DEVELOPMENTAL DYNAMICS 208:505–515 (1997) Chemotaxis of Skeletal Muscle Satellite Cells RICHARD BISCHOFF* Department of Anatomy and Neurobiology, Washington University School of Medicine, Saint Louis, Missouri ABSTRACT Migration of myogenic cells occurs extensively during both embryogenesis and regeneration of skeletal muscle and is important in myoblast gene therapy, but little is known about factors that promote chemotaxis of these cells. We have used satellite cells from adult rats purified by Percoll density gradient centrifugation to test growth factors and wound fluids for chemotactic activity in blind-well Boyden chambers. Of a variety of growth factors tested only hepatocyte growth factor (HGF) and transforming growth factor-beta (TGF-b) exhibited significant chemotactic activity. The dose-response curves for both of these factors was bell-shaped with maximum activity in the 1–10 ng/ml range. Checkerboard analysis of TGF-b showed that chemotaxis occurred only in response to a positive concentration gradient. An extract of rat platelets also exhibited chemotactic activity for satellite cells. Half-maximal activity of this material was about 3 mg/ml, and there was no evidence of inhibition of migration at high concentrations. Checkerboard analysis of platelet extract exhibited evidence of both chemotaxis and chemokinesis, or increase in random motility of cells. Inhibition experiments showed that most, but not all, of the chemotactic activity in platelet extract could be blocked with a neutralizing antibody to TGF-b. A saline extract of crushed muscle was found to contain both mitogenic and motogenic factors for satellite cells. The two activities were present in different fractions after heparin affinity chromatography. We propose that the proliferation and migration of satellite cells during regeneration is regulated by overlapping gradients of several effector molecules released at the site of muscle injury. These molecules may also be useful for enhancing the dispersion of injected myoblasts during gene therapy. Dev. Dyn. 208:505–515, 1997. r 1997 Wiley-Liss, Inc. Key words: skeletal muscle; satellite cell; regeneration; platelet; TGF-b; HGF INTRODUCTION Skeletal muscle precursor cells are capable of migration throughout life. Early in development, myogenic cells emigrate from the somites into the limb buds where they proliferate and fuse to form the primary myotubes of the appendicular muscles (Chevallier et al., 1977; Christ et al., 1977). This migration is induced r 1997 WILEY-LISS, INC. by a signal arising from the mesoderm of the lateral body wall (Hayashi and Ozawa, 1995) and is dependent upon attachment of the cells to fibronectin present in the extracellular matrix that forms the substratum for migration (Brand-Saberi et al., 1993). Once muscle differentiation is underway, most myogenic cells become enclosed by the myotube basal lamina; however, these satellite cells are capable of limited migration between myotubes. During the early postnatal period in rat development, proliferating satellite cells labeled in situ with a retrovirus subsequently give rise to progeny, some of which move across basal laminae and fuse with neighboring myofibers (Hughes and Blau, 1990). The myofiber basal lamina becomes thicker and stronger with age (Kovanen et al., 1988), and most satellite cells become quiescent, so it is likely that little migration occurs in the adult under normal conditions. After muscle trauma, however, there is abundant evidence for the directed migration of satellite cells in response to injury. During the free grafting of larger muscles, satellite cells migrate from the central necrotic area toward the periphery (Schultz et al., 1988), and later, during revascularization, they migrate back toward the center where they undergo myogenesis (Phillips et al., 1987). Satellite cells also migrate from the viable half to the killed half of a longitudinally split muscle autograft (Phillips et al., 1990). Localized muscle trauma apparently produces factors that stimulate chemotaxis of satellite cells from distant sites (Watt et al., 1994). Focal crush injury at one end of a muscle results in the activation and movement of satellite cells beneath the basal lamina from distant uninjured tissue toward the crush site (Schultz et al., 1985). If the injury stimulus is perpendicular to the direction of the myofibers, some cells are able to cross the basal lamina of uninjured myofibers and migrate transversely through the muscle toward the injury (Klein-Ogus and Harris, 1983). In addition to migrating within the muscle, satellite cells are even capable of moving between adjacent muscles under certain conditions. Experiments using genetic variants of metabolic enzymes to identify donor and host tissue in adult mice demonstrated that grafted regenerating muscle becomes invaded by myogenic cells migrating from neighboring muscles (Watt et al., *Correspondence to: Richard Bischoff, Ph.D., Department of Anatomy and Neurobiology, Box 8108, Washington University School of Medicine, 660 S. Euclid Ave., St. Louis, MO 63110. Received 4 October 1996; Accepted 27 December 1996 506 BISCHOFF 1987). Similar experiments carried out in adult rats, however, failed to show satellite cell migration from adjacent muscles (Ghins et al., 1984), and it was suggested that the thicker epimysium of rat muscle may block invasion (Partridge, 1991). This was confirmed by the demonstration that regenerating rat muscle would be invaded by satellite cells from an adjacent muscle only if a ‘‘bridge’’ of tissue was created between the two muscles by damaging the epimysium (Schultz et al., 1986). There is evidence that migration of myogenic cells between muscles depends upon the presence of soluble chemotactic factors, perhaps released during muscle regeneration or inflammation (Moens et al., 1996). Taken together, these studies demonstrate the extensive migratory capacity of satellite cells, but little is known of the chemotactic signals involved. Venkatasubramanian and Solursh (1984) reported that quail limb mesoderm cells respond chemotactically to plateletderived growth factor (PDGF), but the cells were tested after their migration from the somites and were not a pure population. More recently, Daston et al. (1996) have shown that the transcription factor Pax-3 is necessary for emigration of myogenic cells from the somites in mouse embryos. This factor has been implicated in regulating the expression of the c-met-encoded receptor tyrosine kinase, also essential for myogenic cell emigration (Bladt et al., 1995). Mice bearing null mutations of this allele are devoid of limb muscles owing to failure of migration of muscle precursor cells from the somites. The ligand for the c-met receptor is hepatocyte growth factor (HGF; Bottaro et al., 1991), and the forced expression of HGF in the nervous system induces ectopic formation of skeletal muscle (Takayama et al., 1996). These studies indicate that early migration of myogenic cells, probably in response to a chemoattractant, plays an essential role in the origin of appendicular muscles. This chemotactic responsiveness may persist in satellite cells of adult muscle, the descendants of embryonic myoblasts. The present study was undertaken to test several defined growth factors and components of wound fluid for chemotactic activity towards muscle satellite cells. Results show that HGF and TGF-b are potent chemoattractants. Chemotactic activity is also present in extracts of platelets and crushed muscle. RESULTS Preliminary experiments were carried out to determine the optimum incubation time (Fig. 1A) and satellite cell concentration (Fig. 1B) for chemotaxis. A saline extract of crushed skeletal muscle from adult rats was used as the chemoattractant (Robertson et al., 1993; see below). Based upon these results, chemotaxis chambers were loaded with 1,500 satellite cells/mm2 filter surface and incubated for 7 hr. Because growth factors are released during muscle regeneration and may serve as chemoattractants, we Fig. 1. Effect of cell density and incubation time on chemotactic response of satellite cells to 0.5 mg/ml crushed muscle extract. A: Chemotactic chambers were incubated for 1–7 hr before fixation and staining. Cells were counted on both sides of the filter. B: Various concentrations of satellite cells were placed in the upper well and incubated for 6 hr. Cells were scraped from the upper surface of the filter before fixation, and only migrated cells were counted. tested the response of satellite cells to a range of concentration of purified growth factors (Table 1). Several isoforms of PDGF were tested because this has been reported to be a chemoattractant for chick embryo cells (Venkatasubramanian and Solursh, 1984) and mouse myoblasts (Robertson et al., 1993). Although basic fibroblast growth factor (bFGF) and PDGF are mitogens for satellite cells (DiMario and Strohman, 1988; Allen and Boxhorn, 1989; Allen and Rankin, 1990; Yablonka-Reuveni et al., 1990; Jin et al., 1991), they exhibited, as did epidermal growth factor (EGF), only weak chemotactic activity. PDGF-AB at the highest concentration (2,000 ng/ml) induced migration to only twice the level of the negative control value, as compared with the almost sevenfold stimulation of embryo extract, used as a positive control. HGF and TGF-b, however, displayed significant chemotactic activity, amounting to 5–6 times that of the negative control wells. Both factors displayed evidence of bell- CHEMOTAXIS OF SATELLITE CELLS 507 TABLE 1. Migration of Satellite Cells in Response to Various Growth Factors Attractant None 5% chick EE PDGF-AA PDGF-AA PDGF-AA PDGF-BB PDGF-BB PDGF-BB PDGF-AB PDGF-AB PDGF-AB bFGF bFGF bFGF EGF EGF EGF HGF HGF HGF TGF-b TGF-b TGF-b Concentration (ng/ml) — — 12 25 100 12 25 100 125 500 2,000 1 10 100 10 100 1,000 1 10 100 2.5 5.0 25.0 Cells crossing filter (%)a 6.7 6 1.7 45.0 6 2.6* 3.5 6 0.8 8.7 6 2.8 10.6 6 2.4 4.9 6 1.8 4.7 6 1.0 7.0 6 2.1 4.6 6 0.9 3.4 6 1.6 12.1 6 2.9 8.8 6 2.2 5.2 6 0.8 5.6 6 2.9 9.0 6 2.3 9.6 6 1.7 5.8 6 1.5 23.0 6 2.4* 36.4 6 1.2* 3.2 6 0.8 18.2 6 2.6* 29.6 6 2.3* 8.2 6 1.2 aThese results are representative of at least two experiments with similar results. The results are expressed as the mean 6 standard error of the mean. *Values significantly different from the negative control (P , .05). Fig. 2. Chemotactic response of satellite cells to increasing concentrations of TGF-b. shaped dose-response curves, and this was confirmed by testing a wider range of TGF-b concentrations (Fig. 2). Maximum stimulation of migration occurred at 5 ng/ml for TGF-b, and chemotaxis was suppressed to control level by 50 ng/ml. A bell-shaped dose-response curve is characteristic of various purified chemoattractants and cell types (Falk et al., 1980; Lucas et al., 1988; Adelman-Grill et al., 1990; Phillips et al., 1991). To examine the response of satellite cells to TGF-b in more detail, a checkerboard analysis was carried out (Zigmond and Hirsch, 1973). In this procedure, the concentration of chemoattractant in wells is varied both Fig. 3. Checkerboard analysis of chemotaxis for TGF-b. The concentration of TGF-b was varied in both the upper (vertical columns) and lower (horizontal rows) chambers. The diagonal column of gray boxes represents increasing levels of identical concentrations in both chambers. Values in the boxes are the percent cells migrating across the filter. The SEM was less than 20% in all cases; the individual error values were omitted for clarity. above and below the filter to establish positive, negative, and null concentration gradients. Chemotaxis is observed in response to a positive concentration gradient, whereas migration in the presence of a negative or zero gradient of test substance indicates increased random motility of cells, or chemokinesis. This analysis showed that migration occurred only in response to a positive concentration gradient; there was no evidence of chemokinesis (Fig. 3). Inhibition of migration in some instances of high concentration gradient reflects the bell-shaped dose-response curve observed previously. Muscle injury leads to the formation of wound fluid that contains a wide variety of substances released from the injured tissue, blood, and inflammatory cells. Previous work has shown that a saline extract of crushed adult muscle is a potent mitogen for myogenic cells (Kardami et al., 1985; Bischoff, 1986; Chen and Quinn, 1992; Haugk et al., 1995), and this material was also tested for chemotactic activity (Fig. 4). Crude muscle extract stimulated chemotaxis in a dosedependent manner with about the same half-maximal activity, 0.4 mg/ml, as that found for the mitogenic effect (Bischoff, 1986). There was no evidence of inhibition at the highest concentration tested, 1 mg/ml. Previous work has shown that the satellite cell mitogen in muscle extract is a heparin-binding growth factor (Bischoff, 1989; Bischoff, 1990b). When the extract was partially purified by heparin affinity chromatography, 508 BISCHOFF reduced, but did not abolish, the activity of platelet extract (Fig. 8). At the highest concentration tested, treatment with anti-TGF-b produced 65% inhibition of chemotaxis as compared with control antibody. DISCUSSION Fig. 4. Chemotactic response of satellite cells to increasing concentrations of a crude extract of crushed adult rat muscle. chemotactic activity was found to reside in the non– heparin-binding fraction (Fig. 5), whereas all mitogenic activity for satellite cells is contained in the heparinbinding fraction that elutes from the column with 1 M NaCl (Bischoff, 1989; Bischoff, 1990b). Although an extensive dose-response analysis of the non–heparinbinding fraction was not done, the maximum activity (20 µg/ml) of the several concentrations tested appears to represent at least a 20-fold purification as compared with the crude extract. Another component of wound fluid arises from the secretion of platelets that are activated during injury and trigger the clotting mechanism to promote hemostasis. Physical trauma leads to extravasation of blood into the tissue interstices and is often accompanied by widespread bruising. To test for chemotactic factors, platelets were isolated from fresh rat blood and washed, and secretory products were collected after exposure to thrombin. The platelet extract exhibited chemotactic activity for satellite cells with a half-maximal activity of about 3 µg/ml, and no inhibition of migration was observed up to 30 µg/ml, the highest concentration tested (Fig. 6). Checkerboard analysis of platelet extract exhibited evidence of both chemotaxis and chemokinesis (Fig. 7). Increased cell migration occurred whenever there was a positive gradient across the filter, but migration also occurred in the presence of a negative or null gradient at high extract concentration. This suggests that platelet extract stimulates both chemotaxis and chemokinesis of satellite cells in a concentrationdependent manner. Because platelets are a major source of TGF-b, it is possible that this factor is responsible for the chemotactic activity of platelet extract. To examine this, we pretreated platelet extract with an antibody capable of neutralizing the activity of TGF-b. Both platelet extract and neutralizing antibody were tested at two concentrations. The antibody treatment completely eliminated chemotactic activity of purified TGF-b and greatly Migration of myogenic cells from nearby viable muscle to a site of injury provides an important means of augmenting the population of precursor cells during regeneration. A variety of muscle trauma conditions stimulate migration including ischemia (Phillips et al., 1987; Schultz et al., 1988), thermal injury (Morgan et al., 1987; Phillips et al., 1990), crushing (Schultz et al., 1985; Watt et al., 1994), and snake venom toxin (KleinOgus and Harris, 1983). Little is known, however, concerning factors that guide this migration. The findings described here demonstrate that satellite cells are induced to migrate in response to concentration gradients of several soluble factors known to be released during muscle injury. These factors may play a role in the recruitment of satellite cells from distant sites after injury. The most interesting chemoattractant identified in this study is TGF-b, because of extensive prior studies on the roles of TGF-b in regulating myogenesis. The TGF-bs comprise a family of at least five closely related homodimeric proteins first purified from platelets but also present in many cells and tissues (Roberts and Sporn, 1991; Massagúe et al., 1992). The effects of TGF-b on myogenic cells in vitro are complex and depend upon concentration, medium composition, and cell type. With satellite cells grown in serum-containing medium, the most consistent effect observed is suppression of proliferation. Both rat and pig satellite cells exhibit 50–60% inhibition of proliferation at concentrations of TGF-b in the 0.1–0.5 ng/ml range (Allen and Boxhorn, 1987; Pampusch et al., 1990; Cook et al., 1993). Bovine satellite cells are highly sensitive to TGF-b; maximum inhibition occurs at 0.001 ng/ml, and mitotic inhibition diminishes at higher concentrations (Blachowski et al., 1993). In these studies, the cells were already in the cell cycle after stimulation by serum mitogens, but TGF-b is also able to prevent the G0 = G1 transition of quiescent satellite cells. Bischoff (1990b) reported that 0.05 ng/ml TGF-b blocks proliferation of satellite cells on single, viable myofibers stimulated with an extract of crushed muscle. Inhibition of proliferation by TGF-b may augment its role in chemotactic recruitment of satellite cells. Because motility is suppressed in mitotic cells, antiproliferative factors would increase the efficiency of directed migration of satellite cells to the site of injury. The response of satellite cells to TGF-b is concentration dependent, but the level of TGF-b after muscle injury has apparently not been investigated. Cromack et al. (1987) collected wound fluid from chambers implanted subcutaneously in rats and found about 1 CHEMOTAXIS OF SATELLITE CELLS Fig. 5. Chemotactic response to various fractions of crushed muscle extract. The crude extract was passed through a heparin-Sepharose affinity column and separated into an unbound fraction and a bound fraction eluted from the column with 1 M NaCl. Control chambers 509 contained culture medium only, and the extract fractions were tested at a range of concentrations. Asterisks indicate values significantly different from the control (P , .05). Fig. 6. Chemotactic response of satellite cells to increasing concentrations of a saline extract of thrombin-stimulated rat platelets. ng/ml TGF-b at 3 days, the earliest time measured. The level reached a peak of about 20 ng/ml at 7 days. Because the chambers were put in place with minimal tissue damage, the levels of TGF-b would be higher in trauma situations involving extensive platelet secretion. TGF-b is also secreted by inflammatory cells, such as macrophages (Massagúe et al., 1992), and Robertson et al. (1993) have reported that macrophage secretory products induce a chemotactic response in muscle precursor cells. TGF-b may also arise from the injured muscle tissue itself. Yamazaki et al. (1994) have re- Fig. 7. Checkerboard analysis of chemotaxis for platelet extract. The concentration of extract was varied in both the upper (vertical columns) and lower (horizontal rows) chambers. The diagonal column of gray boxes represents increasing levels of identical concentrations in both chambers. Comparison of the 150 µg/ml row and column shows that migration is stimulated at high concentration of platelet extract with positive, null, and negative concentration gradients. Values in the boxes are the percent cells migrating across the filter. The SEM was less than 20% in all cases. 510 BISCHOFF Fig. 8. Inhibition of the chemotactic response to TGF-b and platelet extract by treatment with an antibody to TGF-b. Platelet extract or TGF-b was incubated 1 hr with neutralizing antibody to TGF-b (5 or 50 µg/ml) or control IgG (100 µg/ml) before testing in chemotactic chambers. Asterisks indicate values significantly different from the control antibody for each chemoattractant (P , .05). ported that TGF-b is homogeneously localized in necrotic myofibers in Duchenne muscular dystrophy. Production of TGF-b from all sources would lead to a concentration gradient originating at the site of injury. In certain injuries, such as blunt trauma, the extensive hematoma extending some distance from the center of impact would spread the release of TGF-b from extravasated platelets even further. Because of the bell-shaped dose-response curve (Fig. 2), satellite cells migrating centripetally from uninjured areas may reach a point at which the concentration of TGF-b inhibits further migration. Concentrations of TGF-b that inhibit chemotaxis are comparable to the maximum levels found in wound fluid (Cromack et al., 1987). Satellite cells proliferate at the site of injury during muscle regeneration, but it is not clear how the antimitotic effects of TGF-b are overcome, thereby allowing the cells to enter the mitotic cycle. One possibility is that the anti-mitotic effects of TGF-b are diminished at higher concentrations, as has been demonstrated for bovine satellite cells (Blachowski et al., 1993). Alternatively, because the response of myogenic cells to TGF-b depends upon the presence of other peptide growth factors (Allen and Boxhorn, 1989; Zentella and Massagúe, 1992; Blachowski et al., 1993; Cook et al., 1993), the migrating cells may encounter a pro-mitotic environ- ment as they approach the site of injury. According to this model (Fig. 9), injury-related mitogens, such as those present in crushed muscle extract (Bischoff, 1986, 1990a), are localized to the wound site, whereas TGF-b is more diffuse owing to widespread platelet extravasation. The mitogenic and motogenic behavior of migrating satellite cells may be determined by overlapping concentration gradients of various effectors. The present study has also provided the first evidence that HGF, which has been implicated as a chemoattractant for myogenic cells in the early embryo (Bladt et al., 1995; Takayama et al., 1996), may also carry out this function in satellite cells during muscle regeneration. We found that migration was stimulated by HGF in the 1–10 ng/ml range and inhibited at higher concentration. Jennische et al. (1993) have reported that mRNA for HGF is first expressed in regenerating muscle 3 days after injury, the earliest time measured, and declines again to barely detectable levels by a week after injury. Although concentrations of HGF were not measured, its location and time of expression are consistent with its function as a chemoattractant. HGF may also be involved in other processes during regeneration. Allen et al. (1995) demonstrated that HGF acts as a mitogen for rat satellite cells in culture, primarily by shortening the lag time before quiescent CHEMOTAXIS OF SATELLITE CELLS Fig. 9. Hypothetical model for the regulation of satellite cell behavior after injury by overlapping gradients of mitogenic and motogenic factors. Injury results in the local release of mitogen, such as crushed muscle extract (CME), from the damaged myofibers, thereby producing a steep gradient centered on the injury. Widespread platelet extravasation induced by the inflammatory cascade results in a broad gradient of motogen, such as TGF-b. Satellite cells distant from the injury migrate up the TGF-b gradient until high concentration near the injury inhibits further movement. The high mitogen concentration at the injury site is able to overcome the inhibition of proliferation by TGF-b. cells enter the cell cycle after stimulation by serum factors. This study also showed that quiescent satellite cells express mRNA for c-met, the receptor for HGF. Besides stimulating muscle cell movement and growth, HGF may aid regeneration as an angiogenic factor. HGF is a potent chemoattractant for endothelial cells in vitro and promotes neovascularization in the cornea (Bussolini et al., 1993; Grant et al., 1993). The present results also suggest an approach for improving the efficiency of myoblast transfer therapy for neuromuscular disease. Since the demonstration by Lipton and Schultz (1979) that implanted satellite cells expanded in culture become incorporated into the host myofibers, many studies have used this technique as a method for introducing normal genes into diseased muscle to correct the genetic defect. Although encouraging results have been obtained in dystrophic mice (Karpati et al., 1989; Partridge et al., 1989; Morgan et al., 1990; Kinoshita et al., 1994), the method has not yet been effective in humans (Karpati et al., 1993; Law et al., 1993; Tremblay et al., 1993; Mendell et al., 1995; Morandi et al., 1995). A significant limitation is the failure of implanted cells to spread much beyond the 511 injection site (Hughes and Blau, 1990; Satoh et al., 1993; Rando et al., 1995). Fan et al. (1996) have reported that implanted mouse myoblasts do not migrate more than 0.1 mm, or approximately the diameter of two myofibers, and therefore the fraction of host myofibers that incorporate the donor cells is small. Morgan et al. (1996) have estimated that 2 3 106 myoblast injections would be required to generate new muscle in an adult human. Because implanted myoblasts are more readily incorporated into regenerating muscle (Partridge et al., 1989; Karpati, 1992; Watt et al., 1994; Huard et al., 1994), attempts have been made to increase the efficiency of transplant therapy by injuring the host muscle before or during myoblast implantation. This has increased migration in some cases (Morgan et al., 1987; Watt et al., 1994; Huard et al., 1994; Vilquin et al., 1995), but not others (Satoh et al., 1993; Rando and Blau, 1994; Fan et al., 1996), and the reason underlying the different results is not apparent. Injury to host muscle is precluded in human therapy owing to ethical considerations, but administration of motogenic factors, along with implanted myoblasts, may be used to increase dispersion of the cells. We have shown that platelet extract contains factors that greatly increase the random motility of satellite cells; identification and therapeutic use of these factors may aid in the dispersion of implanted myoblasts. EXPERIMENTAL PROCEDURES Satellite Cells Satellite cells were isolated from the hindleg muscles, tibialis anterior, and extensor digitorum longus of 1.5to 2-month-old male Sprague-Dawley rats (Harlan, Indianapolis, IN) (Bischoff and Heintz, 1994). After removal of epimysium and internal connective tissue, the muscles were minced with scissors and incubated with constant agitation for 1 hr at 37°C in 1% Pronase (Calbiochem, San Diego, CA) in Earle’s balanced salt solution (EBSS). Cells were released from the tissue fragments by vigorous trituration in Eagle’s minimal essential medium (MEM) containing 10% horse serum. Undissociated fragments were removed and triturated in fresh volumes of medium for a total of 3 cycles until no identifiable myofibers remained during microscopic examination. Pooled cells were filtered through a 10 µm Nitex filter (TETCO, New York, NY), centrifuged at 1,000g for 2 min and resuspended in 1.5 ml MEM with 10% horse serum before fractionation by density gradient centrifugation (Yablonka-Reuveni et al., 1987; Morgan, 1988). The cells were layered over a step gradient of Percoll (Pharamacia, Piscataway, NJ) consisting of 3.5 ml each of 70, 50, and 35% Percoll in MEM in a 16 3 125-mm screw cap culture tube and centrifuged with the brake off at 1,250g in a horizontal rotor for 20 min at room temperature. Cells were collected from the 50–70% interface, corresponding to a density of 1.087 g/ml as measured with density marker beads (Sigma, St. Louis, MO), washed in EBSS by centrifugation, and 512 BISCHOFF Fig. 10. Phase-contrast micrographs of cultured satellite cells from the 50–70% Percoll fraction after density gradient centrifugation. A: Cells after 3-day culture in medium containing 10% serum and 5% embryo extract. Most cells exhibit the plump, spindle-shaped morphology of myogenic cells, and some have begun to align themselves. B: Cells after an additional 2 days in medium containing 1% serum to promote differentiation. The cells have fused to form multinucleated, branching myotubes. Bar 5 100 µm. suspended in growth medium consisting of MEM with 10% horse serum, 5% chick embryo extract, and 1% antibiotic-antimycotic mixture. The cells were grown for 3 days in gelatin-coated tissue culture dishes with twice daily medium changes and removed from dishes for experiments by 5 min incubation in trypsin (100 µg/ml, 1:250, Sigma). Muscles from one rat yielded 2.3 6 0.4 3 106 cells after 3 days growth (average of four experiments). The Percoll-purified fraction consisted of greater than 98% myogenic cells based upon morphol- ogy (Fig. 10A) and differentiation to form multinucleated myotubes (Fig. 10B). Chemotaxis Assay Chemotaxis experiments were carried out in blindwell Boyden-type chambers with ten 7-mm-diameter wells per plate (Neuro Probe, Cabin John, MD). Standard polycarbonate filters with 8 µm holes (Neuro Probe; holes 5 6% surface area) were washed thoroughly, treated with bovine fibronectin, 25 µg/ml DW 513 CHEMOTAXIS OF SATELLITE CELLS for 30 min, and dried. Of several extracellular matrix proteins tested, fibronectin gave the best results in terms of attachment, migration, and absence of drop-off from the lower surface of the filter. In use, the bottom wells were filled with test substance, a filter was added, and the top wells were filled with a suspension of satellite cells. The chambers were incubated at 37°C in 95% air and 5% CO2. After allowing for migration, the filters were removed, fixed in alcohol:formalin:acetic acid (20:2:1), stained in hematoxylin, and mounted on slides. Both a negative control, consisting of medium alone, and a positive control, consisting of either crushed muscle extract or chick embryo extract, was carried out for each experiment. Drop-off of cells from the lower surface of the filter, monitored by examining medium from the bottom chamber in a hemocytometer, was negligible. In early experiments, cells were scraped from the top of the filter before counting, but for most of the experiments reported here cells were counted on both surfaces of the filter. Cells were counted using an ocular grid at a magnification of 5003. The culture medium for chemotaxis consisted of MEM with 0.5% bovine albumin and 1% rat serum derived from plateletfree plasma prepared from fresh rat blood. Tissue Extracts Muscle extract was prepared by crushing whole leg muscles of adult rats with forceps and incubating the muscles in EBSS at 4°C for 1 hr (Bischoff, 1986). The supernatant was concentrated about tenfold by ultrafiltration (YM10, Amicon, Beverly, MA) and used in chemotaxis experiments as crude extract. Affinity chromatography on a column of heparin-agarose (Sigma) was used to purify the crude extract as described (Bischoff, 1989; Bischoff, 1990b). Embryo extract was prepared from 11-day-old chick embryos; other tissue culture reagents were from Sigma. Protein concentration was measured with the microbiuret method (Munkres and Richards, 1965) using bovine albumin as a standard. Statistical Analysis The data presented are from single experiments with each factor tested in duplicate or triplicate wells and are representative of at least two experiments. Statistical analysis of individual experiments was carried out by comparing sample means with a two-tailed t-test using StatS (Spreadware, Palm Desert, CA) operating within Excel (Microsoft, Redmond, WA). ACKNOWLEDGMENTS Plasma and Platelet Extract Rats were anesthetized with pentobarbital, and platelets and serum were obtained as described (Catalfamo and Dodds, 1989). The abdominal aorta was exposed, and blood was withdrawn using a 21-gauge needle and collected in a syringe containing enough sodium citrate to give a final concentration of 0.38%. The blood was centrifuged at 550g for 2 min at room temperature, and the platelet-rich supernatant plasma was collected. Centrifugation was repeated 3 times, the platelet-rich plasma was pooled, and an aliquot was diluted 1:100 for counting in a hemocytometer. The platelets were pelleted by centrifugation at 1,800g for 15 min, and the supernatant was used to prepare serum from the platelet-free plasma. The platelets from the pellet were washed twice with EBSS by centrifugation and suspended in EBSS to give 1.5 3 106 platelets/µl. Thrombin (0.5 U/ml) was added to stimulate the platelets and cause the secretion of the alpha granules. After 10 min, the preparation was centrifuged at 1,800g, and the supernatant was retained as platelet extract for testing as chemoattractant. Antibodies and Peptides For inhibition experiments, pan-specific TGF-b neutralizing rabbit IgG (R&D Systems, Minneapolis, MN) was mixed with either platelet extract or TGF-b and incubated for 1 hr at room temperature before testing in chemotactic assays. Controls consisted of nonimmune rabbit IgG (Sigma). Purified growth factors were obtained from both R&D Systems or Sigma and gave comparable results regardless of source. I thank C. Heintz and R. Frederickson for expert assistance. This work was supported by grants from the Muscular Dystrophy Association. REFERENCES Adelman-Grill, B.C., Wach, F., Cully, Z., Hein, R., and Kreig, T. (1990) Chemotactic migration of normal dermal fibroblasts towards epidermal growth factor and its modulation by platelet-derived growth factor and transforming growth factor-beta. Eur. J. Cell Biol. 51:322–326. Allen, R.E., and Boxhorn, L.K. (1987) Inhibition of skeletal muscle satellite cell differentiation by transforming growth factor-beta. J. Cell. Physiol. 133:567–572. Allen, R.E., and Boxhorn, L.K. (1989) Regulation of skeletal-muscle satellite cell-proliferation and differentiation by transforming growth factor-beta, insulin-like growth factor-I, and fibroblast growthfactor. J. Cell. Physiol. 138:311–315. Allen, R.E., and Rankin, L.L. (1990) Regulation of satellite cells during skeletal-muscle growth and development. Proc. Soc. Exp. Biol. Med. 194:81–86. 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. Bischoff, R. (1986) A satellite cell mitogen from crushed adult muscle. Dev. Biol. 115:140–147. Bischoff, R. (1989) Analysis of muscle regeneration using single myofibers in culture. Med. Sci. Sports Exerc. 21:S164–S172. Bischoff, R. (1990a) Cell cycle commitment of rat muscle satellite cells. J. Cell Biol. 111:201–207. Bischoff, R. (1990b) Control of satellite cell proliferation. In: ‘‘Myoblast Transfer Therapy,’’ Griggs, R., and Karpati, G. (eds). New York: Plenum, pp 147–158. Bischoff, R., and Heintz, C. (1994) Enhancement of skeletal muscle regeneration. Dev. Dyn. 201:41–54. Blachowski, S., Motyl, T., Orzechowski, A., Grzelkowska, K., and Interewicz, B. (1993) Comparison of metabolic effects of EGF, TGF-a, and TGF-b1 in primary culture of fetal bovine myoblasts and rat L6 myoblasts. Int. J. Biochem. 25:1571–1577. Bladt, F., Riethmacher, D., Isenmann, S., Aguzzi, A., and Birchmeier, 514 BISCHOFF C. (1995) Essential role for the c-met receptor in the migration of myogenic presursor cells into the limb bud. Nature 376:768–771. Bottaro, D.P., Rubin, J.S., Faletto, D.L., Kmiecik, T.E., Vande Woude, G.F., and Aaronson, S.A. (1991) Identification of the hepatocyte growth factor receptor as the c-met proto-oncogene produce. Science 251:802–804. Brand-Saberi, B., Krenn, V., Grim, M., and Christ, B. (1993) Differences in the fibronectin-dependence of migrating cell populations. Anat. Embryol. (Berl.) 187:17–26. Bussolini, F., Di Renzo, M.F., Ziche, M., Bocchietto, E., Olivero, M., Naldini, L., Gaudino, G., Tamagnone, L., Coffer, A., and Comoglio, P.M. (1993) Hepatocyte growth factor is a potent angiogenic factor which stimulates endothelial cell motility and growth. J. Cell Biol. 119:629–641. Catalfamo, J.L., and Dodds, W.J. (1989) Isolation of platelets from laboratory animals. Methods Enzymol. 169:27–34. Chen, G., and Quinn, L.S. (1992) Partial characterization of skeletal myoblast mitogens in mouse crushed muscle extract. J. Cell. Physiol. 153:563–574. Chevallier, A., Kieny, M., and Mauger, A. (1977) Limb-somite relationship: origin of the limb musculature. J. Embryol. Exp. Morphol. 41:245–258. Christ, B., Jacob, B.J., and Jacob, M. (1977) Experimental analysis of the origin of the wing musculature in avian embryos. Anat. Embryol. (Berl.) 150:171–186. Cook, D.R., Doumit, M.E., and Merkel, R.A. (1993) Transforming growth factor-beta, basic fibroblast growth factor, and plateletderived growth factor-BB interact to affect proliferation of clonally derived porcine satellite cells. J. Cell. Physiol. 157:307–312. Cromack, D.T., Sporn, M.B., Roberts, A.B., Merino, M.J., Dart, L.L., and Norton, J.A. (1987) Transforming growth factor-b levels in rat wound chambers. J. Surg. Res. 42:622–628. Daston, G., Lamar, E., Olivier, M., and Goulding, M. (1996) Pax-3 is necessary for migration but not differentiation of limb muscle precursors in the mouse. Development 122:1017–1027. DiMario, J., and Strohman, R.C. (1988) Satellite cells from dystrophic (mdx) mouse muscle are stimulated by fibroblast growth-factor in vitro. Differentiation 39:42–49. Falk, W., Goodwin, R.H., and Leonard, E.J. (1980) A 48-well micro chemotaxis assembly for rapid and accurate measurement of leucocyte chemotaxis. J. Immunol. Methods 33:239–247. Fan, Y., Maley, M., Beilharz, M., and Grounds, M. (1996) Rapid death of injected myoblasts in myoblast transfer therapy. Muscle Nerve 19:853–860. Ghins, E., Colson-Van, S.M., and Marechal, G. (1984) The origin of muscle stem cells in rat triceps surae regenerating after mincing. J. Muscle Res. Cell Motil. 5:771–722. Grant, D.S., Kleinman, H.K., Goldberg, I.D., Bhargava, M.M., Nickoloff, B.J., Kinsella, J.L., Polverini, P., and Rosen, E.M. (1993) Scatter factor induces blood vessel formation in vivo. Proc. Natl. Acad. Sci. U.S.A. 90:1937–1941. Haugk, K.L., Roeder, R.A., Garber, M.J., and Schelling, G.T. (1995) Regulation of muscle cell proliferation by extracts from crushed muscle. J. Anim. Sci. 73:1972–1981. Hayashi, K., and Ozawa, E. (1995) Myogenic cell migration from somites is induced by tissue contact with medial region of the presumptive limb mesoderm in chick embryos. Development 121: 661–669. Huard, J., Verreault, S., Roy, R., Tremblay, M., and Tremblay, J.P. (1994) High efficiency of muscle regeneration after human myoblast clone transplantation in scid mice. J. Clin. Invest. 93:586–599. Hughes, S.M., and Blau, H.M. (1990) Migration of myoblasts across basal lamina during skeletal muscle development. Nature 345:350– 353. Jennische, E., Ekberg, S., and Matejka, G.L. (1993) Expression of hepatocyte growth factor in growing and regenerating rat skeletal muscle. Am. J. Physiol. 265:C122–C128. Jin, P., Sejersen, T., and Ringertz, N.R. (1991) Recombinant plateletderived growth factor-BB stimulates growth and inhibits differentiation of rat L6 myoblasts. J. Biol. Chem. 266:1245–1249. Kardami, E., Spector, D., and Strohman, R.C. (1985) Selected muscle and nerve extracts contain an activity which stimulates myoblast proliferation and which is distinct from transferrin. Dev. Biol. 112:353–358. Karpati, G. (1992) Approaches to the introduction of normal alleles into skeletal muscle fibers for therapeutic purposes in Duchenne muscular dystrophy. In: ‘‘Duchenne Muscular Dystrophy: Animal Models and Genetic Manipulation,’’ Kakulas, B.A., McHowell, J., and Roses, A.D. (eds). New York: Raven Press, pp 223–236. Karpati, G., Pouliot, Y., Zubrzyckagaarn, E., Carpenter, S., Ray, P.N., Worton, R.G., and Holland, P. (1989) Dystrophin is expressed in mdx skeletal-muscle fibers after normal myoblast implantation. Am. J. Pathol. 135:27–32. Karpati, G., Ajdukovic, D., Arnold, D., Gledhill, R.B., Guttmann, R., Holland, P., Koch, P.A., Shoubridge, E., Spence, D., Vanasse, M., Watters, G.V., Abrahamowicz, M., Duff, C., and Worton, R.G. (1993) Myoblast transfer in Duchenne muscular-dystrophy. Ann. Neurol. 34:8–17. Kinoshita, I., Huard, J., and Tremblay, J.P. (1994) Utilization of myoblasts from transgenic mice to evaluate the efficacy of myoblast transplantation. Muscle Nerve 17:975–980. Klein-Ogus, C., and Harris, J.B. (1983) Preliminary observations of satellite cells in undamaged fibres of the rat soleus muscle assaulted by a snake-venom toxin. Cell Tissue Res. 230:671–676. Kovanen, V., Suominen, H., Risteli, J., and Risteli, L. (1988) Type IV collagen and laminin in slow and fast skeletal muscle in rats— effects of age and life-time endurance training. Coll. Relat. Res. 8:145–153. Law, P.K., Goodwin, T.G., Fang, Q.W., Deering, M.B., Duggirala, V., Larkin, C., Florendo, J.A., Kirby, D.S., Li, H.J., Chen, M., Cornett, J., Li, L.M., Shirzad, A., Quinley, T., Yoo, T.J., and Holcomb, R. (1993) Cell transplantation as an experimental treatment for Duchenne muscular-dystrophy. Cell Transplantation 2:485–505. Lipton, B.H., and Schultz, E. (1979) Developmental fate of skeletal muscle satellite cells. Science 205:1292–1294. Lucas, P.A., Price, P.A., and Caplan, A.I. (1988) Chemotactic response of mesenchymal cells, fibroblasts and osteoblast-like cells to bone Gla protein. Bone 9:319–323. Massagúe, J.S., Cheifetz, S., Laiho, M., Ralph, D.A., Weiss, F.M.B., and Zentella, A. (1992) Transforming growth factor-b. Cancer Surveys 12:81–103. Mendell, J.R., Kissel, J.T., Amato, A.A., King, W., Signore, L., Prior, T.W., Sahenk, Z., Besson, S., Mcandrew, P.E., Rice, R., Nagaraja, H., Stephens, R., Lantry, L., Morris, G.E., and Burghes, A.H.M. (1995) Myoblast transfer in the treatment of Duchenne’s muscular dystrophy. N. Engl. J. Med. 333:832–838. Moens, P.D.J., Colson Van-Schoor, M., and Maréchal, G. (1996) Lack of myoblasts migration between transplanted and host muscles of mdx and normal mice. J. Muscle Res. Cell Motil. 17:37–43. Morandi, L., Bernasconi, P., Gebbia, M., Mora, M., Crosti, F., Mantegazza, R., and Cornelio, F. (1995) Lack of mRNA and dystrophin expression in DMD patients three months after myoblast transfer. Neuromuscular Disorders 5:291–295. Morgan, J.E. (1988) Myogenicity in vitro and in vivo of mouse muscle cells separated on discontinuous Percoll gradients. J. Neurol. Sci. 85:197–207. Morgan, J.E., Coulton, G.R., and Partridge, T.A. (1987) Muscle precursor cells invade and repopulate freeze killed muscles. J. Muscle Res. Cell Motil. 8:386–396. Morgan, J.E., Hoffman, E.P., and Partridge, T.A. (1990) Normal myogenic cells from newborn mice restore normal histology to degenerating muscles of the mdx mouse. J. Cell Biol. 111:2437– 2449. Morgan, J.E., Fletcher, R.M., and Partridge, T.A. (1996) Yields of muscle from myogenic cells implanted into young and old mdx hosts. Muscle Nerve 19:132–139. Munkres, K.D., and Richards, F.M. (1965) The purification and properties of malate dehydrogenase. Arch. Biochem. Biophys. 109: 466–479. Pampusch, M.S., Hembree, J.R., Hathaway, M.R., and Dayton, W.R. (1990) Effect of transforming growth-factor-beta on proliferation of L6 and embryonic porcine myogenic cells. J. Cell. Physiol. 143:524– 528. CHEMOTAXIS OF SATELLITE CELLS Partridge, T.A. (1991) Invited review—myoblast transfer—a possible therapy for inherited myopathies. Muscle Nerve 14:197–212. Partridge, T.A., Morgan, J.E., Coulton, G.R., Hoffman, E.P., and Kunkel, L.M. (1989) Conversion of mdx myofibers from dystrophinnegative to dystrophin-positive by injection of normal myoblasts. Nature 337:176–179. Phillips, G.D., Lu, D., and Carlson, B.M. (1987) Survival of myogenic cells in freely grafted rectus femoris and extensor digitorum longus muscles. Am. J. Anat. 180:365–372. Phillips, G.D., Hoffman, J.R., and Knighton, D.R. (1990) Migration of myogenic cells in the rat extensor digitorum longus muscle studied with a split autograft model. Cell Tissue Res. 262:81–88. Phillips, G.D., Schilb, L.A., Fiegel, V.D., and Knighton, D.R. (1991) An angiogenic extract from skeletal muscle stimulates monocyte and endothelial cell chemotaxis in vitro. Proc. Soc. Exp. Biol. Med. 197:458–464. Rando, T.A., and Blau, H.M. (1994) Primary mouse myoblast purification, characterization, and transplantation for cell-mediated gene therapy. J. Cell Biol. 125:1275–1287. Rando, T.A., Pavlath, G.K., and Blau, H.M. (1995) The fate of myoblasts following transplantation into mature muscle. Exp. Cell Res. 220:383–389. Roberts, A.B., and Sporn, M.B. (1991) The transforming growth factor-bs. In: ‘‘Peptide Growth Factors and Their Receptors I,’’ Sporn, M.B. and Roberts, A.B. (eds). New York: Springer-Verlag, pp 419–472. Robertson, T.A., Maley, M.A.L., Grounds, M.D., and Papadimitriou, J.M. (1993) The role of macrophages in skeletal muscle regeneration with particular reference to chemotaxis. Exp. Cell Res. 207:321– 331. Satoh, A., Huard, J., Labrecque, C., and Tremblay, J.P. (1993) Use of fluorescent latex microspheres (FLMS) to follow the fate of transplanted myoblasts. J. Histochem. Cytochem. 41:1579–1582. Schultz, E., Jaryszak, D.L., and Valliere, C.R. (1985) Response of satellite cells to focal skeletal muscle injury. Muscle Nerve 8:217– 222. Schultz, E., Jaryszak, D.L., Gibson, M.C., and Albright, D.J. (1986) Absence of exogenous satellite cell contribution to regeneration of frozen skeletal muscle. J. Muscle Res. Cell. Motil. 7:361–367. 515 Schultz, E., Albright, D.J., Jaryszak, D.L., and David, T.L. (1988) Survival of satellite cells in whole muscle transplants. Anat. Rec. 222:12–17. Takayama, H., LaRochelle, W.J., Anver, M., Bockman, D.E., and Merline, G. (1996) Scatter factor/hepatocyte growth factor as a regulator of skeletal muscle and neural crest development. Proc. Natl. Acad. Sci. U.S.A. 93:5866–5871. Tremblay, J.P., Malouin, F., Roy, R., Huard, J., Bouchard, J.P., Satoh, A., and Richards, C.L. (1993) Results of a triple blind clinical-study of myoblast transplantations without immunosuppressive treatment in young boys with Duchenne muscular-dystrophy. Cell Transplantation 2:99–112. Venkatasubramanian, K., and Solursh, M. (1984) Chemotactic behavior of myoblasts. Dev. Biol. 104:428–433. Vilquin, J.T., Wagner, E., Kinoshita, I., Roy, R., and Tremblay, J.P. (1995) Successful histocompatible myoblast transplantation in dystrophin-deficient mdx mouse despite the production of antibodies against dystrophin. J. Cell Biol. 131:975–988. Watt, D.J., Morgan, J.E., Clifford, M.A., and Partridge, T.A. (1987) The movement of muscle precursor cells between adjacent regenerating muscles in the mouse. Anat. Embryol. (Berl.) 175:527–536. Watt, D.J., Karasinski, J., Moss, J., and England, M.A. (1994) Migration of muscle cells. Nature 368:406–407. Yablonka-Reuveni, Z., Quinn, L.S., and Nameroff, M. (1987) Isolation and clonal analysis of satellite cells from chicken pectoralis muscle. Dev. Biol. 119:252–259. Yablonka-Reuveni, Z., Balestreri, T.M., and Bowen Pope, D.F. (1990) Regulation of proliferation and differentiation of myoblasts derived from adult-mouse skeletal-muscle by specific isoforms of PDGF. J. Cell Biol. 111:1623–1629. Yamazaki, M., Minota, S., Sakurai, H., Miyazono, K., Yamada, A., Kanazawa, I., and Kawai, M. (1994) Expression of transforming growth factor-beta-1 and its relation to endomysial fibrosis in progressive muscular dystrophy. Am. J. Pathol. 144:221–226. Zentella, A., and Massagúe, J. (1992) Transforming growth-factor-beta induces myoblast differentiation in the presence of mitogens. Proc. Natl. Acad. Sci. U.S.A. 89:5176–5180. Zigmond, S.H., and Hirsch, J.G. (1973) Leucocyte locomotion and chemotaxix: new methods for demonstration of cell-derived chemotactic factor. Exp. Med. 137:387–410.