A previous study revealed that segments of bowel grafted between the neural tube and somites of a younger chick host embryo would induce a unilateral increase in cellularity of the host’s neural tube. The current experiments were done to test the hypotheses that muscle tissue in the wall of the gut is responsible for this growth-promoting effect and that the spinal cord enlargement is the result of a mitogenic action on the neuroepithelium. Fragments of skeletal (E8–15) or cardiac muscle (E4–14) were removed from quail embryos and grafted between the neural tube and somites of chick host embryos (E2). Both skeletal and cardiac muscle grafts mimicked the effect of bowel and induced an increase in cell number as well as a unilateral enlargement of the region of the host’s neural tube immediately adjacent to the grafts. The growth-promoting effect of muscle-containing grafts was restricted to the neural tube itself and was not seen in proximate dorsal root or sympathetic ganglia. The action of the grafts of muscle was neither species-nor class-specific, since enlargement of the neural tube was observed following implantation of fetal mouse skeletal muscle into quail hosts. Grafts of skeletal muscle or gut increased the number of cells taking up [3H]thymidine in the host’s neuroepithelium as early as 9h following implantation of a graft. The increase in the number of cells entering the S phase of the cell cycle preceded the increase in cell number. These observations demonstrate that muscle-containing tissues can increase the rate of proliferation of neuroepithelial cells when these tissues are experimentally placed together.

Tissue interactions play a critical role in development. It has been possible to gain insight into the nature of these interactions in the case of neural crest derivatives by experimentally changing the spatial relations of developing organs through the technique of back-transplantation into the neural crest migration pathways of a younger host embryo (Le Douarin et al. 1978; Erickson et al. 1980; Le Lièvre et al. 1980; Schweizer et al. 1983). In their novel environment in the host embryo many of the transplanted cells recover properties of their ancestors and display characteristic migratory and proliferative capacities. In addition, effects of the graft on surrounding host tissues are also revealed (Rothman et al. 1987).

In previous back-transplantation experiments, done to permit the developing gut of avian embryos to interact with the developing neural tube (Rothman et al. 1987), segments of bowel were placed between the neural tube and somites of 2 day (E2) host embryos. These experiments revealed that musculoconnective tissues of the gut induced a unilateral enlargement of the spinal cord or brain that was restricted in location to the site of the grafts. In contrast to the neural tube, neither sympathetic nor spinal ganglia were affected by enteric transplants. The effect of the bowel on the neuroepithelium lacked species- and even class-specificity; however, the growth-promoting action of the gut could not be mimicked by sham operations or by grafts of developing ciliary ganglion, lung, pancreas, mesonephros or rudiment of the eye. The effect of the gut on the neural tube thus showed organ specificity. The type of cell within the bowel wall that produced the responsible factor(s) was not identified, although gut devoid of endoderm or neural crest cells showed the mitogenic activity. It was proposed that the musculoconnective tissue of the bowel produces a short-range diffusible factor that induces mitosis in the neuroepithelial cells of the neural tube.

To characterize further the source of the growth-promoting activity residing in the gut wall, we tested the ability of types of muscle, other than that included in the bowel, to promote the growth of the neuroepithelium. We report here that both skeletal and cardiac muscles strongly promote the growth of the neural tube. Experiments were also done with [3H]thymidine (3H-TdR) to determine whether the increase in the volume of the neural tube that was observed in proximity to grafts of bowel or muscle was the result of a unilateral increase in the mitotic activity of neuroepithelial cells. A mitogenic effect of muscle-containing grafts was demonstrated. This effect was found to be exerted across species and class boundaries.

Fertilized chick (white Leghorn; staged according to Ham-burger and Hamilton [H-H], 1951) and quail eggs (Coturnix coturnix japónica; staged according to Zacchei [Z], 1961) were incubated at 37 °C in a forced air incubator. Eggs were obtained from commercial sources. Mice (CD-I) were obtained from Charles River and bred at Columbia. The time of the appearance of a vaginal plug was designated as day 0 of gestation (EO).

Grafts were made as previously described (Rothman et al. 1987). Essentially, the dorsomedial portions of one to three somites were removed from recipient chick embryos in order to create a space in which to insert grafted tissue and to prevent the development of a vertebra between the graft and the neural tube. Segments of presumptive duodenum (E4 and E8; stage 18 and 25 [Z]), pectoral, epaxial, or cardiac muscle, kidney, or liver (E4–15; stage 18–32 [Z]), from quail embryos were back-transplanted between the neural tube and somites of chick embryos at different axial levels (stages 9–15 [H-H]; n=69 for skeletal muscle, n=42 for gut; Fig. 1). The host embryos were then returned to the incubator until E3–10. Grafts of murine bowel (presumptive duodenum; removed at E9; n=12) or skeletal muscle (triceps, or epaxial muscle; removed at E14–15; n=26) were transplanted into quail embryos (stages 7–12; [Z]). The nuclei of mouse and quail cells can easily be distinguished from one another in Feulgen-stained sections.

Fig. 1.

An illustration showing the procedure used for grafting segments of muscle into chick or quail host embryos. Although a small fragment of the dorsomedial part of the somites (s) was removed, the neural tube (nt) and neural crest (nc) were not disturbed. Grafts (g) of quail or mouse skeletal (pectoral or epaxial), cardiac muscle, or bowel were then inserted between the neural tube and somites.

Fig. 1.

An illustration showing the procedure used for grafting segments of muscle into chick or quail host embryos. Although a small fragment of the dorsomedial part of the somites (s) was removed, the neural tube (nt) and neural crest (nc) were not disturbed. Grafts (g) of quail or mouse skeletal (pectoral or epaxial), cardiac muscle, or bowel were then inserted between the neural tube and somites.

Host embryos were killed by decapitation, fixed with Camoy’s or Zenker’s solution and embedded in wax. Sections (7–10 μm) were mounted on glass slides and stained with the Feulgen procedure (Le Douarin, 1973). When cell proliferation was to be investigated, [3H]TdR (10 μCi; [specific activity, 20–80Ci mmol−1, New England Nuclear], diluted 1:1 with Tyrode’s solution; GIBCO), ‘was pipetted directly onto the surface of host embryos immediately following the placement of a graft. These embryos were then returned to the incubator and maintained for an additional 24 h. Alternatively, the isotope was administered to some embryos 4–24 h following surgery. The eggs were then sealed with Scotch™ brand cellophane tape and returned to the incubator for an additional 2h. Host embryos were fixed with Camoy’s solution and processed as described above. Sections were stained with the Feulgen procedure, dehydrated and then dried with ether. The slides were coated with Ilford L4 photographic emulsion and maintained at 4°C in a light-proof box for 24–72 h. They were then developed with Kodak D19 developer. Sections of embryos that did not receive [3H]TdR were stained and processed for radioautography to rule out the effects of chemography.

The number of [3H]TdR-labeled cells and the total number of cells were counted on the operated and control sides of each section of the neural tube in which the graft was visible. In addition, the areas of each section of cord were measured. Both the counts and the measurements of cross-sectional area utilized computer-assisted morphometry (Leitz-Bioquant II; R&M Biometrics, Nashville, TN). Measurements were compared in each section between the operated and nonoperated sides of the spinal cord; thus the nonoperated side served as a paired control for each measurement of the effect of the graft.

Back-transplantation of quail embryonic skeletal muscle

Myotubes had already formed in the tissue removed for grafting at the developmental age of the donor embryos from which pectoral or epaxial skeletal muscles were obtained. The grafts survived well in the hosts and were identified in 69 host embryos. Two days after implantation, the grafts had the appearance of a dense, highly cellular mass of fibromuscular tissue (Fig. 2). By 5 days, however, the back-transplanted muscle became reorganized and displayed the typical morphology of skeletal muscle. In all of the grafts, examined 2–8 days following implantation, in which a close apposition between the grafted muscle and the neural tube was maintained, a unilateral enlargement of the spinal cord was observed on the side that faced the graft (Figs 2, 3). In .contrast, in those instances in which a somite developed between the grafted muscle and the spinal cord (n=13), no spinal cord enlargement was seen (Fig. 4). Moreover, when the grafts of skeletal muscle were placed next to the dorsal aorta instead of into the space between the somite and neural tube, enlargement of the spinal cord again failed to occur.

Fig. 2.

A graft of 8 day quail pectoral muscle was placed between somites 10–12 and the neural tube of a 12-somite chick host embryo. The host embryo was fixed at E4.5. Sections were Feulgen stained. (A) A lateral enlargement of the spinal cord is seen on the side of the graft (g). Note the distorted position of the spinal cord in relationship to the graft (×80). (B) The quail tissue is illustrated at higher magnification. Melanocytes are present in this particular graft of quail muscle (×330).

Fig. 2.

A graft of 8 day quail pectoral muscle was placed between somites 10–12 and the neural tube of a 12-somite chick host embryo. The host embryo was fixed at E4.5. Sections were Feulgen stained. (A) A lateral enlargement of the spinal cord is seen on the side of the graft (g). Note the distorted position of the spinal cord in relationship to the graft (×80). (B) The quail tissue is illustrated at higher magnification. Melanocytes are present in this particular graft of quail muscle (×330).

Fig. 3.

A graft (g) of 9-day quail pectoral muscle was inserted between neural tube and somites 10–12 of a 12-somite chick host embryo. The host was fixed at E5. Sections were Feulgen stained. (A) There is a unilateral enlargement of the dorsal horn of the spinal cord (×80). (B) A section close to the one in 3(A) is seen at higher magnification. The graft (g) containing quail nuclei appears as a dense mass of muscular tissue (×330).

Fig. 3.

A graft (g) of 9-day quail pectoral muscle was inserted between neural tube and somites 10–12 of a 12-somite chick host embryo. The host was fixed at E5. Sections were Feulgen stained. (A) There is a unilateral enlargement of the dorsal horn of the spinal cord (×80). (B) A section close to the one in 3(A) is seen at higher magnification. The graft (g) containing quail nuclei appears as a dense mass of muscular tissue (×330).

Fig. 4.

A graft (g) of pectoral muscle from a 9-day quail embryo was intended to be implanted between the neural tube and the two last somites of a 12-somite chick embryo. The host was fixed at E4. Sections were Feulgen stained. In fact, somitic tissue (s) intervened between the graft and the neural tube. Enlargement of the spinal cord failed to occur (×220).

Fig. 4.

A graft (g) of pectoral muscle from a 9-day quail embryo was intended to be implanted between the neural tube and the two last somites of a 12-somite chick embryo. The host was fixed at E4. Sections were Feulgen stained. In fact, somitic tissue (s) intervened between the graft and the neural tube. Enlargement of the spinal cord failed to occur (×220).

The pattern of enlargement of the spinal cord in response to implants of skeletal muscle was very similar to that previously found for grafts of gut (Rothman et al. 1987); however, there were some differences. As was also true of bowel, both gray and white matter increased in area on the operated side following grafts of skeletal muscle (the mean enlargement of the experimental over the contralateral side was 12,363 ±2,272μm2 per section; P<0. 001; n=58 hemi-sections from 5 embryos). As was previously found for the response of the neural tube to grafts of gut, the unilateral expansion in area following back-transplantation of skeletal muscle was due to an increased number of cells on the operated side (the mean increase in cell number on the operated side over that of the contralateral side was 212±42 cells per section; P<0. 001). The effect of the grafts on the neural tube was generally limited in scope to the region in which donor muscle cells were found; however, hypertrophy of the spinal cord occasionally was seen one to two segments rostral to the appearance of the graft. As was true after back-transplantation of gut, no effects of skeletal muscle grafts were noted on the side of the neural tube contralateral to the back-transplanted tissue (in comparison with segments just rostral or caudal to the grafts). Moreover, no change was observed in the size of dorsal root or sympathetic ganglia adjacent to the backgrafts. In contrast to enteric back-transplants, which often fused with the neural tube (Rothman et al. 1987), fusion did not occur between transplants of skeletal muscle and the adjacent spinal cord, although nerve fibers, occasionally extended from the ventral roots into the graft of muscle. Moreover, although in two cases quail cells from a muscle implant were found inside the meninges, no muscle-derived quail cells ever entered the gray matter of the spinal cord in the host embryos. As noted previously (Rothman el al. 1987) sham operations, consisting of removing the dorsomedial regions of somites 9–14, did not increase the size of the developing neural tube on the side of the operations (in two embryos, the ratios of the experimental to the control side were 0. 99±0. 01 [n=70 hemisections] and 0. 95±0. 01 [n=82 hemisections]).

In order to study the timing of the reaction of the neural tube to backgrafts of skeletal muscle, embryos were examined at relatively short survival times after surgery. 24 h following a back-transplant of skeletal muscle, a unilateral increase in the number of neuroepithelial cells and in the thickness of the neuroepithelium was observed on the side of the neural tube that faced the graft. As was previously noted, 24h after back-transplantation of bowel, both the number of cells and the thickness of the neuroepithelium were significantly increased when compared with the contralateral side.

Back-transplantation of quail embryonic cardiac muscle

In order to determine whether the neuroepithelium reacts similarly to cardiac as well as to skeletal muscle and smooth muscle (present in back-transplants of gut), experiments were done in which segments of quail cardiac muscle (E4–14) were back-transplanted into chick host embryos. A unilateral enlargement of the spinal cord was again obtained in response to cardiac muscle. The pattern of enlargement resembled that found after grafts of skeletal muscle or gut and was due to an increase in the number of cells in the neural tube on the side of the graft. The mean increase in cell number on the operated side (24 h after grafting) over that of the contralateral side was 92±12 cells per section; P<0. 002. Again, both gray and white matter were enlarged. The growth-promoting activity of skeletal muscle and the wall of the bowel on the neuroepithelium thus appear to be shared by cardiac muscle.

Back-transplantation of quail embryonic tissues that contain little muscle

In contrast to gut wall, skeletal muscle primordia and heart, all of which contain abundant muscle as well as connective tissue, back-transplants of kidney (mesonephros) and liver failed to cause spinal cord enlargement. Kidney and liver contain connective tissue but no muscle, except for the smooth muscle found in the walls of blood vessels. Previous experiments demonstrated that a growth-promoting effect on the neuroepithelium is also absent from other organ rudiments that contain little muscle, including ciliary ganglion, lung, pancreas, or rudiment of the eye (Rothman et al. 1987). It thus appears that back-transplants of embryonic tissue that are rich in muscle (smooth, skeletal or cardiac) promote the growth of the neuroepithelium, while those that are poor in muscle do not.

Mitogenic effect of gut wall and skeletal muscle on the neuroepithelium

In order to determine whether the growth-promoting action of back-transplanted muscle on the neuroepithelium is to increase the rate of cell proliferation, host embryos were incubated with [3H]TdR either immediately after placement of the grafts, or 2 h prior to killing the embryos. Quail or mouse gut and skeletal muscle were used to provoke the spinal cord reaction. When [3H]TdR was applied immediately after implantation of a graft of quail skeletal muscle virtually all cells of the chick host neuroepithelium on both sides were radioautographically labeled 24 h later; however, since there were more neuroepithelial cells on the operated than on the nonoperated side, there were also more [3H]TdR-labeled cells on the side of the neural tube that faced the graft (the ratio of the number of [3H]TdR labeled cells on the operated side to that of the contralateral side was 2. 4+0. 5; P<0. 001). Since [3H]TdR remains in the closed system of the egg for an extended period of time, these experiments indicate that each of the cells of the neuroepithelium passed through or entered the S phase of the cell cycle at some time during the 24 h following grafting and that more cells did so on the experimental than the control side. Because all of the neuroepithelial cells were labeled, however, these experiments neither established that the grafts exerted a mitogenic action on the neuroepithelium, nor, if there was a mitogenic action, when it occurred. [3H]TdR was therefore applied at 4h, 7h, 15h, and 22 h following surgery and the embryos were killed 2 h later. This modification of the timing of the administration of [3H]TdR permitted the number of cells entering the S phase of the cell cycle to be estimated at each of these times. For each section in which the graft was visible, the total number of cells and the number of cells radioautographically labeled by [3H]TdR were determined ipsilateral and contralateral to the graft. As in previous experiments (Rothman et al. 1987), the contralateral side served as a paired control to which the operated side was compared. No significant increase in cell number was observed between experimental and control sides or in the absolute number or proportion of cells radioautographically labeled by [3H]TdR in the neural tubes of chick embryos killed 6h following a graft of quail skeletal muscle (Fig. 5; Table 1). Actually, at this early time, a small but significant decrease was observed in the number of cells labeled by [3H]TdR on the operated side, as compared to the contralateral side of the spinal cord (Table 1). By 9h, however, both the number of [3H]TdR-labeled cells and their proportion of the total cell population on the operated side of the spinal cord were significantly increased over those of the control (Table 1) although the total number of cells on the operated side had not yet increased. The number and proportion of cells labeled by [3H]TdR continued to be greater on the operated than the contralateral side at 17 h (Fig. 6; Table 1) and 24 h (Fig. 7; Table 1). By 17 h and 24 h following insertion of grafts, significant increases in total cell numbers were found on the experimental side as well (Table 1). At 24 h the increase in number of cells was greater than the increase in [3H]TdR-labeled cells. As a result, at 24h, the proportion of cells on the operated side that were labeled by [3H]TdR, though still significantly elevated, was less than it had been earlier. Similar results were obtained for grafts of quail gut (E4 or E8) or skeletal muscle (E8) 24 h after implantation (Table 2). These observations confirm that back-transplants do, in fact, increase the rate of proliferation of the neuroepithelial cells of the neural tube. They also indicate that the effects of the grafts become manifest by between 6 and 9h following surgery. As would be expected from a stimulus to cell proliferation the effect on [3H]TdR uptake precedes a detectable increase in cell number.

Table 1.

Effects of back-transplants of quail skeletal muscle or gut on the neural tube of host embryos

Effects of back-transplants of quail skeletal muscle or gut on the neural tube of host embryos
Effects of back-transplants of quail skeletal muscle or gut on the neural tube of host embryos
Table 2.

Comparison of the effects of back-transplants of quail skeletal muscle and gut and murine skeletal muscle on the neural tube of chick host embryos

Comparison of the effects of back-transplants of quail skeletal muscle and gut and murine skeletal muscle on the neural tube of chick host embryos
Comparison of the effects of back-transplants of quail skeletal muscle and gut and murine skeletal muscle on the neural tube of chick host embryos
Fig. 5.

6h following insertion of a graft of pectoral muscle. A graft (g) of pectoral muscle from an E8 quail embryo was placed between the neural tube and somites 13–14 of a 14-somite chick host embryo. 4h later [3H]TdR was applied to the surface of the embryo. The embryo was allowed to survive for an additional 2h. Sections were Feulgen stained. There is no increase in the total number of cells, or in the number of cells that have become radioautographically labeled, in the neuroepithelium on the side adjacent to the graft. Note that the cells of the graft (exhibiting the quail nucleus) are virtually unlabeled (×470).

Fig. 5.

6h following insertion of a graft of pectoral muscle. A graft (g) of pectoral muscle from an E8 quail embryo was placed between the neural tube and somites 13–14 of a 14-somite chick host embryo. 4h later [3H]TdR was applied to the surface of the embryo. The embryo was allowed to survive for an additional 2h. Sections were Feulgen stained. There is no increase in the total number of cells, or in the number of cells that have become radioautographically labeled, in the neuroepithelium on the side adjacent to the graft. Note that the cells of the graft (exhibiting the quail nucleus) are virtually unlabeled (×470).

Fig. 6.

17 h following insertion of a graft of bowel. A graft (g) of duodenum from a E4 quail embryo was placed between the neural tube and the last somite of a 16-somite chick host embryo. Fifteen h later [3H]TdR was applied to the surface of the embryo. The embryo was allowed to survive for an additional 2h. Sections were Feulgen stained. There is a significant increase both in the total number of cells and in the number of cells that have become radioautographically labeled in the neuroepithelium on the side adjacent to the graft (×560).

Fig. 6.

17 h following insertion of a graft of bowel. A graft (g) of duodenum from a E4 quail embryo was placed between the neural tube and the last somite of a 16-somite chick host embryo. Fifteen h later [3H]TdR was applied to the surface of the embryo. The embryo was allowed to survive for an additional 2h. Sections were Feulgen stained. There is a significant increase both in the total number of cells and in the number of cells that have become radioautographically labeled in the neuroepithelium on the side adjacent to the graft (×560).

Fig. 7.

24 h following insertion of a graft of skeletal muscle. A graft (g) of pectoral muscle from an E8 quail embryo was placed between the neural tube and somites 12–14 of a 15-somite chick host embryo. 22 h later [3H]TdR was applied to the surface of the embryo. The embryo was allowed to survive for an additional 2h. Sections were Feulgen stained. There is a significant increase both in the total number of cells and in the number of cells that have become radioautographically labeled in the neuroepithelium on the side adjacent to the graft (×420).

Fig. 7.

24 h following insertion of a graft of skeletal muscle. A graft (g) of pectoral muscle from an E8 quail embryo was placed between the neural tube and somites 12–14 of a 15-somite chick host embryo. 22 h later [3H]TdR was applied to the surface of the embryo. The embryo was allowed to survive for an additional 2h. Sections were Feulgen stained. There is a significant increase both in the total number of cells and in the number of cells that have become radioautographically labeled in the neuroepithelium on the side adjacent to the graft (×420).

The effect of back-transplants of mouse skeletal muscle on the neuroepithelium, previously observed for implants of murine bowel (Rothman et al. 1987), was analyzed. Grafts of pectoral or epaxial muscle from mouse embryos (E14–15) were placed between the somites and neural tube of quail host embryos which were then incubated, with [3H]TdR, 22 h after surgery and killed 2h later. In these experiments, murine and quail cells could easily be distinguished from one another with the Feulgen stain. Back-transplanted segments of murine skeletal muscle, like those of quail, induced a unilateral enlargement of the host’s spinal cord (Fig. 8; Table 2). The growth-promoting effect of muscle on the neuroepithelium thus appears to be present in the muscle of both birds and mammals.

Fig. 8.

25 h following insertion of a graft of skeletal muscle. (A) A graft (g) of triceps muscle from an E14 mouse embryo was placed between the neural tube and somites 11–13 of a 14-somite quail host embryo. 23 h later 3H-TdR was applied to the surface of the embryo. The embryo was allowed to survive for an additional 2h. Sections were Feulgen stained. There is a significant increase both in the total number of cells and in the number of cells that have become radioautographically labeled in the neuroepithelium on the side adjacent to the graft (×300). (B) Cells in the graft of murine skeletal muscle (framed in A; arrows) are illustrated at higher magnification (×1450).

Fig. 8.

25 h following insertion of a graft of skeletal muscle. (A) A graft (g) of triceps muscle from an E14 mouse embryo was placed between the neural tube and somites 11–13 of a 14-somite quail host embryo. 23 h later 3H-TdR was applied to the surface of the embryo. The embryo was allowed to survive for an additional 2h. Sections were Feulgen stained. There is a significant increase both in the total number of cells and in the number of cells that have become radioautographically labeled in the neuroepithelium on the side adjacent to the graft (×300). (B) Cells in the graft of murine skeletal muscle (framed in A; arrows) are illustrated at higher magnification (×1450).

We have previously demonstrated that a fragment of avian (quail or chick) or mammalian (mouse) bowel wall, when back-transplanted between the somites and neural tube of a chick or quail host embryo (E2), induce a unilateral increase in the number of cells in the region of the neural tube adjacent to the graft (Rothman et al. 1987). The effect occurred when either early or late embryonic or fetal tissues were back-transplanted. Since the hypertrophy of the neural tube was still observed when enteric grafts were used in which both endoderm and neural crest cells were absent, the growth-promoting action of the bowel must have been produced by the musculoconnective tissue in the wall of the gut. The effects of the enteric backgrafts were limited to the neural tube. No increase in the growth of the neural crest-derived dorsal root and sympathetic ganglia adjacent to the grafts was seen. The purpose of the present study was, first, to determine whether the growth-promoting activity exhibited by musculoconnective tissue in the bowel wall was shared by other types of muscle and, second, to ascertain whether the increase in cellularity of the neural tube, induced by enteric grafts, is due to an increase in the proliferation of cells of the neuroepithelium rather than to a decline in cell death.

In order to investigate the effect on the spinal cord of other types of muscle (the gut contains smooth muscle), skeletal and cardiac muscle from quail embryos or mouse fetuses were back-transplanted, between the neural tube and somites of host embryos, in the same manner as was previously used for grafting segments of donor gut. Both skeletal and cardiac muscle exhibited growth-promoting activity on the neuroepithelium of the host embryos. The activity of each of these muscles was similar in magnitude to that previously found for gut. Again, as was seen when segments of bowel were backgrafted, no enlargement of the adjacent dorsal root or sympathetic ganglia was observed after back-trans-plantation of skeletal or cardiac muscle. The effects of the grafts of muscle-containing tissue thus do not affect these crest derivatives; however, whether or not the crest-derived precursors of enteric neurons or glia respond to muscle has not yet been ascertained. When segments of muscle were placed near the neural tube, only the hemisection of neural tube facing the graft became enlarged. The opposite side was thus used as a control. Comparison of the grafted with the contralateral side permitted each embryo to act as its own control, thereby increasing the statistical power of comparisons. Moreover, in confirmation of previous observations (Rothman et al. 1987), the growth-promoting action of skeletal muscle, cardiac muscle, and gut was not mimicked by backgrafts of liver, lung, kidney or peripheral ganglia (Le Douarin et al. 1978; Erickson et al. 1980; Le Lièvre et al. 1980; Ayer-Le Lièvre and Le Douarin, 1982; Schweizer et al. 1983; Rothman et al. 1987 ; Fontaine-Pérus et al. 1988). It is concluded that skeletal and cardiac muscle share the growth-promoting activity of the smooth-muscle-containing bowel. Although grafts of muscle, as well as bowel, contain elements of connective tissue, the failure to induce neural tube enlargement by organs that contain connective tissue but little muscle strongly suggests that the growth-promoting effect resides in muscle itself.

Experiments done with [3H]TdR indicated that the increase in number of cells adjacent to a graft of muscle-containing tissue follows an increase in the number and proportion of neuroepithelial cells entering the S-phase of the cell cycle. Thus, the proportion of neuroepithelial cells labeled by [3H]TdR was increased as early as 7–9 h following placement of a graft, at which time the number of cells was the same on the operated and contralateral sides of the host’s neural tube. An increase in cell number was not observed until 15-17 h after grafting. The rate of mitotic activity of neuroepithelial cells remained elevated for at least as long as 24 h after exposure of the neural tube to muscle, although the proportion of cells entering S was lower at 24 h than at 17 h. These observations indicate that the primary effect of grafts of muscle-containing tissue is to enhance the rate of proliferation of neuroepithelial cells. The unilateral increase in the cellularity of the neural tube follows from this effect. Although we have not yet established when the proliferative effect no longer occurs, a limit in the capacity of the neuroepithelium to respond to the growth-promoting muscle factor may exist since preliminary experiments in which grafts of muscle were implanted at E5–6 did not affect the size of the spinal cord.

It is important to emphasize the fact that in these experiments the mitogenic activity of back-transplanted muscle-containing tissues is exerted at a time when the neurons (Oppenheim et al. 1989) and non-neuronal cells of the spinal cord are being generated. Production of neuroblasts in the ventral spinal cord of chicks has been estimated to stop at about E5.5 (Hollyday and Hamburger, 1977), although 95 % of these neurons are bom on or before the end of E4 (Hamburger, 1948; Corliss and Robertson, 1963). Glial cell proliferation continues at a significant rate much later in life. More neurons are generated than ultimately survive and the final number is established after many neurons are eliminated by natural cell death. The critical period when surviving motor neurons become target-dependent has been determined to begin at E5.5-6 (Ham-burger, 1977; Slack et al. 1983; Landmesser and Morris, 1975; Tosney and Landmesser, 1985). It has been estimated that approximately 40 % of the motor neurons die between E6 and E10 (Hamburger, 1975; Oppenheim et al. 1978; Oppenheim and Majors-Willard, 1978; Laing, 1982). This timing suggests that an interference with natural cell death probably is not responsible for the effects of these grafts. Experiments are now in progress to determine whether the number of cells on the experimental side of the spinal cord remains elevated after the period of normal cell death.

When embryos, in which bowel (Rothman et al. 1987) or muscle grafts, implanted at E2, were examined at later developmental stages, the appearance of the enlarged spinal cord on the operated side was extremely pronounced. The phenotype(s) expressed by cells in the affected segments of spinal cord at these later developmental stages remains to be established. Also remaining to be established is the relevance, with respect to normal development, of the ability of muscle to exert a mitogenic effect on the neuroepithelium. Conceivably, a factor arising from the myotomal part of the somitic mesoderm might affect the rate of cell division in spinal cord anlagen. This possibility has not yet been critically tested. Finally, whether or not the as yet unidentified substance(s) in muscle-containing grafts that is(are) responsible for the mitogenic effect on the neuroepithelium is identical or similar to the muscle-derived factor(s) that have been found to mediate cell survival and neurite outgrowth by spinal cord neurons in vitro (Henderson et al. 1981; Dohrmann et al. 1986) also remains to be investigated.

The authors would like to thank Mmes Martine Bontoux and Ginette Brochard, Ms Edith Abreu and Ms Greta Katzauer for excellent technical assistance and Mr Jean-Valéry Coumans for assisting in the morphometric analysis of the data. The work, done at the Faculté des Sciences de Nantes was supported by the Centre National de la Recherche Scientifique and by the Association Française contre les myopathies, at the Institut d’Embryologie Cellulaire et Molé-culaire by the Centre National de la Recherche Scientifique, the Institut National de la Santé et de la Recherche Médicale, the Fondation pour la Recherche Médicale and the Ligue Nationale Française contre le Cancer. At Columbia University the work was supported by NIH grants HD 20470, HD 21032, NS 15547, and NSF INT 84-13816.

Ayer-Le Lievre
,
C. C.
and
Le Douarin
,
N. M.
(
1982
).
The early development of cranial sensory ganglia and the potentialities of their component cells studied in quail-chick chimaeras
.
Devi Biol.
94
,
291
310
.
Corliss
,
C. E.
and
Robertson
,
G. G.
(
1963
).
The pattern of mitotic density in the early chick neural epithelium
.
J. exp. Zool
.
153
,
125
140
.
Dohrmann
,
U.
,
Edgar
,
D.
,
Sendtner
,
M.
and
Thoenen
,
H.
(
1986
).
Muscle-derived factors that support survival and promote fiber outgrowth from embryonic chick spinal motor neurons in culture
.
Devi Biol
.
118
,
209
221
.
Erickson
,
C. A.
,
Tosney
,
K. W.
and
Weston
,
J. A.
(
1980
).
Analysis of migratory behavior of neural crest and fibroblastic cells in embryonic tissues
.
Devi Biol
.
77
,
142
156
.
Fontaine-Pérus
,
J.
,
Chanconie
,
M.
and
Le Douarin
,
N. M.
(
1988
).
Developmental potentialities in the nonneuronal population of quail sensory ganglia
.
Devi Biol
.
128
,
359
375
.
Hamburger
,
V.
(
1948
).
The mitotic patterns in the spinal cord of the chick embryo and their relation to histogenetic processes
.
J. comp. Neurol
.
88
,
221
284
.
Hamburger
,
V.
(
1975
).
Cell death in the development of the lateral motor column of the chick embryo
.
J. comp. Neurol
.
160
,
535
546
.
Hamburger
,
V.
(
1977
).
The developmental history of the motor neuron
.
Neurosci. Res. Program Bull
.
15
,
1
37
.
Hamburger
,
V.
and
Hamilton
,
H. L.
(
1951
).
A series of normal stages in the development of the chick embryo
.
J. Embryol. exp. Morph
.
88
,
49
92
.
Henderson
,
C. F.
,
Huchet
,
M.
and
Changeux
,
J. P.
(
1981
).
Neurite outgrowth from embryonic chicken spinal neurons is promoted by media conditioned by muscle cells
.
Proc. natn. Acad. Set. U.S.A
.
78
,
2625
2629
.
Hollyday
,
M.
and
Hamburger
,
V.
(
1977
).
An autoradiographic study of the formation of the lateral motor column in the chick embryo
.
Brain Res
.
132
,
197
208
.
Laing
,
N. G.
(
1982
).
Timing of motoneuron death in the brachial and lumbar regions of the chick embryo
.
Dev. Brain Res
.
5
,
181
186
.
Landmesser
,
L.
and
Morris
,
D. G.
(
1975
).
The development of functional innervation in the hind limb of the chick embryo
.
J. Physiol
.
249
,
301
326
.
Le Douarin
,
N. M.
(
1973
).
A Feulgen-positive nucleolus
.
Expl Cell Res
.
77
,
459
468
.
Le Douarin
,
N. M.
,
Teillet
,
M. A.
,
Ziller
,
C.
and
Smith
,
J.
(
1978
).
Adrenergic differentiation of cells of the cholinergic ciliary and Remak ganglia in avian embryo following in vivo transplantation
.
Proc. natn. Acad. Sci. U.S.A
.
75
,
2030
2034
.
Le Lievre
,
C. S.
,
Schweizer
,
G. G.
,
Ziller
,
C. M.
and
Le Douarin
,
N. M.
(
1980
).
Restrictions of developmental capabilities in neural crest cell derivatives as tested by in vivo transplantations experiments
.
Devi Biol
.
77
,
362
378
.
Oppenheim
,
R. W.
,
Chu-Wang
,
L-W.
and
Maderdrut
,
J. L.
(
1978
).
Cell death of moto neurons in the chick embryo spinal cord. III. The differentiation of moto neurons prior to their induced degeneration following limb-bud removal
.
J. comp. Neurol
.
177
,
87
112
.
Oppenheim
,
R. W.
,
Cole
,
T.
and
Prevette
,
D.
(
1989
).
Early regional variations in motoneuron numbers arise by differential proliferation in the chick embryo spinal cord
.
Devi Biol
.
133
,
468
474
.
Oppenheim
,
R. W.
and
Maiors-Willard
,
C.
(
1978
).
Neuronal death in the brachial spinal cord of the chick is unrelated to the loss of polyneuronal innervation in wing muscle
.
Brain Res
.
154
,
148
152
.
Rothman
,
T. P.
,
Gershon
,
M. D.
,
Fontaine-Pérus
,
J. C.
,
Chanconie
,
M.
and
Le Douarin
,
N. M.
(
1987
).
The effect of back-transplants of the embryonic gut wall on growth of the neural tube
.
Devi Biol
.
124
,
331
346
.
Schweizer
,
G.
,
Ayer-Le Lievre
,
C.
and
Le Douarin
,
N. M.
(
1983
).
Restrictions of developmental capacities in the dorsal root ganglia during the course of development
.
Cell Diff
.
13
,
191
200
.
Slack
,
J. R.
,
Hopkins
,
W. G.
and
Pockett
,
S.
(
1983
).
Evidence for a motor nerve growth factor
.
Muscle and Nerve
6
,
243
252
.
Tosney
,
R. W.
and
Landmesser
,
L. T.
(
1985
).
Development of the major pathways for neurite outgrowth in the chick hindlimb
.
Devi Biol
.
109
,
193
214
.
Zacchei
,
A. M.
(
1961
).
Lo sviluppo embrionale della quaglia giapponese (Cotumix cotumix japónica T.e.S.) Arch
.
Ital. Anat. Embnol
.
66
,
36
62
.