© 1989 by The Society for Integrative and Comparative Biology
Mitotic Domains Partition Fly Embryos, Reflecting Early Cell Biological Consequences of Determination in Progress1
Friday Harbor Laboratories, University of Washington Friday Harbor, Washington 98250
Department of Zoology, University of Washington Seattle, Washington 98195
We link classical observational developmental biology with modern techniques and concepts of cell biology, focusing on early development in the fruit fly, Drosophila melanogaster. In contrast to the first thirteen synchronous nuclear division cycles following fertilization of the egg, cell division in the fourteenth cycle partitions the embryo into mitotic domains. These are bilaterally symmetric clusters of cells, spatially and temporally patterned, invariantly, embryo-to-embryo. Cells in a given mitotic domain share the same mitotic schedule, different from that in neighboring domains. In addition, cells in at least some mitotic domains share distinct attributes such as special cell morphologies, spindle orientations, morphogenetic movement behaviors, and eventual differentiated tissue fates. We argue that a mitotic domain differentiates its control of the fundamental cell cycle process as a consequence of its having embarked on a developmental pathway different from that of neighboring domains. In embryos of flies with certain mutations at earlyacting genetic loci, perturbations of the normal pattern of mitotic domains forecast the final mutant phenotype. Mitotic domains are visible by non-invasive light microscopic observation of live developing embryos, or by staining fixed embryos with fluorescently labeled antibodies to microtubules—methods that will likely work well on embryos of phylogenetically diverse type. We document the methodological and phylogenetic generality of mitotic domains with micrographs of Calliphora (blow-fly) embryos; the homology to Drosophila is obvious. We present a theoretical framework for thinking about the process of embryonic cell determination as a gradual dynamical process and argue that, to learn most about determination, we should correlate the earliest perturbations we can inflict with the earliest phenotypic consequences we can assess: perturbations of the mitotic domain pattern serve this purpose perfectly.