Centromere identity and function is determined by the specific localization of CenH3 (reviewed in [1-7]). Several mechanisms regulate centromeric CenH3 localization, including proteasome-mediated degradation that, both in budding yeast and Drosophila, regulates CenH3 levels and prevents promiscuous misincorporation throughout chromatin [8, 9]. CenH3(CENP-A) proteolysis has also been reported in senescent human cells [10] or upon infection with herpes simplex virus 1 [11]. Little is known, however, about the actual mechanisms that regulate CenH3 proteolysis. Recent work in budding yeast identified Psh1 as an E3-ubiquitin ligase that mediates degradation of CenH3(Cse4p) [12, 13], but E3-ligases regulating CenH3 stability in metazoans are unknown. Here, we report that the F box protein partner of paired (Ppa), which is a variable subunit of the main E3-ligase SCF [14-17], mediates CenH3(CID) stability in Drosophila. Our results show that Ppa depletion results in increased CenH3(CID) levels. Ppa physically interacts with CenH3(CID) through the CATD(CID) that, in the fly, mediates Ppa-dependent CenH3(CID) stability. Altogether, these results strongly suggest that, in Drosophila, SCF(Ppa) regulates CenH3(CID) proteolysis. Interestingly, most known SCF complexes are inactive when, at mitosis, de novo CenH3(CID) deposition takes place at centromeres, suggesting that, in Drosophila, CenH3(CID) deposition and proteolysis are synchronized events.

Barrier-to-Autointegration Factor (BAF) is a conserved nuclear envelope (NE) component that binds chromatin and helps its anchoring to the NE. Cycles of phosphorylation and dephosphorylation control BAF function. Entering mitosis, phosphorylation releases BAF from chromatin and facilitates NE-disassembly. At mitotic exit, PP2A-mediated dephosphorylation restores chromatin binding and nucleates NE-reassembly. Here, we show that in Drosophila a small fraction of BAF (cenBAF) associates with centromeres. We also find that PP4 phosphatase, which is recruited to centromeres by CENP-C, prevents phosphorylation and release of cenBAF during mitosis. cenBAF is necessary for proper centromere assembly and accurate chromosome segregation, being critical for mitosis progression. Disrupting cenBAF localization prevents PP2A inactivation in mitosis compromising global BAF phosphorylation, which in turn leads to its persistent association with chromatin, delays anaphase onset and causes NE defects. These results suggest that, together with PP4 and CENP-C, cenBAF forms a centromere-based mechanism that controls chromosome segregation and mitosis progression.

BackgroundCentromere identity is determined epigenetically by deposition of CenH3, a centromere-specific histone H3 variant that dictates kinetochore assembly. The molecular basis of the contribution of CenH3 to centromere/kinetochore functions is, however, incompletely understood, as its interactions with the rest of centromere/kinetochore components remain largely uncharacterised at the molecular/structural level.Principal FindingsHere, we report on the contribution of Drosophila CenH3CID to recruitment of BubR1, a conserved kinetochore protein that is a core component of the spindle attachment checkpoint (SAC). This interaction is mediated by the N-terminal domain of CenH3CID (NCenH3CID), as tethering NCenH3CID to an ectopic reporter construct results in BubR1 recruitment and BubR1-dependent silencing of the reporter gene. Here, we also show that this interaction depends on a short arginine (R)-rich motif and that, most remarkably, it appears to be evolutionarily conserved, as tethering constructs carrying the highly divergent NCenH3 of budding yeast and human also induce silencing of the reporter. Interestingly, though NCenH3 shows an exceedingly low degree of conservation, the presence of R-rich motives is a common feature of NCenH3 from distant species. Finally, our results also indicate that two other conserved sequence motives within NCenH3CID might also be involved in interactions with kinetochore components.ConclusionsThese results unveil an unexpected contribution of the hypervariable N-domain of CenH3 to recruitment of kinetochore components, identifying simple R-rich motives within it as evolutionary conserved structural determinants involved in BubR1 recruitment.