Bub3 activation and inhibition of the APC/C

Accurate chromosome segregation depends on the precise temporal regulation of the anaphase-promoting complex/cyclosome (APC/C), an E3 ubiquitin ligase. For the onset of anaphase, the APC/C binds its co-activator Cdc20 and targets substrates for ubiquitination and proteasomal degradation.³ One substrate, securin, binds and inhibits separase, a protease that cleaves cohesin. The degradation of securin releases separase, promoting sister chromatid separation and anaphase onset. The activity of the APC/C is regulated throughout the cell cycle by several mechanisms including phosphorylation, sub-cellular localization, binding with different co-activators, and inhibition by the spindle checkpoint. In our recent study, we show that, in budding yeast, the spindle checkpoint protein Bub3 has a previously unknown role in activating the APC/C by facilitating the binding of APC/C and Cdc20.⁷

If kinetochores are not attached to spindle microtubules, the spindle checkpoint delays cells in metaphase by inhibiting APC/C^Cdc20 substrate ubiquitination, allowing additional time to correct errors in attachment.² We were surprised to find that cells that lack the spindle checkpoint protein Bub3 are slower to progress into anaphase.⁷ Our expectation was that cells lacking Bub3 would have either a normal or a faster anaphase onset due to the absence of spindle checkpoint activity. Furthermore, cells lacking another spindle checkpoint protein Bub1 also had a metaphase delay, but cells lacking Mad2 and Mad3 did not. Since all 4 proteins are required for spindle checkpoint signaling, these results suggest that Bub1 and Bub3 have an additional role in regulating the temporal progression of mitosis that is separate from their activity in signaling the spindle checkpoint.

Bub1 and Bub3 also have a known function in ensuring accurate chromosome segregation by recruiting Sgo1 to the kinetochore.⁵ Sgo1 then recruits other regulators for kinetochore biorientation. Cells that lack Bub1 or Bub3 are known to have an increase in chromosome mis-segregation, leading to aneuploid cells that have an extra or missing chromosome.⁶ These cells are often either prominently delayed in the cell cycle or dead,⁴ so we wanted to ensure that we were not including the slow-growing aneuploid cells in our analysis. We used time-lapse microscopy to monitor the first divisions of newly germinated bub3Δ cells sporulated from a bub3Δ heterozygote and timed the metaphase duration of normal divisions that did not produce aneuploid cells.⁷ We find that the euploid bub3Δ cells have a delay in anaphase onset, ruling out aneuploidy as a cause of the delay. Unfortunately, we could not perform the same analysis on bub1Δ cells due to poor spore viability of the bub1Δ heterozygote.

Since APC/C activity transitions cells from metaphase to anaphase, we analyzed the binding of the APC/C with its co-activator Cdc20 in wildtype, bub1Δ, bub3Δ, and sgo1Δ cells.⁷ Surprisingly, using co-immunoprecipitation, we found that bub3Δ cells but not bub1Δ or sgo1Δ cells have impaired binding of Cdc20 and the APC/C. These results suggest that although bub1Δ and bub3Δ cells both have a delay in anaphase onset, the cause of the delay may be different. In bub3Δ cells, the delay is likely due to less APC/C bound to its activator Cdc20. In support of this model, overexpression of Cdc20 suppressed the anaphase onset delay in bub3Δ but not in bub1Δ or sgo1Δ cells.

The kinetochore localization of Bub3 is important for normal APC/C^Cdc20 activity. A Bub3 mutant that failed to localize to the kinetochore was delayed in anaphase onset and also had impaired binding between Cdc20 and APC/C.⁷ Immunofluorescence of chromosome spreads showed that Bub3 and Cdc20 co-localized at the kinetochore, suggesting that Bub3 could interact with Cdc20 at the kinetochore to facilitate the binding of APC/C and Cdc20.

In summary, our results suggest that Bub3 activates APC/C^Cdc20 at the kinetochore to promote anaphase onset.⁷ Although this activity seems contradictory with Bub3's role of delaying anaphase onset during spindle checkpoint activation, the roles are consistent when considering work from previous studies. During spindle checkpoint signaling, the mitotic checkpoint complex (a complex of Bub3, Mad2, and Mad3) prevents APC/C^Cdc20 from ubiquitinating its substrates, however, APC/C^Cdc20 is still active to auto-ubiquitinate Cdc20.³ An in vitro study showed that during checkpoint activation, Bub3 promotes the binding of Cdc20 and APC/C for its auto-ubiquitination.¹ We propose a model in which kinetochore localized Bub3 facilitates the binding of Cdc20 and the APC/C in metaphase. (Fig. 1) When all kinetochores are properly attached to microtubules, this role of Bub3 allows the normal activity of APC/C^Cdc20 to ubiquitinate its substrates for timely progression into anaphase; in the presence of unattached kinetochores, Bub3 still facilitates the binding of APC/C and Cdc20 for auto-ubiquitination and also functions in the inhibitory mitotic checkpoint complex to prevent APC/C^Cdc20 substrate ubiquitination. These findings suggest that the spindle checkpoint protein Bub3 not only acts in a surveillance system but also coordinates the temporal regulation of cell-cycle events.

Figure 1.

Model of Bub3 regulation of APC/C^Cdc20 activity when the spindle checkpoint is either inactive or active.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.