Morphological evidence is presented that definitive mesoderm formation in Xenopus is best understood as extending to the end of the neurula phase of development. A process of recruitment of cells from the deep neurectoderm layers into mesodermal position and behaviour, strictly comparable with that already agreed to occur around the internal blastoporal ‘lip’ during gastrula stages, can be shown to continue at the posterior end of the presumptive body pattern up to stage 20 (earliest tail bud).
Spatial patterns of incidence of mitosis are described for the fifteen hours of development between the late gastrula and stage 20–22. These are related to the onset of new cell behaviours and overt cyto-differentiations characterizing the dorsal axial pattern,which occur in cranio-caudal and then medio-lateral spatial sequence as development proceeds. A relatively abrupt cessation of mitosis, among hitherto asynchronously cycling cells,precedes the other changes at each level in the presumptive axial pattern. The widespread incidence of cells still in DNA synthesis, anterior to the last mitoses in the posterior-to-anteriordevelopmental sequence of axial tissue, strongly suggests that cells of notochord and somites in their prolonged, non-cycling phase are G2-arrested, and thus tetraploid. This is discussed in relation to what is known of cell-cycle control in other situations.
Best estimates for cell-cycle time in the still-dividing, posterior mesoderm of the neurula lie between 10 and 15 h. The supposition of continuing recruitment from neurectoderm can resolve an apparent discrepancy whereby total mesodermal cell number nevertheless contrives to double over a period of approximately 12 h during neurulation when most of the cells are leaving the cycle.
Because of pre-existing evidence that cells maintain their relative positions (despite distortion)during the movements that form the mesodermal mantle, the patterns presented in this paper can be understood in two ways: as a temporal sequence of developmental events undergone by individual, posteriorly recruited cells as they achieve their final positions in the body pattern, or alternatively as a succession of wavefronts with respect to changes of cellstate, passing obliquely across the presumptive body pattern in antero-posterior direction. These concepts are discussed briefly in relation to recent ideas about pattern formation in growing systems.
Null mutations of NT-3 andBaxaffect trigeminal ganglion cell number but not brainstem barrelette pattern formation
