FIRST BLOOD-VESSELS IN THE AVIAN NEURAL-TUBE ARE FORMED BY A COMBINATION OF DORSAL ANGIOBLAST IMMIGRATION AND VENTRAL SPROUTING OF ENDOTHELIAL-CELLS

We studied the early pattern of neural tube (NT) vascularization in qu ail embryos and chick-quail chimeras. Angioblasts appeared first in th e dorsal third at Hamburger and Hamilton (HH) stage 19 as single, migr ating cells. Their distribution did not correspond to a segmental patt ern. After this initial dorsal immigration, endothelial sprouts invade d the NT on either side of the floor plate (HH stage 21). These cells remained continuous with their arterial vascular sources, connected to the venous perineural vascular plexus at HH-stage 22, and formed the first perfused vessels of the NT at HH-stage 23. The same pattern of a ngiotrophic vascularization was observed in a craniocaudal sequence st arting caudal to the rhombencephalic NT. Extremely long filopodia were observed on sprouting cells, extending toward the central canal and t he mantle layer. The exclusively extraneuroectodermal origin of angiob lastic cells was demonstrated with chick-quail chimeras. Following rep lacement of quail NT by chick NT graft, angioblast and sprout distribu tion in chimeras was the same as in controls. We conclude that the NT receives its first blood vessels by a combination of two different pro cesses, dorsal immigration of isolated migrating angioblastic cells an d ventral sprouting of endothelial cells, which derive from perfused v essels. The dorsal invasive angioblasts contribute to the developing i ntraneural vascular plexus after having traversed the neural tube. The initial distribution of blood vessels within the neuroepithelium corr esponds to intrinsic random motility of angioblastic cells; a more reg ular pattern is seen later. The floor plate apparently prohibits conne ctions between sprouts in both NT sides, whereas in the dorsal NT, suc h a separating effect on the migrating angioblasts does not exist. (C) 1996 Academic Press, Inc.