A Candidate Molecule for the Matrix Assembly Receptor to the N-terminal29-kDa Fragment of ~ i ~ r ~ n e c t i n in Chick yob blast^*

Myoblast surface proteins with binding activity to- ward the N-terminal 29-kDa fragment of fibronectin were identified by two different experimental tech- niques: one involves radioi~ination of the cell surface proteins, followed by solubilization with Triton X-100 and affinity purification on a Sepharose column conjugated with the 29-kDa fragment, and the other involves cross-linking of the 29-kDa fragment to the cells meta- bolically labeled with fmS]methionine, followed by im-munoprecipitation with anti-29-kDa IgG. Both a p proaches revealed that primary cultures of chick myoblasts contain the 66- and 48-kDa proteins that bind to the 24kDa fragment. These binding proteins were then purified to apparent homogeneity by two succes- sive chromatographies of the solubilized extracts of 12-day-old embryonic muscle on wheat germ agglutinin- agarose and 29-kDa fragment-Sepharose columns. However, the 48-kDa protein was found to be derived from contaminating fibroblasts upon immunoblot analysis of the myogenic cell lines, rat LSE63 and mouse C2A3, and cultured fibroblasts using the antibody raised against the 66-kDa protein. Anti-66-Wa IgG inhibited the binding of

nectin molecule. And the binding of fibronectin to cell surface receptors of the integrin family has well been documented (4- 6). In addition to the RGD site, fibronectin has been shown to contain a distinct cell-binding site that also appears essential for matrix assembly. The second site is located within the Nterminal 29-kDa f r a~e n t of fibronectin, which contains the first five repeats of the type I homologous domain (7-11).
Recently, a 67-kDa protein that binds specifically to the 29-kDa fragment of fibronectin has been purified from the insoluble fractions of human U93'7 cells and rat peritoneal macrophages (12). This protein may represent a unique macrophage surface binding protein for the N terminus of fibronectin, and yet i s unlikely to function in matrix assembIy because macrophages do not produce matrices. A few other proteins have been reported to interact with the N-terminal domain of fibronectin. McDonald and eo-workers (13) have identified a 150-kDa protein complex by cross-linking the 29-kDa fragment to the surface of fibroblasts. An additional 18-kDa fibroblast protein with N-terminal fibronectin binding activity has also been identified 114). However, the latter two proteins have not yet been purified.
A prominent event in the differentiation of skeletal muscle cells is the fusion of mononucleated myoblasts into multinucleated myotubes. This process accompanies various cellular events, such as a direct physical interaction of plasma membrane (151, reorganization of cytoskeletons (16, 17) and extracellular matrices (18). A number of reports have suggested that fibronectin, a cell surface glycoprotein, exerts a profound effect on myoblast fusion (for a review, see Ref. 19). For example, treatments of exogenous fibronectin to rat myoblast cultures have been shown to block the membrane fusion (20). It has also been demonstrated that the treatment of antibodies directed to integrin inhibits the fusion process (21). In addition, we have recently demonstrated that the level of fibronectin in chick embryonic myoblasts decreases during the course of myogenic differentiation, and this decrease is closely correlated with the fall in the extent of binding of the 29-kDa Ebronectin fragment to the cell surface (22).
In an attempt to elucidate the role of Ebronectin and the surface binding of its 29-kDa f r a~e n t in the regulation of myogenic different~at~on of cultured myoblasts, we have purified a protein of 66 kDa that specifically binds to the N-terminal fragment of fibronectin from the insoluble fractions of the cells. In addition, we provide several lines of evidence that the 66-kDa protein is a potential candidate for fibronectin matrix assembly receptor.
The 66-kDa ~a t~i x Assemb~y Na1251 using Iodo-Beads (Pierce). Antibody against the 29-kDa fragment was prepared by injecting the polypeptide into albino rabbits. To prepare antibody against the 29-kDa fragment-binding protein (Le. the 66-kDa protein; see below), proteins eluted from a Sepharose column conjugated with the 29-kDa fragment were electrophoresed on 10% (wiv) polyacrylamide slab gels in the presence of sodium dodecyl sulfate (SDS) (24). After briefly staining the gels with Coomassie R-250, the protein bands corresponding to 66 kDa were cut out, minced, and injected. IgGs were purified by protein A-Sepharose column chromatography 125). Fab fragments of the anti-66-kDa IgGs were prepared as described (26). Anti-f31-integrin antiserum was kindly provided by Dr. A. F. Horwitz (University of Illinois).
The 29-kDa f r a~e n t -a~n i t y column was prepared by coupling the purified 29-kDa ~l~e p t i d e s to ~NBr-activated Sepharose (27). Approximately 5 mgiml of the peptides were covalently attached to the column. Fluore~ein-labeled fibronectins were prepared by incubating 10 pglml of fluorescein isothiocynate (FITC) and 2 mg/ml of fibronectin in 4 m~ NaH2C0,/Na2HC03, pH 9.0, containing 0.15 M NaCl. After incubating the mixture overnight at 4 "C, the FITC-labeled proteins were further purified by chromatography on a Sephadex G-25 column equilibrated with phosphate-bdered saline.
Celt Cultures-Myoblasts from breast muscle of 12-day-old chick embryos were prepared as described previously (22). The cells were plated at a density of 5 x lo5 celldml in Eagle's minimal essential medium (MEM) containing 10% (vlvf horse serum, 108 (v/v) chick embryo extracts, and 1% (v/v) antibiotics. One day after the cell seeding, the culture medium was changed with the same medium but containing 2% embryo extract.
A myogenic clone of L8E63 rat skeletal myoblasts was grown in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% horse serum as described (28). C2A3 myoblasts, subcloned from the C2 line of mouse satellite cells (29), were grown in DMEM containing 10% fetal bovine serum. Cultures of skeletal muscle fibroblasts and skin fibroblasts were obtained from 12-day-old chick embryos as described (30,31). The cells were cultured in DMEM containing 5% fetal bovine serum. The cells used for experiments were between passage 3 and 5. All cells were cultured in a humidified incubator in an atmosphere of 95% air and 5% C02 at 37 "C. ~a d~o Z a b e~~n g of Cell Surface Pro~eins-Myoblasts were cultured for 24 h and detached from the plates by treating 3 m~ EGTA for 15 rnin. After centrifugation, the pelleted cells were resuspended in phosphatebuffered saline containing 2 m~ pbenylmethylsulfonyl fluoride, 1 mx ~-ethyimaleimide (NEM), and 2 m M EDTA. The cells were surface labeled for 3 min with 1 mCi of Na12&I and 0.2 mg/mI of lactoperoxidase/ IO8 cells. The cultured myobIasts were also metabolically labeled by incubation for 12 h with 5 pCi/ml of 135SJmethionine. ~m~u n o c~~~e u l Analysis of the 29-kRu F r u~~e n t -b~n d i~ Proteins-In order to immunoprecipitate the 29-kDa fragment-binding proteins, the myoblasts labeled with [32"S~methionine were incubated with 20 pg/ml of the 29-kDa fragment for 1 h at 37 "C. After the incubation, the cells were washed three times with 10 m M HEPES, pH 7.6, containing 120 m~ NaCl, 5 m~ KCl, and 1.5 m v MgSO,. The cells were treated with 50 PM dithiobis(succinimidy1 propionate) for 15 min at 4 "C for cross-linking the 29-kDa fragment to its binding proteins on the cell surface. They were then washed three times with 10 m M Tris-HC1, pH 7.5, containing 0.15 M NaCl and solubilized by sonication in the presence of 1% iv/v) Triton X-100. To eliminate nonspecific binding activity, if any, the solubilized extracts were treated with preimmune serum for 1 h at room temperature and then passed through a protein ASepbarose column. The samples were centrifuged for 5 min, and the resulting supematants were added with anti-29-kDa fragment antiserum and incubated a t 4 "C overnight. The immune complexes were precipitated by adding aliquots of protein A-Sepharose suspension. The precipitates were solubilized with 2% (w/v) SDS, electrophoresed as above, and autoradio~ap~ed.
For the ligand-blotting analysis (32), the 66-and 48-kDa proteins obtained during puri~cation (see below) were electrophoresed in the presence of SDS. The proteins in the gels were then transferred to nitrocellulose papers. The papers were incubated with 10 m M HEPES, pH 7.6, containing 0.14 M NaCl, 1 m? CaC12, 1 m~ LMgCl,, 1% Triton X-100, and 3% (w/v) bovine serum albumin and then with the same buffer containing 50 ng/mI of the 1Z51-labeled 29-kDa fragment. They were then dried and autoradiographed. P u r i~c u~~o n of the 29-kDa Frag~ent~bjnding Proteins-Breast muscles from 12-day-old chick embryos were homogenized in cold "isbuffered saline (TBS; 50 m M "is-HC1, pH 7.6, 150 mi NaCl, 1 m~ CaCl,, and 1 m M MgC121 containing 8.5% (wfv) sucrose using a Dounce homogenizer, Cell debris was removed by centrifugation for 10 min at Receptor in Chick ~~o~~u s t s 300 X g, and the resulting supernatants were again centrifuged for 20 rnin at 20,000 xg. The pellet was resuspended in 10 r n~ HEPES, pH 7.6, containing 8.5% sucrose and 1 m M EDTA and layered on a 8.5121/40/54% discontinuous sucrose gradient. The sample was centrifuged at 100,000 x g for 2 h using a SW60 rotor (Beckman). The materials at the interface of 21 and 40% sucrose were collected and solubilized by incubation for 1 h on ice in 10 m~ HEPES, pH 7.6, containing 1% Triton X-100, 1.4 M NaCl, 1 m M CaCl,, and 1 m M MgCl2. Insoluble materials were removed by centrifugation at 24,000 x g for 30 min. The supernatants were then loaded onto a wheat germ agglutinin-agarose column equilibrated with TBS containing 0.5% Triton X-100. After extensively washing the column with the same buffer, proteins were eluted with the buffer containing 0.5 M ~-acetylglucosamine. The eluates were then loaded onto a Sepharose column conjugated with the 29-kDa fragment. The column was washed with 10 volumes of TBS containing 0.5% Triton X-100 and then eluted with the same buBer containing 4 M urea. Elution of the 29-kDa fra~ent-binding proteins was assessed by the ligand-blotting anatysis.
Binding of Radiolabeled 29-kDa Frugment and Fibronectin-All binding assays were performed with myoblasts cultured in DMEM containing 10% fibronectin-depleted horse serum. The cells were incubated with 322"I-29-kDa fragment or lWI-fibronectin at 37 "C for appropriate periods. The cells were then rinsed four times with ice-cold Eagle's balanced salt solution, and their extracts were prepared by resuspending sequentially in 20 m~ Tris-HCl, pH 8.3, containing 1% (w/v) deoxycholate 2 m M pheny~methyls~fonyl fluoride and 2 m~ NEM and the Tris buffer containing 4% SDS (7). In certain cases, the cells were solubilized only with 4% SDS. The radioactivity in the extracts was determined using a gamma counter.
Fluorescence Microscopy-Chick myoblasts were grown for 30 h on coverslips and rinsed twice with Eagle's balanced salt solution. When assayed for the incorporation of exogenous fibronectin into matrices, the cultures were incubated for 3 h with 200 pg/ml of FITC-labeled fibronectin, rinsed with DMEM, and further incubated in DMEM containing 10% horse serum for the next 24 h. For determination of the binding of fibronectin to cell surface, incubations with FITC-labeled fibronectin were performed only for 15 rnin. After the incubations, the cells were fixed with 3.58 (v/v) paraformaldehyde for 30 min, mounted with glycerol, and observed under a fluorescence microscope (Nikont. When assayed for the ability of the cells by themselves to make ~bronectin matrices, the cultures were incubated in D *~E M containing 10% horse serum as above but in the absence of FITC-labeled fibronectin. After the incubation, they were k e d and treated with anti-fibronectin rabbit IgG and then with goat anti-rabbit IgG conjugated with FITC at room temperature for 1 h. To determine whether the 66-kDa protein eo-localizes with the site of fibronectin matrix formation, chick myoblasts cultured for 30 h were incubated for 30 min with FITC-labeled fibronectin. The cells were fixed and treated with the anti-66-kDa IgG and then with anti-rabbit IgG conjugated with tetramethyl rhodamine isothiocyanate (TRITC) (Sigma) at room temperature for 1 h. The double-labeled samples were observed as above.

I~e n t~~~a~~~n of the 29-kDa F r a g~e n~-~~~~~n g
Protein+" identify the 29-kDa fragment-binding proteins, chick myoblasts were radioiodinated using lactoperoxidase and solubilized with 1% Triton X-100. The solubilized proteins were loaded onto a Sepharose column conjugated with the N-terminal 29-kDa fragment of fibronectin, competitively eluted with soluble 29-kDa fragments, electrophoresed in the presence of SDS, and autoradio~aphed. Fig. 1A shows that two distinct proteins are capable o f binding to the 29-kDa fragment. These proteins have apparent molecular masses of 66 and 48 ma under both the reducing and nonreducing conditions. Therefore, it i s unlikely that these polypeptides are covalently linked with each other.
To clarify further the presence of the 29-kDa fragment-binding proteins on the surface of chick myoblasts, the cells that had been metabolically labeled with [35Slmethionine were cross-linked to the 29-kDa fragment using dithiobisfsuccinirnidyl propionate^ and solubilized with 1% Triton X-100. The solubilized proteins were immunoprecipitated using anti-29-kDa IgG and protein A-Sepharose. Precipitates were resus-

R . the cells labeled with
[""Slmethionine were cross-linked with the 29-kI)a fragment, soluhilized by 1%. Triton X-100, and immunoprecipitated using anti-29-kDa antiserum as descrihed in the text. The precipitates were suhjected to electrophoretic analysis as above but only under reducing conditions. pended in 2% SDS and treated with 10% (v/v) 2-mercaptoethanol to break the cross-links. The samples were then electrophoresed in the presence of SDS. Fig. 1B again shows that two proteins with sizes of 66 and 48 kDa can bind to the 29-kDa fragment. These results clearly indicate that the insoluble preparations from chick myoblast cultures contain two polypeptides that specifically interact with the 29-kDa fragment of fibronectin.
Purification of the 29-kDa Fragment-hinding Proteins-In order to isolate the 29-kDa fragment-binding proteins, membrane proteins were obtained from breast muscle tissues of 12-day-old chick embryos and solubilized with 1% Triton X-100. The solubilized proteins were chromatographed on a wheat germ agglutinin-agarose column. Proteins bound to the column were eluted with N-acetylglucosamine and subjected to the ligand-blotting analysis using I29-29-kDa fragment of fibronectin. As shown in Fig. 2, the proteins that could interact with the affinity column and with the I29-29-kDa fragment had molecular masses of 66 and 48 kDa. These results clearly suggest that the proteins fractionated by this procedure are identical to the 29-kDa fragment-binding proteins in cultured chick myoblasts, which were identified by the affinity chromatography and the cross-linking experiments (see Fig. 1). These results also suggest that the 29-kDa fragment-binding proteins are glycoproteins.
To purify further the 29-kDa fragment-binding proteins, the fractions containing the 66-and 48-kDa polypeptides were pooled and loaded onto a Sepharose column conjugated with the 29-kDa fragment. Proteins bound to the column were eluted with 4 M urea and electrophoresed in the presence of SDS. As shown in Fig. 3 A , only two bands of 66 and 48 kDa were evident upon silver-staining. Furthermore. the peak fraction from the column strongly interacted with '251-29-kDa frag- ments as revealed by the ligand-blotting analysis (Fig. 33. lane a ). In addition, this binding could he ahnlishrd upnn treatment of a n excess of nonradioactive 29-kDa f r a p r n t s I lane h J. Thus, it appears clear that the purified 66-and 4H-kDa proteins are the specific 29-kDa fragment-binding proteins in the myoblast culture preparations.
Myohlast-specific Binding of the 66-kDa Pmtein to the 29-kDa Fragment-In order to determine the relationship hetween the 66-and 48-kDa proteins, antibodies dirrcted only to the 66-kDa protein were prepared a s descrihed undrr "Experimental Procedures." The purified 29-kDa fragmrnt-binding proteins were then subjected to immunohlot analysis using anti-66-kDa antiserum. As shown in Fig. 4  prepared as described under "Experimental Procedures."A, the purified 29-kDa fragment-binding proteins from the 29-kDa fragment-affinity column (fraction numbers [2][3][4][5] in Fig. 3 A ) were pooled, and aliquots of them raised possibilities that the 48-kDa polypeptide could be the proteolytic cleavage product of the 66-kDa protein and that either of two proteins might be derived from other contaminating cells, such as fibroblasts, because primary cultures of chick myoblasts could not be completely devoid of fibroblast contamination.
To test these possibilities, cell lysates were prepared from two myogenic cell lines, mouse C2A3 and rat L8E63, and subjected to immunoblot analysis (33). As shown in Fig. 4B, anti-66-kDa antiserum interacted only with the 66-kDa protein in both the cells. On the other hand, the same antiserum reacted solely with the 48-kDa protein when the lysates of chick embryonic fibroblasts were analyzed. Therefore, it appears clear that the 48-kDa protein originates from contaminating fibroblasts in the primary culture of chick myoblasts and the 66-kDa protein is a unique myoblast surface binding protein for the N-terminal 29-kDa fragment of fibronectin. However, the chemical basis for the interaction of anti-66-kDa antiserum with the 48-kDa protein remains unclear.
The 66-kDa Protein Is Distinct from p,-Integrin-To determine whether the 66-kDa protein is distinct from the integrintype receptor, myoblast lysates were prepared, incubated with anti-p,-integrin antiserum, and precipitated by adding protein A-Sepharose. Both the resulting supernatant and pellet fractions were electrophoresed in the presence of SDS and incubated with the anti-&-integrin antiserum or anti-66-kDa antiserum and then with anti-rabbit IgG conjugated with horseradish peroxidase. As shown in Fig. SA, the immunoprecipitation procedure almost completely eliminated from the cell lysates the 120-kDa protein that could interact with anti-plintegrin antiserum. On the other hand, the 29-kDa fragmentbinding proteins that can interact with anti-66-kDa antiserum remained in the supernatant fractions (Fig. 5B). These  IgG. After the incubation, the cells were solubilized with 4% SDS. Fig. 6 shows that anti-66-kDa IgG inhibits the binding of the 29-kDa fragment in a dose-dependent manner unlike the preimmune IgG. These results further support our finding that the 66-kDa protein is responsible for the interaction with the N-terminal 29-kDa fragment of fibronectin. We then examined the effect of anti-66-kDa IgG on the binding of fibronectin and its incorporation into fibronectin matrices. The cells cultured for 30 h were incubated for 3 h with '2sI-fibronectin and anti-66-kDa IgG. After the incubation, proteins were extracted by treating the cells with 1% deoxycholate and then with 4% SDS. Fibronectin bound to cell surface is known to be extractable by 1% deoxycholate, whereas the protein molecules that were assembled into extracellular matrices can be extracted with 4% SDS but not by 1% deoxycholate (7). As shown in Fig. 7, anti-66-kDa IgG revealed little or no effect on the binding of fibronectin on the cell surface, but markedly inhibited the assembly of the protein into matrices.
To determine whether the exogenously added fibronectins can indeed be incorporated into fibronectin matrices, fluorescence microscopic studies were performed using the cultured myoblasts. As shown in Fig. 8A, incubation with FITC-labeled fibronectin resulted in the formation of an extensive extracellular meshwork that completely surrounded the cells. Furthermore, a similar pattern of meshwork was formed by the cells incubated without the exogeneously added fibronectin, as analyzed by the sequential treatment of anti-fibronectin IgG and anti-rabbit IgG conjugated with FITC (Fig. 8B). These results clearly show that cultured myoblasts can make fibronectin matrices by themselves as well as they can incorporate exogenous fibronectin into matrices. We then examined whether the fibronectin matrices co-localize with the 66-kDa protein. The cells cultured for 30 h were double-labeled by sequential treatments of FITC-labeled fibronectin, anti-66-kDa Fab fragments, and anti-rabbit IgG conjugated with TRITC. Fig. 8, C and D, show that both the labels are localized to the cell surface and particularly concentrated at the edges of the surface where fibronectin appears to incorporate into matrices, suggesting that the 66-kDa protein co-localizes with the site of fibronectin matrix formation.
In order to clarify further the involvement of the 66-kDa protein in fibronectin matrix assembly, the cells were incubated for different periods with FITC-labeled fibronectin in the presence of Fab fragments of anti-66-kDa protein IgG or preimmune I&. As shown in Fig. 9, A and B, the cells incubated for 15 min with anti-66-kDa Fab fragments were capable of binding with FITC-labeled fibronectin as well as those incubated with preimmune Fab fragments. Thus, it appears likely that the 66-kDa protein is not involved in the initial binding of fibronectin to the cell surface. When the cells were incubated for 3 h, however, anti-66-kDa Fab fragments completely blocked the ~nco~oration of FITC-~a~eled ~b r o n~t i n into matrices (Fig. W ) , unlike the preimmune Fab Ragments (Fig. 9C). These results strongly suggest that the interaction of the N-terminal 29-kDa domain of fibronectin with the 66-kDa protein is necessary for the matrix assembly. DISCUSSION We have previously demonstrated that the N-terminal 29-kDa fragment of fibronection binds to the surface of cultured chick myoblasts with an apparent dissociation constant of 1.4 x 10" M (22). The density of the binding site on the cell surface has also been estimated to be approximately 3.4 x 105/cell cultured for 30 h, In the present studies, we demonstrate the presence of a myoblast surface protein that is responsible for the binding to the 29-kDa fragment of fibronectin. This protein with an apparent molecular mass of 66 kDa can be not only cross-linked to the 29-kDa fragment but also purified by the affinity column that i s covalently conjugated with the fragments.
Noteworthy is, however, the finding that an additjonal 48-kDa protein also interacts with the 29-kDa fragment and can be co-purified from the myoblast cultures. Furthermore, the 48-kDa protein cross-reacts with antibody raised against the 66-kDa protein. Therefore, we initially suspected whether the 48-kDa protein is derived by proteolytic cleavage of the 66-kDa protein. However, we could demonstrate that the 48-kDa protein is derived from contaminating fibroblasts in the primary culture of chick myoblasts because only the 66-kDa protein is found in the myogenic cell lines of mouse C2.43 and rat L8E63, while the 48-kRa protein is exclusively found in the cultures of chick embryonic muscle and skin fibroblasts. Nevertheless, it remains totally unclear how the antibody directed against the 66-kDa protein can cross-react with the 48-kDa protein, although it is tempting to speculate that the surface proteins may contain a consensus sequence for the specific interaction with the N-terminal 29-kDa fragment of fibronectin.
Several lines of evidence suggest that the 66-kDa protein is a potential candidate for fibronectin matrix assembly receptor. 1) Anti-66-kDa IgG strongly inhibited the interaction of lZ5I-29-kDa fragment of fibronection, but showed little or no effect on the initial binding of fibronectin to the surface of cultured myoblasts. In addition, the immunoprecipitat~on experiment revealed that the 66-kDa protein does not cross-react with the anti-~~-integrin antiserm. Since antibodies against the &-subunit are known to co-precipitate the a5-subunit, the 66-kDa protein appears to be distinct at least from the asPl-integrin. These findings suggest that the binding of fibronectin in the presence of anti-66-kDa IgG is mediated by the interaction of the cell-adhesive domain of the protein molecule with integrin but not by the interaction of the 29-kDa fragment with the 66-kDa protein. 2)Anti-fiG-kDa I& inhibited the incorporation of fibronectin to deoxycholate-insoiuble fraction in a dose-dependent manner, in contrast to its insignificant effect on the initial cell surface binding. ~cKeown-Longo and Mosher (7) have previously shown that, using pulse-chase experiments, fibronectin binds initially in the deoxycholate-soluble fraction (cell surface fraction) and is transferred to the deoxycholateinsoluble fraction (extracellular matrix). Therefore, it appears that anti-66-kDa IgG prevents the assembly of fibronectin into matrices by blocking the interaction of the N-terminal 29-kDa fragment of fibronectin with the 66-kDa receptor. 3) Perhaps the most compelling evidence for involvement of the 66-kDa protein in the matrix assembly of fibronectin are the fluorescence microscopic studies demonstrating that the incubation time-dependent incorporation of ~ITC-labeled fibronectin can be completely prevented by treatment of the anti-66-kDa Fab fragments to the cultured myoblasts. In addition, the 66-kDa protein appears to co-localize with the site of fibronectin matrix formation. Therefore, fibronectin matrix assembly appears to be mediated by the interaction of the N-terminal type I domain supplemented with 10% fibronectin-depleted horse serum. When assayed for the incorporation into fibronectin matrices, the cells were incubated for 3 h with FITC-labeled fibronectin in the presence of Fab fragments of preimmune I& (C) or anti-66-kDa IgG 0 ) . The cells were washed with DMEM containing 10%. horse serum and further cultured for the next 24 h. The cultures were then fixed and observed a s in Fig. 8. Bar,20 pm. of fibronectin with its 66-kDa receptor on the surface of CUIbly (35,36). In addition, a 14-kDa fragment containing the first tured myoblasts.
two type I11 repeats of fibronectin was shown to inhibit fibro-A number of reports have suggested that other regions in nectin matrix assembly (37). The presence of multiple cellfibronectin also involve in matrix assembly. Monoclonal anti-binding regions in fibronectin has been explained by the mulbodies directed to the RGD-containing cell-adhesive domain of tistep model for fibronectin matrix assembly, proposed by fibronectin were found to inhibit matrix assembly (34). The Schwarzbauer et al. (38). Similar models have also been proantibodies against the integrin that binds to the RGD site in posed by McDonald (10) and Mosher et al. (39). This model fibronectin were also found to inhibit fibronectin matrix assem-involves four sequential events: capture of fibronectin mol-The 66-kDa Matrix Assembly Receptor in Chick Myoblasts 7657 ecules from the environment, translocation of the captured fibronectin to the growing end of fibronectin fiber, alignment of fibronectins, and, finally, covalent stabilization of the fibronectin fibril. Therefore, it is possible that the 66-kDa protein is involved in the later event in the first step for fibronectin capture, since the anti-66-kDa IgG inhibits insignificantly the initial binding of fibronection but remarkably the incorporation of fibronectin into deoxycholate-insoluble matrices.
The role(s) of fibronectin and its assembly into extracellular matrices on the regulation of differentiation of cultured myoblasts remains unknown. However, a number of reports have suggested the involvement of fibronectin in the myogenic process (for review, see Ref. 19). Myoblasts have been reported to adhere to fibronectin (40) and extend along the oriented fibrils of fibronectin (41). I t was also reported that fibronectin-coated substratum promotes fusion although it does not distinguish between effects on adhesion and on fusion per se (42). Also reported was that the amount of cell surface fibronectin decreases after the fusion of myoblasts into myotubes (43). In addition, we have recently demonstrated that the decrease in the level of fibronectin during myoblast fusion is closely correlated with the gradual loss of the binding activity of the cell surface with the 29-kDa fragment (22) and that the expression of the 66-kDa protein falls dramatically during the fusion proce m 2 However, more studies are necessary for understanding the precise action mode of fibronectin and the mechanism by which the expression of the 66-kDa receptor is regulated during the differentiation of chick embryonic myoblasts.