The team therefore generated mutant embryos that lack even the maternal supply. These embryos' centrosomes seemed normal, but the dividing nuclei failed to disperse evenly in the embryo and instead remained crowded at the anterior end.
This nuclear spreading is known to depend on the contraction of cortical actin, which occurs in pulses between rapid cycles of mitosis in the embryo. Actin is contracted by myosin, but, in the CP190 mutant, myosin was not at the cortex, where it is normally localized during nuclear spreading.
Constitutively active myosin rescued the defect, but a CP190 mutant that lacks the microtubule and centrosome binding domain did not. Thus, CP190 must be able to bind to the centrosome and microtubules to induce proper myosin localization and activity in response to the nuclear division.
Why is CP190 activity essential in early embryo but not anywhere else? Raff speculates that, in large cells such as the young fly embryo, CP190 might be necessary to carry a signal from the replicating centrosomes along microtubules to cortical myosin and actin. Perhaps in smaller cells, such as those found in older embryos, diffusion suffices when CP190 is absent.