The transition from metaphase to anaphase within the cell cycle is a no-return process. It is initiated when the protease separase cleaves sister chromatid cohesion 1 / RAD21 homolog (SCC1), a subunit of the ring-shaped protein cohesin that holds the sister chromatids together. Once chromosome cohesion is released, the sister chromatids are pulled towards the spindle poles. To avoid segregation errors, human separase is held inactive by securin and cyclin-dependent kinases (CDK)1-cyclin B1 until the chromosomes are properly attached to the spindle microtubules. The formation of the complex separase-CDK1-cyclin B1 requires the separase residue serine 1126 to be phosphorylated (pS1126^separase). pS1126^separase is suggested to be dephosphorylated by the phosphatase protein phosphatase 2 (PP2A). Besides PP2A, also the cohesin-associated protein sororin is suggested to regulate separase and was shown to bind to separase in vivo.

This PhD thesis addresses the questions how separase recognizes its substrate SCC1 and how separase is regulated through interactions with PP2A and sororin.

Among the two previously reported SCC1 cleavage sites, we validate the first cleavage site and correct the second cleavage site. Consistent with in vivo studies with human SCC1 we show that SCC1 cleavage is stimulated if it is phosphorylated. We also identify conserved separase substrate phosphorylation sites that interact with five phosphate-binding pockets on separase. Furthermore, we observe that the cohesin subunit stromal antigen 2 (SA2) promotes SCC1 cleavage by recruiting SCC1 to separase.

Moreover, we show that securin and sororin compete with each other for binding to separase, and that sororin can only bind to securin-free separase. It was previously proposed that the conserved substrate pocket of the regulatory subunit of PP2A binds to the PP2A substrate motif on the flexible insert 2 of separase. Then, PP2A would dissociate from active separase, because active separase undergoes autocleavage, destroying the PP2A substrate motif. Supporting this hypothesis, we can only form a complex of PP2A and inactive separase that cannot do autocleavage. We observe that the main interaction between separase and PP2A is mediated by the conserved substrate pocket of PP2A. Overall, our study demonstrates for the first time that PP2A and sororin bind to separase in vitro.