Abstract
Two mechanisms safeguard the bipolar attachment of chromosomes in mitosis. A correction mechanism destabilizes erroneous attachments that do not generate tension across sister kinetochores [1]. In response to unattached kinetochores, the mitotic checkpoint delays anaphase onset by inhibiting the anaphase-promoting complex/cyclosome (APC/C(Cdc20)) [2]. Upon satisfaction of both pathways, the APC/C(Cdc20) elicits the degradation of securin and cyclin B [3]. This liberates separase triggering sister chromatid disjunction and inactivates cyclin-dependent kinase 1 (Cdk1) causing mitotic exit. How eukaryotic cells avoid the engagement of attachment monitoring mechanisms when sister chromatids split and tension is lost at anaphase is poorly understood [4]. Here we show that Cdk1 inactivation disables mitotic checkpoint surveillance at anaphase onset in human cells. Preventing cyclin B1 proteolysis at the time of sister chromatid disjunction destabilizes kinetochore-microtubule attachments and triggers the engagement of the mitotic checkpoint. As a consequence, mitotic checkpoint proteins accumulate at anaphase kinetochores, the APC/C(Cdc20) is inhibited, and securin reaccumulates. Conversely, acute pharmacological inhibition of Cdk1 abrogates the engagement and maintenance of the mitotic checkpoint upon microtubule depolymerization. We propose that the simultaneous destruction of securin and cyclin B elicited by the APC/C(Cdc20) couples chromosome segregation to the dissolution of attachment monitoring mechanisms during mitotic exit.