INITIAL AXIAL LEVEL-DEPENDENT DIFFERENCES IN SIZE OF AVIAN DORSAL-ROOT GANGLIA ARE IMPOSED BY THE SCLEROTOME

We have recently shown that there is very early variation in dorsal ro ot ganglia (DRG) size based on their axial position. From the time of their condensation at Stage (St.) 20 (Embryonic Day 3), before the ons et of apoptosis, the ganglia in brachial segments 14 and 15 are more t han 80% larger on the average than those in cervical 5 and 6. This dif ference in volume is due to increased numbers of cells in the brachial DRG. In addition, the rostrocaudal length was found to be significant ly greater for brachial ganglia, and the greater length of the brachia l ganglia was found to be correlated with a greater length of brachial than cervical sclerotomes. It was therefore proposed that the differe nce in DRG size at the time of gangliogenesis is likely to arise from colonization by a larger initial number of neural crest cells of the l onger adjacent rostral sclerotomes in brachial somites. In the present work, we have performed two types of experiments to test this hypothe sis. First, we have performed heterotopic grafts of segmental plate me soderm from cervical to brachial levels and vice versa. In all of thes e grafts, the sclerotomes developed with a rostrocaudal extent (length ) corresponding to their level of origin in the donor embryo. DRG that formed in the grafted mesoderm attained a length appropriate to that of ganglia developing in segments of the axial level of the dorter mes oderm, not that of the host. Second, we have estimated proliferation o f DRG cells at St. 20 using HNK-1/bromodeoxyuridine double-stain immun ocytochemistry. The percentage of cells in S-phase at both brachial an d cervical levels of the neuraxis in newly formed ganglia was observed to be the same. These two lines of evidence strongly argue that the i nitial difference in size between DRG at different axial levels is not intrinsic, but rather is imposed by the mesodermal microenvironment i n which they develop, as is the case for DRG segmentation. This is in contrast to what may occur in the hindbrain, where determination of rh ombomere identity and neural crest segmentation are thought to be intr isic to the nervous system. (C) 1995 Academic Press, Inc.