Cohesin is needed for bipolar mitosis in human cells

Abstract

Multi-polar mitosis is strongly linked with aggressive cancers and it is a histological diagnostic of tumor-grade. However, factors that cause chromosomes to segregate to more than two spindle poles are not well understood. Here we show that cohesins Rad21, Smc1 and Smc3 are required for bipolar mitosis in human cells. After Rad21 depletion, chromosomes align at the metaphase plate and bipolar spindles assemble in most cases, but in anaphase the separated chromatids segregate to multiple poles. Time-lapse microscopy revealed that the spindle poles often become split in Rad21-depleted metaphase cells. Interestingly, exogenous expression of non-cleavable Rad21 results in multi-polar anaphase. Since cohesins are present at the spindle poles in mitosis, these data are consistent with a non-chromosomal function of cohesin.

Let's huddle to prevent a muddle: centrosome declustering as an attractive anticancer strategy

“…In support of this idea, all elements of the cohesin complex are found in the centrosome, and cohesin depletion results in centriole disengagement. [37][38][39][40] Furthermore, knockdown of an isoform of the cohesin-binding protein, SGO1 (shugoshin), causes centriole splitting. 41 Intriguingly, experimental evidence also suggests that cleaved cohesin promotes continued centriole engagement, 34,38,42 perhaps by delivering 'engagement factors' to the centrosome to preserve centriole cohesion until G1.…”

“…[37][38][39][40] Furthermore, knockdown of an isoform of the cohesin-binding protein, SGO1 (shugoshin), causes centriole splitting. 41 Intriguingly, experimental evidence also suggests that cleaved cohesin promotes continued centriole engagement, 34,38,42 perhaps by delivering 'engagement factors' to the centrosome to preserve centriole cohesion until G1. 38 Clearly, more data are awaited to clarify how centrioles remain engaged for the appropriate segment of the cell cycle and subsequently part ways in an exquisitely timed manner.…”

An update on cohesin function as a ‘molecular glue’ on chromosomes and spindles

“…In this special issue of Cell Cycle, three papers from the Clarke group further confirmed that the mammalian cohesin subunit, Scc1/Rad21, indeed localized at the spindle poles/centrosomes through very detailed live imaging and RNAi depletion assays. [43][44][45] Beauchene et al found that Rad21 is required for centrosome integrity in human cells independently of its role in chromosome cohesion. 43 Giménez-Abián et al showed that the cohesin protein Rad21 localizes to centrosomes in a manner that is dependent upon known regulators of sister chromatid cohesion as well as regulators of centrosome function.…”

“…These new data are consistent with a non-chromosomal function of cohesin. 44 In addition, Nakamura et al discovered that centrosomes harbor the protein Aki1 (Akt kinase-interacting protein 1), which promotes centriole integrity. Interestingly, Aki1 interaction with cohesin subunit Scc1 prevents centrioles from splitting too soon; Aki1 also binds another cohesin component, SA-2.…”

“…2). The specific role of the mitotic Rae1-SMC1 complex in vivo is further shown by the overexpression of SMC1, which caused aberrant multi-polar spindles, and by the depletion of SMC1 by RNAi, Given the recent flood of cohesin papers, 21,[49][50][51][52][53][54][55] in particular the new area of cohesin subunits on spindle poles/centrosomes; [35][36][37]41,[43][44][45][46] with nearly every finding in the field being made simultaneously by multiple labs, we may expect to see rapid progress in cohesin research related to spindle pole/centrosome biology. It will be a great future challenge to understand when and how different functional connections arise between components of cohesin during evolution.…”