Yeast response to prolonged activation of the spindle assembly checkpoint

Faithful chromosome segregation during mitosis is fundamental for cell viability and genome stability. For a correct division, all kinetochores must be attached to the mitotic spindle and cohesion must be timely removed. Anaphase is triggered by the Anaphase Promoting Complex bound to its regulatory subunit Cdc20 (APC-Cdc20) that polyubiquitylates securin (Pds1 in budding yeast), whose role is to maintain inactive the protease separase (Esp1 in budding yeast) until anaphase onset. Once active, separase cleaves cohesin, thus triggering sister chromatid separation. Separase also promotes cyclinB proteolysis and mitotic exit due to its involvement in the Cdc14-early anaphase release (FEAR) pathway that promotes a partial activation of the Cdc14 phophatase, which is in turn key for CDK inactivation and mitotic exit. Cdc14 is maintained inactive throughout most of the cell cycle bound to its inhibitor Net1/Cfi1 and trapped in the nucleolus. At the beginning of anaphase Cdc14 is released from the nucleolus into the nucleus by the FEAR pathway; subsequently, Cdc14 is released also in the cytoplasm by the MEN (Mitotic Exit Network) pathway. In this way Cdc14 is fully active and can trigger mitotic exit by cyclinB-CDK inactivation. The Spindle Assembly Checkpoint (SAC) is a surveillance mechanism conserved in all eukaryotic organisms that ensures the correct segregation of the genetic material. In fact, it inhibits the metaphase to anaphase transition until all kinetochores are properly attached to the mitotic spindle by inactivating the APC-Cdc20 complex, thus providing the time for error correction. Cells do not arrest indefinitely upon SAC activation. After a variable period of time cells escape from the metaphase arrest also in the presence of a damaged mitotic spindle or faulty kinetochore attachments to spindle microtubules. This process is referred to as adaptation or mitotic slippage and is often involved in the resistance to chemotherapeutic compounds that target the mitotic spindle. In spite of its importance, the adaptation process is still little known. Within this context, the goals of my Ph.D. were: (1) to characterize the molecular mechanisms underlying SAC adaptation and (2) to search for factors involved in this process. For these purposes we used the yeast Saccharomyces cerevisiae as a model organism. (1) We characterized the adaptation process in either the presence or the absence of mitotic spindle perturbations. We depolymerized spindles by using two different drugs that alter microtubule dynamics, i.e. nocodazole and benomyl, whereas we induced SAC hyperactivation without spindle damage by overproducing Mad2 (GAL1-MAD2 cells), one of the key proteins for SAC signal generation and maintenance. We observed that in all the conditions cells are able to adapt, but with different kinetics. In particular, cells adapt faster in benomyl, while in nocodazole and with high levels of Mad2 cells need more time to slip out of mitosis. The few data available about SAC adaptation in higher eukaryotes indicate that SAC adaptation is accompanied by chromatid separation, a decrease in mitotic CDK activity and mitotic exit. Indeed, like in mammalian cells, yeast securin and cyclinB are degraded and sister chromatids are separated during adaptation. In addition, cyclinB stabilization, as well as Cdc20 and Cdc5 (polo kinase) inactivation, markedly delay adaptation, while the only yeast CKI (Sic1) is not involved in this process. Finally, when yeast cells adapt the SAC is likely to be turned off, as shown by the disassembly of the Mad1/Bub3 checkpoint complex. (2) To search for factors involved in SAC adaptation, we performed a genetic screen using GAL1-MAD2 cells. In particular, we screened for mutants that would remain arrested for prolonged times in mitosis upon MAD2 overexpression. We identified Rsc2, a non-essential component of the RSC chromatin remodelling complex, as a regulator of SAC adaptation in yeast. We demonstrated that RSCRsc2 is involved in fine tuning mitotic exit during the unperturbed cell cycle. Its activity becomes particularly important in conditions that would activate the SAC, as it contributes to cyclinB degradation. In the absence of Rsc2 Net1 phosphorylation and the early anaphase release of Cdc14 from the nucleolus are impaired, whereas expression of a dominant allele of CDC14 that loosens Net1 inhibition (CDC14TAB6-1) is sufficient to restore mitotic exit in conditions where Rsc2 becomes essential for this process. We further demonstrated that the ATPase activity of RSC is required for mitotic exit regulation, suggesting that its chromatin-remodelling activity is involved in this process. By studying possible genetic interactions between the RSC2 deletion and FEAR or MEN mutations, we found that RSC2 deletion confers synthetic lethality or sickness to MEN but not to FEAR mutants. Altogether, our data suggest that RSCRsc2 is a novel component of the FEAR pathway. Finally, we demonstrated that Rsc2 interacts in vivo and in vitro with the polo kinase Cdc5, which controls mitotic exit at different levels. Since RSC binds to acetylated histone tails, it is possible that histone transacetylases are also involved in SAC adaptation. We tested if the SAGA (Spt-Ada-Gcn5 Acetyltransferase) complex is involved in SAC adaptation by deleting ADA2 or GCN5 in yeast. Indeed, SAGA seems involved in adaptation, although the contribution of Ada2 and Gcn5 in the process differs depending on the conditions used to activate the SAC. Finally, since we found that upon treatment with benomyl (a microtubule destabilizer) cells adapt dividing nuclei, we wondered if SAC adaptation could be linked to the presence of cytoplasmic microtubules that are still partially detectable in these conditions. We therefore asked whether motor proteins and microtubule regulators are involved in mitotic slippage. Indeed, we found that in the absence of Kip2 and Bik1, which specifically bind to cytoplasmic microtubules, cells divide nuclei and exit mitosis slower than wild type cells, demonstrating that cytoplasmic microtubules and associated proteins could accelerate SAC adaptation. In conclusion, SAC adaptation is a very complex process whose timing probably depends on the interplay between different mechanisms. An important aim for a complete comprehension of this process, as well as for the development of new and more efficient cancer therapies, will be to identify novel factors implicated in adaptation and clarify how their function might be linked to one another.