To examine whether the expression pattern of muscle-specific protein in embryonic muscle tissues at the different developmental stages depends on the types of myogenic cells, chicken breast muscle (pectoralis major) tissues of 10-day, 14-day, and 18-day old embryos (E10, E14, and E18, respectively) were grafted on chorio-allantoic membrane (CAM) of 9-day-old chicken embryos and cultured for 12 days at the longest, since the chorio-allantoic grafting is useful in pursuing muscle-cell lineage during muscle differentiation. The muscle fiber formation and expression of troponin T (TnT) isoforms in grafts were investigated by histological and immunohistochemical methods with anti-fast-muscle-type and anti-slow-muscle-type TnT (rabbit sera). In grafts of E10 breast muscle, most muscle fibers continued to develop without degeneration and TnT isoform expression of the fibers changed from the concomitant expression of fast-muscle-type (Ftype) and slow-muscle-type (S-type) to the expression of F-type only. In grafts of E14 and E18 breast muscles, the muscle fibers first degenerated with pyknotic nuclei and hyaline cytoplasm, and then new muscle fibers expressing F-type TnT isoforms were formed by the fusion of basophilic cells. The new muscle fibers in grafts of E14 muscles were different from those of E18 ones in that the former also expressed S-type TnT isoforms. In this paper, developmental stage-dependent TnT isoform expression of the embryonic breast muscles grafted on CAM is discussed in connection with cell-type difference of grafted muscle cells.
Myoblasts (embryonic and fetal ones) and satellite cells are primary sources in muscle development and regeneration. The two types of cells are distinguishable on the basis of their histological characters. Satellite cells are situated adjacent to myofibers underneath the basement membrane of the fibers (Mauro, 1961; Schultz and McCormick, 1994) and possess more condensed chromatins and higher nucleus-cytoplasm ratio than myoblasts (Carlson, 1973; Schultz, 1976; Allbrook, 1981; Bodine-Fowler, 1994). Based on these characters, satellite cells in chicken breast muscle (pectoralis major) are revealed to appear from around the 18th day after incubation (Armand and Kieny, 1984). The aims of this study are to establish chorio-allantoic grafting of skeletal muscle tissues from younger chicken embryos and to confirm the different expression pattern of muscle specific protein in embryonic breast muscle tissues grafted on the CAM. For these aims, we have tried chorio-allantoic grafting of breast muscle tissues of E10, E14, and E18 and observed the differentiation processes of the grafted muscle tissue by hematoxylin-eosin staining, and immunostaining with anti-fast-muscle-type (anti-F-type) and anti-slow-muscle-type (anti-S-type) TnT (rabbit sera), because TnT isoforms are very good markers for investigating muscle-type differentiation (Nakamura et al., 1989; Yao et al., 1992). In this experiment, E10 and E14 breast muscle tissues used as donors are supposed to contain myoblasts and early myotubes formed by fusion of the myoblasts, while E18 ones are satellite cells and developing myotubes by the myoblast fusion.
Chorio-allantoic membrane (CAM) of chicken embryos has been widely used for the culture of various organs and tissues (Willier and Rawles, 1931; Hunt, 1932; Bradley, 1970; Bonner, 1975; Newgreen et al., 1980; Andrew and Rawdon, 1987). Results of our previous study indicated that breast muscle (pectoralis major) tissues from 18-day-old chicken embryos and 1-day, 20-day, and 180-day old chickens could be cultured on the CAM of chicken embryos (Nakada et al., 1998), since muscle regeneration by satellite cells was observed to have occurred in most muscle grafts following degeneration. The chorio-allantoic grafting of skeletal muscle tissues has advantages as follows: 1) it induces muscle regeneration in the grafts; 2) effects of functional innervation to the graft are avoided during muscle regeneration; 3) contamination of the grafts with exogenous muscle cells is avoided; 4) the grafts are supplied with natural and sufficient nutrition from blood of CAM. Therefore, the chorio-allantoic grafting is considered useful in pursuing muscle-cell lineage during differentiation. However, it is not yet clear whether muscle tissues from younger chicken embryos can be cultured on the CAM by the same procedures as the previous ones.
Changes of myosin and troponin isoforms in muscle regeneration were reported to be remarkably similar to those in muscle development (Cerny and Bandman, 1987; Gorza et al., 1983; Toyota and Shimada, 1984), and regenerating myofibers induced by cold injury were considered to have reverted to an embryonic state as judged by immunological reactivities of antibodies against F-type and cardiac-muscle-type TnT isoforms (Toyota and Shimada, 1984). The expression of S-type TnT isoforms, however, were not studied in their experiments.
In our previous study, breast muscle fibers of E18 or older ones grafted on CAM first degenerated and then new muscle fibers were regenerated expressing only F-type TnT isoforms (Nakada et al., 1998). On the other hand, breast muscle fibers expressed not only F-type but also S-type TnT isoforms in normal development (Mashima et al., 1996). Therefore, the breast muscle fibers in regeneration induced by chorio-allantoic grafting were different from those in normal development in that the former did not express S-type TnT at all even at the earlier stage of regeneration. It is not clear whether the difference was caused by some effect from the CAM or by the different type of cells which formed muscle fibers during regeneration and muscle development.
In this work, results show that embryonic breast muscle tissues can be cultured on the CAM, although cystic formation has occurred with E10 and E14 muscles. The difference of troponin T isoform expression between muscle development and regeneration induced by chorio-allantoic grafting has been considered to be caused by the difference of the type of cells, that is, either myoblasts or satellite cells.
MATERIALS AND METHODS
Fertilized eggs of white leghorn (Gallus domesticus (L)) were obtained from commercial sources.
Chicken eggs incubated for 9 days were used as hosts. Breast muscle (pectoralis major) pieces, about 5 mm in diameter, excised from E10, E14, and E18 were used as donor tissues. Host egg shell was cut to open a window (1 × 1 cm) and the shell membrane was stripped to expose the host CAM on the upper side, as the egg was held in the horizontal position. Donor tissues were placed onto the position of the CAM, where blood vessels bifurcated. After the operation, the shell window was sealed in place with masking tape and the egg was incubated for 12 days at the longest.
Grafts were recovered on 1, 2, 4, 6, 8, 10, and 12 days after the operation and fixed in Bouin's solution. They were washed in 70% ethyl alcohol for one day, dehydrated, cleared, and impregnated with graded butyl alcohol solutions. Then, they were embedded in Paraplast (Oxford Labware) and sectioned serially at 7 μm. The sections were stained with Mayer's hematoxylin for 30 min and restained with alcohol eosin solution for 6 min. Some sections were used for immunostaining as well.
Serial sections of grafts were subjected to indirect immunostaining with anti-F-type TnT and anti-S-type TnT (rabbit sera) as the first antibodies. Rhodamine(TRITC)-conjugated goat anti-rabbit IgG(H+L) was used as the second antibody. The anti-F-type TnT and the anti-S-type TnT recognize chicken F-type and S-type isoforms only, respectively (Yao et al., 1992, 1994). Breast muscles of 8-day old embryos (E8), E10, E14, and E18 were used to study normal development by the same procedures as mentioned above.
Cystic formation in chorio-allantoic grafting of embryonic breast muscle tissues
In the grafts of E10 and E14 breast muscle tissues, cystic formation was observed from the 1st day and the 2nd day after grafting, respectively, and the volume of cysts increased for the culture period (Fig. 1a, b). No cysts were observed in grafts of E18 breast muscle tissue (Fig. 1c). To examine the muscle fiber formation in the cyst, hematoxylin and eosin staining was performed with sections of cystic E14 muscle grafts. Apparently normal muscle tissue was seen in the cyst by the 4th day after grafting (Fig. 1d), and then the tissue was dissociated into muscle fibers or cells by the 8th day (Fig. 1e, f). In the grafts of E10 muscle, the dissociation of muscle tissue occurred as early as the 4th day after grafting (not shown).
Morphological changes of CAM after the muscle grafting are shown schematically in Fig. 2. The CAM consists of three distinct cellular layers (Fig. 2a): outer ectodermal layer (ecto), inner endodermal layer (endo), and mesodermal layer (meso) between the two layers. The ectodermal layer contacting with a graft (oval shape) became thicker soon after grafting (Fig. 2b) and then the graft was progressively engulfed into the CAM by the outgrowth of the ectodermal layer over the graft (Fig. 2c, d). In graftings of E10 and E14 breast muscle tissues (the left side in Fig. 2), the ectodermal layer under the graft fragmented (Fig. 2e) and a cyst was formed by bulging of the endodermal layer (Fig. 2g). While in the E18 grafting (the right side in Fig. 2), the ectodermal layer contacting with the inner surface of the graft became thinner (Fig. 2d) and disappeared completely (Fig. 2f). The graft engulfed into the mesodermal layer of the CAM was stably reared in the place until the host embryo hatched out (Fig. 2h).
Histological changes of embryonic breast muscle tissues grafted on CAM
In grafts of E10 breast muscle, muscle tissues survived on CAM (Fig. 3a-c), that is, no sign of muscle degeneration, such as the appearance of pyknotic nuclei and hyaline cytoplasm, was detected in the grafted muscles (Fig. 3a) except for the injured area which underwent excision from donor embryos. Surviving muscle tissue was completely dissociated into muscle fibers or cells on the 2nd to 8th days after grafting (Fig. 3b, c).
E14 breast muscle tissues grafted on CAM degenerated with pyknotic nuclei and hyaline cytoplasm on the 1st day (Fig. 3d). Then, mononucleated basophilic cells which were stained with hematoxylin appeared among the degenerating myofibers and early myotubes were formed by the fusion of the basophilic cells on the 2nd day (Fig. 3e). The newly formed myotubes developed into myofibers with eosinophilic cytoplasm on the 8th day (Fig. 3f), although dissociation of muscle tissue, which led to cystic formation, began from the 6th day (Fig. 1e, f).
Histological changes of E18 breast muscle tissues grafted on CAM were similar to those of the E14 ones. Degeneration of the grafted E18 breast muscle was observed on the 1st day (Fig. 3g). Basophilic cells fused to form myotubes on the 2nd day (Fig. 3h). The myotubes developed into myofibers with striations on the 4th day (Fig. 3i). The myofibers were maintained stably until the host embryos hatched out, and neither cystic formation nor dissociation of muscle tissue occurred in this grafting.
Expression of troponin T isoforms in embryonic breast muscle tissues grafted on CAM
To compare the TnT isoform expression in grafts on CAM with that in normal breast muscle development, serial sections of E8, E10, E14, and E18 breast muscle tissues were firstly examined by staining them with anti-F-type TnT and anti-S-type TnT. Anti-F-type TnT stained breast muscle tissues of E8, E10, E14, and E18, while anti-S-type TnT stained only those of E10 and E14 (data not shown, summarized in Fig. 5a).
Next, we examined grafted muscle tissues by the same immunohistochemical methods. E10 breast muscle tissues grafted on CAM expressed both F- and S-type TnT isoforms on the 2nd day after grafting (Fig. 4a, b). On the 8th day, dissociated muscle fibers were stained only with anti-F-type TnT (Fig. 4c), and no fibers stained with anti-S-type TnT (Fig. 4d). Thus, changes of TnT isoform expression from both Fand S-types to F-type only were observed in E10 breast muscle tissues grafted on CAM (Fig. 5b).
In grafted E14 breast muscle tissues, newly formed myotubes were stained on the 2nd day with anti-F-type TnT (Fig. 4e), but not with anti-S-type TnT (Fig. 4f). The myotubes developed into myofibers with striations, which were stained with both anti-F-type (Fig. 4g) and anti-S-type TnT on the 8th day (Fig. 4h). Thus, the TnT isoforms expressed in the newly formed muscle fibers changed from F-type only to both Fand S-types (Fig. 5c).
In grafts of E18 breast muscle tissues, newly formed myotubes and myofibers were stained with anti-F-type TnT (Fig. 4i, k), but not with anti-S-type TnT (Fig. 4j, l). No myotubes and myofibers stained with anti-S-type TnT were observed at any stages of grafting. The stainability was maintained until the 12th day after grafting. Thus, changes of TnT isoform expression in E18 breast muscle tissues grafted on CAM differed from those in E14 ones in that the S-type TnT isoforms were not expressed (Fig. 5c, d), although the histological changes of grafted E18 breast muscle tissues (Fig. 3g-i) were similar to those of grafted E14 ones (Fig. 3d-f).
Chorio-allantoic grafting is one of tissue culture methods to show the ability of self-differentiation of grafts. Some authors reported failure of muscle development and its maintenance in the chorio-allantoic grafting, since they found fatty degeneration in most grafts of the muscle tissue of embryonic limb (Hunt, 1932; Bradly, 1970). The present results indicated that embryonic breast muscle tissues could be cultured on the CAM, although cystic formation occurred in grafts of E10 and E14 breast muscle tissues.
In histochemical changes, the muscle degeneration did not occur in E10 breast muscle fibers grafted on CAM except for the injured area which underwent the excision from the donor embryo (Fig. 3a, b). On the other hand, E14 and E18 breast muscle fibers degenerated soon after grafting and new muscle fibers were regenerated (Fig. 3d, e, g, h). In the former case, most muscle fibers were not probably injured by the excision, since muscle fibers in the E10 breast muscle tissue were shorter than those in E14 and E18. In the latter case, all muscle fibers had been injured by the excision, because of the long fiber length, and the muscle degeneration occurred uniformly in the E14 and E18 grafts.
Muscle dissociation was observed in E10 and E14 grafts which formed a cyst. As a cyst grew in volume, the muscle dissociation proceeded immediately. In grafts of E10 and E14 breast muscle tissues, fragmented ectoderm pieces derived from CAM contacting with the graft remained in the cyst (Fig. 2e, g). On the other hand, the ectodermal layer in grafts of E18 ones disappeared completely soon after grafting and no cystic formation occurred (Fig. 2d, f). Thus, the fragmentation and later remaining of CAM's ectodermal layer were phenomena specific to cystic formation. Furthermore, cysts were not observed when E14 breast muscle tissues were grafted in preliminary experiments on a small area of CAM which had been treated by 0.5% trypsin in Locke's solution to digest its ectodermal layer (not shown). Taking these data together, the ectodermal layer of CAM was supposed to participate somehow in the cystic formation.
Changes of TnT isoform expression in breast muscle regeneration induced by chorio-allantoic grafting were previously reported to be different from those in normal development of the original breast muscle, that is, regenerating muscle cells on the CAM expressed only F-type TnT isoforms (Nakada et al., 1998), while developing ones in the original position expressed not only F-type but also S-type TnT isoforms (Mashima et al., 1996). There are two possibilities for understanding the difference of TnT isoform expression between breast muscle regeneration induced by chorio-allantoic grafting and normal development of the original breast muscle. One possibility is that functional innervation is essential for reversion of myogenic cells to the embryonic state, so that only F-type isoforms were expressed in the graft on CAM, in which no functional innervation was expected. If this is the case, embryonic breast muscle tissues grafted on CAM should express only F-type TnT isoforms. The other possibility is that two types of cells which are different in TnT isoform expression are participating in muscle development and regeneration on CAM, that is, myoblasts in the former and satellite cells in the latter. In this case, not only F-type but also S-type TnT isoforms should be expressed in some grafted muscle tissues which contained myoblasts. Actually, both F-type and S-type TnT isoforms were expressed in E10 and E14 grafts, although only F-type TnT isoforms were expressed in E18 grafts in this experiment and older grafts in the previous one (Nakada et al., 1998).
According to the papers so far published, satellite cells in chicken breast muscle (pectoralis major) begin to predominate on around Embryonic Day 18 and represent virtually all mononucleated myogenic cells present at hatching time (Armand and Kieny, 1984; Hartley et al., 1992). Newly formed muscle fibers in chorio-allantoic grafting of breast muscle tissues from E18 and older are considered to be formed by proliferation and following fusion of satellite cells, and the fibers in the grafts from E10 and E14 are by those of myoblasts. Since the types of cells are different it is quite reasonable that different patterns of TnT isoform expression were observed.
Our explanation for the results in this experiment is as follows. In E10 breast muscle grafts, myotubes formed by fusion of myoblasts survived in operation of culture, so that the expression of F-type and S-type TnT isoforms continued, and then ceased the expression of the latter at later stages as in normal development of original breast muscle (Fig. 5b). In E14 grafts, most myotubes might be injured and die by operation, and pyknotic nuclei might be observed in the tissue (Fig. 3d). Since there were no satellite cells in the E14 breast muscle, it is the myoblasts which survived, divided, fused, and differentiated into muscle fibers, and expressed only F-type TnT isoforms first and both F-type and S-type ones later, following the TnT isoform expression pattern of early normal development of original breast muscle (Fig. 5c). These results from E10 and E14 breast muscle grafts suggested that younger embryonic breast muscle tissues grafted could express TnT isoforms in the same manner as that in normal development of the original muscle, that is, the start and stopping of S-type TnT isoform expression seemed to be fixed in breast muscle tissues of younger embryos. In E18 grafts, only satellite cells can survive and express only F-type TnT isoforms, since they were determined so during the embryonic stages. Therefore, we considered that satellite cells in chicken breast muscle were different from myoblasts in that they did not express S-type TnT isoforms. This explanation is well compatible with the report by Feldman and Stockdale (1991) that satellite cells cloned from chicken breast muscle formed muscle fibers which expressed exclusively fast-muscletype myosin heavy chain.
Our conclusions are as follows. 1) Embryonic breast muscle tissues can be cultured on CAM, although cystic formation following dissociation of muscle tissue is observed in grafts from younger embryonic muscle tissues. 2) E10 and E14 breast muscle tissues grafted on CAM were different from E18 one in that S-type TnT isoforms were expressed in the former, but not in the latter, suggesting that myoblasts in the former and satellite cells in the latter are participating. 3) The S-type TnT isoform expression in chicken breast muscle as observed during embryonic myogenesis is probably a specific character of embryonic breast muscle since the start and stopping of S-type TnT isoform expression were found to have occurred in breast muscle grafts from E14 and E10, respectively. 4) Chorio-allantoic grafting is useful technique for studying muscle differentiation and regeneration.