice with PBS, fixed for 5 min in methanol, stained with hematoxylin for 5 min, and rinsed with water. Nuclei were scored under light microscopy. Cells were considered to be fused only if there was clear cytoplasmic continuity and at least three nuclei were present in each myotube. Each value represents the average of at least 10 randomly selected fields with over 300 in the total number. RNA isolation. RNA was isolated from the cells by using TRIZOL Reagent as described in the manufacturer’s instruction. All samples were above 1.7 in 260/280 absorbance ratio. Northern blot. 30 g of RNA was electrophoresed in a 1.5% denaturing formaldehyde-formamide gel, and blotted onto Zeta filters. RNA was hybridized with SPARC cDNA probe overnight at 43°C in 120 mM Na 2HPO 4 (pH 7.2), 50% formamide, 250 mM NaCl, 7% SDS, and 1 mM EDTA. SPARC cDNA was labeled with 32P-␣ -dCTP using Rediprime DNA labeling system (Amersham Life Science). The membrane was washed successively for 30 min in each of the following solutions; 2⫻ SSC/0.1% SDS, 0.5⫻ SSC/0.1% SDS and 0.1⫻ SSC/0.1% SDS, followed by the analysis using a Fuji Bas 1500 Bio-Imaging Analyzer (Fuji Photo Film). Mouse SPARC cDNA (a 916 bp fragment) (15) used as a probe was amplified by revere transcription-PCR with RNA prepared from differentiated C2C12 cells. The muscle creatine kinase (MCK) and glyceraldehyde-3phosphate dehydrogenase (GAPDH) cDNA fragments were also made by PCR from chicken myoblasts, and used as probes for Northern blot (16, 17). 28S and 18S rRNAs were stained with ethidium bromide and used as internal controls.
FIG. 1. Representative differential display of cDNAs and Northern blot analysis for SPARC mRNA from the primary culture of chicken myoblasts. (A) DDRT-PCR was carried out using RNA isolated from the cells grown for 12 and 48 h as described under Materials and Methods. 35 S-labeled PCR products were separated on a 6% polyacrylamide gel with 7 M urea. The bands with differential (filled triangle, MD-23) and no differential expression (open triangles) are shown. (B) Northern blot analysis was performed using RNAs from the primary culture of myoblasts grown from 12 to 72 h in the medium containing a low concentration of embryo extracts. Chick MD-23 (SPARC fragment) and MCK (muscle creatine kinase) cDNAs were used as probes. The amount of RNA was verified using GAPDH (glyceraldehyde-3 phosphate dehydrogenase) mRNA as an internal control. Differential display reverse transcription (DDRT)-PCR and 5⬘ rapid amplification of cDNA ends (5⬘ RACE). Two oligonucleotides were used in this study; 5⬘-GCGCGGATCC(T 10)GT-3⬘ as a 3⬘-(Trich)-primer, 5⬘-CCATGGATCCTATGGTAAAGGG-3⬘ as a 5⬘-arbitrary primer. The underline indicates the BamHI restriction site. Total RNAs were isolated from the chick myoblasts cultured for 12 h and 48 h in the medium containing 2% chick embryo extract. RNA was reversibly transcribed, and amplified by a Perkin-Elmer 2400 thermocycler by 40 cycles (30 s at 94°C, 2 min at 40°C, and 30 s at 72°C). Differentially expressed products were ascertained on gel, eluted, and reamplified by the same set of primers. The reamplified product was subcloned into pGEM-T vector (Promega), sequenced, and used as a probe for Northern blot to confirm the differential expression. When searched by BLAST, there was no sequence homologous to initial clone of MD-23 (150 bp). To obtain a full length of this gene, 5⬘ RACE (18) was performed using 5⬘ RACE kit (Version 2.0, GIBCO BRL) as described in the manufacturer’s instruction. Two sequential antisense primers, 5⬘-TAACAGAG ACTCATTGGGTTGC-3⬘ and 5⬘GGCCCAGGGAACATTCAGAGG-3⬘ were designed from the sequence of the cDNA fragment obtained by DDRT-PCR (19), and used for PCR reaction. The PCR product was subcloned into pGEM-T vector, and sequenced by using an automatic DNA sequencer (Perkin Elmer). The full length of SPARC cDNA (2,021 bp) including 896 bp of coding region was obtained. It turned out that its sequence was exactly identical to that of chick SPARC cDNA (Accession No. L24906) deposited in GenBank, and that the initial clone, MD-23 was the 3⬘ noncoding region.
RESULTS Identification of SPARC involved in chick skeletal muscle differentiation by DDRT-PCR method. A mRNA differential display method was utilized to search for
631
Vol. 271, No. 3, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. Northern blot analysis for SPARC mRNA in mouse myoblast C2C12 cells cultured in GM or DM. C2C12 cells were grown in GM for the cell proliferation or induced to differentiate in DM. Total RNAs were isolated from C2C12 cells grown for 2 or 4 days in GM or DM. Northern blot analysis was carried out using mouse SPARC cDNA as a probe. 28S and 18S rRNAs were used as internal controls. Upon three separate experiments, SPARC mRNA was found to increase 2.3-fold at day 2 and 3-fold at day 4 in DM, but to remain unchanged in GM.
new genes implicated in skeletal muscle formation. Myoblasts from the breasts of 12 day-old chick embryos underwent dramatic changes in morphology and gene expression between 24 and 60 h after the culture in the medium containing a low concentration of embryo extracts. Accordingly, cells at 12 h were harvested as undifferentiated myoblasts and cells at 48 h were as differentiating myoblasts. RNAs from these cells were extracted, and applied to the PCR-based differential screening. One of results from the experiment is shown in Fig. 1A, in which the amount of MD-23 increases at 48 h while the other several genes remain unchanged. A full-length of cDNA for MD-23 was cloned by 5⬘ RACE method. The amplified product (2021 bp) was subcloned into a plasmid vector and was sequenced (data not shown). It turned out that the cloned cDNA was exactly identical to the chicken SPARC cDNA deposited in the sequence databases (15). By Northern blot, with this cDNA as a probe, the expression of SPARC mRNA was examined during the differentiation of myoblasts from the breast muscles of 12 day-old chicks (Fig. 1B). SPARC mRNA began to increase around 36 h and kept increasing up to 72 h, and this change was in parallel with the morphological changes of myoblasts. Expression of muscle creatine kinase (MCK) was also increased during this period as reported elsewhere (20). These results suggest that SPARC may be involved in myoblast differentiation.
Expression of SPARC mRNA is up-regulated during C2C12 myoblast differentiation. Because the primary culture of chick myoblast could be contaminated with other kinds of cells, mainly fibroblast, we examined the expression of SPARC mRNA in the myogenesis using mouse myoblast C2C12 cells (Fig. 2). It has been known that mouse myoblast C2C12 cells proliferate in serum-rich growth medium (GM), and that when cultured at a high cell density in mitogen-deficient medium (DM), they withdraw from the cell cycle, express muscle-specific genes, and fuse to form multinucleated myotube (21). But the differentiation process is much slower in C2C12 cells than in the primary culture of chick myoblasts: C2C12 myoblast fusion begins to occur 2 days after the culture in DM, and about 20 –30% of them fuses to form multinucleated myotubes at day 4 (see Fig. 4A). As in the primary culture of chick myoblast, the expression of SPARC mRNA was upregulated in the differentiation of C2C12 myoblast (Fig. 2). The amount of SPARC mRNA in this cell increased about 3-fold at 4 days in DM, while it remained unaltered in GM. This result confirms the upregulation and possible involvement of SPARC in myoblast differentiation. Addition of anti-SPARC antibody into DM inhibits the myoblast fusion. SPARC is a secreted glycoprotein interacting with the extracellular matrix (ECM) and other components of cells. In order to examine the role(s) of SPARC in myogenesis, SPARC present in the culture medium was inactivated by the addition of an anti-SPARC antibody to DM (Fig. 3). Formation of the SPARC and antibody complex could inactivate SPARC in culture medium, which might be either originated from serum or secreted from the cells. Four days after the medium change with normal DM, the cells underwent significant differentiation, forming myotube (Fig. 3B). On the other hand, the cells treated with the anti-SPARC antibody scarcely formed the myotube. In addition, most of these cells showed polygonalshape rather than bipolar shape, which is typical at the onset stage of myoblast differentiation (Figs. 3A and 3C). Therefore it seems likely that SPARC plays a critical role in the early stage of myoblast differentiation.
FIG. 3. Effect of anti-SPARC antibody on differentiation of C2C12 myoblast. C2C12 cells were proliferated in GM (A) and then cultured for 4 days in DM without (B) or with the addition of 10 g/ml of the monoclonal anti-SPARC antibody (C). The medium containing anti-SPARC antibody or glycerol was replaced every 12 h. Bar, 250 m. 632
Vol. 271, No. 3, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
level by substituting with DM containing no EGTA, the expression of SPARC mRNA started to increase again. Moreover, this reversible effect of EGTA is well-correlated with its effect on myoblast fusion (see Fig. 4A). The influx of Ca 2⫹ into the cytosol appears to be required for myoblast fusion (22). It was shown that the treatment with A23187 stimulated the fusion of both chick and rat myoblasts (24). To test whether the expression of SPARC mRNA in C2C12 myoblast differentiation is affected by cytoplasmic Ca 2⫹ level, A23187 was added to DM (Fig. 5). The expression of SPARC mRNA was rapidly and remarkably increased by the treatment with the drug: the expression of SPARC was enhanced 6- to 7-fold within 1 h in the presence of 0.5 M A23187. These results again suggest the idea that SPARC may play an important role in myogenesis, and that the expression of SPARC mRNA may be controlled by Ca 2⫹-dependent pathway. DISCUSSION In the present paper, we demonstrate that SPARC is involved in myoblast differentiation. The expression of SPARC mRNA was markedly up-regulated during myogenesis, and the treatment with an anti-SPARC antibody completely prevented myoblasts (Figs. 1–3). The treatment with EGTA, which is known to inhibit the fusion of myoblasts, reversibly blocked the up-
FIG. 4. Effect of EGTA on the myoblast fusion and the expression of SPARC mRNA during myogenesis. C2C12 cells were proliferated in GM and then cultured for 8 days in DM with (closed circle) or without the addition of 1.4 mM EGTA (open circle). To test the reversibility, cells were kept for 4 days in EGTA-containing medium and then grown in normal medium (open rectangle). At the indicated day, the cells were stained with hematoxylin to score fusion index (A) or harvested to examine the expression of SPARC mRNA by Northern blot (B). Cells were considered to be fused only if at least three nuclei were present in each myotube. For Northern blot, 30 g of total RNA was used, and 28S and 18S rRNAs were used as internal controls. The amount of SPARC mRNA was averaged from three separate experiments. Values are shown as mean ⫾ standard deviation.
Effect of EGTA or A23187 on the expression of SPARC mRNA during C2C12 myoblast differentiation. Myoblast fusion could be blocked by the deprivation of Ca 2⫹ in the culture medium, and rapidly recovered upon the re-addition of Ca 2⫹ (22, 23) (Fig. 4A). We examined the effect of the Ca 2⫹ level in the medium on the expression of SPARC mRNA of C2C12 myoblast (Fig. 4B). When Ca 2⫹ in DM was deprived by the addition of EGTA, the expression of SPARC mRNA remained at the basal level up to 8 days. In addition, when the medium was added with Ca 2⫹ to a normal
FIG. 5. Effect of A23187 on the expression of SPARC mRNA in C2C12 myoblasts. C2C12 cells were cultured for 12 h in DM and then treated with 0.5 M or 1.0 M A23187 for 30 min or 1 h. Total RNA was isolated and Northern blotting was carried out. In the bottom graph, the amount of SPARC mRNA from cells treated with A23187 for 0.5 h (gray bar) or 1 h (black bar) was compared with that from untreated cells (white bar), using three separate experiments. Values are shown as means ⫾ standard deviation.
633
Vol. 271, No. 3, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
regulation of SPARC mRNA expression (Fig. 4). Last, A23187, which has been known to stimulate myoblast differentiation, rapidly and dramatically increased the level of SPARC transcript (Fig. 5). The myogenesis occurs through a series of steps [1]. After withdrawal from the cell cycle, cells become bipolar and then align along their long axes. The aligned plasma membranes come in close apposition and local membrane fusion occurs to form single multinucleated myotubes. When SPARC in culture media was inactivated by treatment with an anti-SPARC antibody, the cells no longer could fuse and showed polygonal-shape rather than bipolar shape (Fig. 3). Therefore, it appears that SPARC plays a role in the fusion or the step preceding fusion during myoblast differentiation. The expression of SPARC is significantly increased in a variety of tissues undergoing remodeling (3). Multiple functions of SPARC have been known; inhibition of cell spreading and disassembly of focal adhesions, inhibition of cell proliferation, and regulation of expression of various ECMs and proteinases (2). These functions of SPARC all apparently seem to be relevant to myogenesis of embryonic skeletal muscle, because this process encompasses the dramatic remodeling processes such as withdrawal from the cell cycle, formation of bipolar shape cells, spreading and adherence of the cells to each other, membrane fusion, and formation of multinucleated myotube (1). For example, some isoforms of PDGF (platelet-derived growth factor) enhanced the proliferation of myoblasts and partly inhibited their differentiation (25). It was also shown in other tissues that SPARC could interact with some isoforms of PDGF and inhibit the binding of PDGF to its receptors (2, 26). Therefore, it is possible that SPARC may inhibit PDGF in stimulating the proliferation of myoblasts. Expression of muscle-specific genes is regulated primarily by muscle-specific transcription factors, MyoD family (27). Although the treatment with EGTA blocked the myoblast fusion, it did not interfere with the induction of muscle-specific mRNAs such as for muscle creatine kinase and acetylcholine esterase receptors (28, 29). Ca 2⫹ is required for the fusion process, but not for the function of MyoD family. Since the expression of SPARC mRNA was neither influenced by EGTA treatment, it seems that MyoD family does not control the expression of SPARC. It was shown here that the treatments with A23187 to myoblasts markedly induced the expression of SPARC mRNA: the expression of SPARC mRNA was enhanced 6- to 7-fold within 1 h (Fig. 5). So far, there were few reports on the mechanism how the expression of SPARC gene was regulated. This study should be the first report demonstrating that the expression of SPARC mRNA could be controlled by cytosolic Ca 2⫹dependent pathway.
ACKNOWLEDGMENT This study was supported by the academic research fund of Ministry of Education, Republic of Korea (BSRI-97-4409).
REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.
15.
16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29.
634
Wakelam, M. J. (1985) Biochem. J. 228, 1–12. Motamed, K. (1999) Int. J. Biochem. Cell Biol. 31, 1363–1366. Lane, T. F., and Sage, E. H. (1994) FASEB J. 8, 163–173. Buck, C. A., and Horwitz, A. F. (1987) Annu. Rev. Cell Biol. 3, 179 –205. Clark, P., Coles, D., and Peckham, M. (1997) Exp. Cell Res. 230, 275–283. Chung, C. Y., and Kang, M. S. (1990) J. Cell Physiol. 142, 392– 400. Gullberg, D., Sjoberg, G., Velling, T., and Sejersen, T. (1995) Exp. Cell Res. 220, 112–123. Cossu, G., Kelly, R., Tajbakhsh, S. D., Donna, S., Vivarelli, E., and Buckingham, M. (1996) Development 122, 429 – 437. Pownall, M. E., Strunk, K. E., and Emerson, C. P., Jr. (1996) Development 122, 1475–1488. Sage, H., Vernon, R. B., Decker, J., Funk, S., and Iruela-Arispe, M. L. (1989) J. Histochem. Cytochem. 37, 819 – 829. Porter, P. L., Sage, E. H., Lane, T. F., Funk, S. E., and Gown, A. M. (1995) 43, 791– 800. Schwarzbauer, J. E., and Spencer, C. S. (1993) Mol. Biol. Cell 4, 941–952. Kim, H. S., Lee, I. H., Chung, C. H., Kang, M. S., and Ha, D. B. (1992) Dev. Biol. 150, 223–230. Blau, H. M., Pavlath, G. K., Hardeman, E. C., Chiu, C. P, Silberstein, L., Webster, S. G., Miller, S. C., and Webster, C. (1985) Science 230, 758 –766. Bassuk, J. A., Iruela-Arispe, M. L., Lane, T. F., Benson, J. M., Berg, R. A., and Sage, E. H. (1993) Eur. J. Biochem. 218, 117– 127. Myer, A., Wagner, D. S., Vivian, J. L., Olson, E. N., and Klein, W. H. (1997) Dev. Biol. 185, 127–138. Chinni, C., de Niese, M. R., Tew, D. J., Jenkins, A. L., Bottomley, S. P., and Mackie, E. J. (1999) J. Biol. Chem. 274, 9169 –9174. Frohman, M. A., Dush, M. K., and Martin, G. R. (1988) Proc. Natl. Acad. Sci. USA 85, 8998 –9002. Liang, P., and Pardee, A. B. (1992) Science 257, 967–971. Morris, G. E., and Cole, R. J. (1977) FEBS Lett. 79, 183–187. Blau, H. M., Chiu, C. P., and Webster, C. (1983) Cell 32, 1171–1180. David, J. D., See, W. M., and Higginbotham, C. A. (1981) Dev. Biol. 82, 297–307. Shainberg, A., Yagil, G., and Yaffe, D. (1969) Exp. Cell Res. 58, 163–167. Schollmeyer, J. E. (1986) Exp. Cell Res. 162, 411– 422. Ye, W. V., McFarland, D. C., Gilkerson, K. K., and Pesall, J. E. (1996) Cytobios 88(352), 53– 62. Jendraschak, E., and Sage, E. H. (1996) Semin. Cancer Biol. 7, 139 –146. Emerson, C. P. (1990) Curr. Opin. Cell Biol. 2, 1065–1075. Endo, T., and Nadal-Ginard, B. (1987) Cell 49, 515–526. Hu, J. S., and Olson, E. N. (1990) J. Biol. Chem. 265, 7914 –7919.
Involvement of SPARC in in Vitro Differentiation of Skeletal Myoblasts Won Jin Cho,* Eun Ju Kim,* Soo Jung Lee,* Han Do Kim,* Hae Ja Shin,† and Woon Ki Lim* ,1 *Department of Molecular Biology, College of Natural Sciences, Pusan National University, Pusan 609-735, Korea; and †Department of Environmental Engineering, Dongseo University, Pusan 616-716, Korea
Received April 8, 2000
SPARC (secreted protein acidic and rich in cysteine) is an extracellular Ca 2ⴙ-binding glycoprotein associated with the morphogenesis and remodeling of various tissues. Here, involvement of SPARC in the myogenesis of skeletal myoblasts was investigated in vitro. First, the differential expression of SPARC mRNA during the myogenesis was initially identified by a differential display reverse transcription (DDRT)-PCR method. The expression of the SPARC gene was significantly up-regulated during the differentiation of C2C12 mouse myoblasts. Second, the treatment with anti-SPARC antibody almost completely prevented the differentiation of myoblasts. Third, the treatment with EGTA, a Ca 2ⴙ chelator that is known to inhibit the fusion of C2C12 myoblasts, reversibly inhibited the up-regulation of SPARC gene expression. On the other hand, the treatment with A23187, a Ca 2ⴙ ionophore, rapidly and dramatically increased the level of SPARC transcript. Taken together, these results suggest that SPARC may play a critical role(s) in the morphological change of myoblasts, and that the expression of SPARC gene may be controlled by Ca 2ⴙ-dependent pathway in myogenesis. © 2000 Academic Press Key Words: SPARC; myogenesis; Ca 2ⴙ; morphogenesis; C2C12 myoblast; DDRT-PCR.
The differentiation of myoblasts into skeletal muscle is accompanied by the dramatic changes of cellular morphology such as the formation of bipolar myoblast and multinucleated myotube (1). Although our understanding of the molecular basis of myoblast differentiation has appreciably increased in recent years compared to that for other tissues, its many aspects still remain to be elucidated; for example, morphological remodeling. SPARC (secreted protein acidic and rich in cysteine, also known as osteonectin or BM 40) is a 1 To whom correspondence should be addressed at Department of Molecular Biology, College of Natural Sciences, Pusan National University, Jang Jeon-dong, Keum Jeong District, Pusan 609-735, Korea. Fax: 82-51-513-9258. E-mail: [email protected].
0006-291X/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
specific class of extracellular matrix (ECM)-associated components that regulate interactions between cells and ECM (2). It is composed of three structural domains; acidic, follistatin-like, and C-terminal extracellular Ca 2⫹-binding domain. One or more of these are implicated in the regulation of cell adhesion, proliferation and matrix synthesis/turnover. Therefore, SPARC is important in tissue remodeling and repair for developing embryos and adults as well as in invasiveness for angiogenesis (3). On the other hand, ECM proteins such as laminin and fibronectin have been reported to play a crucial role in skeletal muscle differentiation (4 –7). Recently, several lines of evidence have demonstrated that extracellular factors essential for myogenesis influence the initial specification and patterning events as well as the terminal differentiation, and that ECM is important as a part of the signaling mechanism in myogenesis (8, 9). In addition, it has previously been reported that the expression of SPARC is significantly high in some muscle tissues from fetuses and adults of mice and humans (10, 11). There is also a report that appropriate levels of SPARC are essential for normal muscle function in C. elegance (12). However, involvement of SPARC in myogenesis has not been clearly addressed yet. In the present paper, we demonstrate that SPARC plays critical roles in myoblast differentiation. The expression of SPARC gene was markedly up-regulated during in vitro myogenesis, and SPARC protein was required for myogenesis. The regulation of SPARC gene expression was influenced by both intra- and extra-cellular Ca 2⫹ concentrations in the same manner that the fusion of cells was affected by Ca 2⫹ level. Thus, it appears likely that one or more functions of SPARC may be involved in the various steps of skeletal myogenesis. MATERIALS AND METHODS Materials. Eagle’s minimum essential medium (MEM), Dulbecco’s MEM (DMEM), fetal bovine serum (FBS), horse serum, antibiotics,
630
Vol. 271, No. 3, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
adapter primer, 5⬘ RACE system (Version 2.0), and TRIZOL reagent were purchased from GIBCO BRL (Gaithersburg, MD). Zeta-probe membrane was from Bio-Rad (Hercules, CA). Monoclonal antibody anti-osteonectin (SPARC) was purchased from Haematologic Technologies Inc. (Burlington, VT). Rediprime DNA labeling system, 35 S-␣-dATP, and 32P-␣-dCTP were obtained from Amersham Co. (Piscataway, NJ). All primers were obtained from Korea Basic Science Center (Daejeon, Republic of Korea). MMLT reverse transcriptase and pGEM-T vector were from Promega (Madison, WI). Taq polymerase was purchased from Takara (Japan). All other reagents including A23187 were purchased from Sigma (St. Louis, MO). C2C12 myoblast cell line was kindly provided by Prof. M. S. Kang (Seoul National University, Korea). Myoblast cultures. Primary culture of myoblasts from the breast muscles of 12 day-old chick embryos was performed as described elsewhere (13). The myoblasts were plated at a concentration of 5 ⫻ 10 5 cells/ml on collagen-coated dishes in MEM containing 10% horse serum, 10% chick embryo extract, and 1% penicillin/streptomycin at 37°C in an atmosphere of 5% CO 2. The culture medium was changed 24 h after plating with the same medium but containing 2% chick embryo extract. Mouse myoblast C2C12 cells were grown in DMEM containing 10% FBS (growth medium, GM) for the proliferation, and induced to differentiate at 80 –90% confluence by being transferred into DMEM with 2% horse serum (differentiation medium, DM) (14). Thereafter medium was changed every other day. Treatment with antibody, ionophore or EGTA. Monoclonal antibody anti-osteonectin (SPARC) (10 g/ml) was directly added into DM for C2C12. Because anti-SPARC antibody was kept stored in 50% glycerol, control cells were treated with the same amounts of 50% glycerol. The media containing anti-SPARC antibody or glycerol was replaced every 12 h during the differentiation period (up to 4 days). A23187 (1 mM) was dissolved in dimethyl sulfoxide (DMSO), stored at ⫺20°C, and added at 0.5 or 1 M into media 12 h after the cells were switched to DM. To examine the effect of lowered [Ca2⫹] on the expression of SPARC, EGTA at 1.4 mM was added into medium. EGTA below 1.4 mM was not fully effective in blocking fusion, whereas EGTA at 1.8 mM had a tendency to make cells detached from the substratum (data not shown). The culture medium was recovered to normal Ca2⫹ level by substituting with the medium containing no EGTA. Measurement of cell fusion. Cultured cells were rinsed twice with PBS, fixed for 5 min in methanol, stained with hematoxylin for 5 min, and rinsed with water. Nuclei were scored under light microscopy. Cells were considered to be fused only if there was clear cytoplasmic continuity and at least three nuclei were present in each myotube. Each value represents the average of at least 10 randomly selected fields with over 300 in the total number. RNA isolation. RNA was isolated from the cells by using TRIZOL Reagent as described in the manufacturer’s instruction. All samples were above 1.7 in 260/280 absorbance ratio. Northern blot. 30 g of RNA was electrophoresed in a 1.5% denaturing formaldehyde-formamide gel, and blotted onto Zeta filters. RNA was hybridized with SPARC cDNA probe overnight at 43°C in 120 mM Na 2HPO 4 (pH 7.2), 50% formamide, 250 mM NaCl, 7% SDS, and 1 mM EDTA. SPARC cDNA was labeled with 32P-␣ -dCTP using Rediprime DNA labeling system (Amersham Life Science). The membrane was washed successively for 30 min in each of the following solutions; 2⫻ SSC/0.1% SDS, 0.5⫻ SSC/0.1% SDS and 0.1⫻ SSC/0.1% SDS, followed by the analysis using a Fuji Bas 1500 Bio-Imaging Analyzer (Fuji Photo Film). Mouse SPARC cDNA (a 916 bp fragment) (15) used as a probe was amplified by revere transcription-PCR with RNA prepared from differentiated C2C12 cells. The muscle creatine kinase (MCK) and glyceraldehyde-3phosphate dehydrogenase (GAPDH) cDNA fragments were also made by PCR from chicken myoblasts, and used as probes for Northern blot (16, 17). 28S and 18S rRNAs were stained with ethidium bromide and used as internal controls.
FIG. 1. Representative differential display of cDNAs and Northern blot analysis for SPARC mRNA from the primary culture of chicken myoblasts. (A) DDRT-PCR was carried out using RNA isolated from the cells grown for 12 and 48 h as described under Materials and Methods. 35 S-labeled PCR products were separated on a 6% polyacrylamide gel with 7 M urea. The bands with differential (filled triangle, MD-23) and no differential expression (open triangles) are shown. (B) Northern blot analysis was performed using RNAs from the primary culture of myoblasts grown from 12 to 72 h in the medium containing a low concentration of embryo extracts. Chick MD-23 (SPARC fragment) and MCK (muscle creatine kinase) cDNAs were used as probes. The amount of RNA was verified using GAPDH (glyceraldehyde-3 phosphate dehydrogenase) mRNA as an internal control. Differential display reverse transcription (DDRT)-PCR and 5⬘ rapid amplification of cDNA ends (5⬘ RACE). Two oligonucleotides were used in this study; 5⬘-GCGCGGATCC(T 10)GT-3⬘ as a 3⬘-(Trich)-primer, 5⬘-CCATGGATCCTATGGTAAAGGG-3⬘ as a 5⬘-arbitrary primer. The underline indicates the BamHI restriction site. Total RNAs were isolated from the chick myoblasts cultured for 12 h and 48 h in the medium containing 2% chick embryo extract. RNA was reversibly transcribed, and amplified by a Perkin-Elmer 2400 thermocycler by 40 cycles (30 s at 94°C, 2 min at 40°C, and 30 s at 72°C). Differentially expressed products were ascertained on gel, eluted, and reamplified by the same set of primers. The reamplified product was subcloned into pGEM-T vector (Promega), sequenced, and used as a probe for Northern blot to confirm the differential expression. When searched by BLAST, there was no sequence homologous to initial clone of MD-23 (150 bp). To obtain a full length of this gene, 5⬘ RACE (18) was performed using 5⬘ RACE kit (Version 2.0, GIBCO BRL) as described in the manufacturer’s instruction. Two sequential antisense primers, 5⬘-TAACAGAG ACTCATTGGGTTGC-3⬘ and 5⬘GGCCCAGGGAACATTCAGAGG-3⬘ were designed from the sequence of the cDNA fragment obtained by DDRT-PCR (19), and used for PCR reaction. The PCR product was subcloned into pGEM-T vector, and sequenced by using an automatic DNA sequencer (Perkin Elmer). The full length of SPARC cDNA (2,021 bp) including 896 bp of coding region was obtained. It turned out that its sequence was exactly identical to that of chick SPARC cDNA (Accession No. L24906) deposited in GenBank, and that the initial clone, MD-23 was the 3⬘ noncoding region.
RESULTS Identification of SPARC involved in chick skeletal muscle differentiation by DDRT-PCR method. A mRNA differential display method was utilized to search for
631
Vol. 271, No. 3, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. Northern blot analysis for SPARC mRNA in mouse myoblast C2C12 cells cultured in GM or DM. C2C12 cells were grown in GM for the cell proliferation or induced to differentiate in DM. Total RNAs were isolated from C2C12 cells grown for 2 or 4 days in GM or DM. Northern blot analysis was carried out using mouse SPARC cDNA as a probe. 28S and 18S rRNAs were used as internal controls. Upon three separate experiments, SPARC mRNA was found to increase 2.3-fold at day 2 and 3-fold at day 4 in DM, but to remain unchanged in GM.
new genes implicated in skeletal muscle formation. Myoblasts from the breasts of 12 day-old chick embryos underwent dramatic changes in morphology and gene expression between 24 and 60 h after the culture in the medium containing a low concentration of embryo extracts. Accordingly, cells at 12 h were harvested as undifferentiated myoblasts and cells at 48 h were as differentiating myoblasts. RNAs from these cells were extracted, and applied to the PCR-based differential screening. One of results from the experiment is shown in Fig. 1A, in which the amount of MD-23 increases at 48 h while the other several genes remain unchanged. A full-length of cDNA for MD-23 was cloned by 5⬘ RACE method. The amplified product (2021 bp) was subcloned into a plasmid vector and was sequenced (data not shown). It turned out that the cloned cDNA was exactly identical to the chicken SPARC cDNA deposited in the sequence databases (15). By Northern blot, with this cDNA as a probe, the expression of SPARC mRNA was examined during the differentiation of myoblasts from the breast muscles of 12 day-old chicks (Fig. 1B). SPARC mRNA began to increase around 36 h and kept increasing up to 72 h, and this change was in parallel with the morphological changes of myoblasts. Expression of muscle creatine kinase (MCK) was also increased during this period as reported elsewhere (20). These results suggest that SPARC may be involved in myoblast differentiation.
Expression of SPARC mRNA is up-regulated during C2C12 myoblast differentiation. Because the primary culture of chick myoblast could be contaminated with other kinds of cells, mainly fibroblast, we examined the expression of SPARC mRNA in the myogenesis using mouse myoblast C2C12 cells (Fig. 2). It has been known that mouse myoblast C2C12 cells proliferate in serum-rich growth medium (GM), and that when cultured at a high cell density in mitogen-deficient medium (DM), they withdraw from the cell cycle, express muscle-specific genes, and fuse to form multinucleated myotube (21). But the differentiation process is much slower in C2C12 cells than in the primary culture of chick myoblasts: C2C12 myoblast fusion begins to occur 2 days after the culture in DM, and about 20 –30% of them fuses to form multinucleated myotubes at day 4 (see Fig. 4A). As in the primary culture of chick myoblast, the expression of SPARC mRNA was upregulated in the differentiation of C2C12 myoblast (Fig. 2). The amount of SPARC mRNA in this cell increased about 3-fold at 4 days in DM, while it remained unaltered in GM. This result confirms the upregulation and possible involvement of SPARC in myoblast differentiation. Addition of anti-SPARC antibody into DM inhibits the myoblast fusion. SPARC is a secreted glycoprotein interacting with the extracellular matrix (ECM) and other components of cells. In order to examine the role(s) of SPARC in myogenesis, SPARC present in the culture medium was inactivated by the addition of an anti-SPARC antibody to DM (Fig. 3). Formation of the SPARC and antibody complex could inactivate SPARC in culture medium, which might be either originated from serum or secreted from the cells. Four days after the medium change with normal DM, the cells underwent significant differentiation, forming myotube (Fig. 3B). On the other hand, the cells treated with the anti-SPARC antibody scarcely formed the myotube. In addition, most of these cells showed polygonalshape rather than bipolar shape, which is typical at the onset stage of myoblast differentiation (Figs. 3A and 3C). Therefore it seems likely that SPARC plays a critical role in the early stage of myoblast differentiation.
FIG. 3. Effect of anti-SPARC antibody on differentiation of C2C12 myoblast. C2C12 cells were proliferated in GM (A) and then cultured for 4 days in DM without (B) or with the addition of 10 g/ml of the monoclonal anti-SPARC antibody (C). The medium containing anti-SPARC antibody or glycerol was replaced every 12 h. Bar, 250 m. 632
Vol. 271, No. 3, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
level by substituting with DM containing no EGTA, the expression of SPARC mRNA started to increase again. Moreover, this reversible effect of EGTA is well-correlated with its effect on myoblast fusion (see Fig. 4A). The influx of Ca 2⫹ into the cytosol appears to be required for myoblast fusion (22). It was shown that the treatment with A23187 stimulated the fusion of both chick and rat myoblasts (24). To test whether the expression of SPARC mRNA in C2C12 myoblast differentiation is affected by cytoplasmic Ca 2⫹ level, A23187 was added to DM (Fig. 5). The expression of SPARC mRNA was rapidly and remarkably increased by the treatment with the drug: the expression of SPARC was enhanced 6- to 7-fold within 1 h in the presence of 0.5 M A23187. These results again suggest the idea that SPARC may play an important role in myogenesis, and that the expression of SPARC mRNA may be controlled by Ca 2⫹-dependent pathway. DISCUSSION In the present paper, we demonstrate that SPARC is involved in myoblast differentiation. The expression of SPARC mRNA was markedly up-regulated during myogenesis, and the treatment with an anti-SPARC antibody completely prevented myoblasts (Figs. 1–3). The treatment with EGTA, which is known to inhibit the fusion of myoblasts, reversibly blocked the up-
FIG. 4. Effect of EGTA on the myoblast fusion and the expression of SPARC mRNA during myogenesis. C2C12 cells were proliferated in GM and then cultured for 8 days in DM with (closed circle) or without the addition of 1.4 mM EGTA (open circle). To test the reversibility, cells were kept for 4 days in EGTA-containing medium and then grown in normal medium (open rectangle). At the indicated day, the cells were stained with hematoxylin to score fusion index (A) or harvested to examine the expression of SPARC mRNA by Northern blot (B). Cells were considered to be fused only if at least three nuclei were present in each myotube. For Northern blot, 30 g of total RNA was used, and 28S and 18S rRNAs were used as internal controls. The amount of SPARC mRNA was averaged from three separate experiments. Values are shown as mean ⫾ standard deviation.
Effect of EGTA or A23187 on the expression of SPARC mRNA during C2C12 myoblast differentiation. Myoblast fusion could be blocked by the deprivation of Ca 2⫹ in the culture medium, and rapidly recovered upon the re-addition of Ca 2⫹ (22, 23) (Fig. 4A). We examined the effect of the Ca 2⫹ level in the medium on the expression of SPARC mRNA of C2C12 myoblast (Fig. 4B). When Ca 2⫹ in DM was deprived by the addition of EGTA, the expression of SPARC mRNA remained at the basal level up to 8 days. In addition, when the medium was added with Ca 2⫹ to a normal
FIG. 5. Effect of A23187 on the expression of SPARC mRNA in C2C12 myoblasts. C2C12 cells were cultured for 12 h in DM and then treated with 0.5 M or 1.0 M A23187 for 30 min or 1 h. Total RNA was isolated and Northern blotting was carried out. In the bottom graph, the amount of SPARC mRNA from cells treated with A23187 for 0.5 h (gray bar) or 1 h (black bar) was compared with that from untreated cells (white bar), using three separate experiments. Values are shown as means ⫾ standard deviation.
633
Vol. 271, No. 3, 2000
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
regulation of SPARC mRNA expression (Fig. 4). Last, A23187, which has been known to stimulate myoblast differentiation, rapidly and dramatically increased the level of SPARC transcript (Fig. 5). The myogenesis occurs through a series of steps [1]. After withdrawal from the cell cycle, cells become bipolar and then align along their long axes. The aligned plasma membranes come in close apposition and local membrane fusion occurs to form single multinucleated myotubes. When SPARC in culture media was inactivated by treatment with an anti-SPARC antibody, the cells no longer could fuse and showed polygonal-shape rather than bipolar shape (Fig. 3). Therefore, it appears that SPARC plays a role in the fusion or the step preceding fusion during myoblast differentiation. The expression of SPARC is significantly increased in a variety of tissues undergoing remodeling (3). Multiple functions of SPARC have been known; inhibition of cell spreading and disassembly of focal adhesions, inhibition of cell proliferation, and regulation of expression of various ECMs and proteinases (2). These functions of SPARC all apparently seem to be relevant to myogenesis of embryonic skeletal muscle, because this process encompasses the dramatic remodeling processes such as withdrawal from the cell cycle, formation of bipolar shape cells, spreading and adherence of the cells to each other, membrane fusion, and formation of multinucleated myotube (1). For example, some isoforms of PDGF (platelet-derived growth factor) enhanced the proliferation of myoblasts and partly inhibited their differentiation (25). It was also shown in other tissues that SPARC could interact with some isoforms of PDGF and inhibit the binding of PDGF to its receptors (2, 26). Therefore, it is possible that SPARC may inhibit PDGF in stimulating the proliferation of myoblasts. Expression of muscle-specific genes is regulated primarily by muscle-specific transcription factors, MyoD family (27). Although the treatment with EGTA blocked the myoblast fusion, it did not interfere with the induction of muscle-specific mRNAs such as for muscle creatine kinase and acetylcholine esterase receptors (28, 29). Ca 2⫹ is required for the fusion process, but not for the function of MyoD family. Since the expression of SPARC mRNA was neither influenced by EGTA treatment, it seems that MyoD family does not control the expression of SPARC. It was shown here that the treatments with A23187 to myoblasts markedly induced the expression of SPARC mRNA: the expression of SPARC mRNA was enhanced 6- to 7-fold within 1 h (Fig. 5). So far, there were few reports on the mechanism how the expression of SPARC gene was regulated. This study should be the first report demonstrating that the expression of SPARC mRNA could be controlled by cytosolic Ca 2⫹dependent pathway.
ACKNOWLEDGMENT This study was supported by the academic research fund of Ministry of Education, Republic of Korea (BSRI-97-4409).
REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.
15.
16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29.
634
Wakelam, M. J. (1985) Biochem. J. 228, 1–12. Motamed, K. (1999) Int. J. Biochem. Cell Biol. 31, 1363–1366. Lane, T. F., and Sage, E. H. (1994) FASEB J. 8, 163–173. Buck, C. A., and Horwitz, A. F. (1987) Annu. Rev. Cell Biol. 3, 179 –205. Clark, P., Coles, D., and Peckham, M. (1997) Exp. Cell Res. 230, 275–283. Chung, C. Y., and Kang, M. S. (1990) J. Cell Physiol. 142, 392– 400. Gullberg, D., Sjoberg, G., Velling, T., and Sejersen, T. (1995) Exp. Cell Res. 220, 112–123. Cossu, G., Kelly, R., Tajbakhsh, S. D., Donna, S., Vivarelli, E., and Buckingham, M. (1996) Development 122, 429 – 437. Pownall, M. E., Strunk, K. E., and Emerson, C. P., Jr. (1996) Development 122, 1475–1488. Sage, H., Vernon, R. B., Decker, J., Funk, S., and Iruela-Arispe, M. L. (1989) J. Histochem. Cytochem. 37, 819 – 829. Porter, P. L., Sage, E. H., Lane, T. F., Funk, S. E., and Gown, A. M. (1995) 43, 791– 800. Schwarzbauer, J. E., and Spencer, C. S. (1993) Mol. Biol. Cell 4, 941–952. Kim, H. S., Lee, I. H., Chung, C. H., Kang, M. S., and Ha, D. B. (1992) Dev. Biol. 150, 223–230. Blau, H. M., Pavlath, G. K., Hardeman, E. C., Chiu, C. P, Silberstein, L., Webster, S. G., Miller, S. C., and Webster, C. (1985) Science 230, 758 –766. Bassuk, J. A., Iruela-Arispe, M. L., Lane, T. F., Benson, J. M., Berg, R. A., and Sage, E. H. (1993) Eur. J. Biochem. 218, 117– 127. Myer, A., Wagner, D. S., Vivian, J. L., Olson, E. N., and Klein, W. H. (1997) Dev. Biol. 185, 127–138. Chinni, C., de Niese, M. R., Tew, D. J., Jenkins, A. L., Bottomley, S. P., and Mackie, E. J. (1999) J. Biol. Chem. 274, 9169 –9174. Frohman, M. A., Dush, M. K., and Martin, G. R. (1988) Proc. Natl. Acad. Sci. USA 85, 8998 –9002. Liang, P., and Pardee, A. B. (1992) Science 257, 967–971. Morris, G. E., and Cole, R. J. (1977) FEBS Lett. 79, 183–187. Blau, H. M., Chiu, C. P., and Webster, C. (1983) Cell 32, 1171–1180. David, J. D., See, W. M., and Higginbotham, C. A. (1981) Dev. Biol. 82, 297–307. Shainberg, A., Yagil, G., and Yaffe, D. (1969) Exp. Cell Res. 58, 163–167. Schollmeyer, J. E. (1986) Exp. Cell Res. 162, 411– 422. Ye, W. V., McFarland, D. C., Gilkerson, K. K., and Pesall, J. E. (1996) Cytobios 88(352), 53– 62. Jendraschak, E., and Sage, E. H. (1996) Semin. Cancer Biol. 7, 139 –146. Emerson, C. P. (1990) Curr. Opin. Cell Biol. 2, 1065–1075. Endo, T., and Nadal-Ginard, B. (1987) Cell 49, 515–526. Hu, J. S., and Olson, E. N. (1990) J. Biol. Chem. 265, 7914 –7919.