Centrosomes and the Organization of the Actin/Myosin Cytoskeleton in Early Fly Embryos: The Role of Scrambled, Nuf, and CP190
As described in Section 13.2.3, there is strong evidence to suggest that centrosomes influence the behavior of the actin/myosin cytoskeleton in the early fly embryo. 3-D reconstructions of the actin, myosin and microtubule networks in fixed embryos before and after various drug treatments have suggested that actin and myosin
II filaments can be transported along microtubules from the minus ends at centrosomes to the plus ends . This transport, together with assumed differential affinities of actin and myosin for the cell cortex and a regulation of these properties during the cell cycle, could explain the distribution and interdependencies of the cytoskeletal networks. Studies on the Drosophila Scrambled (Sced) protein, however, suggest a different model of how centrosomes could influence actin . Sced localizes to both centrosomes and to the pseudo-cleavage furrows that normally surround the spindles during mitosis in the syncytial embryos. Mutations in sced do not interfere with actin cap formation in interphase, but block the recruitment of actin to the pseudo-cleavage furrows in mitosis. Surprisingly, injecting microtubule depolymerizing drugs into embryos does not disrupt the localization of Sced or the organization of actin in caps or furrows. These observations suggest that Sced is essential to help centrosomes recruit actin into the furrow and that centrosomes can influence actin recruitment even in the absence of microtubules.
Studies on the protein Nuclear fall out (Nuf) have suggested that centrosomes may also direct the recruitment of membranes to both pseudo-cleavage furrows and the furrows that invaginate around the nuclei during cellularization . Nuf is recruited to centrosomes specifically during prophase. In nuf mutant embryos, actin is not properly recruited to the furrows, even though other furrow components are recruited. Nuf is a member of the arfophilin family, which has been implicated in regulating membrane trafficking. Moreover, Nuf is required to recruit the membrane-associated protein Discontinuous actin hexagon (Dah) to furrows, providing a potential mechanism whereby centrosomes could guide the transport of membrane components to the furrow [90, 91].
One of the most studied Drosophila centrosomal proteins, CP190, has an unexpected role in regulating myosin in the early fly embryo. CP190 was the first centrosomal protein to have its cDNA cloned in Drosophila  and it is widely used as a centrosomal marker. CP190, together with its binding partner CP60, cycles between the nucleus in interphase and the centrosomes in mitosis, although CP60 levels do not peak at the centrosome until anaphase/telophase [93-95]. Both proteins interact directly with microtubules in vitro, and the ability of CP60 to bind microtubules is abolished when it is phosphorylated by cdc2 (also known as Cdk1)/cyclin B kinase. These observations suggested that CP190 and CP60 were involved in regulating microtubule behavior specifically during late stages of mitosis, when cdc2/cyclin B levels are in decline. A mutation in the cp190 gene, however, does not detectably affect centrosomal microtubules or any aspect of mitosis, even though CP190 and CP60 are no longer detectable at centrosomes in the mutant cells .
The cp190 mutation, however, is lethal . The lethality can be rescued by the expression of a deleted form of CP190 (CP190DM) that no longer binds to centrosomes or microtubules, demonstrating that CP190 must have some critical function (possibly in the nucleus) that is independent of its ability to bind to centro-somes or microtubules. In early embryos that lack CP190 function, axial expansion fails (S. Chodagam, W. Whitfield, and J. Raff, unpublished observations). As described in Section 13.2.2, axial expansion is an actin/myosin-dependent process that spreads the nuclei throughout the early embryo [22, 23]. Recent studies have shown that cycles of myosin II accumulation occur in the region of the embryo cortex that is directly above the migrating nuclei, even before the nuclei reach the cortex . These cycles stimulate a cortical contraction that appears to drive axial expansion. In cp190 mutant embryos, the cycles are severely diminished, and axial expansion fails. Moreover, this failure can be rescued by the expression of a consti-tutively activated myosin light chain, strongly suggesting that CP190 is somehow involved in regulating myosin function in the early embryo. Although the axial expansion defect of cp190 mutant embryos is rescued by the expression of CP190, it is not rescued by the expression of CP190DM, suggesting that the association of CP190 with centrosomes and microtubules is required to regulate myosin II function (S. Chodagam, W. Whitfield, and J. Raff, unpublished observations).
These observations serve as a salutary lesson for those of us who rely on biochemical techniques to identify proteins of interest. The CP190/CP60 complex had many properties suggestive of a role in regulating centrosomal microtubules during mitosis, but the genetic analysis reveals that, in vivo, this is not the case.
Was this article helpful?