I sat in on the “Mitosis and Meiosis” minisymposium on Tuesday morning. Co-chair Duane Compton (Dartmouth Medical School) kicked things off with a presentation of his lab’s recent studies on how microtubule stability effects accurate chromosome segregation. Kinetochores attach to dynamic microtubules during mitosis. Last year, Compton and colleagues showed that this microtubule instability is an important feature that allows errors in kinetochore-microtubule attachment to be corrected. Cancer cells have more stable attachments, leading to more frequent chromosome mis-segregation, a phenotype that can be induced in normal cells by depleting microtubule destabilizing proteins such as the kinesin MCAK. However, Compton’s group has recently shown that cells are unable to progress through mitosis if their kinetochore-microtubule attachments are too unstable – the microtubules turnover too fast to satisfy the mitotic checkpoint. A molecular switch involving the proteins CLASP1, astrin and Kif2b ensures that kinetochore-microtubule attachments have the right levels of stability to accurately proceed through mitosis: the destabilizing kinesin Kif2b binds to kinetochores in early mitosis to promote microtubule turnover and error correction. But Kif2b is displaced by astrin on bioriented kinetochores during metaphase, stabilizing the microtubule attachments to allow checkpoint inactivation and mitotic progression.
Kinetochore-microtubule attachments are also stabilized by the tension generated when sister chromatids are correctly attached to opposite spindle poles. This is at least partly due to the fact that, under tension, outer kinetochore proteins are pulled away from Aurora B kinase in the inner kinetochore, reducing their phosphorylation state, which allows them to bind microtubules more stably. Andrew Powers (from Chip Asbury’s lab at the University of Washington) presented data suggesting that tension also directly strengthens kinetochore-microtubule attachments. Powers and colleagues purified native kinetochore complexes (lacking Aurora B) from budding yeast and followed their binding to dynamic microtubules in vitro. By attaching the kinetochore complexes to beads and using an optical trap, the researchers applied a pulling force to the in vitro kinetochore-microtubule attachments and found that increased force led to more stable kinetochore-microtubule associations, possibly because tension seems to slow microtubule disassembly rates. Kinetochore-microtubule attachments therefore act somewhat like the catch bonds formed between integrins and extracellular matrix molecules, where increased force also generates a more stable association (though Powers is careful to point out that phosphoregulation by Aurora B also plays an important role at kinetochores). Powers was deservingly awarded the Norton B. Gilula Memorial award – sponsored by the Rockefeller University Press and the Journal of Cell Biology – for his work.
Finally, Julien Dumont - now with his own lab at the Insitut Curie - presented his postdoctoral work with Arshad Desai, revealing that chromosomes can segregate without the aid of kinetochores in C. elegans oocytes. Female worm meioses occur in the absence of centrosomes, and Dumont et al. found that kinetochores are also dispensable for segregation, although they do help orient the chromosomes. So how are chromosomes pulled apart in these divisions? The answer seems to be that they aren’t – they’re pushed apart instead. Dumont showed that, during anaphase, microtubules form in between separating sets of chromosomes rather than on either side of them where the spindle poles would normally be. A ring-shaped assembly of proteins such as Aurora B and the microtubule-binding protein CLASP also forms in this central region and, in the absence of CLASP, chromosomes aren’t pushed apart and segregated. One remaining question is how these central microtubules attach to chromosomes to push them in opposite directions.
Image © Schmidt et al., 2010.