Compromising the activity of the spindle checkpoint permits mitotic exit in the presence of unattached kinetochores and, consequently, greatly increases the rate of aneuploidy in the daughter cells. The metazoan checkpoint mechanism is more complex than in yeast in that it requires additional proteins and activities besides the classical Mads and Bubs. Among these are Rod, Zw10, and Zwilch, components of a 700 Kdal complex (Rod/Zw10) that is required for recruitment of dynein/dynactin to kinetochores but whose role in the checkpoint is poorly understood. The dynamics of Rod and Mad2, examined in different organisms, show intriguing similarities as well as apparent differences. Here we simultaneously follow GFP-Mad2 and RFP-Rod and find they are in fact closely associated throughout early mitosis. They accumulate simultaneously on kinetochores and are shed together along microtubule fibers after attachment. Their behavior and position within attached kinetochores is distinct from that of BubR1; Mad2 and Rod colocalize to the outermost kinetochore region (the corona), whereas BubR1 is slightly more interior. Moreover, Mad2, but not BubR1, Bub1, Bub3, or Mps1, requires Rod/Zw10 for its accumulation on unattached kinetochores. Rod/Zw10 thus contributes to checkpoint activation by promoting Mad2 recruitment and to checkpoint inactivation by recruiting dynein/dynactin that subsequently removes Mad2 from attached kinetochores.

Mad2 has a key role in the spindle-assembly checkpoint (SAC) - the mechanism delaying anaphase onset until all chromosomes correctly attach to the spindle. Here, we show that unlike every other reported case of SAC inactivation in metazoans, mad2-null Drosophila are viable and fertile, and their cells almost always divide correctly despite having no SAC and an accelerated 'clock', which is caused by premature degradation of cyclin B. Mitosis in Drosophila does not need the SAC because correct chromosome attachment is achieved very rapidly, before even the cell lacking Mad2 can initiate anaphase. Experimentally reducing spindle-assembly efficiency renders the cells Mad2-dependent. In fact, the robustness of the SAC may generally mask minor mitotic defects of mutations affecting spindle function. The reported lethality of other Drosophila SAC mutations may be explained by their multifunctionality, and thus the 'checkpoint' phenotypes previously ascribed to these mutations should be considered the consequence of eliminating both the checkpoint and a second mitotic function.

SUMMARYIn female meiosis, chromosome missegregations lead to the generation of aneuploid oocytes and can cause the development of trisomies or infertility. Because mammalian female meiosis I is error prone, the full functionality of control mechanisms, such as the spindle assembly checkpoint (SAC), has been put into question. The SAC monitors the correct orientation, microtubule occupancy and tension on proteinaceous structures named kinetochores. Although it has been shown previously that the SAC exists in meiosis I, where attachments are monopolar, the role of microtubule occupancy for silencing the SAC and the importance of certain essential SAC components, such as the kinase Mps1, are unknown in mammalian oocytes. Using a conditional loss-of-function approach, we address the role of Mps1 in meiotic progression and checkpoint control in meiosis I. Our data demonstrate that kinetochore localization of Mps1 is required for the proper timing of prometaphase and is essential for SAC control, chromosome alignment and aurora C localization in meiosis I. The absence of Mps1 from kinetochores severely impairs chromosome segregation in oocyte meiosis I and, therefore, fertility in mice. In addition, we settle a long-standing question in showing that kinetochore-microtubule attachments are present in prometaphase I at a time when most of the SAC protein Mad2 disappears from kinetochores.