Identification and Purification of an Intrinsic Human Muscle Myogenic Factor That Enhances Muscle Repair and Regeneration

Archives of Biochemistry and Biophysics Vol. 384, No. 2, December 15, pp. 263–268, 2000 doi:10.1006/abbi.2000.2100, available online at http://www.idealibrary.com on

Identification and Purification of an Intrinsic Human Muscle Myogenic Factor That Enhances Muscle Repair and Regeneration Ming Li,* Kaiming Chan,* ,1 Dongqing Cai,* Pingchun Leung,* Chunyiu Cheng,* Kwongman Lee,† Kenneth Kaho Lee‡ *Department of Orthopaedics and Traumatology, 5th Floor, Clinical Building, Prince of Wales Hospital, The Chinese University of Hong Kong, Shatin, NT, Hong Kong; †Lee Hysan Clinical Research Laboratory, 7th Floor, Clinical Building, Prince of Wales Hospital, The Chinese University of Hong Kong, Shatin, NT, Hong Kong; and ‡Department of Anatomy, Faculty of Medicine, The Chinese University of Hong Kong, Shatin, NT, Hong Kong

Received April 7, 2000, and in revised form August 21, 2000

The limited ability of damaged muscle to regenerate after gross injuries is a major clinical problem. To date, there is no effective therapeutic treatment for muscle injuries. In the present study, we have examined the ability of crude and fractionated human skeletal muscle extracts to promote myogenic cell proliferation and differentiation. It was found that the crude muscle extract could significantly stimulate BrdU incorporation in C2C12 myogenic cell line. In addition, the extract also promoted myogenic cell alignment and fusion. Using electrophoresis techniques, in conjunction with in vitro refolding technique, a protein with molecular weight of approximately 40 kDa was identified that could produce the same effects as the crude muscle exdtract. We also tested the ability of semipurified (30 –50 kDa) muscle extract to promote muscle repair in adult rats. Surgical intervention was used to induce muscle damage in the tibialis anterior. The semipurified muscle extract (fraction H) was injected subcutaneously over the tibialis anterior for a period of 5 days. It was found that the damaged muscle fibers were replaced by newly regenerated muscle fibers. These newly regenerated fibers originated from the fusion of differentiated satellite cells as revealed by BrdU-labeling analysis. In contrast, the injury site of muscles treated with BSA control protein contained mainly fibroblasts. © 2000 Academic Press

Key Words: myogenic factor; myogenic cell proliferation, muscle healing.

1 To whom correspondence and reprint requests should be addressed. Fax: 852-2646-3020. E-mail: [email protected].

0003-9861/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.

Muscle damage can be caused by trauma, crush injury, excessive exercise, and disuse (1, 2). Skeletal muscle possesses a limited ability to regenerate after injury (2– 4). However, when a substantial quantity of muscles is damaged the muscles are not normally replaced. In most situations, severely destroyed muscle fibers are replaced by fast growing fibrous tissue, which compromises the function of the affected muscles (5, 6). At present, there is no effective therapeutic measure that can be taken to promote skeletal muscle regeneration after injury (6 –9). Previous studies have indicated that certain hormones, growth factors, and transcription factors, such as bFGF and MyoD have the ability to enhance satellite cell division or differentiation, however, these factors only act on the satellite cells that are already active, but not on quiescent satellite cells (6 –11). Moreover, bFGF could also inhibit satellite cell differentiation (10, 11). There are indications from previous studies that quiescent satellite cells could be committed to enter cell proliferation cycle in response to unidentified growth factors released from injured myofibers (2– 4). These factors could enhance both the proliferation and differentiation of satellite cells. However, these factors have still not been clearly identified and purified in animal studies (2– 4). For human, it is not known whether human muscle possesses myogenic activities as found in animals. Muscle contains a small population of progenitor cells called satellite cells. These cells have been shown to be the only source of myogenic stem cells in adult skeletal muscle (12). Under normal condition, satellite cells are held in the quiescent phase of the cell cycle. However, they can reenter the cell proliferation cycle in 263

264

LI ET AL.

response to muscle injury, possibly under the influence of yet unidentified myogenic factors released from injured muscle fibers (13, 14). This healing process is, however, far from sufficient normally (14). It is generally assumed that there are certain endogenous myogenic factors, released following muscle fiber damage, that are involved in triggering and regulating satellite cell production. They play a major role in determining the outcome of muscle repair and regeneration after injury. However, the levels of the myogenic factors released from the damaged muscle fibers may be too low to effectively trigger off sufficient satellite cells to enter the cell cycle. Effective muscle repair and functional restoration require a suitable environment that can promote myogenic cell proliferation, differentiation and incorporation/fusion. The discovery and purification of intrinsic myogenic factors that could creat a microenvironment favorable for muscle regeneration (through manipulating and amplifying the myogenic factors) would provide an effective means in the management of muscle injuries. Therefore, in the present study, we aimed at identifying and isolating myogenic factors from human muscles that could create the microenvironment required for efficient muscle repair. MATERIAL AND METHODS Preparation of human muscle extract. Human muscle extract (HME) was prepared from the soleus muscles of four 21- to 35-yearold female volunteers, with consent and human ethic approval from the Chinese University of Hong Kong. The muscle biopsies were directly transferred into 100 ml ice-cold phosphate-buffered saline (PBS) and crushed in several places using a forceps. The crushed muscles in PBS were incubated for 2 h at 4°C with gentle stirring on a shaker. The supernatant of the sample was recovered by centrifugation at 20,000g for 1 h. The concentration of proteins in the extract was adjusted to approximately 10 mg/ml. All of the above procedures were performed at 4°C, and the final preparation was immediately stored at ⫺70°C to prevent the synthesis of de novo proteins. Identification and purification of HME fractions containing myogenic activity. The myogenic activity of crude HME was firstly confirmed using C2C12 cells. The HME was solubilized in PBS (10 mg/ml, pH 7.4) and then fractionated by isoelectric focusing electrophoresis (IEF, pH 3–9). The proteins separated by IEF were excised from the gel into 20 fractions designated as fraction A-T. Each of the excised gel slices was then put into a separate dialysis tubing and each tube was filled with 150 ␮l serum-free Dulbecco’s modified Eagle’s minimal essential medium (DMEM) with antibiotics (15, 16). Each gel slice inside the tube was mashed and dialyzed against DMEM medium for 72 h at 4°C. The dialysing medium was changed four times. After dialysis, the mashed gel was removed by centrifugation and the supernatants of all 20 fractions were recovered. The protein concentrations of all protein fractions were determined. All 20 fractions were tested for their ability to stimulate C2C12 cell proliferation. After the active myogenic fraction was identified, more of the same fraction was prepared. The prepared myogenic fraction was further purified using SDS–PAGE (17). The SDS–PAGE gels were rinsed with dideoxy H 2O and stained with Coomassie blue R-250 dissolved in 50% methanol and 50% H 2O (containing no acetic acid). The resolved protein bands in SDS–PAGE were excised and recovered by electro-elution (Bio-Rad Model 422 Electro-Eluter). The

proteins recovered were refolded using dialysis method as described below. Each of the refolded protein fractions was tested for its ability to enhance myogenic cell growth. The fractions that possess the desired biological activities were collected and used in subsequent in vivo experiments. Refolding of extracted proteins. Three part ice-cold acetone was added to one part of the fractionated HME extracts. The mixtures were allowed to precipitate for 2 h at ⫺80°C and then centrifuged at 20,000g for 30 min at 4°C. The acetone supernatant was poured off and the tubes inverted to drain. The precipitate was solubilized in 6 M guanidine HCl, 0.1 M Tris–HCl buffer, pH 8.0, 1 mM dithiothreitol, and 0.1 mM EDTA. The dissolved protein samples were poured into dialysis tubings separately and dialyzed against 0.2 M Tris–HCl buffer containing 0.2 M arginine and 0.1 M heparin, pH 9.0, for 72 h at 4°C. The buffer was changed four times. After dialysis, the samples were transferred into clean tubes and centrifuged at 20,000g for 20 min at 4°C. The supernatants were recovered and the protein concentrations of the recovered solutions were determined. The refolded samples were then dialyzed against DMEM and used for in vitro assessments. In vitro bioassay for myogenic activity in muscle extract. C2C12 myogenic cell line (18) was used to examine the biological activity of proteins extracted from human muscles. To surpress the fast proliferation rate of C2C12 cells cultured in the presence of 10% fetal bovine serum (FBS) and to exclude other overlap effects contributed by other growth factors present in FBS, the minimum requirement for maintaining C2C12 viability was determined. In the presence of serum-free DMEM, the C2C12 cell could only proliferate at relative low level and remain viable for up to 14 days. However, a parallel culture of C2C12 cells was also conducted in the presence of DMEM with 5% FBS. The C2C12 cells were cultured in 96-well culture dishes. The biological effects of (1) human muscle extract, (2) fractionated fractions of human muscle extracts, and (3) purified myogenic factors were tested. Then 0.01– 0.4 mg/ml of these different extracts were introduced into the cultures. The same concentration of bovine serum albumin was used for the controls. All cultures were maintained at 37°C and 5% CO 2 for 36 h. To determine the extent of cell proliferation, 10 ␮M of 5-bromo-2⬘-deoxyuridine (BrdU)-labeling solution (Boehringer-Mannheim) was added to each of the culture. The treated cultures were harvested 24 h later. After the labeling, the cultured C2C12 cells were fixed and the incorporation of BrdU was quantified. Briefly, the cultures were incubated with anti-BrdU-POD solution for 90 min at room temperature and then washed thoroughly in PBS. The washed cultures were then incubated with an appropriate amount of MTP substrate at room temperature until color development was sufficient for photometric detection (5–30 min). As a negative control, no BrdU antibodies were added for immunolabeling. The incorporated BrdU inside the newly synthesized DNA was quantified by measuring the absorbance at 450 nm wavelength using a microplate reader. The absorbance readings directly indicate the extent of myogenic cell proliferation. In vivo evaluation of HME on muscle healing. Since the amount of purified 40-kDa protein fraction was very limited and only sufficient for in vitro assessment, the active semipurified HME (fraction H, 30 –50 kDa) was used for in vivo experiment. The effects of fraction H on injured muscle were tested on a rat model (female, N ⫽ 6, average weight 150 g) according to the International Experimental Animal Standard (19). The rats were subjected to strain-induced muscle injury created by a transection, with surgical scissors, on the tibialis anterior. Three rats were injected beneath the epidermis of tibialis anterior with fraction H (total protein: 0.2 mg) 1 h after strain-induced injuries. Three other rats were injected with an equivalent amount of bovine serum albumin (BSA) as control. The injections were repeated daily for 5 days. For the third and fourth injections, BrdU (2 mg/100 g body wt) was injected along with fraction H or BSA. All rats were sacrificed 4 days after the last injection.

MYOGENIC FACTOR ENHANCING MUSCLE REPAIR

265

FIG. 1. (A) Showing two IEF gel slices of human muscle extract. The top gel was stained with Coommassie blue R-250. The proteins contained in the muscle extract were resolved (pH 3–9) into 20 major fractions (A–T). The bottom unstained gel slice was cut and recovered individually with reference to the stained top gel. (B) The 20 fractions (A–T) isolated from IEF gels were tested for their ability to stimulate C2C12 cell proliferation. Fraction H contained proteins that actively stimulated BrdU-incorporation in C2C12 cells.

The tibialis anterior was removed and processed for paraffin sectioning and BrdU immunohistochemical staining.

RESULTS

Bioassay-guided isolation of the myogenic factor. The mitogenic activity of myogenic factors was determined by the extent of BrdU uptake in C2C12 cell cultures. The absorbance for BrdU uptake was measured in an ELISA reader at 450 nm. The reading produced by per milligram of crude HME was set as a full unit for the convenience of calculation. The HME was firstly fractionated according to surface charge differences of native proteins found in the HME using preparative IEF (pH 3–9) electrophoresis. The HME was resolved into 20 fractions (labeled as fraction A-T) by IEF (Fig. 1A) and the proteins inside the fractions were recovered using diffusion-dialysis techniques. The purity of all recovered fractions was confirmed by SDS–PAGE. The ability of all recovered fractions to enhance C2C12 cell proliferation was determined. The results of our bioassay-guided isolation strategy showed that the myogenic activity of HME was mainly contained within fraction H (Fig. 1B), corresponding to polypeptides in the range of 30 –50 kDa as confirmed by SDS–PAGE (Fig. 2). This fraction could significantly stimulate C2C12 cell proliferation by as much as fivefold when compared with control cultures (Fig. 1B). Maximal response was observed at approximately 0.2 mg/ml of the fractionated proteins (Fig. 3). It was also

shown that this active fraction H not only caused a significant increase in proliferation rate of C2C12 cells, but also enhanced C2C12 cell fusion to form wellaligned myotubes. This feature can be used to distinguish its specific activity from that in bFGF and other growth factors since bFGF was reported to inhibit the differentiation of myoblast in culture (10, 11). Fraction H was further separated using a preparative gradient SDS–PAGE (Fig. 4A). The individually resolved protein bands were isolated from the gels and pooled accordingly so that there were sufficient quantities of proteins for refolding and analysis. The individually recovered proteins were refolded as described

FIG. 2. SDS–PAGE (Coommassie blue R-250 stained) of resolved protein bands selected from fractions A–L in the IEF. M is the molecular weight marker. Fraction H possesses the highest myogenic activity and is composed mainly of proteins with molecular weights of 30 –50 kDa.

266

LI ET AL.

DISCUSSION

FIG. 3. BrdU uptake by C2C12 cells cultured in DMEM with (OO ■ ) and without (OO } ) 5% fetal bovine serum treated with different concentrations of fraction H. 0.2 mg/ml of the extract was sufficient to elicit maximal proliferative response from C2C12 cells as indicated by a sharp increase in BrdU uptake as compared with the BSA control cultures.

in the methodology, so that the proteins could reconstitute their native structures and biological functions. Activity assay on all refolded protein fractions revealed that a protein fraction (Fig. 4B, fraction H4) with molecular weight about 40 kDa (Fig. 5) could dramatically enhance the proliferation and differentiation of C2C12 in culture (Fig. 6). However, we were unable to identify the 40-kDa protein by protein sequencing because of N-terminal blockage. Presently, we are harvesting and accumulating sufficient amount of this 40-kDa myogenic factor for creation of fresh N-termini by specific protease digestion. In vivo functional evaluation of the semipurified myogenic factor. Based on the information obtained from the C2C12 experiments, the specific activity of the semipurified 30- to 50-kDa myogenic fraction (fraction H) was further tested in a rat model. The experimental rats were strained to produce gross muscle injury. After treatment with our fraction H, the results showed that the damaged muscle fibers were replaced by newly regenerated muscle fibers (Figs. 7B and 7D). These newly regenerated fibers originated from the fusion of differentiated satellite cells as revealed by BrdU labeling analysis (Fig. 7D). The contour of the newly fused satellite cells was still recognizable in the newly formed muscle fibers (Figs. 7B and 7D), with BrdU-positively stained nuclei incorporated into the fibers in all directions (Fig. 7D). In contrast, injury site of control muscle treated with BSA contained mainly fibroblasts (Fig. 7A). For multifocal injuries, the muscle extract also stimulated satellite cell proliferation and fusion into the affected myofibers (Fig. 7C).

Muscle injury is very common in clinical practice. The limited capacity of damaged skeletal muscle fibers to regenerate after injury has always been a major problem. To date, there is no effective therapeutic treatment for improving muscle healing (6 –9). Studies on mouse and rat have indicated that muscles contain a basal level of myogenic activity, which could induce satellite cell proliferation and differentiation (2– 4, 20 –22). Although skeletal muscles possess this capacity it is insufficient to promote major muscle repair following trama. Therefore, fast growing fibroblasts normally replace the injured muscle fibers. Our animal experiments suggested that human skeletal muscles might also contain a basal level of muscle myogenic factor. In addition, these studies also infer the possibility of manipulating the myogenic factors to

FIG. 4. The resolved protein bands (H1–H6) of fraction H by SDS– PAGE (A) were recovered and refolded, so that the proteins could reconstitute their native structures and biological functions. Bioassay using these refolded protein fractions revealed that protein fraction (H4) could significantly enhance BrdU-incorporation in C2C12 cells.

MYOGENIC FACTOR ENHANCING MUSCLE REPAIR

create a micro-environment favourable for muscle healing. Before the myogenic factor present in animals could be considered for use in human clinical trials, it is of importance to explore whether there is an equivalent in human muscles. According to literature, the myogenic factor in animals has still not been specifically identified (2– 4, 10 –14), and it is still not known whether human muscle also contains a similar basal levels of intrinsic myogenic factors. In our present study, we have demonstrated that human muscles contained myogenic factors that can trigger C2C12 in vitro and quiescent satellite cells in vivo to proliferate and enhance muscle repair in adult rats. The myogenic factor is approximately 40 kDa in size which is different from bFGF (18 kDa). In addition, the ability of our myogenic factor to enhance both proliferation and differentiation of myogenic cells is different to bFGF, since bFGF inhibits the differentiation of C2C12 cells (10, 11). The growth stimulative effect of our myogenic factor was observed in vivo in strain-induced skeletal muscle injuries in adult rats. It was shown that the growthstimulating effect of the myogenic factor was not the effect of increased nutrition since the BSA injected rats could not elicit a similar effect. The increase in the number of newly synthesized nuclei present per muscle fiber after myogenic factor treatment indicated that satellite cells were the main target of this myogenic factor. Based on our findings, we propose that human muscle, like those in animals, also contains a basal level of myogenic factor and during muscle injury it is probably released to stimulate muscle healing. However, the concentration of the myogenic factor released following injuries probably too low to stimulate major muscle repair. In order to create a microenvironment favorable for muscle healing at the injury site, higher levels of the myogenic factor are required. This favorable condition could be attained by injecting exogenous myogenic factor into the injury site to compensate for the

267

FIG. 6. The H4 protein fraction was tested for its ability to stimulate C2C12 cell proliferation. Representative appearance of C2C12 cells cultured in the absence of serum for 2 (A) to 3 days (B). The presence of the protein fraction H4 in the culture medium dramatically enhanced cell proliferation and the production of myotubes (arrows), as observed after 2 (C) or 3 (D) days of culture. FIG. 7. Appearance of strain injured tibialis anterior following treatment with fraction H for 5 days. N indicates the appearance of normal muscle fibers. (A) In the BSA control, a mass of fibrous tissue (F) is normally formed at the damaged muscle site. (B) For injured muscles treated with protein fraction H, the proliferated satellite cells (S) have fused together to form myotubes (R). Fus shows satellite cells fusing to form myotubes. (C) BrdU immunolabeled section: in focal muscle injuries, fraction H also stimulates satellite cell proliferation and fusion as shown by BrdU-positively labeled nuclei at the fusion sites (FR). (D) BrdU immunolabeled section: Numerous BrdU-positive nuclei were found forming myotube (R) to replace the damaged muscle fiber (arrows).

insufficient level of this factor released during the natural healing process. The results of our in vivo experiments strongly supported this hypothesis. The identification and purification of this human muscle myogenic factor will in future provide a possible therapeutic treatment for muscle injuries in human. REFERENCES

FIG. 5. Demonstrates that fraction H4 was composed mainly of a 40-kDa protein as shown by P on the SDS–PAGE gel. M is the molecular weight marker, ME is the crude muscle extract.

1. Armstrong, R. B. (1990) Med. Sports Exerc. 22, 429 – 435. 2. Carlson, B. M., and Faulkner, J. A. (1983) Med. Sci. Sports Exer. 15, 187–198. 3. Plaghki, L. (1985) J. Physiol. Paris 80, 51–110. 4. Grounds, M. D. (1991) Pathol. Res. Pract. 187, 1–22.

268

LI ET AL.

5. Inman, V. T., Alson, H. J., and Todd, F. (1981) Human Walking. Williams and Wilkins, Baltimore, MD. 6. Nikolaou, P. K., Macdonald, B. L., Glisson, R. R., Seaber, A. V., and Garrett, W. E., Jr. (1987) Am. J. Sports Med. 21, 517–522. 7. Almekinders, L. C. (1991) Med. Sci. Sports Exer. 23(Suppl.), 5710. 8. Almekinders, L. C., and Gilbert, J. A. (1986) Am. J. Sports Med. 14, 303–308. 9. Reddy, A. S., Reedy, M. K., Best, T. M., Seaber, A. V., and Garrett, W. E., Jr. (1993) Med. Sci. Sports Exer. 25, 321–327. 10. Kardami, E., Spector, D., and Strohman, R. C. (1985) Proc. Natl. Acad. Sci. USA 82, 8044 – 8047. 11. Bischoff, R. (1990) J. Cell Biol. 111, 201–207. 12. Bischoff, R. (1994) in Myology (Engel, A. G., and Franzini-Armstrong, C., Eds.), 2nd ed., pp. 97–133, Hill, New York. 13. Mezzogiorno, A., Coletta, M., Zani, B. M., Cossu, G., and Molinaro, M. (1993) Mech. Ageing Dev. 70, 35– 44.

14. Houenou, L. J., Haverkamp, L. J., McManaman, J. L., Oppenheim, R. W. (1991) Development 2(Suppl.), 149 –155. 15. Semsarian, C., Wu, M. J., Ju, Y. K., Marciniec, T., Yeoh, T., Allen, D. G., Harvey, R. P., Graham, R. M. (1999) Nature 400, 576 –581. 16. McArdle, A., Maglara, A., Appleton, P., Watson, A. J. M., Grierson, I., Jackson, M. J. (1999) Lab. Invest. 79, 1069. 17. Scha¨gger, H., and Von Jagow, G. (1987) Anal. Biochem. 166, 368 –379. 18. Yaffe, D., and Saxel, O. (1977) Nature 270, 725–727. 19. Bedford, T. G. (1979) J. Appl. Physiol. 47, 1278 –1283. 20. Chan, K. M., Li, M., Cheng, S. H., Lee, S., and Li, J. X. (1998) Med. Sci. Sport Exer. 30, 5. 21. Bischoff, R. (1989) Med. Sci Sports Exer. 21, 164 –172. 22. Friden, J., Sjostrom, M., and Ekblom, B. (1983) Int. J. Sports Med. 4, 170 –176.