Extracellular matrix histone H1 binds to perlecan, is present in regenerating skeletal muscle and stimulates myoblast proliferation

The identification of proteins that bind to the glycosaminoglycan (GAG)chains of proteoglycans has allowed a greater understanding of the wide variety of functions that these macromolecules exert on cell behavior(Bernfield et al., 1999;Iozzo, 1998). Especially relevant examples are proteins that bind to heparin, an analogue of heparan sulfate, which is the GAG attached to most membraneanchored and basal lamina proteoglycans. Functional heparin-binding domains have been identified in constituents of the extracellular matrix (ECM)(Barkalow and Schwarzbauer,1991; Kouzi-Koliakos et al.,1989), heparin-binding growth factors(Bernfield et al., 1999) and proteins present on the surface of bacteria, parasites and viruses(Rostand and Esko, 1997). In addition to these examples, a number of diverse extracellular proteins display affinity for heparan sulfate, including proteins involved in morphogenesis,cell adhesion, proliferation, migration, tissue remodeling and blood coagulation (Bernfield et al.,1999).

We are interested in the role that heparan sulfate proteoglycans play during skeletal muscle formation. In this process, committed myogenic progenitor cells, or myoblasts, are maintained in a proliferative,undifferentiated state until appropriate signals trigger their conversion to multinucleated myotubes. The differentiation of muscle cells is controlled in a negative manner by specific heparan-sulfate-binding mitogens, such as basic fibroblast growth factor (FGF-2), hepatocyte growth factor and transforming growth factor-β (Brunetti and Goldfine, 1990; Heino and Massague, 1990; Takayama et al., 1996).

Sodium chlorate, a specific inhibitor of proteoglycan sulfation, has previously been shown to decrease the deposition and assembly of ECM components in cultured C₂C₁₂ myoblasts that are induced to differentiate. This results in the inhibition of cell fusion(Melo et al., 1996;Olwin and Rapraeger, 1992;Osses and Brandan, 2001) and FGF-2-dependent suppression of myoblast differentiation(Larraín et al., 1998). We have also shown that the expression of heparan sulfate proteoglycans during the terminal skeletal muscle differentiation of C₂C₁₂cells is highly regulated. The expression of glypican, a glycosylphosphatidylinositolanchored heparan sulfate proteoglycan that is also released into the ECM, increases (Brandan et al., 1996; Campos et al.,1993), whereas expression of perlecan, an ECM proteoglycan, and the transmembrane proteoglycans syndecan-1 and -3 decreases during differentiation (Fuentealba et al.,1999; Larraín et al.,1997a; Larraín et al.,1997b). Furthermore, heparan sulfate proteoglycans may regulate muscle physiology. We have demonstrated that expression of syndecan-1(Larraín et al., 1998)and syndecan-3 (Fuentealba et al.,1999) inhibit the differentiation of cultured myoblasts in an FGF-2-dependent manner. On the other hand, the presence of perlecan in muscle basement membrane has been suggested to participate in cell-ECM interactions(Villar et al., 1999),neuromuscular junction formation (Jacobson et al., 2001) and muscle regeneration(Gulati et al., 1983).

In the present study we searched for endogenous skeletal muscle cell ligand(s) for heparan sulfate proteoglycans that had not previously been described. We did this on the premise that knowledge of such proteins might aid in elucidating the role(s) that these macromolecules exert in muscle physiology. Here, we demonstrate that an extracellularly occurring histone H1 colocalizes with the heparan sulfate proteoglycan perlecan in the ECM of myotube cultures and in regenerating skeletal muscle. In vitro assays show that histone H1 stimulates myoblast proliferation via a heparan-sulfate-dependent mechanism. We propose that this extracellular heparan sulfate proteoglycan-binding protein may play a functional role during muscle regeneration.

The C₂C₁₂ cell line was purchased from ATCC,Manassas, VA. The JIE7 antibody (against Escherichia coliβ-galactosidase) developed by T. L. Mason and J. A. Partaledis and the F1.652 antibody (against embryonic myosin) developed by Helen Blau were obtained from the Developmental Studies Hybridoma Bank developed under the auspices of the NICHD and maintained by the University of Iowa, Department of Biological Sciences, Iowa City, IA. Dulbecco's modified Eagle's (DMEM)/Ham's F12 medium (1:1), chicken embryo extract, horse serum, fetal calf serum and Hank's balanced salt solution were obtained from Life Technologies, Inc.,Boston, MA. Creatine kinase (CK) and lactate dehydrogenase (LDH) assay kits were from Valtek, Santiago, Chile. Heparin, chondroitin sulfate, dermatan sulfate, N-desulfated heparin, heparin-agarose, laminin, fibronectin,fluorescein-isothiocyanate-conjugated and tetramethylrhodamine-5(and 6)-isothiocyanate-conjugated anti IgG were from Sigma, St Louis, MO. Histone H1 and anti-human histone H1 monoclonal antibody (clone AE-4) were from Upstate Biotechnology, Lake Placid, NY. The C17 polyclonal anti-histone H1 antibody was from Santa Cruz Biotechnology, Inc., Santa Cruz, CA. SuperSignal ULTRA Chemiluminiscent substrate was from Pierce, Rockford, IL. Macro-Prep DEAE was from Bio-Rad, Hercules, CA. Cell death detection ELISA^PLUSkit was from Roche, Mannheim, Germany. Heparitinase and anti-Δ-heparan sulfate antibody were from Seikagaku, Tokyo, Japan.[³⁵S]H₂SO₄ carrier-free (1050-1600 Ci/mmol)and [methyl-³H]thymidine (79 Ci/mmol) were obtained from NEN Life Science Products, Boston, MA. Tissue Freezing Medium gel was from Triangle Biomedical Sciences, Dr Durham, NC. Cellulosenitrate was from Schleicher and Shuell, Dassel, Germany.

The mouse skeletal muscle cell line C₂C₁₂ was grown and induced to differentiate as previously described(Larraín et al.,1997b). In some experiments, cells were cultured and maintained during differentiation in the presence of sodium chlorate at a final concentration of 30 mM (Melo et al.,1996). MST cells were grown as previously described(Montgomery et al., 1992).

For sequential extractions, C₂C₁₂ myotubes were rinsed twice with phosphate buffered saline (PBS) and harvested by scraping in the same buffer. The cells were homogenized and a PBS-soluble extract was obtained by centrifuging for 10 minutes at 13,000 g and 4°C. The resulting pellet was homogenized in TX-100 buffer (0.05 M Tris-HCl, pH 7.4,0.5% Triton X-100, 0.15 M NaCl) and centrifuged under the same conditions. The final pellet was homogenized in TX-100/KCl solution (0.05 M Tris-HCl, pH 7.4,0.5% Triton X-100, 0.5 M KCl). For heparin displacement of extracellular proteins from intact muscle cells, myoblasts or differentiating myoblasts were washed three times for 10 minutes with cold PBS containing 0.1 mM CaCl₂ and 1 mM MgCl₂ under gentle agitation at 4°C. The cells were then incubated in 2 mg/ml heparin in the same buffer for 20 minutes. The PBS-heparin extract was clarified by centrifugation at 13,000 g for 2 minutes at 4°C. Acid extraction of nuclear proteins was carried out essentially as described previously(Schmiedeke et al., 1989).

For ligand blot assays, [³⁵S]heparin was biosynthetically labeled and isolated from MST cells as previously described(Montogomery et al., 1992). Proteins in the various muscle cell extracts were separated by SDS-PAGE and electrotransferred onto nitrocellulose membranes at 100 V for 4 hours. Membranes were then blocked with 5% bovine serum albumin (BSA) in PBS for 1 hour and incubated overnight with 2×10⁶ cpm/ml[³⁵S]heparin at 4°C under gentle agitation. After washing the membranes, heparin-binding proteins were then visualized by autoradiography.

Myotubes were subjected to the sequential extraction described above. TX-100/KCl extracts were dialyzed against 0.05 M NaCl, 0.05 M Tris-HCl, pH 7.8 and loaded onto a 0.5 ml heparin-agarose column pre-equilibrated in 0.3 M NaCl, 0.05 M Tris-HCl, pH 7.8, and 1 mM PMSF(Brandan and Inestrosa, 1984). After rinsing the column extensively with equilibration buffer, bound material was eluted using a linear gradient of 0.3 to 1 M NaCl in 0.05 M Tris-HCl, pH 7.8. Total, unbound and eluted fractions were analyzed by SDS-PAGE followed by Coomassie blue staining.

Identification of p33/30 was carried out at HHMI Biopolymer and W. M. Keck Biotechnology Resource Laboratory, Yale University, CT, USA. Briefly,molecular masses of in-gel digested p33 peptides were obtained by matrix-assisted laser desorption ionization mass spectrometry (MALDI-MS)carried out on a Research Grade, VG Tofspec SE instrument. Resulting peptides were further purified on a C-18 reverse-phase HPLC column and a single peptide subjected to conventional Edman degradation for amino-acid sequencing using an Applied Biosystems sequencer. Immunoblots were developed as previously described (Brix et al., 1998)using anti-histone H1 antibody and visualized by enhanced chemiluminescence(ECL).

Histone H1 (0.5 mg) was separated by SDS-PAGE and electrotransferred onto a nitrocellulose sheet. After brief staining with 0.5% Ponceau S red, 5% acetic acid, histone H1-containing nitrocellulose pieces were excised and then blocked with 5% dry nonfat milk, 0.05 M Tris-HCl, pH 7.4, 0.15 M NaCl. A proteoglycanenriched fraction obtained from myotube-conditioned media was partially purified on a 1 ml DEAE column pre-equilibrated in 0.15 M NaCl, 0.05 M Tris-HCl, pH 7.4 and eluted with 1 M NaCl prepared in the same buffer. The nitrocellulose immobilized-histone H1 was incubated overnight with the dialyzed proteoglycan-containing fraction in 2% BSA at 4°C and washed with the same buffer. Heparan sulfate proteoglycans bound to immobilized histone-H1 were then eluted with 8 M urea, 0.05 M Tris-HCl, 0.15 M NaCl, pH 7.4. Total,unbound and bound fractions were dialyzed and treated with heparitinase as described (Brandan et al.,1996). The core proteins of heparan sulfate proteoglycans were visualized by western blot using the anti-Δ-heparan sulfate monoclonal antibody and polyclonal antiperlecan specific antisera, as previously described (Fuentealba et al.,1999; Larraín et al.,1997a).

The tibialis anterior muscle of adult hamsters were injected intramuscularly, under ketamine/xylazine anesthesia, with 0.25 ml of an aqueous solution of BaCl₂ (1.2% w/v)(Caldwell et al., 1990). Four and five days later the injected muscle area was dissected out of the anesthetized animal and the tissue quickly frozen in isopentane previously cooled in liquid nitrogen. Frozen tissue was embedded in Tissue Freezing Medium gel and a series of 6 μm cryostat sections was obtained.

Immunofluorescence microscopy of C₂C₁₂ cultures and skeletal muscle cryosections was carried out essentially as described previously (Riquelme et al.,2001). In some experiments, 95% ethanol/5% acetic acid was used for fixation and permeabilization of cells.

Myoblast proliferation was assayed by measuring [³H]thymidine incorporation, as described previously(Currie et al., 1997). For some assays, skeletal muscle ECM was obtained after detaching myotubes differentiated during six days by incubating with 10 mM EDTA in PBS at 37°C for 10 minutes, as described previously(Andress, 1995). Anti-histone H1 (C17) or anti-β galactosidase antibodies were diluted 1:100 for an overnight preincubation of the ECM-containing dishes and 1:200 through the proliferation assay.

Adhesion assays were performed as described previously(Chernousov et al., 1996). DNA and protein content determinations were performed as described previously(Riquelme et al., 2001). Cell death was determined by measuring histone-complexed DNA fragments (mono and oligonucleosomes) in conditioned media (necrosis) and cell extracts(apoptosis) using specific ELISA assays as described by the supplier.

In order to identify heparin-binding proteins in C₂C₁₂ cultures, proteins that were soluble in PBS,TX-100 and TX-100/KCl were obtained by sequential extraction and subjected to ligand blot assays using [³⁵S]heparin as a probe. As shown inFig. 1A, two proteins of 33 and 30 kDa (named p33/30 hereafter), which were present only in the TX-100/KCl extract, showed significant [³⁵S]heparin binding. Similar ligand blot assays using chondroitin and dermatan sulfate GAG chains as competitors suggested that p33/30 bound specifically to heparin(Fig. 1C).

To determine whether p33/30 was present in the extracellular space, intact C₂C₁₂ cell monolayers on various days of differentiation were incubated with a heparin solution. The solubilized material was analyzed by [³⁵S]heparin ligand blot assay. As shown inFig. 1B heparin was able to solubilize p33/30 from intact cells. In control experiments CK, LDH and DNA were poorly detected in the heparin extracts under these conditions(Table 1), demonstrating that the permeability of the cell monolayers was unaffected by heparin treatment. Furthermore, as shown in Fig. 1B, the binding of [³⁵S]heparin to heparin-solubilized p33/30 increased during skeletal muscle differentiation. These data strongly suggest that p33/30 binds to heparan sulfate, is present on the cell surface or in the extracellular space of muscle cell cultures, and increases during differentiation.

In order to purify p33/30, TX-100/KCl-soluble proteins obtained from differentiated myotubes were subjected to heparin-agarose chromatography. Proteins were eluted by a linear NaCl gradient, analyzed by SDS-PAGE and stained with Coomassie blue. Fig. 2A shows a 30 and 33 kDa doublet that was almost completely retained by the column and was eluted between 0.55 and 0.75 M NaCl. The identity of this p33/30 doublet was determined by MALDI-MS analysis of heparin affinity-purified fractions. The highest probability scores were obtained with histone H1 from different species (data not shown). Moreover, N-terminal sequencing of an HPLC-purified tryptic peptide from p33 showed highest identity with different subclasses of histone H1 from at least four species(Fig. 2B). This identification was confirmed by western blot analysis using an anti-histone H1 monoclonal antibody. As shown in Fig. 2C,an immunoreactive 33/30 kDa doublet was detected in the fractions shown above to contain p33/30 by the [³⁵S]heparin ligand blot assay. The high specificity of the antibody for histone H1 was confirmed by specific staining of a nuclear-acid-soluble protein fraction containing nuclear skeletal muscle histones (Fig. 2C). Furthermore, like p33/30, histone H1 was specifically solubilized from myoblasts by heparin, unlike other GAGs(Fig. 2D). Collectively, these data allow the identification of p33/30 as a subclass of histone H1.

To study a possible interaction between histone H1 and heparan sulfate proteoglycan(s), a solid-phase assay was performed by incubating a partially purified total proteoglycan fraction obtained from C₂C₁₂myotube-conditioned media with immobilized histone H1. Total, unbound and histone-H1-bound proteoglycans were analyzed by western blot using anti-Δ-heparan sulfate monoclonal antibody, which recognizes heparan sulfate proteoglycan core proteins after heparitinase treatment(Fig. 3A), and a polyclonal anti-perlecan-specific antibody (Fig. 3B). Although several core proteins of heparan sulfate proteoglycans, including glypican, were detected with the anti-Δ-heparan sulfate antibody in the total sample, only perlecan was observed to bind to the immobilized histone H1. In a negative control experiment, heparan sulfate proteoglycans did not bind to a blank nitrocellulose sheet (data not show). These in vitro data strongly suggest that histone H1 binds specifically to perlecan.

To evaluate the extracellular localization of histone H1 in skeletal muscle cells cultured in vitro, dual immunofluorescent staining was performed on non-permeabilized C₂C₁₂ myotubes. As shown inFig. 4 (left and middle panels), anti-histone H1 antibody (H1) produced an extracellular filamentous staining pattern that colocalized extensively with the heparan sulfate proteoglycans perlecan (per) and glypican (gly). Interestingly, as is the case with perlecan, the extracellular histone H1 distribution was restricted to the ECM in myotubes, whereas glypican was also present on the plasma membrane. Control experiments in permeabilized myotubes showed, as expected,anti-histone H1 antibody staining primarily in cell nuclei (pH1).

To investigate the relation between heparan sulfate proteoglycans present in the ECM and the extracellular localization of histone H1,C₂C₁₂ cultures were treated with sodium chlorate, a metabolic inhibitor of heparan sulfate synthesis(Fig. 4, right panel). Chlorate treatment resulted in fewer and shorter myotubes and a strong disorganization of the ECM (Melo et al.,1996), as revealed by phase-contrast microscopy as well as anti-perlecan (per) and anti-glypican (data not shown) staining. Under these conditions, the immunoreactivity of extracellular histone H1 was strongly diminished, suggesting that heparan sulfate proteoglycans are essential for its extracellular localization in the ECM.

The extracellular localization of histone H1 was evaluated in hamster skeletal muscle that had been induced to regenerate after barium chloride injection. As observed by hematoxilyn/eosin staining(Fig. 5A), four days after the injection, abundant mononucleated cells — probably both inflammatory and muscle precursor cells — were encountered together with degenerating myofibers and remnants of the original basement membranes. Newly formed centrally nucleated myotubes could be observed at this stage, but they were more evident and abundant five days after the injection.Fig. 5B shows that in normal adult skeletal muscle (ctrl) only nuclear histone H1 staining was observed. However, in regenerating muscle at day four after barium chloride injection, a pool of extracellular histone H1 was detected in a filamentous pattern between myofibers and mononucleated cells (H1, insert) that does not colocalize with nuclei stained with Hoechst 33258 (Ho, insert). Histone H1 extracellular staining colocalized with anti-perlecan antibody staining in the basement membrane of degenerated fibers (per, insert). At the fifth day of regeneration, extracellular histone H1 staining was weaker and was found mostly associated with the basal lamina surrounding the newly formed regenerating myotubes, where again it colocalized with perlecan. Similar myotubes can be detected by indirect immunofluorescence anti-embryonic myosin(em), a transient early skeletal muscle differentiation marker in similar regenerating foci of adjacent cryosections. No staining was observed with this antibody in control skeletal muscle (data not shown).

Taken together, these results demonstrate that a pool of extracellular histone H1 binds specifically to perlecan in skeletal muscle ECM.

To determine whether cell lysis could account for the presence of histone H1 in the ECM, LDH activity was assayed in conditioned media collected over 2 day intervals following induction of cell differentiation.Fig. 6A shows that LDH was released into the medium during the 6 days after induction of differentiation,with the highest level of LDH activity detected within the first two days. In these experiments, necrotic and apoptotic cell death were also evaluated using ELISA assays. As shown in Fig. 6B, necrosis was the main cause of cell death and showed a similar temporal pattern to LDH release during differentiation, whereas very little apoptotic cell death was detected.

The cumulative LDH release, calculated using the data inFig. 6A(Fig. 6A, insert), closely paralleled the accumulation of extracellular heparin-displaced histone H1 detected during the differentiation of C₂C₁₂ cells(Fig. 6C,Fig. 1B). These findings support the possibility of cell death as the source of the pool of histone H1 in the ECM of skeletal muscle cells.

To study the possible functions of extracellular histone H1, myoblasts were plated on dishes coated with histone H1 purified from myoblasts by heparin-affinity chromatography. Myoblast attachment and spreading was monitored by phase contrast microscopy.Fig. 7A (upper panel) shows that 2 hours after plating myoblasts on histone H1-coated dishes assumed a flattened polygonal morphology characteristic of C₂C₁₂myoblasts. For comparison, laminin, fibronectin and BSA were also used as attachment substrates. Myoblasts were well spread on laminin but not BSA. Interestingly, 21 hours after plating (Fig. 7A, lower panel), there was a large increase in the number of cells on histone-H1-containing dishes. Laminin showed a similar apparent increase in cell number, whereas fibronectin and BSA failed to produce such an effect. This result led us to examine a possible role for extracellular histone H1 in myoblast proliferation.

Cell proliferation, assayed by measuring [³H]thymidine incorporation, was seen to increase upon addition of exogenous histone H1 in a time- and concentration-dependent manner, reaching a maximal value at 10μg/ml (Fig. 7B,C). To evaluate the involvement of heparan-sulfate-like molecules in the proliferative effect observed for histone H1, myoblasts were incubated with increasing concentrations of histone H1 in the absence or presence of heparin.Fig. 7D shows that histone H1-dependent [³H]thymidine incorporation was increased in cells that were also treated with heparin. Moreover, the incorporation of[³H]thymidine induced by histone H1 was almost completely abolished in cells treated with sodium chlorate (data not shown). To directly evaluate the function of histone H1 externalized by C₂C₁₂ cells,[³H]thymidine incorporation was determined on myoblasts seeded on myotube ECM, which were prepared by detaching the myotube monolayer. This histone H1-containing ECM (data not shown) was previously incubated with anti-histone H1 (C17) and the antibody was maintained during the assay. As a control, an unrelated anti-β galactosidase antibody was also used. Anti-histone H1 antibody caused a 40% reduction of the proliferative effect induced by the myotube ECM (Fig. 7E). Collectively, these results strongly suggest that endogenously released histone H1, which is bound to the ECM, stimulates myoblast proliferation via a mechanism dependent on heparan sulfate chains.

The results presented in this study provide novel evidence that a pool of histone H1, a nuclear protein bound to linker DNA connecting adjacent nucleosomes in eukaryotic chromatin, becomes associated with the ECM of a skeletal muscle cell line and regenerating skeletal muscle. The amount of this extracellular histone H1, which colocalized with heparan sulfate proteoglycans, increases during muscle differentiation in culture. Moreover,our results demonstrate that histone H1 binds to and colocalizess specifically with perlecan, an ECM heparan sulfate proteoglycan expressed by muscle cells(Larraín et al., 1997a;Murdoch et al., 1994). We also showed that histone H1 promoted the proliferation of myoblasts via a mechanism that was stimulated by exogenous heparin, which acts as a functional and structural analogue of heparan sulfate chains.

Previous studies have shown that nucleosome-complexed core histones, namely H2A, H2B, H3 and H4 (Schmiedeke et al.,1989) as well as histone-like proteins(Kohnke-Godt and Gabius,1991), may bind to basement membranes and particularly to cell surface heparan sulfate proteoglycans(Watson et al., 1999). Extracellular nucleosomes have been linked either to the formation of auto-antibodies in the pathogenesis of systemic lupus erythematosus(Schmiedeke et al., 1989) or the stimulation of the synthesis of immunoglobulins and other immunomodulatory factors (Emlen et al.,1992).

Our data unequivocally demonstrate that the extracellular 33/30 kDa doublet, which was displaced from intact cell monolayers by heparin, is a subclass of histone H1, as determined by MALDI-MS, N-terminal sequencing and immunological analyses. During this investigation, extracellular histone H1 was never found complexed to DNA or other core histones.

Histone H1 is expressed in a wide variety of isoforms(Schulze et al., 1994) and shows different degrees of post-translational modification(Spencer and Davie, 1999). Nuclear histone H1 functions not only as a structural protein but also acts as a positive and negative regulator of gene transcription(Wolffe et al., 1997). Indeed,genes controlling muscle differentiation are known to be selectively repressed by histone H1 (Steinbach et al.,1997). Cytoplasmic pools of histone H1 have also been described in mammalian cells (Zlatanova et al.,1990), and an extracellular histone H1 was identified at the cell surface of a macrophage cell line, where it acts as a thyroglobulin-binding protein that mediates its endocytosis via a mechanism that is inhibited by heparin (Brix et al., 1998). Moreover, addition of histone H1 to different cell types has been correlated with cell division (Kundahl et al.,1981), differentiation(Okabe-Kado et al., 1981) and cytotoxicity (Class et al.,1996). These studies have led to the notion that histone H1 may act as a multifunctional protein, which is consistent with the findings described in this study.

The present data suggest that among several GAGs, histone H1 binds specifically to heparan sulfate chains. This interaction seems to be dependent on GAG sulfation, given that N-desulfated heparin failed to solubilize histone H1 from myoblasts. More specifically, a biochemical approach showed that,among the whole population of heparan sulfate proteoglycans externalized by myotubes, histone H1 exclusively binds to perlecan. Immunohistological studies in non-permeabilized myotubes showed that histone H1 colocalizes extensively with perlecan in the ECM, even though glypican is also present on the cell surface. Similarly, a filamentous pattern of extracellular histone H1 was observed in skeletal muscle regeneration foci. This pattern was also observed for the immunostaining of heparan sulfate proteoglycan core proteins (data not shown), and the specific colocalization of histone H1 with perlecan suggests that it is deposited in the ECM of regenerating skeletal muscle.

The specific interaction of histone H1 with perlecan is an intriguing finding given that many extracellular ligands bind to the heparan sulfate chains of different proteoglycans with similar affinities(Bernfield et al., 1999;Tumova et al., 2000). Moreover, we have demonstrated the presence of at least perlecan and glypican in the ECM of differentiating myotubes(Brandan et al., 1996;Larraín et al., 1997a). However, there is an increasing body of evidence that suggests a role for specific heparan sulfate sequence motifs in ligand specificity (for a review,see Turnbull et al., 2001). Alternatively, the expression of multiple isoforms of perlecan(Iozzo, 1998) might influence its specific binding to histone H1.

The results of this study point to a functional role for extracellular histone H1 in cell proliferation. We observed a five- to seven-fold increase in [³H]thymidine incorporation in histone-H1-treated muscle cells,which was correlated with an increase in cell number. Moreover, an anti-histone H1 antibody caused a 40% reduction of the proliferation induced by the ECM obtained from differentiated skeletal muscle cells. Heparan sulfate proteoglycans have been extensively characterized as co-receptors of mitogenic factors, with the FGF-heparan sulfate interaction constituting an archetype example (Bernfield et al.,1999). We have previously demonstrated that the myogenic inhibitory activity of FGF-2 is potentiated by the transmembrane heparan sulfate proteoglycans syndecan-1 and -3(Fuentealba et al., 1999;Larraín et al., 1998). Extracellularly occurring heparan sulfate proteoglycans such as perlecan or shed syndecan-1 may have the ability to inhibit signaling by sequestering the growth factor (Aviezer et al.,1994b; Kato et al.,1998). Nevertheless, perlecan has also been described as a positive regulator of FGF-2 activity in some conditions(Aviezer et al., 1994a),suggesting that cellular context may be important in determining functional roles.

As a potential mechanism for the observed effect of histone H1 on myoblast proliferation, the entry of extracellular histone H1 to the nucleus, which can be increased by the addition of heparin(Villeponteau, 1992), has been shown previously to generate new initiation sites for DNA replication(Kundahl et al., 1981). Although our results do not provide direct evidence, it is tempting to suggest that heparan sulfate may physiologically mediate the internalization of histone H1, as heparin addition was found to increase histone-H1-induced myoblast proliferation. In further support of this hypothesis, perlecan has been reported to mediate the internalization of different ligands(Fuki et al., 2000), in addition to its localization on the surface of proliferative myoblasts(Larraín et al.,1997a).

On the basis of the present novel data, we suggest that the presence of histone H1 in the ECM may play a role in muscle physiology. The regenerative capacity of skeletal muscle arises mainly from the activation of a small population of quiescent mononucleated cells, called satellite cells, located between the basal lamina and the sarcolemma of mature muscle fibers(Mauro, 1961). After injury,satellite cells, from which C₂C₁₂ cells are derived(Andres and Walsh, 1996),proliferate and fuse in response to extracellular signals in order to regenerate the damaged muscle fibers. Several well-known heparin-binding mitogenic factors, such as FGF-2, are released from muscle cells following injury (Clarke et al., 1993),and here we provide evidence suggesting that cell death may be involved in histone H1 externalization. Infiltrating leukocytes accumulate early at the site of damage and phagocytose the necrotic tissue. However, empty muscle fiber basement membrane structures and specific molecular components,including type IV collagen, laminin and heparan sulfate proteoglycan, are initially preserved (Gulati et al.,1983) and are repopulated by proliferating myoblasts. The persistence of basement membranes has been shown to accelerate muscle regeneration (Caldwell et al.,1990) and favor the reinnervation at original synaptic sites(Sanes et al., 1978). In this context, our results support the idea that after damage histone H1 may act as an externalized growth signal in regenerating muscle that can be stored by heparan sulfate proteoglycan ECM components, such as perlecan, to subsequently induce the proliferation of satellite cells. Indeed, during inflammatory processes, active growth-factor—heparan-sulfate complexes have been shown to be released from ECM via controlled proteolytic cleavage(Benezra et al., 1993;Whitelock et al., 1996).

In summary, our results demonstrate that histone H1 localizes to the ECM of cultured muscle cells and in regenerating skeletal muscle, where it binds specifically to perlecan. Our results also suggest that extracellular histone H1 may stimulate the proliferation of satellite myoblasts during skeletal muscle regeneration.

The authors are indebted to John R. Hassell for generously providing anti-perlecan antibody, to Hugo Olguin for his preliminary work in this investigation, and to Carlos B. Hirschberg and Nelson Osses for valuable discussion. This work was supported in part by grants from FONDAP N°13980001, FONDECYT 1990151 (to E.B.), 2980049 (to J.P.H.) and 2000113 (to J.C.C.), Fogarty International Research Collaboration Award TW 00093 (to D.J.C. and E.B.) and NIH grant NS21925 (to D.J.C.). The research of E.B. was supported in part by an International Research Scholars grant from the Howard Hughes Medical Institute, a Presidential Chair in Science from the Chilean Government and a FONDAP grant in Biomedicine. The Millennium Institute for Fundamental and Applied Biology (MIFAB) is financed in part by the Ministerio de Planificación y Cooperación (Chile).