Cell lineage of larval and imaginal thoracic anlagen cells of Drosophila melanogaster, as revealed by single-cell transplantations

Abstract

ABSTRACT We have analyzed the cell lineage of larval and imaginal cells in the thoracic ectoderm of the early embryo of Drosophila melanogaster, by homotopic transplantation of single cells in the region of 50 –60% egg length. Single cells were isolated prior to transplantation in an in vitro solution. The donors were ‘enhancer-trap’ lines in which the nuclei of all larval and imaginal cells exhibit a uniformly intense expression of the lacZ gene of E. coli. The transplantations were carried out from the blastoderm to the early gastrula stage, as a rule immediately after the onset of gastrulation (stage 6). It was found that at this time the cells of the thoracic ectoderm are not yet committed to form larval or imaginal structures, as indicated by the presence of clones overlapping all structures formed by the thoracic ectoderm, i.e. the nervous system, the larval epidermis, the tracheae and the imaginal discs. The average size of pure epidermal clones was five cells. In clones overlapping either other larval tissues or imaginal discs, the average number of epidermal cells was between three and four. The mean relative clone size was 1/5 of the size of the total structure for leg imaginal discs and 1/7 for the wing imaginal disc. We therefore infer that the precursors for the leg discs and wing disc on one side together number 22 cells in the blastoderm or early gastrula stage. These cells eventually give rise not only to precursors of the imaginal discs but usually also to larval epidermal and nervous-system cells, because most of the imaginal disc clones (80%) overlap larval tissue. The transplantations were not precisely homotopic; the fact that up to 10 cells were removed from the donor essentially rules out exact homotopy between donor and host sites, because a segment anlage is only about three cells wide. Nevertheless, the clones developed completely normal tissue together with the recipient cells. Although the clones have the capacity to extend over different ectodermal tissues and can include both imaginal discs in a given segment, no clones were found that clearly crossed larval or imaginal segment boundaries. We propose a model in which the segregation of the cells that are to differentiate into the imaginal tissues does not occur until the second postblastodermal mitosis