We have investigated dorsal root ganglion formation, in the avian embryo, as a function of the composition of the paraxial somitic mesoderm. Three or four contiguous young somites were unilaterally removed from chick embryos and replaced by multiple cranial or caudal halfsomites from quail embryos. Migration of neural crest cells and formation of DRG were subsequently visualized both by the HNK-1 antibody and the Feulgen nuclear stain.

At advanced migratory stages (as defined by Teillet et al. Devi Biol. 120, 329 – 347 1987), neural crest cells apposed to the dorsolateral faces of the neural tube were distributed in a continuous, nonsegmented pattern that was indistinguishable on unoperated sides and on sides into which either half of the somites had been grafted. In contrast, ventrolaterally, neural crest cells were distributed segmentally close to the neural tube and within the cranial part of each normal sclerotome, whereas they displayed a nonsegmental distribution when the graft involved multiple cranial half-somites or were virtually absent when multiple caudal half-somites had been implanted.

In spite of the identical dorsal distribution of neural crest cells in all embryos, profound differences in the size and segmentation of DRG were observed during gangliogenesis (E4 – 9) according to the type of graft that had been performed. Thus when the implant consisted of compound cranial half-somites, giant, coalesced ganglia developed, encompassing the entire length of the graft. On the other hand, very small, dorsally located ganglia with irregular segmentation were seen at the level corresponding to the graft of multiple caudal halfsomites. We conclude that normal morphogenesis of dorsal root ganglia depends upon the craniocaudal integrity of the somites.

     
  • DRG

    dorsal root ganglia

  •  
  • E

    embryonic day

  •  
  • PNS

    peripheral nervous system

The segmental pattern of organization of the peripheral nervous system (PNS) arises as a consequence of interactions taking place between neural crest, neural tube and somite cells.

Neural crest cells are the progenitors of dorsal root ganglia (DRG), sympathetic ganglia and Schwann cells (for review see Weston, 1970; Le Douarin, 1982). These cells migrate through defined pathways that are imposed by the metameric organization of the somites (Weston, 1963; Tosney, 1978) and by their dissociation into dermomyotome and sclerotome (Teillet et al. 1987). It is already well established that migration of Schwann cell precursors and outgrowth of motoneuron fibers are restricted to the cranial half of each sclerotome (Keynes & Stern, 1984, 1985; Rickmann et al.1985; Bronner-Fraser, 1986; Newgreen et al. 1986; Loring & Erickson, 1987; Teillet et al. 1987). Moreover, DRG form exclusively opposite the cranial half of each somite and those neural crest cells that originate from the level of the caudal half become redistributed along the craniocaudal axis to localize opposite cranial somitic segments where they participate in the formation of two consecutive ganglia (Teillet et al. 1987). If intrinsic craniocaudal differences within the somitic mesoderm are the basis for segmental arrangement of the various PNS structures, then changing experimentally the craniocaudal order of the somites should impair normal PNS segmentation. In order to put this idea to the test, Stern & Keynes (1987) constructed multiple cranial and multiple caudal compound-somites and studied neurite outgrowth and neural crest cell migration at sclerotomal levels. They found that the zones occupied by spinal nerves and crest cells were much larger in the case of sclerotomes formed from multiple cranial half-somites.

In the present study, we have investigated whether the development of dorsally located structures like the DRG depends upon craniocaudal differences within the somites to the same extent as it was reported for motoneuron fibers and Schwann cells. By implanting multiple cranial or caudal somite halves from quail donors into chick hosts, we demonstrate that the early phases of migration of neural crest cells along the dorsolateral aspect of the neural tube are not affected by changing the craniocaudal composition of the somites. On the other hand, the final distribution of DRG progenitor cells and the morphogenesis of the ganglia are strongly modified by the composition of the paraxial mesoderm.

Embryos of chick (Gallus gallus) and Japanese quail (Coturnix coturnix japónica) were used for this study. Eggs of commercial sources were kept in a humidified incubator at 38 ± 1°C.

Embryo microsurgery

Quail embryos, aged from 12 to 31 somites, were excised from the egg and pinned out on black Sylgard dishes. A drop of 20 % pancreatin (Gibco) in Tyrode buffer (pH 7 · 4) containing 35 μ g ml-1 gentamycin sulfate was locally applied to facilitate dissection. The last two to four segmented somites on both sides of the neural tube were cut into cranial and caudal halves with the aid of sharpened steel microknives. Like somitic halves were pooled in Tyrode buffer until grafting.

Chick embryos were at the 10-to 20-somite stage at the time of implantation. The vitelline membrane was removed at the caudal end of the embryos and a drop of 20 % pancreatin in Tyrode buffer was locally applied. The last three or four segmented somites were unilaterally removed. Complete excision of the mesoderm was achieved by the mild enzymatic treatment which was harmless to the embryos. As illustrated in Fig. 1, the space left after host somite removal was filled with multiple cranial or caudal half-somites from quail donors. Since quail embryos were in most cases older than chicks, the space left by four consecutive host somites was occupied by four equivalent donor halves. Only those embryos showing complete replacement of chick mesoderm by quail tissue (as assessed in each graft by Feulgen staining) were taken for this study.

Fig. 1.

Schematic representation of the operation procedure. Cranial (A) or caudal (B) somitic halves were excised from the last three or four formed somites of quail donors aged from 12 to 31 somites and grafted into the equivalent area of 10-to 20-somite-stage chick embryos. Since donors were older than hosts, the size of donor somitic halves usually corresponded to the size of the entire host somites.

Fig. 1.

Schematic representation of the operation procedure. Cranial (A) or caudal (B) somitic halves were excised from the last three or four formed somites of quail donors aged from 12 to 31 somites and grafted into the equivalent area of 10-to 20-somite-stage chick embryos. Since donors were older than hosts, the size of donor somitic halves usually corresponded to the size of the entire host somites.

29 embryos were implanted with multiple cranial somitic halves; among them 15 embryos were fixed at the 27-to 30-somite stage, corresponding to advanced migratory stage as defined by Teillet et al. (1987), and 14 embryos were fixed at 4 to 9 days of incubation (E4 – 9). 25 embryos were implanted with multiple caudal half-somites. Eleven were fixed at advanced migratory stage and 14 at organogenetic stages (E4 – 9).

Histology and immunocytochemistry

Embryos processed directly for visualization of cell nuclei were fixed in Zenker’s solution for 2—4 h at room temperature, dehydrated in alcohol and embedded in paraffin wax. Serial transverse or frontal sections, 5 to 7 gm thick, were stained according to the Feulgen-Rossenbeck technique to identify quail and chick cells (Le Douarin, 1973). Embryos taken for analysis of neural crest migration were fixed in Bouin’s fluid, and embedded in paraffin wax. To visualize neural crest cells and their derivatives, sections were first stained with the HNK-1 monoclonal antibody (Abo & Balch, 1981; Lipinski et al. 1983; Tucker et al. 1984) essentially as previously described (Kalcheim & Le Douarin, 1986; Teillet et al. 1987). The same sections were then postfixed overnight in Zenker’s fluid and stained according to the Feulgen-Rossenbeck procedure to assess the correspondence between the localization of migrating cells and the grafted somitic tissue.

Analysis of ganglion size

The size of DRG on control and operated sides was measured from serial transverse sections of embryos whose ganglia were first projected on paper with a camera lucida. The surface of projected ganglia was measured using the image analysis system Bioquant IV, and graphically represented as a function of section thickness (each section was 5 μ m thick).

Migration of neural crest cells from the side adjacent to multiple cranial or caudal half-somites

The distribution of chick neural crest cells facing multiple quail half-somites was visualized by staining sections with the HNK-1 antibody and followed by processing of sections for the quail marker to assess the quality of the grafts. Embryos aged from 27 to 30 somites were sectioned either in frontal or transverse planes for evaluation and quantitative estimation of neural crest cell distribution.

In frontal sections (9 cases observed), migration of neural crest cells in close proximity to the dorsal face of the neural tube was longitudinally continuous and nonsegmented after either type of operation as well as on the respective control sides (Figs 2A, 3A). On the other hand, in ventrolateral areas between neural tube and sclerotome and within the sclerotome itself, migrating neural crest cells were distributed differently as a function of the type of grafted mesoderm: on the intact sides, neural crest cells were distributed in a segmental pattern corresponding with each cranial somitic half (Figs 2B, 3B); in embryos grafted with multiple cranial half-somites, neural crest cells appeared continuously distributed along the entire grafted sclerotome, showing no apparent segmental arrangement (Fig. 2B); in embryos grafted with multiple caudal half-somites, the sclerotome area was almost completely devoid of neural crest cells (Fig. 3B). Figs 2 and 3 (panels B and C) illustrate that the continuity or absence of HNK-1 immunoreactive neural crest cells in the sclerotomal region correspond to the areas of the grafts.

Fig. 2.

Migration of neural crest cells opposite a mesoderm composed of multiple cranial somitic halves. Frontal sections through a 28-somite-stage chick embryo grafted at the 15-somite stage over a length corresponding to somites 12 to 15 (four consecutive somites) with four cranial somitic halves derived from a 19-somite-stage quail donor. (A) Section through the dorsal part of the neural tube showing HNK-l-immunoreactive neural crest cells in both control (upper) and operated sides distributed in a continuous pattern laterally to the neural tube. (B) HNK-1-immunolabeling of a more ventral section of the same embryo showing, on the control side, a segmental distribution of neural crest cells opposite each cranial somitic half (rostral is to the left); on the operated side, HNK-l-positive neural crest cells migrate through the grafted sclerotome in a nonsegemented, continuous manner. (C) Feulgen-Rossenbeck staining at the level of section B to show the grafted area. Q, quail graft. Bar, 50 pm.

Fig. 2.

Migration of neural crest cells opposite a mesoderm composed of multiple cranial somitic halves. Frontal sections through a 28-somite-stage chick embryo grafted at the 15-somite stage over a length corresponding to somites 12 to 15 (four consecutive somites) with four cranial somitic halves derived from a 19-somite-stage quail donor. (A) Section through the dorsal part of the neural tube showing HNK-l-immunoreactive neural crest cells in both control (upper) and operated sides distributed in a continuous pattern laterally to the neural tube. (B) HNK-1-immunolabeling of a more ventral section of the same embryo showing, on the control side, a segmental distribution of neural crest cells opposite each cranial somitic half (rostral is to the left); on the operated side, HNK-l-positive neural crest cells migrate through the grafted sclerotome in a nonsegemented, continuous manner. (C) Feulgen-Rossenbeck staining at the level of section B to show the grafted area. Q, quail graft. Bar, 50 pm.

Fig. 3.

Migration of neural crest cells opposite a mesoderm composed of multiple caudal somitic halves. Frontal sections through a 30-somite-stage chick embryo, implanted at the 18-somite stage over a length corresponding to somites 15 to 18 with four caudal somitic halves derived from a 20-somite-stage quail donor. (A) HNK-l-immunoreactive neural crest cells located dorsolaterally with respect to the neural tube, and showing a continuous distribution on both control (lower) and operated sides. (B)HNK-1 immunolabeling of a ventral section showing the absence of neural crest cells in the operated sclerotome, whereas segmentally organized cells are observed on the contralateral side. (C) Feulgen-Rossenbeck staining. Q, quail graft. Bar, 50pm.

Fig. 3.

Migration of neural crest cells opposite a mesoderm composed of multiple caudal somitic halves. Frontal sections through a 30-somite-stage chick embryo, implanted at the 18-somite stage over a length corresponding to somites 15 to 18 with four caudal somitic halves derived from a 20-somite-stage quail donor. (A) HNK-l-immunoreactive neural crest cells located dorsolaterally with respect to the neural tube, and showing a continuous distribution on both control (lower) and operated sides. (B)HNK-1 immunolabeling of a ventral section showing the absence of neural crest cells in the operated sclerotome, whereas segmentally organized cells are observed on the contralateral side. (C) Feulgen-Rossenbeck staining. Q, quail graft. Bar, 50pm.

On embryos sectioned transversely (a total of 17 cases observed), we counted the sections with and without HNK-1 positive cells, discerning the cells situated dorsally or ventrolaterally to the neural tube and within the sclerotome. For dorsal levels, all sections on control sides or on either operated side (multiple cranial or caudal) showed labelled cells, confirming that dorsally located neural crest cells are longitudinally continuous at this stage, regardless of the nature of the adjacent somitic mesoderm. For ventrolateral levels and within the sclerotome, results varied according to the type of the paraxial mesoderm. On control sides, there was a regular alternation of 10 to 17 sections with and without fluorescent cells. On operated sides composed of multiple cranial half-somites, there appeared areas of 35 to 45 consecutive sections showing fluorescent cells. In contrast, when multiple caudal halfsomites were grafted, areas of 22 to 39 consecutive sections lacking labelled cells at this level were found. Surprisingly, in 4 out of these 9 later embryos, few neural crest cells were present between the dermomyo-toine and sclerotome and these were scattered along the grafted area without retaining their normal segmental arrangement.

DRG formation opposite multiple cranial or caudal half-somites

We have studied the pattern of segmentation, size and relative localization of DRG composed of cells with chick phenotype, developing adjacent to a quail mesoderm of either cranial or caudal composition. At E4 – 5 normal DRG are segmentally organized facing each cranial half-somite, and are expanding at the expense of the original cranial sclerotomal tissue.

In the case of cranial half-somitic grafts, DRG appeared normal in cross-section with respect both to their size and position relative to the neural tube (Fig. 4A). However, a serial analysis of transverse sections showed that an abnormally long ganglion had developed facing the entire length of the cranial somitic graft (Fig. 5A). Normal segmentation was essentially lost since no real interganglionic spaces were observed. All transverse sections cut through the grafted region contained ganglion cells (Fig. 5A). However, the ganglion was not of entirely uniform thickness throughout its length; cross-sectional area varied from 0 · 015 mm2 at its maximal expression to about half this value (Fig. 5A). Thus, implantation of multiple cranial half-somites resulted in the formation of a ‘giant’ DRG whose volume was about twice the sum of the volumes of the corresponding DRG on the contralateral side.

Fig. 4.

Dorsal root ganglion formation opposite multiple cranial or multiple caudal somitic halves. (A) Transverse section of a 5-day-old chick embryo implanted at the 12-somite stage at the level of somites 10 to 12 with three rostral somitic halves from a 26-somite-stage quail embryo. Note, on the right side, a DRG with chick cells surrounded by a mesoderm composed exclusively of quail cells. The form of the ganglion and its position in cross-section appear normal when compared to the contralateral side. (B) Transverse section through a 4-day-old chick embryo implanted at the 15-somite stage at the level of somites 13 to 15 with three caudal half-somites from a 17-somite-stage quail embryo. Note, on the right side of the photomicrograph, a small, round, dorsally located DRG compound of chick cells surrounded by a quail mesoderm. Feulgen-Rossenbeck staining. Bar, 40 μm.

Fig. 4.

Dorsal root ganglion formation opposite multiple cranial or multiple caudal somitic halves. (A) Transverse section of a 5-day-old chick embryo implanted at the 12-somite stage at the level of somites 10 to 12 with three rostral somitic halves from a 26-somite-stage quail embryo. Note, on the right side, a DRG with chick cells surrounded by a mesoderm composed exclusively of quail cells. The form of the ganglion and its position in cross-section appear normal when compared to the contralateral side. (B) Transverse section through a 4-day-old chick embryo implanted at the 15-somite stage at the level of somites 13 to 15 with three caudal half-somites from a 17-somite-stage quail embryo. Note, on the right side of the photomicrograph, a small, round, dorsally located DRG compound of chick cells surrounded by a quail mesoderm. Feulgen-Rossenbeck staining. Bar, 40 μm.

Fig. 5.

Serial section analysis of DRG developing adjacent to multiple cranial or caudal somitic halves. (A) Multiple cranial somitic halves. The embryo illustrated is the same as in Fig. 4A. (B) Multiple caudal somitic halves. The embryo shown is the same as in Fig. 4B. DRG surface is plotted as a function of section thickness., normal side;,operated side. Vertical arrows show the limits of the grafted area.

Fig. 5.

Serial section analysis of DRG developing adjacent to multiple cranial or caudal somitic halves. (A) Multiple cranial somitic halves. The embryo illustrated is the same as in Fig. 4A. (B) Multiple caudal somitic halves. The embryo shown is the same as in Fig. 4B. DRG surface is plotted as a function of section thickness., normal side;,operated side. Vertical arrows show the limits of the grafted area.

In the case of caudal half-somitic grafts, DRG appeared smaller than normal in cross-section, were circular in shape as compared to elongated normal ganglia, and dorsally localized (Fig. 4). Serial analysis of transverse sections at E4-5 revealed perturbations of ganglion segmentation. Small ganglia (about one third to one fifth the normal size) were irregularly fragmented or even elongated over the entire length of the graft (Fig. 5B).

DRG and vertebral development opposite multiple cranial or caudal half-somites

On the operated side of the embryos into which compound cranial half-somites had been transplanted, long unsegmented DRG were observed at E8 – 10 along the grafted area (Figs 6A, 8A), essentially similar to the picture obtained at E4 (Figs 4A, 5A). The giant ganglion illustrated in Fig. 6A is in contact with the spinal cord throughout its length, whereas normally segmented DRG alternate with the vertebrae (Fig. 6B).

Fig. 6.

Development of a ‘giant’ DRG facing multiple cranial half-somites. Frontal sections through an 8-day-old chick embryo grafted at the 14-somite stage over a length corresponding to somites 11 to 14 with four cranial halfsomites from a 16-somite-stage quail embryo. (A) Operated side showing a long unsegmented DRG apposed to the spinal cord (Sc). Qv, quail vertebra; Qm, quail muscle. (B) Control side. Zenker fixation, Feulgen-Rossenbeck staining. Bars: (A) 40m. (B) 150 fjm.

Fig. 6.

Development of a ‘giant’ DRG facing multiple cranial half-somites. Frontal sections through an 8-day-old chick embryo grafted at the 14-somite stage over a length corresponding to somites 11 to 14 with four cranial halfsomites from a 16-somite-stage quail embryo. (A) Operated side showing a long unsegmented DRG apposed to the spinal cord (Sc). Qv, quail vertebra; Qm, quail muscle. (B) Control side. Zenker fixation, Feulgen-Rossenbeck staining. Bars: (A) 40m. (B) 150 fjm.

Adjacent to quail mesoderm composed originally of caudal half-somites, E8-10 DRG remained smaller (Fig. 7A) and dorsally located with respect to normal ganglia (Fig. 7C) as was the case at E4 – 5 (Fig. 4B). Fig. 8B illustrates a serial analysis of transverse sections of one such embryo, revealing the presence of two smaller ganglia that formed along the grafted area instead of three normal ganglia present on the contralateral side. Similar observations were made in 4 out of 5 embryos. In addition, the vertebrae were abnormally segmented (Fig. 7A) and DRG were always localized between a portion of vertebra and the spinal cord rather than normally protruding between two adjacent vertebrae (Fig. 7).

Fig. 7.

DRG development facing multiple caudal halfsomites. (A) Frontal section through a 9-day-old chick embryo grafted at the 15-somite stage at the level of somites 12 to 15 with three caudal somitic halves from a 19-somite-stage quail embryo. Note the abnormal segmentation of the ganglia and their reduced size. Observe also the abnormal morphogenesis of the vertebrae. (B) Higher magnification of A. Note the chick DRG surrounded by quail cartilage and muscle cells. (C) Transverse section through a 9-day-old chick embryo implanted at the stage of 15 somites at the level of somites 13 to 15 with four caudal somitic halves from a 15-somite-stage quail embryo. Note the reduced size of the DRG, and its dorsal localization with respect to the spinal cord. Operated sides are to the left. Zenker fixation, Feulgen-Rossenbeck staining. Bars: (A,C) 200pm, (B) 50pm.

Fig. 7.

DRG development facing multiple caudal halfsomites. (A) Frontal section through a 9-day-old chick embryo grafted at the 15-somite stage at the level of somites 12 to 15 with three caudal somitic halves from a 19-somite-stage quail embryo. Note the abnormal segmentation of the ganglia and their reduced size. Observe also the abnormal morphogenesis of the vertebrae. (B) Higher magnification of A. Note the chick DRG surrounded by quail cartilage and muscle cells. (C) Transverse section through a 9-day-old chick embryo implanted at the stage of 15 somites at the level of somites 13 to 15 with four caudal somitic halves from a 15-somite-stage quail embryo. Note the reduced size of the DRG, and its dorsal localization with respect to the spinal cord. Operated sides are to the left. Zenker fixation, Feulgen-Rossenbeck staining. Bars: (A,C) 200pm, (B) 50pm.

Fig. 8.

Serial section analysis of DRG localized opposite multiple cranial or caudal somitic halves.- (A) Multiple cranial somitic halves. A 9-day-old chick embryo was implanted at the stage of 15 somites on the length of the three last-formed somites with four cranial somitic halves excised from a quail embryo of the same age. (B) Multiple caudal somitic halves. The embryo shown is the same as in Fig. 7C. DRG surface is plotted as a function of section thickness., normal side;, operated side. Vertical arrows mark the limits of the graft.

Fig. 8.

Serial section analysis of DRG localized opposite multiple cranial or caudal somitic halves.- (A) Multiple cranial somitic halves. A 9-day-old chick embryo was implanted at the stage of 15 somites on the length of the three last-formed somites with four cranial somitic halves excised from a quail embryo of the same age. (B) Multiple caudal somitic halves. The embryo shown is the same as in Fig. 7C. DRG surface is plotted as a function of section thickness., normal side;, operated side. Vertical arrows mark the limits of the graft.

In both experimental paradigms, morphogenesis of the vertebrae was significantly perturbed, although either somitic half was capable of developing into vertebral cartilage (Figs 6, 7), in agreement with the findings previously reported by Stern & Keynes (1987).

The development of dorsal root ganglia was investigated as a function of the craniocaudal composition of the somites. We have demonstrated the following.

(a) Migration of neural crest cells in dorsolateral regions apposed to the neural tube is independent of the craniocaudal asymmetry of the somites. Neural crest cells appeared as a continuous strip along both dorsal sides of the neural tube, in a nonsegmented fashion regardless of whether they were facing normal adjacent somites or multiple cranial or caudal half-somites. Therefore, early migration of cells in the region corre-spending to the future DRG correlated rather with the organization of the neural tube which is thought to be a nonsegmented structure (Keynes & Stern, 1985).

(b) The patterning of the DRG, in terms of ganglion segmentation, size and relative localization with respect to the adjacent structures, is strongly correlated with craniocaudal somitic asymmetry. Normal metameric distribution of DRG was impaired by changing the alternating craniocaudal nature of adjacent somites; this was achieved by experimentally constructing either multiple crania] or caudal half-somites. The striking result obtained after grafting mesoderm made up of cranial half-somites was the development of a single unsegmented ganglion whose size was at least twice the sum of the sizes of the contralateral DRG (Figs 5A, 6, 8A). An unsegmented pattern of penetration of neural crest cells within the multiple cranial sclerotome, as shown in Fig. 2B and as previously reported by Stern & Keynes (1987), preceded this type of ganglion development. Experiments involving a massive deletion of multiple somites (Lewis et al. 1981; Tosney, 1988) also resulted in a loss of segmentation of the DRG and spinal nerves. However, no significant information about the development of the DRG under these conditions could be drawn from the data since complete deletion of somites along such a considerable length of the neuraxis is likely to deplete neural crest progenitors to different extents. Concerning the same experiments, it remains to be tested whether DRG precursor cells migrate and accumulate in a cell-free space or, alternatively, whether there is a regeneration of cranial-like somitic tissue within the depleted area.

Facing a mesoderm composed of multiple caudal somitic halves, however, the situation concerning dor-sally and ventrally located neural crest derivatives was qualitatively and quantitatively different; whereas the sclerotome was completely devoid of neural crest cells (Fig. 3B, and Stern & Keynes, 1987) and of motor nerves (Stern & Keynes, 1987), DRG were present albeit abnormally segmented and reduced in size. The reasons for this observed type of development may be twofold. First, under normal conditions neural crest cells originating opposite caudal half-somites remigrate to become localized at cranial somitic levels where they participate in ganglion formation (Teillet et al. 1987). Under the present experimental conditions, however, the long mesoderm with caudal properties (usually about four somites long) imposes a lack of immediately adjacent cranial regions into which neural crest cells moving along the longitudinal axis can home. Second, neural crest cells seem unable to expand laterally at the expense of the half-caudal somitic mesoderm. As a result of these two constraints imposed by the grafted caudal moieties, the ganglia remain in a dorsal position during gangliogenesis, their segmentation is perturbed and their final size is smaller than normal. In this connection, we have observed that the ganglia localized just beyond both ends of the grafts are bigger than normal, suggesting a compensatory development compared to the DRG that had developed at the level of the graft (our unpublished results).

Observation of neural crest cell migration and DRG formation in normal embryos (Rickmann et al. 1985; Bronner-Fraser, 1986; Loring & Erickson, 1987; Teillet et al. 1987; Lallier & Bronner-Fraser, 1988), and the results of transplantation studies (Stern & Keynes, 1987; Teillet et al. 1987), provide growing evidence that there must be qualitative differences between cranial and caudal somitic cells with respect to their interaction with neural crest cells. The molecular basis for these interactions is still unclear, although a variety of markers have already been identified that are differentially distributed at early stages between the two somitic halves. For instance, peanut agglutinin binding sites were preferentially localized to the caudal half-sclerotome (Stern et al. 1986), as was a chondroitin sulfate proteoglycan whose initial distribution was diffuse over the sclerotome and became restricted to its caudal part at the time of neural crest cell invasion (Tan et al. 1987). The cranial half-somite was found to express, in a preferential manner, the extracellular matrix proteins cytotactin (Tan et al. 1987) and tenascin (Mackie et al. 1988); likewise the enzymes butyrylcholinesterase and acetylcholinesterase defined a craniocaudal asymmetry at the level of the dermomyotome (Layer et al. 1988). Based on the results discussed above, it becomes clear that, although yet undefined, differential interactions between neural crest cells and somitic cells from either half determine, at least partly, the developmental pattern of DRG.

Moreover, the importance of the central nervous system primordium in early DRG development should be stressed in the present context. As demonstrated by Kalcheim & Le Douarin (1986), the neural tube may provide survival factor(s) for developing ganglion cells, since early separation of the DRG anlage from the neural tube results in death of those neural crest cells remaining disconnected. In further studies to elucidate this question, it was shown that brain-derived neurotrophic factor, in concert with laminin, is one of the active molecules of central nervous system origin that supports survival of developing DRG cells in vivo (Kalcheim et al. 1987; Hofer & Barde, 1988). In vitro, the responsive cells were characterized as being substance P-immunoreactive neurons (Kalcheim & Gendreau, 1988). Moreover, it was recently found that basic fibroblast growth factor, another central nervous system-derived molecule, stimulates survival of neural crest-derived non-neuronal cells both in vivo and in tissue culture (Kalcheim, 1989).

Thus, since the operations described here altered the normal topographical relationship between the neural crest and the neural tube, it is very likely that the final size and morphology of the DRG that formed are a result of interactions involving neural crest, neural tube and either cranial or caudal somitic cells. For instance, the mesoderm corresponding to multiple caudal half somites might interfere with the establishment of stable contacts between neural crest-derived cells and the neural tube, thus impairing ganglion organization in those areas. On the other hand, grafts of the multiple cranial-type mesoderm, being permissive to neural crest cell distribution in a nonsegmented fashion, would facilitate close contact between crest cells and neural tube over a surface wider than the normal area of contact; in this case, the resulting DRG would be bigger than the contralateral ganglia and become coalesced (Figs 5A, 6, 8A). It remains to be tested whether the combined action of cranial half-mesoderm and neural tube merely supports survival of developing DRG cells or actively stimulates their proliferation. Moreover, interesting questions are raised concerning the ontogeny of peripheral and central projections emanating from the ‘giant’ sensory ganglia that develop in the context of multiple cranial half-somites, and about the mechanism by which the final number of cells within these ganglia is regulated.

The authors are indebted to Prof. N. Le Douarin for her continuous support and wish to thank Dr J. Smith for critical reading of the manuscript.

We thank Bernadette Schuler and Chana Carmeli for excellent technical assistance, Bernard Henri and Sophie Gournet for contributing to the illustrations; and Mindy Slupski for preparation of the typescript. This work was supported by research grants from the Muscular Dystrophy Association and the Israel Academy of Sciences and Humanities (to C.K.) and by the Centre National de la Recherche Scientifique, the Institut Nationale de la Santé et de la Recherche Médicale, the Fondation pour la Recherche Médicale Française and the March of Dimes Birth Defects Foundation Basic Research Grant (No. 1-866).

Abo
,
T.
&
Balch
,
C. M.
(
1981
).
A differentiation antigen of human NK and K cells identified by a monoclonal antibody (HNK-1)
.
J. Immunol
.
127
,
1024
1029
.
Bronner-Fraser
,
M.
(
1986
).
Analysis of the early stages of trunk neural crest migration in avian embryos using monoclonal antibody HNK-1
.
Devi Biol
.
115
,
44
55
.
Hofer
,
M. M.
&
Barde
,
Y. A.
(
1988
).
Brain-derived neurotrophic factor prevents naturally occurring neuronal death in vivo
.
Nature, Lond
.
331
,
259
261
.
Kalcheim
,
C.
(
1989
).
Basic fibroblast growth factor stimulates survival of non neuronal cells developing from trunk neural crest
.
Devi Biol (in press)
.
Kalcheim
,
C.
,
Barde
,
Y. A.
,
Thoenen
,
H.
&
Le Douarin
,
N. M.
(
1987
).
In vivo effect of brain-derived neurotrophic factor on the survival of developing dorsal root ganglia
.
EMBO J
.
6
,
2871
2873
.
Kalcheim
,
C.
&
Gendreau
,
M.
(
1988
).
Brain-derived neurotrophic factor stimulates survival and neuronal differentiation in cultured avian neural crest
.
Devi Brain Res
.
41
,
79
86
.
Kalcheim
,
C.
&
Le Douarin
,
N. M.
(
1986
).
Requirement of a neural tube signal(s) for the differentiation of neural crest cells into dorsal root ganglia
.
Devi Biol
.
116
,
451
466
.
Keynes
,
R. J.
&
Stern
,
C. D.
(
1984
).
Segmentation in the vertebrate nervous system
.
Nature, Lond
.
310
,
786
789
.
Keynes
,
R. J.
&
Stern
,
C. D.
(
1985
).
Segmentation and neural development in vertebrates
.
Trends Neurosci
.
8
,
220
223
.
Lallier
,
T. E.
&
Bronner-Fraser
,
M.
(
1988
).
A spatial and temporal analysis of dorsal root and sympathetic ganglion formation in the avian embryo
.
Devi Biol
.
127
,
99
112
.
Layer
,
P. G.
,
Alber
,
R.
&
Rathjen
,
F. G.
(
1988
).
Sequential activation of butyrylcholinesterase in rostral half somites and acetylcholinesterase in motoneurones and myotomes preceding growth of motor axons
.
Development
102
,
387
396
.
Le Douarin
,
N. M.
(
1973
).
A Feulgen-positive nucleolus
.
Expl Cell Res
.
77
,
459
468
.
Le Douarin
,
N. M.
(
1982
).
The Neural Crest
.
Cambridge University Press
,
Cambridge
. Pp.
160
165
.
Lewis
,
J.
,
Chevalier
,
A.
,
Kieny
,
M.
&
Wolfert
,
L.
(
1981
).
Muscle nerve branches do not develop in chick wings devoid of muscle
.
J. Embryol. exp. Morph
.
64
,
211
232
.
Lipinski
,
M.
,
Braham
,
X.
,
Caillaud
,
J. M.
,
Canlu
,
C.
&
Tursz
,
T.
(
1983
).
HNK-1 antibody detects an antigen expressed on neuroectodermal cells
.
J. exp. Med
.
158
,
1775
1780
.
Loring
,
J. F.
&
Erickson
,
C. A.
(
1987
).
Neural crest cell migratory pathways in the trunk of the chick embryo
.
Devi Biol.
121
,
220
236
.
Mackie
,
E. J.
,
Tucker
,
R. P.
,
Haleter
,
W.
,
Chiquet-Ehrismann
,
R.
&
Epperlein
,
H. H.
(
1988
).
The distribution of tenascin coincides with pathways of neural crest cell migration
.
Development
102
,
237
250
.
Newgreen
,
D. F.
,
Scheel
,
M.
&
Kastner
,
V.
(
1986
).
Morphogenesis of sclerotome and neural crest in avian embryos. In vivo and in vitro studies on the role of notochordal extracellular material
.
Cell Tissue Res
.
244
,
299
313
.
Rickmann
,
M.
,
Fawcett
,
J. W.
&
Keynes
,
R. J.
(
1985
).
The migration of neural crest cells and the growth of motor axons through the rostral half of the chick somite
.
J. Embryol. exp. Morph
.
90
,
437
455
.
Stern
,
C. D.
&
Keynes
,
R. J.
(
1987
).
Interactions between somite cells; the formation and maintenance of segment boundaries in the chick embryo
.
Development
99
,
261
272
.
Stern
,
C. D.
,
Sisodiya
,
S. M.
&
Keynes
,
R. J.
(
1986
).
Interactions between neurites and somite cells: inhibition and stimulation of nerve growth in the chick embryo
.
J. Embryol. exp. Morph
.
91
,
209
226
.
Tan
,
S. S.
,
Crossin
,
K. L.
,
Hoffman
,
H.
&
Edelman
,
G. M.
(
1987
).
Asymmetric expression in somites of cytotactin and its proteoglycan ligand is correlated with neural crest cell distribution
.
Proc. natn. Acad. Sci. U.S.A
.
84
,
7977
7981
.
Teillet
,
M.-A.
,
Kalcheim
,
C.
&
Le Douarin
,
N. M.
(
1987
).
Formation of the dorsal root ganglia in the avian embryo: Segmental origin and migratory behavior of neural crest progenitor cells
.
Devi Biol
.
120
,
329
347
.
Tosney
,
K. W.
(
1978
).
The early migration of neural crest cells in the trunk region of the avian embryo. An electron microscopic study
.
Devi Biol
.
62
,
317
333
.
Tosney
,
K. W.
(
1988
).
Proximal tissues and patterned neurite outgrowth at lumbosacral level of the chick embryo: partial and complete deletion of the somite
.
Devi Biol
.
127
,
266
286
.
Tucker
,
G. C.
,
Aoyama
,
H.
,
Lipinski
,
M.
,
Tursz
,
T.
&
Thiery
,
J. P.
(
1984
).
Identical reactivity of monoclonal antibodies HNK-1 and NC-1: Conservation in vertebrates on cells derived from the neural primordium and on some leukocytes
.
Cell Diff
.
14
,
223
230
.
Weston
,
J. A.
(
1963
).
A radioautographic analysis of the migration and localization of trunk neural crest cells in the chick
.
Devi Biol
.
6
,
279
310
.
Weston
,
J. A.
(
1970
).
The migration and differentiation of neural crest cells
.
Adv. Morphog
.
8
,
41
114
.