During the cell cycle, duplicated DNA in S phase is segregated, in the form of chromatids, into two daughter cells in mitosis. The accuracy of chromosome segregation is essential as two daughter cells have the same genetic contents as the mother cell. Two major mechanisms are utilized by the cell to ensure accurate chromosome segregation. First, interactions between the dynamic microtubules and kinetochores, the proteinaceous structures built on centromeres of mitotic chromosomes that act as the attachment site for microtubules, serve as major forces to position each pair of chromosomes to the metaphase plate. Secondly, a surveillance system, known as the mitotic checkpoint, put the anaphase onset on hold until each pair of sister chromosomes are aligned at the metaphase plate and appropriately attached with microtubule plus ends by kinetochores. In the first part (Chapter 2) of this thesis, I illustrate the role of the auto-phosphorylation of BubR1, a mitotic checkpoint protein, in kinetochore-microtubule attachment and the mitotic checkpoint.

Using a phospho-specific antibody against the auto-phosphorylation site identified by mass spectrometry, I demonstrate that kinetochore-associated BubR1 phosphorylates itself in human cells in vivo and that this phosphorylation is dependent on its binding partner, the kinetochore-associated kinesin motor CENP-E. Studies using cells expressing a non-phosphorylatable BubR1 mutant revealed that the CENP-E-dependent BubR1 phosphorylation at unattached kinetochores is important for a full-strength mitotic checkpoint to prevent single chromosome loss. Furthermore, replacing endogenous BubR1 with the non-phosphorylatable BubR1 mutant or depletion of CENP-E, the BubR1 kinase activator, results in metaphase chromosome misalignment and increased incidents of syntelic attachments. Using indirect immunofluorescence, I have discovered a decreased level of Aurora B-mediated Ndc80 phosphorylation at the kinetochore of cells expressing the non-phosphorylatable BubR1 mutant, which might contribute to the alignment defect.

Moreover, expressing a phosphomimetic BubR1 mutant substantially reduces the incidence of polar chromosomes in CENP-E-depleted cells, further supporting a signaling cascade function of CENP-E and BubR1 on the kinetochore. Thus, the state of CENP-E-dependent BubR1 auto-phosphorylation in response to spindle microtubule capture by CENP-E is important for kinetochore functions in achieving accurate chromosome segregation. In the second part (Chapter 3), my colleague and I demonstrate a novel mechanism of mitotic spindle assembly in Xenopus egg extracts and mammalian cells. I show that the MRN (Mre11, Rad50, and Nbs1) complex is required for metaphase chromosome alignment. Consistent with the result of my colleague using Xenopus egg extracts, disruption of MRN function by depleting Mre11 using an inducible shRNA system, or Mre11 inhibitor mirin, triggers a metaphase delay and disrupts the RCC1-dependent Ran-GTP gradient. Addition of mirin to mammalian cells reduces RCC1 association with mitotic chromosomes and changes the confirmation of RCC1.

Thus, the MRN-CtIP pathway contributes to Ran-dependent mitotic spindle assembly by modulating RCC1 chromosome association. In summary, my novel findings have revealed a pair of molecular mechanisms not known previously, which are important to the mitosis field.