BIOLOGICAL OVERVIEW

Gastrulation, the morphologenic process that creates mesoderm and endoderm from a uniform blastula early in development, involves multiple cell movements in Drosophila, unlike vertebrate gastrulation in which ingressing cells move through a single blastopore. During early stages of gastrulation in Drosophila, two populations of cells invaginate. At the posterior pole of the embryo, the posterior midgut invagination brings endodermal precursors into the interior. The ventral furrow {Images] forms along the ventral surface of the central body region of the embryo and internalizes the mesoderm precursors. Both these morphogenic movements are driven by individual cell shape changes within the invagination primordia. Both invaginations are initiated by apical constrictions that transform columnar cells to a wedge- or cone-like shape. As the cells constrict their apices (the outer surface of the cell), the apical plasma membrane folds into many small protrusions, or blebs, rather than decreasing in surface area. This observation suggests that the force that drives apical constriction is generated below the cell membrane. Since cytoplasmic myosin accumulates at the apical ends of cells in the ventral furrow and posterior midgut primordia at the start of cell shape changes, contraction of a cortical actin-myosin network may mediate apical constriction (Costa, 1994 and references).

How are the cell shape changes coordinated during gastrulation so that all cells in the gastrulating primordia behave synchronously? Two genes have been identified that are required specifically for the ventral furrow and the posterior midgut invaginations: the maternal gene concertina and the zygotic gene folded gastrulation (fog) do not seem to be essential for the mechanical aspects of constriction of the apical cytoplasm, but rather are necessary for coordination of cell shape changes throughout the invaginating primordia. Concertina is a G-protein alpha subunit, suggesting a role for cell-cell signaling in invagination (Parks, 1991). It is suggested Fog functions as a secreted signal that activates the G-protein alpha subunit encoded by concertina in neighboring cells, and thus a secreted signal ensures the rapid, orderly progression of constriction initiations from the middle of invagination primordia out toward the margins. Apparently Fog acts through an unidentified G-protein coupled receptor, that is, a 7 pass transmembrane serpentine receptor (Costa, 1994).

Additional components in this pathway have been identified. DRhoGEF2, a guanine nucleotide exchange factor for Rho1 mediates specific cell shape changes in response to the extracellular ligand, Folded gastrulation. fog was expressed ectopically from a huckebein promoter, which is normally active in a subset of cells at the anterior and posterior ends of the embryo. In all the hkb-fog expressing embryos a characteristic transient depression in the dorsal head region can be seen. The surfaces of cells in this depression exhibit membrane blebbing and constrictions closely resembling those normally seen in cells along the ventral furrow in wild-type embryos. In addition, the nuclei of these cells have migrated from an apical to a basal position. In the absence of DRhoGEF2, ectopic Fog expression fails to induce any detectable cell shape changes, despite equivalent levels of fog transgene expression. Together, these results establish a Rho-mediated signaling pathway that is essential for the major morphogenetic events in Drosophila gastrulation (Barrett, 1997).

The central domain of DRhoGEF2 contains a likely phorbol ester-response motif. The homologous domain in Protein kinase C mediates kinase activation in response to diacylglycerol, which is generate by phospholipase C (PLC). Thus is it possible that the GEF activity of DRhoGEF2 is responsive to diacylglycerol. Since PLC-mediated production of diacylglycerol can be promoted by both receptor tyrosine kinase activation and by activation of receptor-coupled heterotrimeric G proteins, it is possible that the nucleotide-exchange activity of DRhoGEF2 is stimulated by signals transduced by both of these types of receptors. The presence of a PDZ domain in DRhoGEF2, suggests that it may interact with additional signaling proteins. Therefore, it appears that the GEF activity of DRhoGEF2 may be regulated by multiple upstream signals (Barrett, 1997).

It is postulated that Fog acts via the G alpha protein Concertina to activate DRhoGEF2, thereby promoting Rho1 activation and consequent actin rearrangements. Significantly, the Drosophila G alpha subunit, Concertina, exhibits the strongest sequence similarity to the mammalian Galpha12 and Galpha13 proteins, which mediate the activation of Rho by LPA. Thus, it appears likely that a Rho-mediated signaling pathway linked to heterotrimeric G proteins has been evolutionarily conserved (Barrett, 1997 and references).

Why do cells of distinct fates (for example, presumptive mesodermal cells or presumptive posterior midgut cells) require a G-protein coupled signal to initiate gastrulation? The answer may lie in the fact that cell commitment to a certain fate does not in itself provide a mechanism for the cytoskeletal modifications that are necessary for invagination. Secretion of a ligand that functions to excite G-protein coupled receptors would be required to signal cytoskeletal changes that drive gastrulation. Such signaling could also act to synchronize cell activity and ensure graded responses by cells somewhat removed from the ventral midline. Cell signaling mechanisms might be required for the rapid completion of the invagination process. Invagination of the ventral furrow is completed in about 10 minutes, and invagination of the posterior midgut invagination takes about 20 minutes (Costa, 1994).

Because gastrulation starts without the G-protein coupled mechanism, it is suggested that two separate mechanisms initiate apical constriction of cells during invagination. One mechanism that does not require fog operates in the most ventral and posterior cells, respectively, of the mesodermal and posterior midgut primordia. A second mechanism, taking place subsequent to the first, requires a secreted ligand and G-protein coupled receptors, as well as fog. Such a mechanism is postulated, since fog mutant flies do, in fact, initiate invagination. It is not unusual in biological systems that what appears to be a unitary process is actually staged. In these cases, a later process, in this case secretion of a ligand, often appears as a cellular response to an earlier process (initiation of gastrulation).

GENE STRUCTURE

Bases in 5' UTR - 849

Bases in 5' UTR - 849

Bases in 3' UTR - 612

PROTEIN STRUCTURE

Amino Acids - 730

Structural Domains

fog encodes a novel protein with a putative signal sequence (functioning to facilitate secretion) but no potential transmembrane domains. There are nine potential N-linked glycosylation sites and the C-terminal three-quarters of the protein is highly enriched in serine and threonine. There are three potential protease cleavage sites in the Fog sequence. Cleavage at the two amino-terminal sites would produce a central peptide of 109 amino acids; this is in the size range of known peptide ligands of G-protein-coupled receptors (Costa, 1994).

date revised: 10 March 98 

Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.

The Interactive Fly resides on the
Society for Developmental Biology's Web server.