Chromatin was known to promote spindle assembly via its effects on the Ran GTPase. RanGTP is abundant near chromosomes due to activation by a Ran–guanosine nucleotide exchange factor localized on chromosomes, but the extent of RanGTP localization and its function remained unknown. “Researchers suspected that long-range spindle effects were being mediated by chromosomes, but they couldn't prove this for certain,” says Karsenti.
Combining mathematical modeling and fluorescent microscopy, Karsenti's group now shows that RanGTP diffusion away from chromosomes creates an interrelated series of short- and long-range protein gradients. The gradient of free RanGTP is very steep, but built upon it are shallower gradients of RanGTP–importin β and RanGTP–importin β–RanBP1. The RanGTP–importin β complex helps build the spindle. But binding of RanBP1 to RanGTP–importin β inactivates the complex.
This inactivation helps shape the RanGTP–importin β gradient. It is the RanGTP–importin β that ultimately promotes two different behaviors: microtubule nucleation at chromosomes; and stabilization of growing tubules within the spindle. It does so by activating the release of nuclear localization signal (NLS)-containing proteins that regulate microtubules. “It is really the free NLS proteins that are important for driving spindle formation,” says Karsenti.
One NLS protein, TPX2, facilitates microtubule nucleation in an all-or-nothing manner that occurs only when the amount of RanGTP–importin β is above a certain threshold—that is, right around chromosomes. Release of another yet-unidentified NLS protein increases linearly with RanGTP–importin β concentrations and promotes microtubule stabilization. Thus, microtubules emanating from centrosomes become increasingly stabilized the closer they get to chromosomes.
The next goal in Karsenti's view is to identify the NLS protein gradients to better understand how they promote local nucleation and biased microtubule growth toward chromosomes.