- Email: [email protected]
Syndecan-3 and Syndecan-4 Specifically Mark Skeletal Muscle Satellite Cells and Are Implicated in Satellite Cell Maintenance and Muscle Regeneration
Developmental Biology 239, 79 –94 (2001) doi:10.1006/dbio.2001.0416, available online at http://www.idealibrary.com on
Syndecan-3 and Syndecan-4 Specifically Mark Skeletal Muscle Satellite Cells and Are Implicated in Satellite Cell Maintenance and Muscle Regeneration D. D. W. Cornelison,* ,1 Mark S. Filla,† ,1 Heather M. Stanley,* Alan C. Rapraeger,† ,2 and Bradley B. Olwin* ,2 *Department of Molecular, Cellular and Developmental Biology, University of Colorado, Boulder, Colorado 80309; and †Department of Pathology and Laboratory Medicine, University of Wisconsin, Madison, Wisconsin 53706
Myogenesis in the embryo and the adult mammal consists of a highly organized and regulated sequence of cellular processes to form or repair muscle tissue that include cell proliferation, migration, and differentiation. Data from cell culture and in vivo experiments implicate both FGFs and HGF as critical regulators of these processes. Both factors require heparan sulfate glycosaminoglycans for signaling from their respective receptors. Since syndecans, a family of cell-surface transmembrane heparan sulfate proteoglycans (HSPGs) are implicated in FGF signaling and skeletal muscle differentiation, we examined the expression of syndecans 1– 4 in embryonic, fetal, postnatal, and adult muscle tissue, as well as on primary adult muscle fiber cultures. We show that syndecan-1, -3, and -4 are expressed in developing skeletal muscle tissue and that syndecan-3 and -4 expression is highly restricted in adult skeletal muscle to cells retaining myogenic capacity. These two HSPGs appear to be expressed exclusively and universally on quiescent adult satellite cells in adult skeletal muscle tissue, suggesting a role for HSPGs in satellite cell maintenance or activation. Once activated, all satellite cells maintain expression of syndecan-3 and syndecan-4 for at least 96 h, also implicating these HSPGs in muscle regeneration. Inhibition of HSPG sulfation by treatment of intact myofibers with chlorate results in delayed proliferation and altered MyoD expression, demonstrating that heparan sulfate is required for proper progression of the early satellite cell myogenic program. These data suggest that, in addition to providing potentially useful new markers for satellite cells, syndecan-3 and syndecan-4 may play important regulatory roles in satellite cell maintenance, activation, proliferation, and differentiation during skeletal muscle regeneration. © 2001 Academic Press
INTRODUCTION Skeletal muscle development in the prospective limb consists of migration of committed myoblasts from limblevel somites into the limb field, colonization of the premuscle masses, patterning of future muscles, and differentiation into fast and slow contractile myofibers. Later, myogenesis in the adult (mediated by adult myoblasts or muscle satellite cells) involves similar processes in the regenerative response to damage or disease. Growth factors 1
D.D.W.C. and M.S.F. contributed equally to this work. To whom correspondence should be addressed. Fax: (303) 4921587 (B.B.O.) or (608) 265-3301 (A.C.R.). E-mail: Bradley.Olwin@ colorado.edu or [email protected]. 2
0012-1606/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.
such as members of the fibroblast growth factor (FGF) family and scatter factor/hepatocyte growth factor (HGF) play diverse and critical roles during development of the vertebrate limb and its musculature, regulating growth, patterning, and cellular differentiation in developing muscle tissue. HGF signaling is required for emigration of myoblasts from the somites into the limb field (Bladt et al., 1995) and for secondary myogenesis (Maina et al., 1996). FGF signaling regulates growth and differentiation of early embryonic primary myoblasts (Flanagan-Steet et al., 2000; Itoh et al., 1996) as well as secondary and adult myoblasts (Flanagan-Steet et al., 2000). FGF and HGF signals are transduced through transmembrane receptor tyrosine kinases: in the case of FGFs, four such receptors with differing affinities for individual FGF
79
80
Cornelison et al.
FIG. 1. Binding of FGF2 to endogenous heparan sulfate in mouse forelimb skeletal muscle. Fixed frozen sections were incubated with 5 nM biotinylated FGF-2 to visualize FGF binding sites (A–E). The dependence of the binding on heparan sulfate was examined by pretreating the sections with heparinase I and II (E). Total heparan sulfate was visualized by staining with antibody 3G10 (F). Stages examined were E14.5 (A), E18 (B), postnatal day 2 (P2) (C, F), and young adult (D). Cartilage (Ca), perichondrium (PC), and skeletal muscle (SkM) are denoted. Bar, 50 m.
family members (FGFR 1– 4) are known (reviewed in Wilkie et al., 1995); the receptor for HGF is c-met (Bottaro et al., 1991). When ligand is bound, receptor dimers transautophosphorylate and activate diverse and as yet poorly understood downstream signaling events. When signaling is
interfered with, either by loss of the ligand, use of dominant-negative receptor mutants, or nonfunctional signaling pathway mutants, the effects in the developing limb and musculature vary by receptor. These include embryonic lethality [loss of FR1 (Deng et al., 1994; Yamaguchi et
Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.
81
Satellite Cell Maintenance and Muscle Regeneration
FIG. 2. Immunolocalization of syndecan-3 in embryonic and adult mouse forelimb skeletal muscle. Sections of forelimb skeletal muscle from embryonic and postnatal mice were stained with polyclonal anti-syndecan-3 antibody (A, B, C, E) or nonspecific IgG (D). Developmental stages shown are E14.5 (A, D), E18 (B), P2 (C), and young adult (E, F). Insets show higher magnification views of developing myotubes. Young adult forelimb skeletal muscle is double stained for syndecan-3 (E) and DAPI (F). Arrowheads in (C), (E), and (F) show localized expression of syndecan-3 in P2 and young adult stages and its correspondence with individual nuclei. (Bar, 50 m).
al., 1994)]; embryonic lethality and failure to form limb buds [loss of FR2 (Xu et al., 1998)]; skeletal overgrowth [loss of FR3 (Colvin et al., 1996)]; no apparent phenotype [loss of FR4 (Weinstein et al., 1998)]; skeletal muscle abnormalities [dominant-negative FR1 in developing muscle (FlanaganSteet et al., 2000; Itoh et al., 1996)]; inhibition of limb and tongue muscle precursor cell migration [loss of c-met (Bladt et al., 1995)]; abnormalities in secondary myogenesis [c-met
mutant (Maina et al., 1996)] and in skeletal muscle regeneration [FGF6 null mutant (Floss et al., 1997)]. FGFs and HGF exhibit an absolute requirement for heparan sulfate in order to transduce an intracellular signal (reviewed in Rapraeger, 2000). It has been suggested that the role of heparan sulfate in the ternary signaling complex may be to facilitate receptor activation, required for intracellular autophosphorylation and signal transduction (Deakin and
Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.
82
Cornelison et al.
Lyon, 1999; Rapraeger et al., 1991; Yayon et al., 1991). Recent x-ray crystallography evidence suggests that heparan sulfate chains are responsible for FGF receptor dimerization and promotion of high affinity FGF binding by contacting both FGF receptor monomers and FGF ligands (Pellegrini et al., 2000; Schlessinger et al., 2000). Developmentally, deleting expression of specific syndecans has produced no detectable phenotype in vivo (Ishiguro et al., 2000; Perrimon and Bernfield, 2000), suggesting that there is functional redundancy between syndecans and possibly other HSPGs as well. During adult life, skeletal myogenesis in mammals is restricted to the activity of muscle satellite cells. These myogenic stem cells, whose exact embryonic origin is as yet unclear, are characterized by their position beneath the basal laminae of mature muscle fibers and their normally quiescent metabolic state. When the local muscle is damaged due to injury or disease, satellite cells are responsible for repair and regeneration of healthy muscle (reviewed in Bischoff, 1994; Schultz and McCormick, 1994). Both FGFs and HGF, along with their cognate receptors, have been strongly implicated as mediators of these functions. Quiescent satellite cells express the c-met receptor as well as FGFR-1 and FGFR-4, and will respond in vivo and in vitro to exogenously added FGF (FGF2 in particular) and HGF; HGF and FGFs are normally present in muscle tissue and appear to be released upon injury and expressed at higher levels during regeneration (Allen et al., 1995; Cornelison et al., 2000; Cornelison and Wold, 1997; Johnson and Allen, 1995; Tatsumi et al., 1998). Here, we present new data on the expression of members of the syndecan family of HSPGs during limb development and adult life, as well as functional studies on the effects of disrupting heparan sulfate during muscle regeneration in single-fiber culture. We find that, in addition to being coexpressed with receptor tyrosine kinases implicated in regulation of satellite cell activity, heparan sulfate appears to be required for proper progression through the satellite cell myogenic program, suggesting that signaling through receptors such as FGF receptors and c-met is also required. The epitopes identified in this work also represent new molecular tools for identification of quiescent and activated satellite cells.
MATERIALS AND METHODS Tissue samples from embryonic, postnatal, or young adult mice were embedded in OCT (Tissue-Tek), sectioned for staining, and fixed in 4% paraformaldehyde. Adult hindlimb samples were fixed in 4% paraformaldehyde prior to embedding and sectioning. Primary antibodies and dilutions used were: mouse monoclonal anti-syndecan-1 (mAb 2E9, courtesy of Dr. Guido David, University of Leuven, Belgium) at 2.5 g/ml; mouse monoclonal antisyndecan-2 [mAb 10H4, courtesy of Dr. Guido David (David, 1993)] at 2.5 g/ml; rabbit polyclonal anti-syndecan-3 (Carey et al., 1992) at 1:50; rabbit polyclonal anti-GST-syndecan-4 extracellular domain (␣-GST-mS4ED) (McFall and Rapraeger, 1997) at 1:50; rabbit
polyclonal anti-c-met (Santa Cruz) at 1:50; rabbit polyclonal antiFGFR-1 (Santa Cruz) at 1:333; rabbit polyclonal anti-laminin (Sigma) at 1:250, and mouse monoclonal anti-hemagglutinin (mAb 12 Ca5) at 2.5 g/ml or rabbit polyclonal anti-GST at 5 g/ml as negative controls. Secondary antibodies were conjugated with Alexa 546 or Alexa 488 (Molecular Probes) used at 1:500; nuclei were visualized with DAPI. When double-labeling with two rabbit polyclonal antibodies was required, the staining was done sequentially, blocking with 10% rabbit serum in PBS between the first and second set of incubations. Myofibers with their associated satellite cells were prepared as described previously (Cornelison and Wold, 1997). Briefly, muscle was dissected from adult mouse hindlimbs and digested with collagenase type I (Worthington) to yield single intact myofibers which were cultured in DMEM supplemented with 10% fetal bovine serum, 5% chick embryo extract (Gibco), and 10 M BrdU. BrdU is routinely added to cultures to facilitate cell cycle studies if desired; however, this measurement only applies to the first cell cycle after activation (at approximately 36 h after harvest in wild-type satellite cells) as all cells are labeled at this point. Additional supplements to the medium for chlorate experiments included 25 mM sodium chlorate, 10 mM sodium sulfate, 25 mM sodium chloride, and 250 g/ml heparin. At the designated time points after harvest, fibers were fixed and stained as described above. Primary antibodies and dilutions used also included mouse monoclonal anti-BrdU (BMB) at 1:10, mouse monoclonal antiMyoD (Novocastra) at 1:10, and mouse monoclonal anti-myogenin [F5D (Cusella-DeAngelis et al., 1992)], neat. Counts of resident satellite cells per 100 myonuclei were done by counting DAPI-stained myofiber nuclei (which can be identified by their characteristic elongated shape) and DAPI-stained satellite cell nuclei coincident with syndecan-4-positive cell outlines. At least 20 myofibers containing at least 5000 myonuclei total were counted per time point per condition. Counts of MyoD-positive satellite cells were done by counting MyoD-positive DAPI-stained nuclei within syndecan-4-positive cell outlines and comparing to counts of total DAPI-stained nuclei within syndecan-4-positive cell outlines.
RESULTS Syndecan Expression in Developing Muscle During development, fetal myoblasts resident in the limbs fuse to form muscle fibers and begin expressing contractile proteins (reviewed in Stockdale, 1992). It is thought that regulation of these differentiation events is controlled in part by the actions of FGF receptors (FlanaganSteet et al., 2000; Itoh et al., 1996; Marcelle et al., 1995). However, in order to form a functional signaling complex, appropriate cell-surface heparan sulfate proteoglycans must also be incorporated into the ligand-receptor complex, thus providing an additional level of regulatory control (reviewed in Zimmermann and David, 1999). While the expression of the four FGF receptors in the limb is known (Marcelle et al., 1994; Orr-Urtreger et al., 1991; Partanen et al., 1991; Stark et al., 1991; Szebenyi et al., 1995; Yamaguchi et al., 1992), the corresponding HSPG expression has not yet been examined as closely, particularly with regard to a possible role in myogenesis. Syndecan-1 has previously been shown to be transiently expressed in the early (⬍14 days) limb bud
Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.
83
Satellite Cell Maintenance and Muscle Regeneration
(Solursh et al., 1990) and syndecan-3 in known to be expressed in the developing limb (Gould et al., 1995). In the chick, the HSPG glypican is expressed in the apical ectodermal ridge (Niu et al., 1996) as well. To investigate the contribution of heparan sulfate to FGF binding in limb skeletal muscle, biotinylated-FGF2 was incubated with frozen sections of mouse forelimb derived from embryonic stages E14.5 and E18, as well as neonatal (P2) and young adult stages. FGF2 binds to numerous sites throughout the cartilaginous matrix of presumptive bone, the surrounding perichondrium, and within the developing skeletal muscle (Fig. 1). Within muscle, the binding is most prominent in a honeycomb pattern that represents the basement membrane surrounding developing and mature myotubes. This is less ordered at E14.5 (Fig. 1A), but becomes highly defined by the young adult stage (⬍60 days old; Fig. 1D). Staining for total heparan sulfate glycosaminoglycan was performed on sections pretreated with heparinases to remove the heparan sulfate chains. Treatment with these enzymes leaves a terminal unsaturated uronic acid on the stub of the glycosaminoglycan chain remaining attached to the proteoglycan core proteins; this is detected by staining with antibody 3G10 specific for these sites. This demonstrates that the FGF binding pattern mirrors the pattern of heparan sulfate distribution (compare Figs. 1C and 1F). Furthermore, prior removal of heparan sulfate by this enzyme treatment blocks the binding of FGF (Fig. 1E). Staining the sections with an antibody to perlecan, the abundant basement membrane heparan sulfate proteoglycan, indicates that it is distributed in the basement membranes with a similar distribution to the majority of the FGF binding (data not shown). Although FGF binding to the abundant heparan sulfate of the basement membrane is demonstrated, binding is also detected in connective tissue between the myotubes, and to localized sites or cells along the periphery of the myotubes (e.g., arrowheads in Fig. 1D). This suggests that specific proteoglycans, such as cell surface proteoglycans, may be localized within these discrete locations. To examine this hypothesis more closely, the sections were stained with antibodies to each of the four syndecans. We observed syndecan-3 staining in skeletal muscle of the forelimb throughout development and in young adult mouse muscle. Expression in the E14.5-day limb shows that syndecan-3 is expressed in cartilage chondrocytes of the developing long bones, in the perichondrium surrounding the cartilage, and in individual myoblasts of the developing skeletal muscle (Fig. 2A). The inset in Fig. 2A shows syndecan-3-positive myoblasts in aggregates prior to their fusion into myotubes. Staining with a nonspecific IgG is negative (shown for E14.5 in Fig. 2D) at all stages examined. By E18 (Fig. 2B), the myoblasts have assembled into presumptive myotubes, but individual myoblasts and a central lumen are still discerned (see inset). Syndecan-3 staining is observed around the entire surface of these myoblasts. Staining of the muscle is more intense than that of the surrounding connective tissue and perichondrium. By neo-
natal day 2, however, the syndecan-3 staining is becoming localized to discrete sites within the muscle (Fig. 2C). Individual myoblasts are no longer seen as fusion into multinucleate myotubes has taken place. Weak syndecan-3 staining is observed around the circumference of the myotubes, but stronger staining is seen at localized sites at the myotube periphery. These sites are seen only on a fraction of the myotubes in cross-section, suggesting that they may represent sites that are scattered intermittently along the length of individual myotubes. Staining of young adult muscle shows that the syndecan-3 staining is even more highly localized than that seen at P2 (Fig. 2E). The proteoglycan is seen in almost a punctate distribution bordering on the basal lamina that encapsulates the myotube. Occasional staining is seen around the entire myotube, but this staining is weak. Double staining with DAPI, to detect the nuclei of the multinucleate myotubes and of individual satellite cells, shows that the syndecan-3 staining corresponds with a subset of the nuclei in the section (Fig. 2F). Expression of syndecan-4 in the developing limb is very similar to that of syndecan-3. Expression is detected in individual chondrocytes of the developing long bone, within the perichondrium and developing skeletal muscle. Syndecan-4 is seen in the developing myoblasts at E14.5, with some cells staining more intensely (Fig. 3A). The proteoglycan is expressed around the lumen and periphery of early myotubes at E18 (Fig. 3B), with staining also observed in occasional cells that stain more brightly than their neighbors; the fate of these cells is not yet known. However, at P2, it is clear that individual cells are present that retain abundant syndecan-4 expression while expression on the entire fused myotube is less intense (Fig. 3C). In the young adult, the expression of syndecan-4 becomes more restricted to these localized sites, although it is not as punctate as that seen for syndecan-3 and some expression remains on the myotubes (Fig. 3E). As seen with syndecan-3, the localized sites of syndecan-4 expression coincide with sites of DAPI staining (Fig. 3F), identifying a subset of nuclei. The expression of syndecan-1 and syndecan-2 in limb skeletal muscle differs from the other two syndecans. At E14.5, syndecan-1 staining is detected weakly on the surfaces of myoblasts in the developing muscle (Fig. 4A). At E18, the staining is seen surrounding the lumen and the basal side of developing myotubes (Fig. 4B) and this staining has diminished or is nonexistent by P2 (Fig. 4C). No staining is seen in the young adult or adult (not shown). Syndecan-1 staining at other sites, such as in the epidermis, is much more intense (not shown). Syndecan-2 expression is seen in connective tissue and the perichondrium, but is not detected in muscle in any of the stages that were examined. Stage E18 is shown as an example (Fig. 4E).
Syndecan-3 and Syndecan-4 Are Expressed in Adult Muscle Satellite Cells The highly localized expression of syndecan-3 and syndecan-4 in neonatal and young adult muscle suggests
Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.
84
Cornelison et al.
FIG. 3. Immunolocalization of syndecan-4 in embryonic and adult mouse forelimb skeletal muscle. Frozen sections were stained with polyclonal anti-syndecan-4 antibody generated against GST-syndecan-4 fusion protein (A, B, C, E), antibody against GST alone (D), or DAPI (F). Developmental stages, arrowheads, and abbreviations are identical to Fig. 2. (Bar, 50 m).
that the expression of these proteoglycans is being restricted to satellite cells. To further test this hypothesis, adult muscles were examined in cross-section and costained for each syndecan and for either the satellite cell marker c-met or laminin to determine the relative position of the cell with respect to the basement membrane and muscle fiber membrane (Fig. 5). Staining for c-met is seen at
a small number of localized sites within the muscle, consistent with the number of satellite cells expected in adult tissue (Fig. 5). These sites are at the periphery of myotubes, also consistent with the possibility that they are localized to the satellite cells. Costaining with either syndecan-3 or syndecan-4 and c-met yields completely overlapping staining patterns (Fig. 5). As c-met is a universally accepted
Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.
85
Satellite Cell Maintenance and Muscle Regeneration
FIG. 4. Immunolocalization for syndecan-1 and syndecan-2 in developing skeletal muscle. Syndecan-1 staining is detected during early muscle development and is diminished by P2. Syndecan-1 is detected by mAb281.2 (A, E14.5); (B, E18); (C, P2). Nonspecific rat IgG is shown in (D) (E14.5). Skeletal muscle staining with syndecan-2, detected with mAb10H4 (E, E18) is negative. Nonspecific mouse IgG is shown in (D) (E18). Abbreviations are shown in Fig. 1. Connective tissue, CT. (Bar, 50 m).
marker for quiescent and activated satellite cells, syndecan-3 and syndecan-4 thus appear to identify satellite cells in adult skeletal muscle. To confirm that the cells staining positively for syndecan-3 and syndecan-4 are localized between the basal lamina and the myofiber membrane, we costained adult muscle tissue sections for syndecan-4 and laminin. The staining patterns clearly demonstrate that syndecan-4-positive cells are sandwiched between the enveloping basal lamina and the myotube proper. DAPI staining shows the presence of a nucleus at the great majority of these sites (Fig. 5).
Staining for either syndecan-3 or syndecan-4 demonstrates that satellite cells express both of these proteoglycans (Fig. 5). Quantifying the number of cell profiles that express syndecan-3 and also have a nucleus, express c-met, or express syndecan-4 shows that the correlation between all four markers is greater than 95% (Table 1).
Coexpression of Syndecans and FGF Receptor 1 The proliferation and differentiation of satellite cells is regulated by numerous factors; among factors thought to be
Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.
86
Cornelison et al.
important for these processes are HGF and FGFs. Signal transduction for both of these growth factors relies on the formation of ternary complexes containing the growth factor, its receptor tyrosine kinase, and heparan sulfate provided by heparan sulfate proteoglycans. To verify that the satellite cells coexpress FGF receptor in addition to syndecans and c-met, neonatal and young adult muscle was stained for FGF receptor 1 (FGFR-1) (Fig. 6). The receptor is expressed in skeletal muscle at both stages in a punctate distribution that colocalizes with syndecan-3 (Fig. 6). At the P2 stage, where the syndecan is not as highly localized as in the young adult, the FGFR-1 is already highly localized (Figs. 6A and 6C). In young adult muscle, syndecan-3 and FGFR-1 staining colocalize exclusively to satellite cells (Figs. 6B and 6D).
Satellite Cells on Isolated Fibers Express Syndecan-3 and Syndecan-4 If the syndecans present on intact satellite cells play a role in mediating satellite cell activation, proliferation, and/or differentiation, the proteoglycans should be expressed on satellite cells in myofiber cultures. We isolated single myofibers, cultured them, and compared staining for the syndecans, c-met, MyoD, and myogenin as a function of satellite cell proliferation. Removal of single myofibers from mouse skeletal muscle activates satellite cells, resulting in upregulation of myogenic regulatory factors such as MyoD (which is detected only rarely in fresh preparations but is prevalent by 24 h of culture) and synchronous division at about 36 and 72 h in culture. Syndecan-3 is coexpressed with c-met in satellite cells on myofibers at 0, 24, 48, and 96 h in culture (Fig. 7A). Similarly, syndecan-4 is also present and coexpressed with c-met (Fig. 7B). Together, these data and the syndecan staining in sections identify syndecan-3 and syndecan-4 as markers of quiescent and proliferating satellite cells. Syndecan-3 (Fig. 7C) and syndecan-4 (data not shown) are expressed prior to detectable expression of MyoD and remain following MyoD induction and proliferation, suggesting a potential role for these factors in satellite cell activation, proliferation, and differentiation.
Disruption of Heparan Sulfate-Mediated Signaling by Addition of Chlorate To determine whether the presence of syndecan-3 or syndecan-4 (or potentially other heparan sulfate proteoglycans) is required for activation or proliferation of satellite cells, myofibers were cultured in medium containing 25 mM sodium chlorate. Sodium chlorate is a competitive inhibitor of ATP sulfurylases (Baeuerle and Huttner, 1986) preventing protein sulfation, and thereby disrupts signaling via FGFRs (Rapraeger et al., 1991) and c-met (Deakin and Lyon, 1999). We have previously used chlorate to inhibit sulfation in a skeletal muscle cell line (MM14) that closely represents satellite cells (Olwin and Rapraeger, 1992; Ra-
praeger et al., 1991). Chlorate effectively reduces sulfation of all glycosaminoglycan chains in myoblast cultures. Restoration of biological activity with soluble heparin is significant, as this indicates that the biological responses observed are likely to be heparan sulfate-specific (Olwin and Rapraeger, 1992; Rapraeger et al., 1991). Although addition of soluble heparin to the culture may restore growth factor function, it will not replace signaling events requiring immobilized heparan sulfate, HSPG-mediated cell adhesion, or syndecan-independent signaling. When muscle fiber cultures were grown in chloratecontaining medium from the time of harvest, the number of associated satellite cells, while unchanged at 24 h of culture (prior to the first division after activation), was significantly decreased at 48 and 96 h, indicating that the presence of chlorate inhibits satellite cell division (Fig. 8A). BrdU labeling indices were less than 100% at 48 h but had recovered by 96 h (data not shown.) The majority of the effect seen at 96 h is probably due to defects in proliferation at earlier stages of the culture, resulting in a lag in cell accumulation. This decrease in cell number was not seen in a sodium chloride control, and could be rescued by addition of heparin (Fig. 8A), indicating that satellite cell proliferation is dependent on heparan sulfate. The number of satellite cell colonies adhering to the tissue culture plate was proportional to the number of fiber-associated cells in all cases, suggesting that the decrease in satellite cell number is due to a decrease in cell division rather than an increase in migration away from the host myofibers (data not shown). When the populations of fiber-associated satellite cells were assessed for expression of the myogenic regulatory factor MyoD, there was an initial increase in the percentage of satellite cells expressing MyoD at 24 h of culture in sodium chlorate, suggesting a switch from proliferation to differentiation-prone cells (Fig. 8B). However, the relative increase in MyoD was gone by 48 h in culture and by 96 h had in fact reversed (Fig. 8B.) Addition of heparin could completely rescue the chlorate-treated cells at both time points, suggesting an involvement of heparan sulfate proteoglycans in MyoD expression that involves heparan sulfate-dependent growth factors. Interestingly, when examined for expression of the myogenic regulatory factor myogenin (which is normally expressed only after the first mitotic division and tends to be associated with commitment to terminal differentiation), satellite cells in both conditions stained similarly (Fig. 8C). This, together with the apparent trend towards recovery by 96 h of culture, would seem to imply that the effects of chlorate treatment are at a maximum during the early stages of activation.
DISCUSSION We have examined the expression of the syndecan family of heparan sulfate proteoglycans, essential components of signaling complexes whose activity has been shown to be
Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.
87
Satellite Cell Maintenance and Muscle Regeneration
TABLE 1 Colocalization of Syndecan-3 with Other Proteins or Markers in Adult Muscle Satellite Cells Protein/marker
Image
No. of S3 ⫹ sites
No. of double ⫹ sites
DAPI
1 2 3 4 5
98 77 81 76 91
96 (98%) 73 (95%) 78 (95%) 72 (95%) 91 (100%) (Avg. 97 ⫾ 2%)
MET
1 2 3
82 94 77
80 (97.5%) 93 (99%) 75 (97%) (Avg. 98 ⫾ 1%)
Syn-4
1 2 3
116 58 68
116 (100%) 56 (96.5%) 68 (100%) (Avg. 99 ⫾ 2%)
Note. Sections from adult forelimb skeletal muscle were doublelabeled as described in Fig. 2. Two images of the same field of view showing either focal syndecan-3 localization or localization of nuclei (DAPI), c-met, or syndecan-4 were acquired. The number of these sites that also stained positive with DAPI or localized c-met or syndecan-4 (double positive) was then determined. The percentage of syndecan-3-positive sites that also were positive for DAPI, c-met, or syndecan-4 are in parentheses.
required for differentiation in myogenic cell lines (Larrain et al., 1997, 1998; Rapraeger et al., 1991). The ligands and receptors with which they interact have also been implicated in satellite cell activation and proliferation. We further examine the effects of interfering with heparan sulfate proteoglycan function in primary satellite cell culture, and, based on these results, suggest a role for these HSPGs in satellite cell function. An initial examination of syndecan expression patterns in adult muscle revealed a limited staining pattern for syndecan-3 and syndecan-4, suggesting they may be restricted to the satellite cells. To elucidate the roles of syndecans in skeletal muscle development and during muscle regeneration, we have assayed for expression of syndecans 1– 4 in embryonic, fetal, postnatal, young adult, and adult mouse muscle sections, as well as in singlemyofiber tissue culture preparations. In the developing animal, syndecan-1 is expressed transiently during embryonic and fetal development but does not appear to be expressed during later myogenesis or in satellite cells. This confirms a previous study of developmental expression for syndecan-1 (Solursh et al., 1990) and in cultured limb bud explants for syndecan-3 (Olguin and Brandan, 2001), but is in apparent contradiction of work suggesting that syndecan-1 is required for myogenic differentiation in the C2 cell line (Larrain et al., 1998). This discrepancy seems
most likely to be due to variations in gene expression and activity which have arisen as artifacts in the tissue culture lines used; both MM14 and C2C12 cells display gene expression profiles which deviate from those seen in primary satellite cells, with the C2 line being more divergent (Cornelison et al., manuscript in preparation). Syndecan-2 was not detectably expressed in skeletal muscle tissue from any age examined and is therefore unlikely to play a direct role in skeletal muscle development or regeneration. In contrast, syndecans 3 and 4 are expressed in myogenic cells in embryonic (ED14), fetal (ED18), neonatal (P2), young adult (⬍60 days), and adult (⬎10 wk) muscle tissue. Of particular interest is their expression in adult muscle, where immunoreactivity is restricted to cells which, based on morphology and position in the intact muscle, appear to be satellite cells. These data suggest a role for syndecan-3 and syndecan-4 in skeletal muscle development as well as skeletal muscle regeneration. When compared with expression of the satellite cell marker c-met, syndecan-3 and syndecan-4 expression correlates extremely well. When sections of adult muscle containing presumably quiescent satellite cells are costained for syndecan-3 and FGFR-1, a complete correlation is also seen. One important implication of these observations is that all of the components necessary for activation of both c-met and FGF receptors by their respective ligands are present on quiescent satellite cells. Taken together, our data suggest that quiescent satellite cells are capable of responding to an FGF or HGF signal without any additional requirements, such as protein synthesis. This supports the hypothesis that HGF and/or FGF(s) constitute the activating signal(s) for quiescent satellite cells in vivo. Despite their importance both as the cells responsible for myogenic regeneration after injury or disease and possibly during aging, and as potential vectors in cell-based therapies, mechanisms involving key processes in satellite cell biology are currently not well understood. A properly functioning satellite cell must remain quiescent in healthy muscle, become activated in response to an insult to the local muscle tissue, proliferate at the site of the injury to produce a pool of replacement myoblasts, differentiate in situ into the required fiber type and size, and possibly return to a quiescent state after the regeneration response has been completed. While studies of myogenesis in the embryo and in tissue culture lines may provide suggestions for the mechanisms in place during adult myogenesis, there are sufficiently substantial differences in gene expression patterns and behavior between satellite cells and embryonic or fetal myoblasts (Cornelison et al., 2000), and particularly between primary satellite cells and myogenic cell lines (Cornelison et al., manuscript in preparation), that examination of primary satellite cells is the optimal method for determining satellite cell-specific gene expression and activity. Based on work done in tissue culture, FGFs are excellent candidates for mediators of these activities; all myoblast cells lines studied respond to exogenously applied FGFs, and at least one transformed satellite cell line
Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.
88
Cornelison et al.
FIG. 5. c-met, laminin, and syndecan staining of adult skeletal muscle. Top and middle panels: sections of adult mouse hindlimb costained for the satellite cell marker c-met (red) and syndecan-3 (green) or syndecan-4 (green), demonstrating that syndecan-3 and syndecan-4 consistently colocalize with c-met on presumptive satellite cells. Lower panel: sections of mouse hindlimb costained for syndecan-4 (green), laminin (red), and DAPI (blue), showing that cells in the satellite position in quiescent muscle express syndecan-4.
(MM14) requires FGF signaling both to support proliferation and to suppress myogenic differentiation (Fedorov et al., 2001; Jones et al., 2001; Kudla et al., 1998; Linkhart et al., 1981; Olwin and Hauschka, 1986). Finally, FGFR-1 staining colocalizes with c-met, syndecan-3, and syndecan-4 in sections of young adult muscle and in satellite cells on intact myofibers (data not shown). When examined in sections from mdx mice (data not shown) and in single-fiber culture, activated satellite cells persist in expression of syndecan-3 and syndecan-4. Together, these data support the hypothesis that FGFs may be involved regulating satellite cell activation, proliferation, and differentiation. Given this, would the effects of FGF signaling be altered in vitro by reducing or eliminating the available heparan sulfate? To address this question, we treated fresh cultures of primary satellite cells resident on their host myofibers with sodium chlorate, which prevents sulfation of GAG chains and renders them unable to interact with either HGF or FGF ligands, thus eliminating signaling from both c-met and FGF receptors (Deakin and Lyon, 1999; Rapraeger et al.,
1991). Although initial numbers of chlorate-treated and untreated satellite cells (which have been freshly harvested and thereby newly activated) are identical at 24 h as should be expected, treated cells display decreased cell counts by 48 h postharvest, a time when all untreated satellite cells have undergone their first mitotic division. This condition persists throughout the first 96 h in culture. This effect is not observed in sodium chloride-treated controls, and can be fully rescued by addition of exogenous heparin. Since chlorate treatment dramatically reduces sulfation of all proteins, recovery with heparin strongly suggests that satellite cell proliferation requires heparan sulfate. Since the proliferation response can be rescued by exogenous heparin, it is likely that loss of heparan sulfate eliminates a response to one or more growth factors. Two potential candidates involved in satellite cell proliferation are HGF and FGF family members, both of which are heparan sulfatedependent. However, HGF activity cannot be restored by soluble heparin in chlorate-treated cells and requires immobilized heparan sulfate (Deakin and Lyon, 1999). Our data therefore suggest that HGF may not be involved in regula-
Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.
89
Satellite Cell Maintenance and Muscle Regeneration
FIG. 6. Colocalization of syndecan-3 and FGFR-1 in neonatal and adult forelimb skeletal muscle. (A, C) Same section of P2 neonate forelimb incubated with anti-syndecan-3 and anti-FGFR-1, respectively. (B, D) Same section of young adult forelimb incubated with anti-syndecan-3 and anti-FGFR-1, respectively. Bar, 20 m; inset magnification is 3.15⫻ greater than larger image.
tion of satellite cell proliferation and FGFs remain highly likely candidates. Thus, the presence of syndecan-3 and/or syndecan-4 on satellite cells appears to be required for a functional response for proliferation that may be mediated via FGFs. These data do not distinguish between the requirement for an initial proliferation signal versus a continuous signal; since the satellite cells were all activated at the same time by the initial harvest, they are roughly synchronized for the first several cell cycles. An analogous state can be induced in the satellite cell line MM14, which requires exogenous FGF to maintain proliferation and repress myogenesis (Clegg et al., 1987). Removal of serum from the culture medium in the presence of FGF maintains the cells in a quiescent, undifferentiated state, similar to the chloratetreated satellite cells. To better determine the effects of HSPG loss on myogenesis in satellite cells on intact myofibers, we examined the expression pattern of the myogenic regulatory factors MyoD and myogenin in untreated and chlorate-treated fiber cultures. Chlorate-treated satellite cells first display an increase in the percentage of MyoDpositive cells over untreated, possibly representing cells
that would otherwise proliferate beginning to differentiate precociously. By 96 h, chlorate-treated cells have begun to recover; however, the percentage of MyoD-positive cells is still less than is seen in untreated cultures. These cells may be “stalled” with no clear signal to either proliferate or differentiate, a hypothesis which is supported by the observation that chlorate-treated cells form less than half the number of myosin-positive colonies on the culture plate when examined at 96 h than either BrdU-only or heparinrescued cultures (data not shown). Similar to the effect on proliferation, changes in MyoD expression can be rescued by adding heparin. This suggests the involvement of heparan sulfate-dependent growth factors in regulation of differentiation as well as regulation of proliferation. Expression of the myogenic regulatory factor myogenin, which is a significantly later event than initiation of MyoD upregulation and is associated with cells entering a differentiative state, was also examined. Under no conditions was any myogenin expression detected at 24 h, implying that the elevated levels of MyoD observed in chloratetreated cells have not led to a premature commitment to terminal differentiation. Additionally, at 96 h, there is only
Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.
90
Cornelison et al.
FIG. 7. Immune staining of fiber-associated satellite cells. Fibers were stained immediately after harvest (0 h), during early activation (24 h postharvest), and after onset of differentiation (48 and 96 h postharvest). (A) Coexpression of syndecan-3 (green) and the satellite cell marker c-met (red) in fiber-associated satellite cells. (B) Coexpression of syndecan-4 (green) and the satellite cell marker c-met (red) in fiber-associated satellite cells. (C) Coexpression of syndecan-3 (green) with MyoD (red) and DAPI (blue).
a slight increase in the percentage of myogenin-positive cells in chlorate-treated vs. BrdU-only cultures. Taken together with the proliferation and MyoD data, this suggests that the major effects of chlorate treatment are on the early events of satellite cell activation and initiation of proliferation, although it should be noted that satellite cells appear to be unique among myogenic cell types in that myogenin-expressing cells retain the capability to divide (Cornelison and Wold, 1997; Maley et al., 1994). At later times, perhaps either the requirements for heparan sulfatemediated signaling are decreased or the cells have successfully scavenged sufficient sulfate present in the environment to ameliorate the effects of chlorate inhibition. When considering these data, it should also be taken into account that, unlike myogenic cell lines or cultured populations of pure primary myoblasts, these cultures also contain the host fibers, which in fact make up the majority of cell mass in the cultures. It has been shown previously
that primary satellite cells in mass culture express MyoD and myogenin with delayed kinetics relative to in situ crush injuries (Maley et al., 1994); satellite cells in singlefiber cultures have an expression profile which is more similar to that seen in crush injuries, expressing MyoD protein at detectable levels by 6 h postharvest but with myogenin expression delayed until closer to 48 h (Cornelison and Wold, 1997). Therefore, the contribution of the fibers in the form of cell– cell interactions, paracrine or secreted growth factors, or other components may affect the behavior of the satellite cells under these conditions but do not fully reconstitute the environment found in vivo. Our data also have several practical implications. The syndecan-3 and syndecan-4 genes may be considered potential markers for both quiescent and activated satellite cells in situ: such markers are rare, and the reagents available for detecting them are not always reliable. Thus, proteins that are expressed exclusively in satellite cells yet universally
Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.
FIG. 7—Continued
92
Cornelison et al.
within the satellite cell population and for which strong and specific antibodies exist represent a powerful new tool for the study of satellite cells. Here, we report the identification of two new such markers, syndecan-3 and syndecan-4. Future experiments will address further the role(s) of these HSPGs in satellite cell activation, proliferation, and differentiation.
ACKNOWLEDGMENTS This work was supported by grants from the NIH (AR39467) and MDA (to B.B.O.) and NIH (GM48850 and HD21881) (to A.C.R.). D.D.W.C. is supported by NIH Training Grant HL07851.
REFERENCES
FIG. 8. Chlorate treatment of fiber-associated satellite cells. (A) Representation of absolute satellite cell number per 100 myonuclei in myofiber cultures. DAPI-stained myofiber nuclei and DAPI-stained satellite cell nuclei colocalizing with syndecan-4 were counted from myofibers at the stated time points after harvest. Additions to medium were: B, 10 M BrdU only; BN, BrdU ⫹ 25 mM NaCl (sodium chloride); BC, BrdU ⫹ 25 mM NaClO 3 (sodium chlorate); BCH, BrdU ⫹ sodium chlorate ⫹ 250 mg/ml heparin. Note that the number of satellite cells at 96 h results from the cumulative proliferation history, and decreased numbers do not necessarily reflect a continued defect in proliferation. No error bars exist because the absolute number of satellite cells per 100 myonuclei is depicted. (B) Representation of percentage of satellite cells in fiber culture express MyoD protein. DAPI-stained satellite cell nuclei within syndecan-4positive cell outlines were examined for immunoreactivity for MyoD at the stated timepoints after harvest. Media additions are as in (A). Error bars represent calculated standard error. (C) Representation of percentage of satellite cells in fiber culture expressing myogenin protein. DAPI-stained satellite cell nuclei within syndecan-4-positive cell outlines were examined for immunoreactivity for myogenin at the stated time points after harvest. Media additions are as in (A). Error bars represent calculated standard error.
Allen, R. E., Sheehan, S. M., Taylor, R. G., Kendall, T. L., and Rice, G. M. (1995). Hepatocyte growth factor activates quiescent skeletal muscle satellite cells in vitro. J. Cell. Physiol. 165, 307–312. Baeuerle, P. A., and Huttner, W. B. (1986). Chlorate: A potent inhibitor of protein sulfation in intact cells. Biochem. Biophys. Res. Commun. 141, 870 – 877. Bischoff, R. (1994). In “Myology” (A. G. Engel and C. FranziniArmstrong, Eds.), pp. 97–118. McGraw-Hill, New York. Bladt, F., Riethmacher, D., Isenmann, S., Aguzzi, A., and Birchmeier, C. (1995). Essential role for the c-met receptor in the migration of myogenic precursor cells into the limb bud. Nature 376, 768 –771. Bottaro, D. P., Rubin, J. S., Faletto, D. L., Chan, A. M. L., Kmiecik, T. E., Woude, G. F. V., and Aaronson, S. A. (1991). Identification of hepatocyte growth factor receptor as the met proto-oncogene product. Science 251, 802– 804. Carey, D. J., Evans, D. M., Stahl, R. C., Asundi, V. K., Conner, K. J., Garbes, P., and Cizmeci-Smith, G. (1992). Molecular cloning and characterization of N-syndecan, a novel transmembrane heparan sulfate proteoglycan. J. Cell Biol. 117, 191–201. Clegg, C. H., Linkhart, T. A., Olwin, B. B., and Hauschka, S. D. (1987). Growth factor control of skeletal muscle differentiation: Commitment to terminal differentiation occurs in G 1 phase and is repressed by fibroblast growth factor. J. Cell Biol. 105, 949 – 956. Colvin, J. S., Bohne, B. A., Harding, G. W., Mcewen, D. G., and Ornitz, D. M. (1996). Skeletal overgrowth and deafness in mice lacking fibroblast growth factor receptor 3. Nat. Genet. 12, 390 –397. Cornelison, D. D. W., Olwin, B. B., Rudnicki, M. A., and Wold, B. J. (2000). MyoD⫺/⫺ satellite cells in single fiber culture are differentiation defective and MRF4 deficient. Dev. Biol. 224, 122–137. Cornelison, D. D. W., and Wold, B. J. (1997). Single-cell analysis of regulatory gene expression in quiescent and activated mouse skeletal muscle satellite cells. Dev. Biol. 191, 270 –283. Cusella-DeAngelis, M. G., Lyons, G., Sonnino, C., DeAngelis, L., Vivarelli, E., Farmer, K., Wright, W. E., Molinaro, M., Bouche, M., Buckingham, M., and Cossu, G. (1992). MyoD, myogenin independent differentiation of primordial myoblasts in mouse somites. J. Cell Biol. 116, 1243–1255. David, G., Bai, X. M., Van der Schueren, B., Manyen, P., Cassiman, J. J., and Van den Berghe, H. (1993). Spatial and temporal changes
Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.
93
Satellite Cell Maintenance and Muscle Regeneration
in the expression of fibroglycan (syndecan-2) during mouse embryonic development. Development 119, 841– 854. Deakin, J. A., and Lyon, M. (1999). Differential regulation of hepatocyte growth factor/scatter factor by cell surface proteoglycans and free glycosaminoglycan chains. J. Cell Sci. 112, 1999 – 2009. Deng, C. X., Wynshaw-Boris, A., Shen, M. M., Daugherty, C., Ornitz, D. M., and Leder, P. (1994). Murine FGFR-1 is required for early postimplantation growth and axial organization. Genes Dev. 8, 3045–3057. Fedorov, Y. V., Rosenthal, R. S., and Olwin, B. B. (2001). Oncogenic ras-induced proliferation requires autocrine fibroblast growth factor 2 signaling in skeletal muscle cells. J. Cell Biol. 152, 1301–1306. Flanagan-Steet, H., Hannon, K., McAvoy, M. J., Hullinger, R., and Olwin, B. B. (2000). Loss of FGF receptor 1 signaling reduces skeletal muscle mass and disrupts myofiber organization in the developing limb. Dev. Biol. 218, 21–37. Floss, T., Arnold, H. H., and Braun, T. (1997). A role for FGF-6 in skeletal muscle regeneration. Genes Dev. 11, 2040 –2051. Gould, S. E., Upholt, W. B., and Kosher, R. A. (1995). Characterization of chicken syndecan-3 as a heparan sulfate proteoglycan and its expression during embryogenesis. Dev. Biol. 168, 438 – 451. Ishiguro, K., Kadomatsu, K., Kojima, T., Muramatsu, H., Tsuzuki, S., Nakamura, E., Kusugami, K., Saito, H., and Muramatsu, T. (2000). Syndecan-4 deficiency impairs focal adhesion formation only under restricted conditions. J. Biol. Chem. 275, 5249 –5252. Itoh, N., Mima, T., and Mikawa, T. (1996). Loss of fibroblast growth factor receptors is necessary for terminal differentiation of embryonic limb muscle. Development 122, 291–300. Johnson, S. E., and Allen, R. E. (1995). Activation of skeletal muscle satellite cells and the role of fibroblast growth factor receptors. Exp. Cell Res. 219, 449 – 453. Jones, N. C., Fedorov, Y. V., Rosenthal, R. S., and Olwin, B. B. (2001). ERK1/2 is required for myoblast proliferation but is dispensable for muscle gene expression and cell fusion. J. Cell Physiol. 186, 104 –115. Kudla, A. J., Jones, N. C., Rosenthal, R. S., Arthur, K., Clase, K. L., and Olwin, B. B. (1998). The FGF receptor-1 tyrosine kinase domain regulates myogenesis but is not sufficient to stimulate proliferation. J. Cell Biol. 142, 241–250. Larrain, J., Carey, D. J., and Brandan, E. (1998). Syndecan-1 expression inhibits myoblast differentiation through a basic fibroblast growth factor-dependent mechanism. J. Biol. Chem. 273, 32288 – 32296. Larrain, J., Cizmeci-Smith, G., Troncoso, V., Stahl, R. C., Carey, D. J., and Brandan, E. (1997). Syndecan-1 expression is downregulated during myoblast terminal differentiation. Modulation by growth factors and retinoic acid. J. Biol. Chem. 272, 18418 – 18424. Linkhart, T. A., Clegg, C. H., and Hauschka, S. D. (1981). Myogenic differentiation in permanent clonal myoblast cell lines: Regulation by macromolecular growth factors in the culture medium. Dev. Biol. 86, 19 –30. Maina, F., Casagranda, F., Audero, E., Simeone, A., Comoglio, P. M., Klein, R., and Ponzetto, C. (1996). Uncoupling of Grb2 from the Met receptor in vivo reveals complex roles in muscle development. Cell 87, 531–542. Maley, M. A., Fan, Y., Beilharz, M. W., and Grounds, M. D. (1994). Intrinsic differences in MyoD and myogenin expression between primary cultures of SJL/J and BALB/C skeletal muscle. Exp. Cell Res. 211, 99 –107.
Marcelle, C., Eichmann, A., Halevy, O., Breant, C., and LeDouarin, N. M. (1994). Distinct developmentalexpression of a new avian fibroblast growth factor receptor. Development 120, 683– 694. Marcelle, C., Wolf, J., and Bronner-Fraser, M. (1995). The in vivo expression of the FGF receptor Frek mRNA in avian myoblasts suggests a role in muscle growth and differentiation. Dev. Biol. 172, 100 –114. McFall, A. J., and Rapraeger, A. C. (1997). Identification of an adhesion site within the syndecan-4 extracellular protein domain. J. Biol. Chem. 272, 12901–12904. Niu, S., Antin, P. B., Akimoto, K., and Morkin, E. (1996). Expression of avian glypican is developmentally regulated. Dev. Dyn. 207, 25–34. Olguin, H., and Brandan, E. (2001). Expression and localization of proteoglycans during limb myogenic activation. Dev. Dyn. 221, 106 –115. Olwin, B. B., and Hauschka, S. D. (1986). Identification of the fibroblast growth factor receptor of Swiss 3T3 cells and mouse skeletal muscle myoblasts. Biochemistry 25, 3487–3492. Olwin, B. B., and Rapraeger, A. (1992). Repression of myogenic differentiation by aFGF, bFGF, and K-FGF is dependent on cellular heparan sulfate. J. Cell Biol. 118, 631– 639. Orr-Urtreger, A., Givol, D., Yayon, A., Yarden, Y., and Lonai, P. (1991). Developmental expression of two murine fibroblast growth factor receptors, flg and bek. Development 113, 1419 – 1434. Partanen, J., Ma¨kela¨, T. P., Eerola, E., Korhonen, J., Hirvonen, H., Claesson Welsh, L., and Alitalo, K. (1991). FGFR-4, a novel acidic fibroblast growth factor receptor with a distinct expression pattern. EMBO J. 10, 1347–1354. Pellegrini, L., Burke, D. F., von Delft, F., Mulloy, B., and Blundell, T. L. (2000). Crystal structure of fibroblast growth factor receptor ectodomain bound to ligand and heparin. Nature 407, 1029 – 1034. Perrimon, N., and Bernfield, M. (2000). Specificities of heparan sulphate proteoglycans in developmental processes. Nature 404, 725–728. Rapraeger, A. C. (2000). Syndecan-regulated receptor signaling. J. Cell Biol. 149, 995–998. Rapraeger, A. C., Krufka, A., and Olwin, B. B. (1991). Requirement of heparan sulfate for bFGF-mediated fibroblast growth and myoblast differentiation. Science 252, 1705–1708. Schlessinger, J., Plotnikov, A. N., Ibrahimi, O. A., Eliseenkova, A. V., Yeh, B. K., Yayon, A., Linhardt, R. J., and Mohammadi, M. (2000). Crystal structure of a ternary FGF-FGFR-heparin complex reveals a dual role for heparin in FGFR binding and dimerization. Mol. Cell 6, 743–750. Schultz, E., and McCormick, K. M. (1994). Skeletal muscle satellite cells. Rev. Physiol. Biochem. Pharmacol. 123, 213–257. Solursh, M., Reiter, R. S., Jensen, K. L., Kato, M., and Bernfield, M. (1990). Transient expression of a cell surface heparan sulfate proteoglycan (syndecan) during limb development. Dev. Biol. 140, 83–92. Stark, K. L., McMahon, J. A., and McMahon, A. P. (1991). FGFR-4, a new member of the fibroblast growth factor receptor family, expressed in the definitive endoderm and skeletal muscle lineages of the mouse. Development 113, 641– 651. Stockdale, F. E. (1992). Myogenic cell lineages. Dev. Biol. 154, 284 –298. Szebenyi, G., Savage, M. P., Olwin, B. B., and Fallon, J. F. (1995). Changes in the expression of FGF receptors mark distinct stages
Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.
94
Cornelison et al.
of chondrogenesis in vitro and during chick limb skeletal patterning. Dev. Dyn. 204, 446 – 456. Tatsumi, R., Anderson, J. E., Nevoret, C. J., Halevy, O., and Allen, R. E. (1998). HGF/SF is present in normal adult skeletal muscle and is capable of activating satellite cells. Dev. Biol. 194, 114 –128. Weinstein, M., Xu, X., Ohyama, K., and Deng, C. X. (1998). FGFR-3 and FGFR-4 function cooperatively to direct alveogenesis in the murine lung. Development 125, 3615–3623. Wilkie, A. O., Morriss-Kay, G. M., Jones, E. Y., and Heath, J. K. (1995). Functions of fibroblast growth factors and their receptors. Curr. Biol. 5, 500 –507. Xu, X., Weinstein, M., Li, C., Naski, M., Cohen, R. I., Ornitz, D. M., Leder, P., and Deng, C. (1998). Fibroblast growth factor receptor 2 (FGFR2)-mediated reciprocal regulation loop between FGF8 and FGF10 is essential for limb induction. Development 125, 753–765.
Yamaguchi, T. P., Conlon, R. A., and Rossant, J. (1992). Expression of the fibroblast growth factor receptor FGFR-1/flg during gastrulation and segmentation in the mouse embryo. Dev. Biol. 152, 75– 88. Yamaguchi, T. P., Harpal, K., Henkemeyer, M., and Rossant, J. (1994). fgfr-1 is required for embryonic growth and mesodermal patterning during mouse gastrulation. Genes Dev. 8, 3032–3044. Yayon, A., Klagsbrun, M., Esko, J. D., Leder, P., and Ornitz, D. M. (1991). Cell surface, heparin-like molecules are required for binding of basic fibroblast growth factor to its high affinity receptor. Cell 64, 841– 848. Zimmermann, P., and David, G. (1999). The syndecans, tuners of transmembrane signaling. FASEB J. 13, S91–S100. Received for publication May Revised August Accepted August Published online September
Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.
23, 14, 14, 26,
2001 2001 2001 2001
Syndecan-3 and Syndecan-4 Specifically Mark Skeletal Muscle Satellite Cells and Are Implicated in Satellite Cell Maintenance and Muscle Regeneration D. D. W. Cornelison,* ,1 Mark S. Filla,† ,1 Heather M. Stanley,* Alan C. Rapraeger,† ,2 and Bradley B. Olwin* ,2 *Department of Molecular, Cellular and Developmental Biology, University of Colorado, Boulder, Colorado 80309; and †Department of Pathology and Laboratory Medicine, University of Wisconsin, Madison, Wisconsin 53706
Myogenesis in the embryo and the adult mammal consists of a highly organized and regulated sequence of cellular processes to form or repair muscle tissue that include cell proliferation, migration, and differentiation. Data from cell culture and in vivo experiments implicate both FGFs and HGF as critical regulators of these processes. Both factors require heparan sulfate glycosaminoglycans for signaling from their respective receptors. Since syndecans, a family of cell-surface transmembrane heparan sulfate proteoglycans (HSPGs) are implicated in FGF signaling and skeletal muscle differentiation, we examined the expression of syndecans 1– 4 in embryonic, fetal, postnatal, and adult muscle tissue, as well as on primary adult muscle fiber cultures. We show that syndecan-1, -3, and -4 are expressed in developing skeletal muscle tissue and that syndecan-3 and -4 expression is highly restricted in adult skeletal muscle to cells retaining myogenic capacity. These two HSPGs appear to be expressed exclusively and universally on quiescent adult satellite cells in adult skeletal muscle tissue, suggesting a role for HSPGs in satellite cell maintenance or activation. Once activated, all satellite cells maintain expression of syndecan-3 and syndecan-4 for at least 96 h, also implicating these HSPGs in muscle regeneration. Inhibition of HSPG sulfation by treatment of intact myofibers with chlorate results in delayed proliferation and altered MyoD expression, demonstrating that heparan sulfate is required for proper progression of the early satellite cell myogenic program. These data suggest that, in addition to providing potentially useful new markers for satellite cells, syndecan-3 and syndecan-4 may play important regulatory roles in satellite cell maintenance, activation, proliferation, and differentiation during skeletal muscle regeneration. © 2001 Academic Press
INTRODUCTION Skeletal muscle development in the prospective limb consists of migration of committed myoblasts from limblevel somites into the limb field, colonization of the premuscle masses, patterning of future muscles, and differentiation into fast and slow contractile myofibers. Later, myogenesis in the adult (mediated by adult myoblasts or muscle satellite cells) involves similar processes in the regenerative response to damage or disease. Growth factors 1
D.D.W.C. and M.S.F. contributed equally to this work. To whom correspondence should be addressed. Fax: (303) 4921587 (B.B.O.) or (608) 265-3301 (A.C.R.). E-mail: Bradley.Olwin@ colorado.edu or [email protected]. 2
0012-1606/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.
such as members of the fibroblast growth factor (FGF) family and scatter factor/hepatocyte growth factor (HGF) play diverse and critical roles during development of the vertebrate limb and its musculature, regulating growth, patterning, and cellular differentiation in developing muscle tissue. HGF signaling is required for emigration of myoblasts from the somites into the limb field (Bladt et al., 1995) and for secondary myogenesis (Maina et al., 1996). FGF signaling regulates growth and differentiation of early embryonic primary myoblasts (Flanagan-Steet et al., 2000; Itoh et al., 1996) as well as secondary and adult myoblasts (Flanagan-Steet et al., 2000). FGF and HGF signals are transduced through transmembrane receptor tyrosine kinases: in the case of FGFs, four such receptors with differing affinities for individual FGF
79
80
Cornelison et al.
FIG. 1. Binding of FGF2 to endogenous heparan sulfate in mouse forelimb skeletal muscle. Fixed frozen sections were incubated with 5 nM biotinylated FGF-2 to visualize FGF binding sites (A–E). The dependence of the binding on heparan sulfate was examined by pretreating the sections with heparinase I and II (E). Total heparan sulfate was visualized by staining with antibody 3G10 (F). Stages examined were E14.5 (A), E18 (B), postnatal day 2 (P2) (C, F), and young adult (D). Cartilage (Ca), perichondrium (PC), and skeletal muscle (SkM) are denoted. Bar, 50 m.
family members (FGFR 1– 4) are known (reviewed in Wilkie et al., 1995); the receptor for HGF is c-met (Bottaro et al., 1991). When ligand is bound, receptor dimers transautophosphorylate and activate diverse and as yet poorly understood downstream signaling events. When signaling is
interfered with, either by loss of the ligand, use of dominant-negative receptor mutants, or nonfunctional signaling pathway mutants, the effects in the developing limb and musculature vary by receptor. These include embryonic lethality [loss of FR1 (Deng et al., 1994; Yamaguchi et
Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.
81
Satellite Cell Maintenance and Muscle Regeneration
FIG. 2. Immunolocalization of syndecan-3 in embryonic and adult mouse forelimb skeletal muscle. Sections of forelimb skeletal muscle from embryonic and postnatal mice were stained with polyclonal anti-syndecan-3 antibody (A, B, C, E) or nonspecific IgG (D). Developmental stages shown are E14.5 (A, D), E18 (B), P2 (C), and young adult (E, F). Insets show higher magnification views of developing myotubes. Young adult forelimb skeletal muscle is double stained for syndecan-3 (E) and DAPI (F). Arrowheads in (C), (E), and (F) show localized expression of syndecan-3 in P2 and young adult stages and its correspondence with individual nuclei. (Bar, 50 m).
al., 1994)]; embryonic lethality and failure to form limb buds [loss of FR2 (Xu et al., 1998)]; skeletal overgrowth [loss of FR3 (Colvin et al., 1996)]; no apparent phenotype [loss of FR4 (Weinstein et al., 1998)]; skeletal muscle abnormalities [dominant-negative FR1 in developing muscle (FlanaganSteet et al., 2000; Itoh et al., 1996)]; inhibition of limb and tongue muscle precursor cell migration [loss of c-met (Bladt et al., 1995)]; abnormalities in secondary myogenesis [c-met
mutant (Maina et al., 1996)] and in skeletal muscle regeneration [FGF6 null mutant (Floss et al., 1997)]. FGFs and HGF exhibit an absolute requirement for heparan sulfate in order to transduce an intracellular signal (reviewed in Rapraeger, 2000). It has been suggested that the role of heparan sulfate in the ternary signaling complex may be to facilitate receptor activation, required for intracellular autophosphorylation and signal transduction (Deakin and
Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.
82
Cornelison et al.
Lyon, 1999; Rapraeger et al., 1991; Yayon et al., 1991). Recent x-ray crystallography evidence suggests that heparan sulfate chains are responsible for FGF receptor dimerization and promotion of high affinity FGF binding by contacting both FGF receptor monomers and FGF ligands (Pellegrini et al., 2000; Schlessinger et al., 2000). Developmentally, deleting expression of specific syndecans has produced no detectable phenotype in vivo (Ishiguro et al., 2000; Perrimon and Bernfield, 2000), suggesting that there is functional redundancy between syndecans and possibly other HSPGs as well. During adult life, skeletal myogenesis in mammals is restricted to the activity of muscle satellite cells. These myogenic stem cells, whose exact embryonic origin is as yet unclear, are characterized by their position beneath the basal laminae of mature muscle fibers and their normally quiescent metabolic state. When the local muscle is damaged due to injury or disease, satellite cells are responsible for repair and regeneration of healthy muscle (reviewed in Bischoff, 1994; Schultz and McCormick, 1994). Both FGFs and HGF, along with their cognate receptors, have been strongly implicated as mediators of these functions. Quiescent satellite cells express the c-met receptor as well as FGFR-1 and FGFR-4, and will respond in vivo and in vitro to exogenously added FGF (FGF2 in particular) and HGF; HGF and FGFs are normally present in muscle tissue and appear to be released upon injury and expressed at higher levels during regeneration (Allen et al., 1995; Cornelison et al., 2000; Cornelison and Wold, 1997; Johnson and Allen, 1995; Tatsumi et al., 1998). Here, we present new data on the expression of members of the syndecan family of HSPGs during limb development and adult life, as well as functional studies on the effects of disrupting heparan sulfate during muscle regeneration in single-fiber culture. We find that, in addition to being coexpressed with receptor tyrosine kinases implicated in regulation of satellite cell activity, heparan sulfate appears to be required for proper progression through the satellite cell myogenic program, suggesting that signaling through receptors such as FGF receptors and c-met is also required. The epitopes identified in this work also represent new molecular tools for identification of quiescent and activated satellite cells.
MATERIALS AND METHODS Tissue samples from embryonic, postnatal, or young adult mice were embedded in OCT (Tissue-Tek), sectioned for staining, and fixed in 4% paraformaldehyde. Adult hindlimb samples were fixed in 4% paraformaldehyde prior to embedding and sectioning. Primary antibodies and dilutions used were: mouse monoclonal anti-syndecan-1 (mAb 2E9, courtesy of Dr. Guido David, University of Leuven, Belgium) at 2.5 g/ml; mouse monoclonal antisyndecan-2 [mAb 10H4, courtesy of Dr. Guido David (David, 1993)] at 2.5 g/ml; rabbit polyclonal anti-syndecan-3 (Carey et al., 1992) at 1:50; rabbit polyclonal anti-GST-syndecan-4 extracellular domain (␣-GST-mS4ED) (McFall and Rapraeger, 1997) at 1:50; rabbit
polyclonal anti-c-met (Santa Cruz) at 1:50; rabbit polyclonal antiFGFR-1 (Santa Cruz) at 1:333; rabbit polyclonal anti-laminin (Sigma) at 1:250, and mouse monoclonal anti-hemagglutinin (mAb 12 Ca5) at 2.5 g/ml or rabbit polyclonal anti-GST at 5 g/ml as negative controls. Secondary antibodies were conjugated with Alexa 546 or Alexa 488 (Molecular Probes) used at 1:500; nuclei were visualized with DAPI. When double-labeling with two rabbit polyclonal antibodies was required, the staining was done sequentially, blocking with 10% rabbit serum in PBS between the first and second set of incubations. Myofibers with their associated satellite cells were prepared as described previously (Cornelison and Wold, 1997). Briefly, muscle was dissected from adult mouse hindlimbs and digested with collagenase type I (Worthington) to yield single intact myofibers which were cultured in DMEM supplemented with 10% fetal bovine serum, 5% chick embryo extract (Gibco), and 10 M BrdU. BrdU is routinely added to cultures to facilitate cell cycle studies if desired; however, this measurement only applies to the first cell cycle after activation (at approximately 36 h after harvest in wild-type satellite cells) as all cells are labeled at this point. Additional supplements to the medium for chlorate experiments included 25 mM sodium chlorate, 10 mM sodium sulfate, 25 mM sodium chloride, and 250 g/ml heparin. At the designated time points after harvest, fibers were fixed and stained as described above. Primary antibodies and dilutions used also included mouse monoclonal anti-BrdU (BMB) at 1:10, mouse monoclonal antiMyoD (Novocastra) at 1:10, and mouse monoclonal anti-myogenin [F5D (Cusella-DeAngelis et al., 1992)], neat. Counts of resident satellite cells per 100 myonuclei were done by counting DAPI-stained myofiber nuclei (which can be identified by their characteristic elongated shape) and DAPI-stained satellite cell nuclei coincident with syndecan-4-positive cell outlines. At least 20 myofibers containing at least 5000 myonuclei total were counted per time point per condition. Counts of MyoD-positive satellite cells were done by counting MyoD-positive DAPI-stained nuclei within syndecan-4-positive cell outlines and comparing to counts of total DAPI-stained nuclei within syndecan-4-positive cell outlines.
RESULTS Syndecan Expression in Developing Muscle During development, fetal myoblasts resident in the limbs fuse to form muscle fibers and begin expressing contractile proteins (reviewed in Stockdale, 1992). It is thought that regulation of these differentiation events is controlled in part by the actions of FGF receptors (FlanaganSteet et al., 2000; Itoh et al., 1996; Marcelle et al., 1995). However, in order to form a functional signaling complex, appropriate cell-surface heparan sulfate proteoglycans must also be incorporated into the ligand-receptor complex, thus providing an additional level of regulatory control (reviewed in Zimmermann and David, 1999). While the expression of the four FGF receptors in the limb is known (Marcelle et al., 1994; Orr-Urtreger et al., 1991; Partanen et al., 1991; Stark et al., 1991; Szebenyi et al., 1995; Yamaguchi et al., 1992), the corresponding HSPG expression has not yet been examined as closely, particularly with regard to a possible role in myogenesis. Syndecan-1 has previously been shown to be transiently expressed in the early (⬍14 days) limb bud
Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.
83
Satellite Cell Maintenance and Muscle Regeneration
(Solursh et al., 1990) and syndecan-3 in known to be expressed in the developing limb (Gould et al., 1995). In the chick, the HSPG glypican is expressed in the apical ectodermal ridge (Niu et al., 1996) as well. To investigate the contribution of heparan sulfate to FGF binding in limb skeletal muscle, biotinylated-FGF2 was incubated with frozen sections of mouse forelimb derived from embryonic stages E14.5 and E18, as well as neonatal (P2) and young adult stages. FGF2 binds to numerous sites throughout the cartilaginous matrix of presumptive bone, the surrounding perichondrium, and within the developing skeletal muscle (Fig. 1). Within muscle, the binding is most prominent in a honeycomb pattern that represents the basement membrane surrounding developing and mature myotubes. This is less ordered at E14.5 (Fig. 1A), but becomes highly defined by the young adult stage (⬍60 days old; Fig. 1D). Staining for total heparan sulfate glycosaminoglycan was performed on sections pretreated with heparinases to remove the heparan sulfate chains. Treatment with these enzymes leaves a terminal unsaturated uronic acid on the stub of the glycosaminoglycan chain remaining attached to the proteoglycan core proteins; this is detected by staining with antibody 3G10 specific for these sites. This demonstrates that the FGF binding pattern mirrors the pattern of heparan sulfate distribution (compare Figs. 1C and 1F). Furthermore, prior removal of heparan sulfate by this enzyme treatment blocks the binding of FGF (Fig. 1E). Staining the sections with an antibody to perlecan, the abundant basement membrane heparan sulfate proteoglycan, indicates that it is distributed in the basement membranes with a similar distribution to the majority of the FGF binding (data not shown). Although FGF binding to the abundant heparan sulfate of the basement membrane is demonstrated, binding is also detected in connective tissue between the myotubes, and to localized sites or cells along the periphery of the myotubes (e.g., arrowheads in Fig. 1D). This suggests that specific proteoglycans, such as cell surface proteoglycans, may be localized within these discrete locations. To examine this hypothesis more closely, the sections were stained with antibodies to each of the four syndecans. We observed syndecan-3 staining in skeletal muscle of the forelimb throughout development and in young adult mouse muscle. Expression in the E14.5-day limb shows that syndecan-3 is expressed in cartilage chondrocytes of the developing long bones, in the perichondrium surrounding the cartilage, and in individual myoblasts of the developing skeletal muscle (Fig. 2A). The inset in Fig. 2A shows syndecan-3-positive myoblasts in aggregates prior to their fusion into myotubes. Staining with a nonspecific IgG is negative (shown for E14.5 in Fig. 2D) at all stages examined. By E18 (Fig. 2B), the myoblasts have assembled into presumptive myotubes, but individual myoblasts and a central lumen are still discerned (see inset). Syndecan-3 staining is observed around the entire surface of these myoblasts. Staining of the muscle is more intense than that of the surrounding connective tissue and perichondrium. By neo-
natal day 2, however, the syndecan-3 staining is becoming localized to discrete sites within the muscle (Fig. 2C). Individual myoblasts are no longer seen as fusion into multinucleate myotubes has taken place. Weak syndecan-3 staining is observed around the circumference of the myotubes, but stronger staining is seen at localized sites at the myotube periphery. These sites are seen only on a fraction of the myotubes in cross-section, suggesting that they may represent sites that are scattered intermittently along the length of individual myotubes. Staining of young adult muscle shows that the syndecan-3 staining is even more highly localized than that seen at P2 (Fig. 2E). The proteoglycan is seen in almost a punctate distribution bordering on the basal lamina that encapsulates the myotube. Occasional staining is seen around the entire myotube, but this staining is weak. Double staining with DAPI, to detect the nuclei of the multinucleate myotubes and of individual satellite cells, shows that the syndecan-3 staining corresponds with a subset of the nuclei in the section (Fig. 2F). Expression of syndecan-4 in the developing limb is very similar to that of syndecan-3. Expression is detected in individual chondrocytes of the developing long bone, within the perichondrium and developing skeletal muscle. Syndecan-4 is seen in the developing myoblasts at E14.5, with some cells staining more intensely (Fig. 3A). The proteoglycan is expressed around the lumen and periphery of early myotubes at E18 (Fig. 3B), with staining also observed in occasional cells that stain more brightly than their neighbors; the fate of these cells is not yet known. However, at P2, it is clear that individual cells are present that retain abundant syndecan-4 expression while expression on the entire fused myotube is less intense (Fig. 3C). In the young adult, the expression of syndecan-4 becomes more restricted to these localized sites, although it is not as punctate as that seen for syndecan-3 and some expression remains on the myotubes (Fig. 3E). As seen with syndecan-3, the localized sites of syndecan-4 expression coincide with sites of DAPI staining (Fig. 3F), identifying a subset of nuclei. The expression of syndecan-1 and syndecan-2 in limb skeletal muscle differs from the other two syndecans. At E14.5, syndecan-1 staining is detected weakly on the surfaces of myoblasts in the developing muscle (Fig. 4A). At E18, the staining is seen surrounding the lumen and the basal side of developing myotubes (Fig. 4B) and this staining has diminished or is nonexistent by P2 (Fig. 4C). No staining is seen in the young adult or adult (not shown). Syndecan-1 staining at other sites, such as in the epidermis, is much more intense (not shown). Syndecan-2 expression is seen in connective tissue and the perichondrium, but is not detected in muscle in any of the stages that were examined. Stage E18 is shown as an example (Fig. 4E).
Syndecan-3 and Syndecan-4 Are Expressed in Adult Muscle Satellite Cells The highly localized expression of syndecan-3 and syndecan-4 in neonatal and young adult muscle suggests
Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.
84
Cornelison et al.
FIG. 3. Immunolocalization of syndecan-4 in embryonic and adult mouse forelimb skeletal muscle. Frozen sections were stained with polyclonal anti-syndecan-4 antibody generated against GST-syndecan-4 fusion protein (A, B, C, E), antibody against GST alone (D), or DAPI (F). Developmental stages, arrowheads, and abbreviations are identical to Fig. 2. (Bar, 50 m).
that the expression of these proteoglycans is being restricted to satellite cells. To further test this hypothesis, adult muscles were examined in cross-section and costained for each syndecan and for either the satellite cell marker c-met or laminin to determine the relative position of the cell with respect to the basement membrane and muscle fiber membrane (Fig. 5). Staining for c-met is seen at
a small number of localized sites within the muscle, consistent with the number of satellite cells expected in adult tissue (Fig. 5). These sites are at the periphery of myotubes, also consistent with the possibility that they are localized to the satellite cells. Costaining with either syndecan-3 or syndecan-4 and c-met yields completely overlapping staining patterns (Fig. 5). As c-met is a universally accepted
Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.
85
Satellite Cell Maintenance and Muscle Regeneration
FIG. 4. Immunolocalization for syndecan-1 and syndecan-2 in developing skeletal muscle. Syndecan-1 staining is detected during early muscle development and is diminished by P2. Syndecan-1 is detected by mAb281.2 (A, E14.5); (B, E18); (C, P2). Nonspecific rat IgG is shown in (D) (E14.5). Skeletal muscle staining with syndecan-2, detected with mAb10H4 (E, E18) is negative. Nonspecific mouse IgG is shown in (D) (E18). Abbreviations are shown in Fig. 1. Connective tissue, CT. (Bar, 50 m).
marker for quiescent and activated satellite cells, syndecan-3 and syndecan-4 thus appear to identify satellite cells in adult skeletal muscle. To confirm that the cells staining positively for syndecan-3 and syndecan-4 are localized between the basal lamina and the myofiber membrane, we costained adult muscle tissue sections for syndecan-4 and laminin. The staining patterns clearly demonstrate that syndecan-4-positive cells are sandwiched between the enveloping basal lamina and the myotube proper. DAPI staining shows the presence of a nucleus at the great majority of these sites (Fig. 5).
Staining for either syndecan-3 or syndecan-4 demonstrates that satellite cells express both of these proteoglycans (Fig. 5). Quantifying the number of cell profiles that express syndecan-3 and also have a nucleus, express c-met, or express syndecan-4 shows that the correlation between all four markers is greater than 95% (Table 1).
Coexpression of Syndecans and FGF Receptor 1 The proliferation and differentiation of satellite cells is regulated by numerous factors; among factors thought to be
Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.
86
Cornelison et al.
important for these processes are HGF and FGFs. Signal transduction for both of these growth factors relies on the formation of ternary complexes containing the growth factor, its receptor tyrosine kinase, and heparan sulfate provided by heparan sulfate proteoglycans. To verify that the satellite cells coexpress FGF receptor in addition to syndecans and c-met, neonatal and young adult muscle was stained for FGF receptor 1 (FGFR-1) (Fig. 6). The receptor is expressed in skeletal muscle at both stages in a punctate distribution that colocalizes with syndecan-3 (Fig. 6). At the P2 stage, where the syndecan is not as highly localized as in the young adult, the FGFR-1 is already highly localized (Figs. 6A and 6C). In young adult muscle, syndecan-3 and FGFR-1 staining colocalize exclusively to satellite cells (Figs. 6B and 6D).
Satellite Cells on Isolated Fibers Express Syndecan-3 and Syndecan-4 If the syndecans present on intact satellite cells play a role in mediating satellite cell activation, proliferation, and/or differentiation, the proteoglycans should be expressed on satellite cells in myofiber cultures. We isolated single myofibers, cultured them, and compared staining for the syndecans, c-met, MyoD, and myogenin as a function of satellite cell proliferation. Removal of single myofibers from mouse skeletal muscle activates satellite cells, resulting in upregulation of myogenic regulatory factors such as MyoD (which is detected only rarely in fresh preparations but is prevalent by 24 h of culture) and synchronous division at about 36 and 72 h in culture. Syndecan-3 is coexpressed with c-met in satellite cells on myofibers at 0, 24, 48, and 96 h in culture (Fig. 7A). Similarly, syndecan-4 is also present and coexpressed with c-met (Fig. 7B). Together, these data and the syndecan staining in sections identify syndecan-3 and syndecan-4 as markers of quiescent and proliferating satellite cells. Syndecan-3 (Fig. 7C) and syndecan-4 (data not shown) are expressed prior to detectable expression of MyoD and remain following MyoD induction and proliferation, suggesting a potential role for these factors in satellite cell activation, proliferation, and differentiation.
Disruption of Heparan Sulfate-Mediated Signaling by Addition of Chlorate To determine whether the presence of syndecan-3 or syndecan-4 (or potentially other heparan sulfate proteoglycans) is required for activation or proliferation of satellite cells, myofibers were cultured in medium containing 25 mM sodium chlorate. Sodium chlorate is a competitive inhibitor of ATP sulfurylases (Baeuerle and Huttner, 1986) preventing protein sulfation, and thereby disrupts signaling via FGFRs (Rapraeger et al., 1991) and c-met (Deakin and Lyon, 1999). We have previously used chlorate to inhibit sulfation in a skeletal muscle cell line (MM14) that closely represents satellite cells (Olwin and Rapraeger, 1992; Ra-
praeger et al., 1991). Chlorate effectively reduces sulfation of all glycosaminoglycan chains in myoblast cultures. Restoration of biological activity with soluble heparin is significant, as this indicates that the biological responses observed are likely to be heparan sulfate-specific (Olwin and Rapraeger, 1992; Rapraeger et al., 1991). Although addition of soluble heparin to the culture may restore growth factor function, it will not replace signaling events requiring immobilized heparan sulfate, HSPG-mediated cell adhesion, or syndecan-independent signaling. When muscle fiber cultures were grown in chloratecontaining medium from the time of harvest, the number of associated satellite cells, while unchanged at 24 h of culture (prior to the first division after activation), was significantly decreased at 48 and 96 h, indicating that the presence of chlorate inhibits satellite cell division (Fig. 8A). BrdU labeling indices were less than 100% at 48 h but had recovered by 96 h (data not shown.) The majority of the effect seen at 96 h is probably due to defects in proliferation at earlier stages of the culture, resulting in a lag in cell accumulation. This decrease in cell number was not seen in a sodium chloride control, and could be rescued by addition of heparin (Fig. 8A), indicating that satellite cell proliferation is dependent on heparan sulfate. The number of satellite cell colonies adhering to the tissue culture plate was proportional to the number of fiber-associated cells in all cases, suggesting that the decrease in satellite cell number is due to a decrease in cell division rather than an increase in migration away from the host myofibers (data not shown). When the populations of fiber-associated satellite cells were assessed for expression of the myogenic regulatory factor MyoD, there was an initial increase in the percentage of satellite cells expressing MyoD at 24 h of culture in sodium chlorate, suggesting a switch from proliferation to differentiation-prone cells (Fig. 8B). However, the relative increase in MyoD was gone by 48 h in culture and by 96 h had in fact reversed (Fig. 8B.) Addition of heparin could completely rescue the chlorate-treated cells at both time points, suggesting an involvement of heparan sulfate proteoglycans in MyoD expression that involves heparan sulfate-dependent growth factors. Interestingly, when examined for expression of the myogenic regulatory factor myogenin (which is normally expressed only after the first mitotic division and tends to be associated with commitment to terminal differentiation), satellite cells in both conditions stained similarly (Fig. 8C). This, together with the apparent trend towards recovery by 96 h of culture, would seem to imply that the effects of chlorate treatment are at a maximum during the early stages of activation.
DISCUSSION We have examined the expression of the syndecan family of heparan sulfate proteoglycans, essential components of signaling complexes whose activity has been shown to be
Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.
87
Satellite Cell Maintenance and Muscle Regeneration
TABLE 1 Colocalization of Syndecan-3 with Other Proteins or Markers in Adult Muscle Satellite Cells Protein/marker
Image
No. of S3 ⫹ sites
No. of double ⫹ sites
DAPI
1 2 3 4 5
98 77 81 76 91
96 (98%) 73 (95%) 78 (95%) 72 (95%) 91 (100%) (Avg. 97 ⫾ 2%)
MET
1 2 3
82 94 77
80 (97.5%) 93 (99%) 75 (97%) (Avg. 98 ⫾ 1%)
Syn-4
1 2 3
116 58 68
116 (100%) 56 (96.5%) 68 (100%) (Avg. 99 ⫾ 2%)
Note. Sections from adult forelimb skeletal muscle were doublelabeled as described in Fig. 2. Two images of the same field of view showing either focal syndecan-3 localization or localization of nuclei (DAPI), c-met, or syndecan-4 were acquired. The number of these sites that also stained positive with DAPI or localized c-met or syndecan-4 (double positive) was then determined. The percentage of syndecan-3-positive sites that also were positive for DAPI, c-met, or syndecan-4 are in parentheses.
required for differentiation in myogenic cell lines (Larrain et al., 1997, 1998; Rapraeger et al., 1991). The ligands and receptors with which they interact have also been implicated in satellite cell activation and proliferation. We further examine the effects of interfering with heparan sulfate proteoglycan function in primary satellite cell culture, and, based on these results, suggest a role for these HSPGs in satellite cell function. An initial examination of syndecan expression patterns in adult muscle revealed a limited staining pattern for syndecan-3 and syndecan-4, suggesting they may be restricted to the satellite cells. To elucidate the roles of syndecans in skeletal muscle development and during muscle regeneration, we have assayed for expression of syndecans 1– 4 in embryonic, fetal, postnatal, young adult, and adult mouse muscle sections, as well as in singlemyofiber tissue culture preparations. In the developing animal, syndecan-1 is expressed transiently during embryonic and fetal development but does not appear to be expressed during later myogenesis or in satellite cells. This confirms a previous study of developmental expression for syndecan-1 (Solursh et al., 1990) and in cultured limb bud explants for syndecan-3 (Olguin and Brandan, 2001), but is in apparent contradiction of work suggesting that syndecan-1 is required for myogenic differentiation in the C2 cell line (Larrain et al., 1998). This discrepancy seems
most likely to be due to variations in gene expression and activity which have arisen as artifacts in the tissue culture lines used; both MM14 and C2C12 cells display gene expression profiles which deviate from those seen in primary satellite cells, with the C2 line being more divergent (Cornelison et al., manuscript in preparation). Syndecan-2 was not detectably expressed in skeletal muscle tissue from any age examined and is therefore unlikely to play a direct role in skeletal muscle development or regeneration. In contrast, syndecans 3 and 4 are expressed in myogenic cells in embryonic (ED14), fetal (ED18), neonatal (P2), young adult (⬍60 days), and adult (⬎10 wk) muscle tissue. Of particular interest is their expression in adult muscle, where immunoreactivity is restricted to cells which, based on morphology and position in the intact muscle, appear to be satellite cells. These data suggest a role for syndecan-3 and syndecan-4 in skeletal muscle development as well as skeletal muscle regeneration. When compared with expression of the satellite cell marker c-met, syndecan-3 and syndecan-4 expression correlates extremely well. When sections of adult muscle containing presumably quiescent satellite cells are costained for syndecan-3 and FGFR-1, a complete correlation is also seen. One important implication of these observations is that all of the components necessary for activation of both c-met and FGF receptors by their respective ligands are present on quiescent satellite cells. Taken together, our data suggest that quiescent satellite cells are capable of responding to an FGF or HGF signal without any additional requirements, such as protein synthesis. This supports the hypothesis that HGF and/or FGF(s) constitute the activating signal(s) for quiescent satellite cells in vivo. Despite their importance both as the cells responsible for myogenic regeneration after injury or disease and possibly during aging, and as potential vectors in cell-based therapies, mechanisms involving key processes in satellite cell biology are currently not well understood. A properly functioning satellite cell must remain quiescent in healthy muscle, become activated in response to an insult to the local muscle tissue, proliferate at the site of the injury to produce a pool of replacement myoblasts, differentiate in situ into the required fiber type and size, and possibly return to a quiescent state after the regeneration response has been completed. While studies of myogenesis in the embryo and in tissue culture lines may provide suggestions for the mechanisms in place during adult myogenesis, there are sufficiently substantial differences in gene expression patterns and behavior between satellite cells and embryonic or fetal myoblasts (Cornelison et al., 2000), and particularly between primary satellite cells and myogenic cell lines (Cornelison et al., manuscript in preparation), that examination of primary satellite cells is the optimal method for determining satellite cell-specific gene expression and activity. Based on work done in tissue culture, FGFs are excellent candidates for mediators of these activities; all myoblast cells lines studied respond to exogenously applied FGFs, and at least one transformed satellite cell line
Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.
88
Cornelison et al.
FIG. 5. c-met, laminin, and syndecan staining of adult skeletal muscle. Top and middle panels: sections of adult mouse hindlimb costained for the satellite cell marker c-met (red) and syndecan-3 (green) or syndecan-4 (green), demonstrating that syndecan-3 and syndecan-4 consistently colocalize with c-met on presumptive satellite cells. Lower panel: sections of mouse hindlimb costained for syndecan-4 (green), laminin (red), and DAPI (blue), showing that cells in the satellite position in quiescent muscle express syndecan-4.
(MM14) requires FGF signaling both to support proliferation and to suppress myogenic differentiation (Fedorov et al., 2001; Jones et al., 2001; Kudla et al., 1998; Linkhart et al., 1981; Olwin and Hauschka, 1986). Finally, FGFR-1 staining colocalizes with c-met, syndecan-3, and syndecan-4 in sections of young adult muscle and in satellite cells on intact myofibers (data not shown). When examined in sections from mdx mice (data not shown) and in single-fiber culture, activated satellite cells persist in expression of syndecan-3 and syndecan-4. Together, these data support the hypothesis that FGFs may be involved regulating satellite cell activation, proliferation, and differentiation. Given this, would the effects of FGF signaling be altered in vitro by reducing or eliminating the available heparan sulfate? To address this question, we treated fresh cultures of primary satellite cells resident on their host myofibers with sodium chlorate, which prevents sulfation of GAG chains and renders them unable to interact with either HGF or FGF ligands, thus eliminating signaling from both c-met and FGF receptors (Deakin and Lyon, 1999; Rapraeger et al.,
1991). Although initial numbers of chlorate-treated and untreated satellite cells (which have been freshly harvested and thereby newly activated) are identical at 24 h as should be expected, treated cells display decreased cell counts by 48 h postharvest, a time when all untreated satellite cells have undergone their first mitotic division. This condition persists throughout the first 96 h in culture. This effect is not observed in sodium chloride-treated controls, and can be fully rescued by addition of exogenous heparin. Since chlorate treatment dramatically reduces sulfation of all proteins, recovery with heparin strongly suggests that satellite cell proliferation requires heparan sulfate. Since the proliferation response can be rescued by exogenous heparin, it is likely that loss of heparan sulfate eliminates a response to one or more growth factors. Two potential candidates involved in satellite cell proliferation are HGF and FGF family members, both of which are heparan sulfatedependent. However, HGF activity cannot be restored by soluble heparin in chlorate-treated cells and requires immobilized heparan sulfate (Deakin and Lyon, 1999). Our data therefore suggest that HGF may not be involved in regula-
Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.
89
Satellite Cell Maintenance and Muscle Regeneration
FIG. 6. Colocalization of syndecan-3 and FGFR-1 in neonatal and adult forelimb skeletal muscle. (A, C) Same section of P2 neonate forelimb incubated with anti-syndecan-3 and anti-FGFR-1, respectively. (B, D) Same section of young adult forelimb incubated with anti-syndecan-3 and anti-FGFR-1, respectively. Bar, 20 m; inset magnification is 3.15⫻ greater than larger image.
tion of satellite cell proliferation and FGFs remain highly likely candidates. Thus, the presence of syndecan-3 and/or syndecan-4 on satellite cells appears to be required for a functional response for proliferation that may be mediated via FGFs. These data do not distinguish between the requirement for an initial proliferation signal versus a continuous signal; since the satellite cells were all activated at the same time by the initial harvest, they are roughly synchronized for the first several cell cycles. An analogous state can be induced in the satellite cell line MM14, which requires exogenous FGF to maintain proliferation and repress myogenesis (Clegg et al., 1987). Removal of serum from the culture medium in the presence of FGF maintains the cells in a quiescent, undifferentiated state, similar to the chloratetreated satellite cells. To better determine the effects of HSPG loss on myogenesis in satellite cells on intact myofibers, we examined the expression pattern of the myogenic regulatory factors MyoD and myogenin in untreated and chlorate-treated fiber cultures. Chlorate-treated satellite cells first display an increase in the percentage of MyoDpositive cells over untreated, possibly representing cells
that would otherwise proliferate beginning to differentiate precociously. By 96 h, chlorate-treated cells have begun to recover; however, the percentage of MyoD-positive cells is still less than is seen in untreated cultures. These cells may be “stalled” with no clear signal to either proliferate or differentiate, a hypothesis which is supported by the observation that chlorate-treated cells form less than half the number of myosin-positive colonies on the culture plate when examined at 96 h than either BrdU-only or heparinrescued cultures (data not shown). Similar to the effect on proliferation, changes in MyoD expression can be rescued by adding heparin. This suggests the involvement of heparan sulfate-dependent growth factors in regulation of differentiation as well as regulation of proliferation. Expression of the myogenic regulatory factor myogenin, which is a significantly later event than initiation of MyoD upregulation and is associated with cells entering a differentiative state, was also examined. Under no conditions was any myogenin expression detected at 24 h, implying that the elevated levels of MyoD observed in chloratetreated cells have not led to a premature commitment to terminal differentiation. Additionally, at 96 h, there is only
Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.
90
Cornelison et al.
FIG. 7. Immune staining of fiber-associated satellite cells. Fibers were stained immediately after harvest (0 h), during early activation (24 h postharvest), and after onset of differentiation (48 and 96 h postharvest). (A) Coexpression of syndecan-3 (green) and the satellite cell marker c-met (red) in fiber-associated satellite cells. (B) Coexpression of syndecan-4 (green) and the satellite cell marker c-met (red) in fiber-associated satellite cells. (C) Coexpression of syndecan-3 (green) with MyoD (red) and DAPI (blue).
a slight increase in the percentage of myogenin-positive cells in chlorate-treated vs. BrdU-only cultures. Taken together with the proliferation and MyoD data, this suggests that the major effects of chlorate treatment are on the early events of satellite cell activation and initiation of proliferation, although it should be noted that satellite cells appear to be unique among myogenic cell types in that myogenin-expressing cells retain the capability to divide (Cornelison and Wold, 1997; Maley et al., 1994). At later times, perhaps either the requirements for heparan sulfatemediated signaling are decreased or the cells have successfully scavenged sufficient sulfate present in the environment to ameliorate the effects of chlorate inhibition. When considering these data, it should also be taken into account that, unlike myogenic cell lines or cultured populations of pure primary myoblasts, these cultures also contain the host fibers, which in fact make up the majority of cell mass in the cultures. It has been shown previously
that primary satellite cells in mass culture express MyoD and myogenin with delayed kinetics relative to in situ crush injuries (Maley et al., 1994); satellite cells in singlefiber cultures have an expression profile which is more similar to that seen in crush injuries, expressing MyoD protein at detectable levels by 6 h postharvest but with myogenin expression delayed until closer to 48 h (Cornelison and Wold, 1997). Therefore, the contribution of the fibers in the form of cell– cell interactions, paracrine or secreted growth factors, or other components may affect the behavior of the satellite cells under these conditions but do not fully reconstitute the environment found in vivo. Our data also have several practical implications. The syndecan-3 and syndecan-4 genes may be considered potential markers for both quiescent and activated satellite cells in situ: such markers are rare, and the reagents available for detecting them are not always reliable. Thus, proteins that are expressed exclusively in satellite cells yet universally
Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.
FIG. 7—Continued
92
Cornelison et al.
within the satellite cell population and for which strong and specific antibodies exist represent a powerful new tool for the study of satellite cells. Here, we report the identification of two new such markers, syndecan-3 and syndecan-4. Future experiments will address further the role(s) of these HSPGs in satellite cell activation, proliferation, and differentiation.
ACKNOWLEDGMENTS This work was supported by grants from the NIH (AR39467) and MDA (to B.B.O.) and NIH (GM48850 and HD21881) (to A.C.R.). D.D.W.C. is supported by NIH Training Grant HL07851.
REFERENCES
FIG. 8. Chlorate treatment of fiber-associated satellite cells. (A) Representation of absolute satellite cell number per 100 myonuclei in myofiber cultures. DAPI-stained myofiber nuclei and DAPI-stained satellite cell nuclei colocalizing with syndecan-4 were counted from myofibers at the stated time points after harvest. Additions to medium were: B, 10 M BrdU only; BN, BrdU ⫹ 25 mM NaCl (sodium chloride); BC, BrdU ⫹ 25 mM NaClO 3 (sodium chlorate); BCH, BrdU ⫹ sodium chlorate ⫹ 250 mg/ml heparin. Note that the number of satellite cells at 96 h results from the cumulative proliferation history, and decreased numbers do not necessarily reflect a continued defect in proliferation. No error bars exist because the absolute number of satellite cells per 100 myonuclei is depicted. (B) Representation of percentage of satellite cells in fiber culture express MyoD protein. DAPI-stained satellite cell nuclei within syndecan-4positive cell outlines were examined for immunoreactivity for MyoD at the stated timepoints after harvest. Media additions are as in (A). Error bars represent calculated standard error. (C) Representation of percentage of satellite cells in fiber culture expressing myogenin protein. DAPI-stained satellite cell nuclei within syndecan-4-positive cell outlines were examined for immunoreactivity for myogenin at the stated time points after harvest. Media additions are as in (A). Error bars represent calculated standard error.
Allen, R. E., Sheehan, S. M., Taylor, R. G., Kendall, T. L., and Rice, G. M. (1995). Hepatocyte growth factor activates quiescent skeletal muscle satellite cells in vitro. J. Cell. Physiol. 165, 307–312. Baeuerle, P. A., and Huttner, W. B. (1986). Chlorate: A potent inhibitor of protein sulfation in intact cells. Biochem. Biophys. Res. Commun. 141, 870 – 877. Bischoff, R. (1994). In “Myology” (A. G. Engel and C. FranziniArmstrong, Eds.), pp. 97–118. McGraw-Hill, New York. Bladt, F., Riethmacher, D., Isenmann, S., Aguzzi, A., and Birchmeier, C. (1995). Essential role for the c-met receptor in the migration of myogenic precursor cells into the limb bud. Nature 376, 768 –771. Bottaro, D. P., Rubin, J. S., Faletto, D. L., Chan, A. M. L., Kmiecik, T. E., Woude, G. F. V., and Aaronson, S. A. (1991). Identification of hepatocyte growth factor receptor as the met proto-oncogene product. Science 251, 802– 804. Carey, D. J., Evans, D. M., Stahl, R. C., Asundi, V. K., Conner, K. J., Garbes, P., and Cizmeci-Smith, G. (1992). Molecular cloning and characterization of N-syndecan, a novel transmembrane heparan sulfate proteoglycan. J. Cell Biol. 117, 191–201. Clegg, C. H., Linkhart, T. A., Olwin, B. B., and Hauschka, S. D. (1987). Growth factor control of skeletal muscle differentiation: Commitment to terminal differentiation occurs in G 1 phase and is repressed by fibroblast growth factor. J. Cell Biol. 105, 949 – 956. Colvin, J. S., Bohne, B. A., Harding, G. W., Mcewen, D. G., and Ornitz, D. M. (1996). Skeletal overgrowth and deafness in mice lacking fibroblast growth factor receptor 3. Nat. Genet. 12, 390 –397. Cornelison, D. D. W., Olwin, B. B., Rudnicki, M. A., and Wold, B. J. (2000). MyoD⫺/⫺ satellite cells in single fiber culture are differentiation defective and MRF4 deficient. Dev. Biol. 224, 122–137. Cornelison, D. D. W., and Wold, B. J. (1997). Single-cell analysis of regulatory gene expression in quiescent and activated mouse skeletal muscle satellite cells. Dev. Biol. 191, 270 –283. Cusella-DeAngelis, M. G., Lyons, G., Sonnino, C., DeAngelis, L., Vivarelli, E., Farmer, K., Wright, W. E., Molinaro, M., Bouche, M., Buckingham, M., and Cossu, G. (1992). MyoD, myogenin independent differentiation of primordial myoblasts in mouse somites. J. Cell Biol. 116, 1243–1255. David, G., Bai, X. M., Van der Schueren, B., Manyen, P., Cassiman, J. J., and Van den Berghe, H. (1993). Spatial and temporal changes
Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.
93
Satellite Cell Maintenance and Muscle Regeneration
in the expression of fibroglycan (syndecan-2) during mouse embryonic development. Development 119, 841– 854. Deakin, J. A., and Lyon, M. (1999). Differential regulation of hepatocyte growth factor/scatter factor by cell surface proteoglycans and free glycosaminoglycan chains. J. Cell Sci. 112, 1999 – 2009. Deng, C. X., Wynshaw-Boris, A., Shen, M. M., Daugherty, C., Ornitz, D. M., and Leder, P. (1994). Murine FGFR-1 is required for early postimplantation growth and axial organization. Genes Dev. 8, 3045–3057. Fedorov, Y. V., Rosenthal, R. S., and Olwin, B. B. (2001). Oncogenic ras-induced proliferation requires autocrine fibroblast growth factor 2 signaling in skeletal muscle cells. J. Cell Biol. 152, 1301–1306. Flanagan-Steet, H., Hannon, K., McAvoy, M. J., Hullinger, R., and Olwin, B. B. (2000). Loss of FGF receptor 1 signaling reduces skeletal muscle mass and disrupts myofiber organization in the developing limb. Dev. Biol. 218, 21–37. Floss, T., Arnold, H. H., and Braun, T. (1997). A role for FGF-6 in skeletal muscle regeneration. Genes Dev. 11, 2040 –2051. Gould, S. E., Upholt, W. B., and Kosher, R. A. (1995). Characterization of chicken syndecan-3 as a heparan sulfate proteoglycan and its expression during embryogenesis. Dev. Biol. 168, 438 – 451. Ishiguro, K., Kadomatsu, K., Kojima, T., Muramatsu, H., Tsuzuki, S., Nakamura, E., Kusugami, K., Saito, H., and Muramatsu, T. (2000). Syndecan-4 deficiency impairs focal adhesion formation only under restricted conditions. J. Biol. Chem. 275, 5249 –5252. Itoh, N., Mima, T., and Mikawa, T. (1996). Loss of fibroblast growth factor receptors is necessary for terminal differentiation of embryonic limb muscle. Development 122, 291–300. Johnson, S. E., and Allen, R. E. (1995). Activation of skeletal muscle satellite cells and the role of fibroblast growth factor receptors. Exp. Cell Res. 219, 449 – 453. Jones, N. C., Fedorov, Y. V., Rosenthal, R. S., and Olwin, B. B. (2001). ERK1/2 is required for myoblast proliferation but is dispensable for muscle gene expression and cell fusion. J. Cell Physiol. 186, 104 –115. Kudla, A. J., Jones, N. C., Rosenthal, R. S., Arthur, K., Clase, K. L., and Olwin, B. B. (1998). The FGF receptor-1 tyrosine kinase domain regulates myogenesis but is not sufficient to stimulate proliferation. J. Cell Biol. 142, 241–250. Larrain, J., Carey, D. J., and Brandan, E. (1998). Syndecan-1 expression inhibits myoblast differentiation through a basic fibroblast growth factor-dependent mechanism. J. Biol. Chem. 273, 32288 – 32296. Larrain, J., Cizmeci-Smith, G., Troncoso, V., Stahl, R. C., Carey, D. J., and Brandan, E. (1997). Syndecan-1 expression is downregulated during myoblast terminal differentiation. Modulation by growth factors and retinoic acid. J. Biol. Chem. 272, 18418 – 18424. Linkhart, T. A., Clegg, C. H., and Hauschka, S. D. (1981). Myogenic differentiation in permanent clonal myoblast cell lines: Regulation by macromolecular growth factors in the culture medium. Dev. Biol. 86, 19 –30. Maina, F., Casagranda, F., Audero, E., Simeone, A., Comoglio, P. M., Klein, R., and Ponzetto, C. (1996). Uncoupling of Grb2 from the Met receptor in vivo reveals complex roles in muscle development. Cell 87, 531–542. Maley, M. A., Fan, Y., Beilharz, M. W., and Grounds, M. D. (1994). Intrinsic differences in MyoD and myogenin expression between primary cultures of SJL/J and BALB/C skeletal muscle. Exp. Cell Res. 211, 99 –107.
Marcelle, C., Eichmann, A., Halevy, O., Breant, C., and LeDouarin, N. M. (1994). Distinct developmentalexpression of a new avian fibroblast growth factor receptor. Development 120, 683– 694. Marcelle, C., Wolf, J., and Bronner-Fraser, M. (1995). The in vivo expression of the FGF receptor Frek mRNA in avian myoblasts suggests a role in muscle growth and differentiation. Dev. Biol. 172, 100 –114. McFall, A. J., and Rapraeger, A. C. (1997). Identification of an adhesion site within the syndecan-4 extracellular protein domain. J. Biol. Chem. 272, 12901–12904. Niu, S., Antin, P. B., Akimoto, K., and Morkin, E. (1996). Expression of avian glypican is developmentally regulated. Dev. Dyn. 207, 25–34. Olguin, H., and Brandan, E. (2001). Expression and localization of proteoglycans during limb myogenic activation. Dev. Dyn. 221, 106 –115. Olwin, B. B., and Hauschka, S. D. (1986). Identification of the fibroblast growth factor receptor of Swiss 3T3 cells and mouse skeletal muscle myoblasts. Biochemistry 25, 3487–3492. Olwin, B. B., and Rapraeger, A. (1992). Repression of myogenic differentiation by aFGF, bFGF, and K-FGF is dependent on cellular heparan sulfate. J. Cell Biol. 118, 631– 639. Orr-Urtreger, A., Givol, D., Yayon, A., Yarden, Y., and Lonai, P. (1991). Developmental expression of two murine fibroblast growth factor receptors, flg and bek. Development 113, 1419 – 1434. Partanen, J., Ma¨kela¨, T. P., Eerola, E., Korhonen, J., Hirvonen, H., Claesson Welsh, L., and Alitalo, K. (1991). FGFR-4, a novel acidic fibroblast growth factor receptor with a distinct expression pattern. EMBO J. 10, 1347–1354. Pellegrini, L., Burke, D. F., von Delft, F., Mulloy, B., and Blundell, T. L. (2000). Crystal structure of fibroblast growth factor receptor ectodomain bound to ligand and heparin. Nature 407, 1029 – 1034. Perrimon, N., and Bernfield, M. (2000). Specificities of heparan sulphate proteoglycans in developmental processes. Nature 404, 725–728. Rapraeger, A. C. (2000). Syndecan-regulated receptor signaling. J. Cell Biol. 149, 995–998. Rapraeger, A. C., Krufka, A., and Olwin, B. B. (1991). Requirement of heparan sulfate for bFGF-mediated fibroblast growth and myoblast differentiation. Science 252, 1705–1708. Schlessinger, J., Plotnikov, A. N., Ibrahimi, O. A., Eliseenkova, A. V., Yeh, B. K., Yayon, A., Linhardt, R. J., and Mohammadi, M. (2000). Crystal structure of a ternary FGF-FGFR-heparin complex reveals a dual role for heparin in FGFR binding and dimerization. Mol. Cell 6, 743–750. Schultz, E., and McCormick, K. M. (1994). Skeletal muscle satellite cells. Rev. Physiol. Biochem. Pharmacol. 123, 213–257. Solursh, M., Reiter, R. S., Jensen, K. L., Kato, M., and Bernfield, M. (1990). Transient expression of a cell surface heparan sulfate proteoglycan (syndecan) during limb development. Dev. Biol. 140, 83–92. Stark, K. L., McMahon, J. A., and McMahon, A. P. (1991). FGFR-4, a new member of the fibroblast growth factor receptor family, expressed in the definitive endoderm and skeletal muscle lineages of the mouse. Development 113, 641– 651. Stockdale, F. E. (1992). Myogenic cell lineages. Dev. Biol. 154, 284 –298. Szebenyi, G., Savage, M. P., Olwin, B. B., and Fallon, J. F. (1995). Changes in the expression of FGF receptors mark distinct stages
Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.
94
Cornelison et al.
of chondrogenesis in vitro and during chick limb skeletal patterning. Dev. Dyn. 204, 446 – 456. Tatsumi, R., Anderson, J. E., Nevoret, C. J., Halevy, O., and Allen, R. E. (1998). HGF/SF is present in normal adult skeletal muscle and is capable of activating satellite cells. Dev. Biol. 194, 114 –128. Weinstein, M., Xu, X., Ohyama, K., and Deng, C. X. (1998). FGFR-3 and FGFR-4 function cooperatively to direct alveogenesis in the murine lung. Development 125, 3615–3623. Wilkie, A. O., Morriss-Kay, G. M., Jones, E. Y., and Heath, J. K. (1995). Functions of fibroblast growth factors and their receptors. Curr. Biol. 5, 500 –507. Xu, X., Weinstein, M., Li, C., Naski, M., Cohen, R. I., Ornitz, D. M., Leder, P., and Deng, C. (1998). Fibroblast growth factor receptor 2 (FGFR2)-mediated reciprocal regulation loop between FGF8 and FGF10 is essential for limb induction. Development 125, 753–765.
Yamaguchi, T. P., Conlon, R. A., and Rossant, J. (1992). Expression of the fibroblast growth factor receptor FGFR-1/flg during gastrulation and segmentation in the mouse embryo. Dev. Biol. 152, 75– 88. Yamaguchi, T. P., Harpal, K., Henkemeyer, M., and Rossant, J. (1994). fgfr-1 is required for embryonic growth and mesodermal patterning during mouse gastrulation. Genes Dev. 8, 3032–3044. Yayon, A., Klagsbrun, M., Esko, J. D., Leder, P., and Ornitz, D. M. (1991). Cell surface, heparin-like molecules are required for binding of basic fibroblast growth factor to its high affinity receptor. Cell 64, 841– 848. Zimmermann, P., and David, G. (1999). The syndecans, tuners of transmembrane signaling. FASEB J. 13, S91–S100. Received for publication May Revised August Accepted August Published online September
Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.
23, 14, 14, 26,
2001 2001 2001 2001