Developmental fate of a distinct class of chick myoblasts after transplantation of cloned cells into quail embryos

In order to study changes in the differentiated states of individual cells or populations of cells during development, one must be able to identify particular cells in some initial state and follow them to a point where the cells or their progeny can be identified as part of a particular tissue or as having attained a differentiated state different from the original state. If a developmental progression of differentiated states is to be fully understood, the sequence of changes that a cell undergoes before it reaches its final differentiated state must be determined. One method of approaching this problem is to remove undifferentiated tissue from embryos and allow it to grow and differentiate in culture as an explant or as dissociated cells. Unfortunately, the exact cellular composition of the original tissue is often not known, so that conclusions are complicated by ignorance of the cell types which give rise to differentiated cells. Because of this difficulty in identifying cell populations during the development of heterogeneous tissue, the ultimate cell source of differentiated tissue usually cannot be determined.

Another method of ascertaining a cell’s fate is transplantation of embryonic tissues between two closely related species which carry stable, definitive histological or biochemical differences in their cells (Le Douarin, 1969, 1973 a, b). This technique is superior to the usual tissue culture methods for identification of the tissue of origin of particular differentiated cell types and has been applied to a number of different problems concerning the developmental origins and fates of cells. But, like tissue culture, tissue transplantation suffers the disadvantage of an unknown degree of cell heterogeneity in the grafted tissue.

This report describes a method in which tissue culture and transplantation are combined and which circumvents the problem of cell heterogeneity. Muscle clones grown in culture from single chick cells are harvested, aggregated, and transplanted into quail embryos. After a period of in vivo growth, the heterospecific, transplant-containing quail tissue is examined in vitro by clonal analysis and the cloned progeny of the original chick cells identified by Feulgen staining.

Mesenchymal cells derived from the limbs of chick embryos younger than Hamburger & Hamilton stage 20 can be grown in mass culture and will occasionally exhibit muscle differentiation (fusion) (Dienstman, Biehl, Holtzer & Holtzer, 1974) but single cells from these ages do not grow as clones. The first clonable cells appear in the chick leg at stage 20 and the first cells giving rise to differentiated muscle clones are found at stage 21 (Bonner & Hauschka, 1974). Muscle-colony-forming cells become an increasingly greater proportion of the total clonable cell population in the chick leg during the next few days of embryonic development until, by day 10 (stage 35), approximately 85 % of the clonable cells differentiate as muscle in conditioned medium (White, Bonner, Nelson & Hauschka, 1975). The muscle-colony-forming cell population is composed of a number of distinct myoblast classes. Chick myoblast classes are distinguished on the bases of clone morphology (Bonner & Hauschka, 1974), culture medium requirements for differentiation (White et al. 1975), and differentiative interactions with the nervous system (Bonner, 1978, 1980). While eventually we hope to examine the developmental fates of all chick myoblast types, the present work is concerned solely with the fate of that particular myoblast type (CMR-I) which is the first to appear during development and which gives rise to morphologically distinctive clones.

White leghorn or White-Rock chicken and Japanese quail eggs were incubated at 38 ± 1 °C. Embryos were staged according to the criteria of Hamburger & Hamilton (1951). All operations were performed using sterile technique.

Techniques used for the production of single-cell suspensions from embryonic skeletal muscle and the conditions of clonal culture have been described previously (Bonner & Hauschka, 1974; White et al. 1975). Briefly, embryonic chick hind limb buds (stages 26–28; 5– 6 days of egg incubation) are removed, the tissue is minced with fine forceps and incubated at 37 °C in the presence of 0·05 % crude collagenase (Worthington CLS) for 10 min. The enzymic reaction is stopped by adding an equal volume of cold, serum-containing fresh medium (FM). Filtration through Nitex (20 μm pore size) removes the remaining tissue pieces and cell aggregates. The cells are pelleted, resuspended in FM and, after dilution, are placed in gelatin-coated 60 mm tissue-culture plates (Falcon 3002) containing approximately 2·5 ml of conditioned medium. The plates are incubated for 13– 14 days at 37 °C in a water-saturated atmosphere of 95 % air and 5 % CO₂.

Fresh medium is 79 % Ham’s F-10 nutrient solution, 1 % penicillinstreptomycin (Stock solution contains 10000 units/ml penicillin G and 0·5mg/ml of streptomycin sulfate), 15% preselected horse serum, and 5% day-12 chick embryo extract (Konigsberg, 1968). Conditioned medium is fresh medium that has been exposed, in 20 ml aliquots, for 24 h to confluent secondary cultures of chick leg muscle cells grown in 100 mm petri dishes (Falcon 3003) (White & Hauschka, 1971; Hauschka, 1972).

At the end of the culture period, living cultures were examined by phase microscopy and the locations of selected CMR-I muscle clones were marked on the culture plate bottom surface. Individual clones were isolated with the aid of ceramic cylinders (Penicylinder, Fisher). The cylinders were ringed with sterile silicon grease at one end and placed over the clone, silicon side down, to effect a tight seal. Collagenase was added to the cylinder for 10 min at 37 °C to dissociate clone cells from the plate. The collagenase and cells were removed, individual clones of the same type pooled when necessary, and mixed with an equal volume of FM. The cells were washed with FM and pelleted in FM-containing microfuge tubes. The microfuge tube containing the pellet and 1 ml FM was left in the incubator overnight, at 37 °C under 95 % air and 5 % CO₂, to allow the cells to adhere and form a tight aggregate.

Quail eggs at 3 days of incubation (equivalent to Hamburger & Hamilton stages 19–21) were prepared for transplantation according to the procedure of Zwilling (1959). Eggs were prepared on the day before transplantation by drilling a small hole in the pointed end of the egg, removing approximately 0·5 ml of albumin with a sterile syringe, and cutting a window in the top of the egg. Cellophane tape covered the window until the eggs were used for transplantation.

Cell aggregates were removed from microfuge tubes and placed in cold, sterile, Puck’s Saline G solution, containing a small amount of Nile Blue, for 3– 5 min. This lightly stains the outer layer of cells and enables the aggregates to be seen during transplantation. After staining, the aggregates were transferred to a dish containing cold Saline G and cut into smaller pieces for transplantation.

Quail embryos were prepared for transplantation by cutting a slit through the membranes covering the right leg bud. A piece of the clone aggregate was then placed in the tip of a drawn-out Pasteur pipette, the pipette tip inserted into the prospective thigh region of the leg bud, and the aggregate gently blown into the leg bud as the pipette was slowly withdrawn.

Four days after transplantation the quail right and left legs were separately prepared for clonal analysis. The untreated left leg served as a control for the chick-cell-containing right leg. No developmental abnormalities due to the transplantation procedure of the quail legs were seen. The experimental right leg of each treated embryo was of similar size and developmental stage as the control left leg. After two weeks of clonal culture the dishes were fixed with 3:1 ethanol : glacial acetic acid for 5 min and rinsed with tap water before staining.

Transplant-containing and control quail embryo legs were removed at various times between 7 and 10 days of incubation (4–7 days post-transplantation), washed in Puck’s Saline G and fixed in a 3:1 (v/v.) mixture of 100% ethanol and glacial acetic acid. Fixed tissues were embedded in Paraplast (Fisher), sectioned at 6 μm and stained with either hematoxylin and eosin, or the Feulgen procedure with a fast green counterstain.

Stained culture dishes were flooded with distilled water to retard fading and individual clones were examined to determine whether they contained quail or chick cells. The locations of all chick clones were marked and the dishes counterstained with Harris’ hematoxylin. The hematoxylin-stained quail and chick clones were then scored as muscle (myotubes present) or as non-fusing (no myotubes); all chick muscle clones were then scored as CMR-I or CMR-II, CMR-III (see Results for definitions of chick muscle clone types). Using these methods one can determine the proportion of all chick clones which are differentiated, and the types of chick muscle clones recovered from the transplanted quail legs.

Sections which had been stained by the Feulgen procedure with a fast green counterstain were examined at 400 × magnification. Quail nuclei characteristically exhibit one or more darkly Feulgen-positive spots while chick nuclei generally stain relatively homogeneously with a few small, lightly Feulgen-positive spots. With experience, the two nucleus types can be distinguished without much difficulty. For quantitative scoring of nuclei a grid reticule was inserted in the microscope ocular and all nuclei appearing within the grid limits were noted as either quail or chick. At 400× magnification, 200 to 400 nuclei were contained within the grid. In some cases, only those nuclei seen to be in myotubes were scored. Since it is difficult in many instances to determine by light microscopy whether or not nuclei are in myotubes, only the best examples of myotube nuclei were scored. Here, both sides of the myotube were visible and the nuclei were clearly in the center of and in the same focal plane as the myotube. In sections cut from older tissue (10- to 12-day-old host quail embryos) cross-striations often aided identification of myotubes. Since the myotubes were generally sectioned at various oblique angles the number of nuclei seen per myotube was small, usually from 2 to 8.

The difference in proportion of nuclei scored as chick between transplantcontaining and pure quail tissue is fairly small. To assure ourselves that bias in scoring nuclei did not affect the data, three persons unfamiliar with the project scored representative fields of transplant-containing and control quail sections. The ratios of percentage chick nuclei in transplant regions to percentage chick nuclei in non-transplant or control regions in these blind scorings were nearly identical to those determined by us.

Developing chick embryo leg skeletal muscle contains a number of different clonable myoblast classes. Some of these, termed CMR, require the presence of conditioned medium for in vitro differentiation. The first class of CMR clonable myoblasts to appear during development is the ‘early embryo’ (Bonner & Hauschka, 1974), or CMR-I class. CMR-I is the only type of clonable myoblast derived from chick legs until about stage 25 ; after stage 25, the proportion of CMR-I declines and two other conditioned-mediumrequiring myoblast types appear - CMR-II and CMR-III. CMR-II and CMR-III clones are morphologically identical to each other but different from CMR-I (White et al. 1975). CMR-I clones are characterized by the presence of short, stubby myotubes containing relatively few (10–20) nuclei, and the myotubes are arranged in an unpatterned, irregular manner. CMR-II and CMR-III muscle clones exhibit myotubes which contain up to hundreds of nuclei each. These myotubes are much longer than those of CMR-I clones and are arranged in swirling patterns.

The stability of the CMR-I clonal morphology was tested by subcloning. Cells derived from young chick embryos were grown for two weeks in primary clonal culture and CMR-I clones identified by phase microscopy. The primary CMR-I clones were isolated with a ceramic cylinder and the cells removed from the plate surface with collagenase. Single cells were then added to fresh petri plates containing conditioned medium for another 2-week period of clonal growth and differentiation. Of the 453 clones recovered after subcloning, 91 % were differentiated, and virtually all the muscle clones were identified as CMR-I. These results show that clonal morphology is a stable indicator of this muscle clone type and that CMR-I myoblasts in culture do not spontaneously change to the more advanced CMR types.

To test the viability of cloned CMR-I cells which had been formed into multicell aggregates two techniques were employed. First, aggregates formed overnight were redissociated to single cells, and the ability of the cells to exclude the vital dye Nigrosin was tested. In three experiments, 70·1 ±1·1 (S.E.M.) % of the cells remained viable and excluded the dye. Second, chick CMR-I cell aggregates were transplanted into quail embryo leg buds for only 12 h before dissociation of the transplant-containing quail legs and inoculation of the cells into culture dishes. During such a short period of transplantation most cells are not expected to divide, since collagenase-dissociated cells generally remain dormant for 6– 12 h after treatment (personal observations on cultured cells). Any cells capable of division immediately after transplantation should do so only once in view of the 10– 12 h generation time of similar cells (Janners & Searls, 1970; Buckley & Konigsberg, 1974; Zalin, 1979). When placed into culture, the chick cells should be essentially the same ones which were transplanted. The cultured cells were fixed at 12 and 24 h post-inoculation to, again, avoid significant expansion of either the chick or quail cell populations by mitotic proliferation. Some dishes were stained by the Feulgen procedure to distinguish chick and quail cells while others were stained with hematoxylin to determine the total number of cells on the dishes and thus the plating efficiency. Culture for 12 or 24 h gave identical results : 38 ± 1·0 (S.E.M.) % of the cells added to the dishes adhered to the plate surface, spread out, and were stained by hematoxylin; of these attached cells, 26·5 + 0·5 (S.E.M.) % were chick. Aggregates of chick CMR-I cells thus contain significant numbers of viable cells, as is also shown by the clonal analysis data below.

Quail legs containing chick CMR-I myoblasts were subjected to clonal analysis four days after transplantation (Table 1 and Fig. 1). Of the 65 chick clones recovered in six experiments, 91 % differentiated as muscle. Of these chick muscle clones, 97 % (57 of 59) were identified as CMR-I. These results indicate that those cells which survive the transplantation and cloning procedures remain CMR-L No difference in percentage muscle colony differentiation was noted between quail clones grown from transplant-containing or contralateral, transplant-free, quail leg cells.

Since clonal analysis did not indicate a significant transition of CMR-I cells to either CMR-II or CMR-III, a likely alternative fate of CMR-I could be to fuse with other chick or quail muscle cells to form myotubes. To investigate these possibilities, transplant-containing quail legs were sectioned and examined, after Feulgen staining, to determine where chick nuclei are found. Sections of control quail tissue exhibit significant proportions of nuclei which do not show the darkly stained nuclear spots characteristic of quail, and, consequently, appear indistinguishable from chick nuclei (Table 2). Therefore, simple observation proved inadequate to locate tissues composed of more-than- background proportions of chick nuclei, and quantitation of chick and quail nuclei proportions in microscope fields was necessary.

Cell aggregates were placed in the prospective dorsal thigh region. Therefore, transplanted and control legs were sectioned longitudinally and the following tissues examined: dorsal thigh muscle, ventral thigh muscle, and calf muscle. The percentages of chick and quail nuclei found in 400 × microscope fields were determined and, in addition, nuclei which were clearly contained in myotubes were scored as chick or quail. Figure 2 contains photographs of representative fields.

From the data in Table 2 it can be seen that the proportion of chick nuclei in microscope fields of transplant dorsal thigh muscle is significantly greater than the proportion seen in any other region, greater even than the ventral thigh muscle, suggesting that the chick cells remain near the site of transplantation and do not migrate away to any great extent. The proportion of chick nuclei in frank myotubes of the dorsal thigh (36·7%) is considerably higher than it is when all nuclei of a field are considered (19·8 %). The presence of chick nuclei in myotubes shows that the CMR-I cells were capable of fusion, and the relatively high proportion of them in myotubes suggests further that fusion is their fate.

Three different fates for CMR-I myoblasts are possible : they may die, they may undergo differentiation and reappear as another cell type, or they may fuse to form myotubes. Cell death as a fate cannot be ruled out by the present work but is unlikely in view of the observation that clonable chick myoblasts are recovered from transplant-containing quail legs. If programmed cell death were their normal fate then the four days spent in vivo after transplantation should be sufficient to allow their death, since the same time span in the chick ends with the nearly complete disappeaiance of clonable CMR-I (White et al. 1975). It should also be noted that the transplanted cells were cultured for two weeks before and for two weeks after transplantation.

The available evidence also suggests that transplanted CMR-I cells do not reappear as a different cell type. Certainly they do not undergo transition to the more advanced clonable myoblast types, CMR-II and CMR-III, as these clone types do not appear at significant levels after transplantation. CMR-I cells also do not reappear as clonable non-muscle cells at a significant frequency since 91 % of the chick clones derived from transplant-containing quail legs were differentiated muscle clones. The possibility that the CMR-I cells reappear as a different cell type which is not clonable cannot be ruled out since such cells would not be taken into account by clonal analysis.

To further investigate the fate of CMR-I cells, Feulgen-stained sections of transplant-containing quail legs were examined. When nuclei within microscope fields were scored as chick or quail the region which contained the highest ratio of chick to quail nuclei was found to be the dorsal thigh muscle, the region into which the transplant was originally placed. Since most of the chick nuclei were not found to be inside obvious myotubes the identity of the mononucleated cells containing the nuclei remains in some doubt. They could be myoblasts or connective tissue cells, especially fibroblasts. These two cell types cannot be distinguished at the light microscope level. Other cell types found in the thigh which might have had chick nuclei but are distinguishable by histological criteria, such as blood vessel endothelium and lipocytes, were ruled out by direct observation while chondrocytes, perichondrial fibroblasts and epidermal epithelial cells were ruled out by quantitative scoring. It is most likely that such chick cells seen in sections are indeed myoblasts because clonal analysis of similar tissue demonstrated that nearly all of the chick clones grown from transplant-containing quail leg cells differentiated as muscle and were therefore derived from myoblasts. The possibility remains, however, that such putative chick fibroblasts may not be clonable.

In order to be more certain that particular nuclei were muscle, only those nuclei which were obviously contained in frank myotubes were scored. Again, the only region which contained a ratio of chick to quail nuclei significantly greater than quail controls was the dorsal thigh muscle. From these observations we conclude that while mononucleated cells containing chick nuclei are not necessarily all myoblasts, at least some of the transplanted chick CMR-I myoblasts retain their myogenic properties and enter myotubes by fusion. Clonal analysis has shown these myoblasts to be CMR-I and, further, that the more advanced myoblast types, CMR-II and CMR-III, do not appear. It is suggested, then, that the developmental fate of CMR-I myoblasts is to form myotubes. The data also suggest that CMR-I myoblasts do not serve as the precursor to CMR-II or CMR-III.

This clone transplantation system should be applicable to determination of the fate of any cell type which will recognizably differentiate in clonal culture and which will retain its ability to differentiate after at least one subclonal passage. This system is advantageous primarily because the donor cells are derived from a single proliferating cell and are phenotypically identical.

This research was supported by a grant (HD-10307) to P.H.B. from the National Institute of Child Health and Human Development.