Exit from mitosis in animal cells is substantially delayed when spindle assembly is inhibited, spindle bipolarity is disrupted, or when a monopolar spindle is formed. These observations have led to the proposal that animal cells have a ‘spindle assembly’ checkpoint for the metaphase-anaphase transition that monitors bipolar spindle organization. However, the existence of such a checkpoint is uncertain because perturbations in spindle organization can produce unattached kinetochores, which by themselves are known to delay anaphase onset. In this study we have tested if cells monitor bipolar spindle organization, independent of kinetochore attachment, by analyzing the duration of mitosis in sea urchin zygotes and vertebrate somatic cells containing multipolar spindles in which all kinetochores are attached to spindle poles. We found that sea urchin zygotes containing tripolar or tetrapolar spindles progressed from nuclear envelope breakdown to anaphase onset with normal timing. We also found that the presence of supernumerary, unpaired spindle poles did not greatly prolong mitosis. Observation of untreated PtK1 cells that formed tripolar or tetrapolar spindles revealed that they progressed through mitosis, on average, at the normal rate. More importantly, the interval between the bipolar attachment of the last monooriented chromosome and anaphase onset was normal. Thus, neither of these cell types can detect the presence of gross aberrations in spindle architecture that inevitably lead to aneuploidy. We conclude that animal cells do not have a checkpoint for the metaphase-anaphase transition that monitors defects in spindle architecture independent of the checkpoint that monitors kinetochore attachment to the spindle. For dividing cells in which spindle microtubule assembly is not experimentally compromised, we propose that the completion of kinetochore attachment is the event which limits the time of the metaphase-anaphase transition.
- © 1997 by Company of Biologists