Cohesin was a long sought-after protein complex: a glue for sister chromatids that is dissolved at the onset of anaphase thanks to cleavage of its SCC1 subunit by the protease separase. In budding yeast, cohesin lived up to its billing as a complex that was destroyed at anaphase onset.
But in humans cells its behavior was less convincing. The vast majority of cohesin comes off human chromosomes in prophase and prometaphase, coincident with the dissociation of sister chromatid arms. Only after a hard look is it apparent that a little cohesin is left at the centromere. Now Jan-Michael Peters and colleagues (Research Institute of Molecular Pathology (IMP), Vienna, Austria) have used a noncleavable SCC1 to show that cleavage of this residual SCC1 is indeed required for correct chromosome segregation and cytokinesis.
“Because so little SCC1 is cleaved in human cells it was unclear whether this would be of physiological relevance,” says first author Silke Hauf. Yet the noncleavable SCC1 induces many errors in chromosome segregation, causes many cleavage furrows to regress, and results in the formation of micronuclei and polyploid cells. Based on the aftermath of the aberrant mitoses, faithful sister chromatid separation is not necessary for mitotic exit or DNA re-replication.
Frank Uhlmann's group (Imperial Cancer Research Fund (ICRF), London, UK), meanwhile, has shown that in yeast separase is doing more than cleaving SCC1. Separase is also cleaving a protein called Slk19, whose cleaved form is necessary to stabilize the anaphase spindle. Uhlmann suggests that the absence of cleaved Slk9 is one reason why triggering anaphase by artificial cleavage of SCC1 results in an unstable spindle. ▪