spin via P-T3-H3 is essential for Aurora Bmediated phosphorylation of kinetochore substrates and for checkpoint function. Forced localization of Aurora B activity at kinetochores or inhibition of phosphatases that oppose Aurora B activity at kinetochores is sufficient to restore the phosphorylation status of substrates or the localization of SAC components. 5-ITu affects Aurora B activity on the outer kinetochore An 80-residue N-terminal segment of Hec1/Ndc80 is important for kinetochoremicrotubule binding and is a well-established prometaphase target of Aurora B. We tested if 5-ITu suppressed the accumulation of the phosphorylated form of Ser44 of Hec1/Ndc80, which depends on Aurora B. Indeed, 5-ITu caused an 50% reduction in the kinetochore levels of P-S44-Hec1. This effect was inferior to that caused by direct inhibition of Aurora B kinase with Hesperadin. However, contrarily to the latter, it was rescued by the CENP-BINCENP construct, suggesting that the accumulation of normal levels of P-S44-Hec1 at kinetochores requires Aurora B activity locally. Effects of 5-ITu on the spindle assembly checkpoint Kinetochore recruitment of SAC proteins is considered important for optimal checkpoint operation. Because 5-ITu appeared to affect kinetochore recruitment of at least two SAC proteins, we asked if cells challenged with 5-ITu were able to 
maintain a checkpoint-dependent arrest in mitosis. We therefore tested the effects of 5-ITu on the SAC in two distinct assays. First, we generated a population of mitotic cells by mitotic shake-off of cycling cells treated with 3.3 M nocodazole for 6 h. Control cells maintained the arrest for at least 11 h. Conversely, cells treated with 5-ITu underwent checkpoint override at times that depended on drug concentration, with effects that became evident between 0.5 and 1.0 M. At 10 M, the adverse effect of 5-ITu on checkpoint duration was as pervasive as that caused by 1 M Hesperadin. As a positive control for these experiments, we also monitored the effects of the Cdk1 inhibitor flavopiridol, which caused mitotic exit in less than 2 h at 1-M or higher concentrations. Second, we asked if 5-ITu was able to override the SAC arrest caused by expression of the CENP-BINCENP chimera in the absence of spindle poisons. This effect, which has been discussed before, is probably caused by alignment errors after relocalization of Aurora B close to the outer kinetochores. Cells expressing the CENP-BINCENP chimera and control cells were first arrested in nocodazole. Nocodazole was then washed out, and the time of anaphase onset was monitored by time-lapse video microscopy. Control cells underwent anaphase at 79 57 min, whereas cells expressing CENP-BINCENP underwent anaphase at 250 148 min, confirming that expression of CENP-BINCENP causes a mitotic arrest in otherwise normally dividing PubMed ID:http://www.ncbi.nlm.nih.gov/pubmed/19834025 cells. We therefore asked if 5-ITu was able to override such arrest. As a control, we used Reversine, a small-molecule ATPcompetitive inhibitor of Mps1. Both Reversine and 5-ITu abrogated the mitotic arrest caused by 277 Characterization of 5-iodotubercidin as Haspin inhibitor De Antoni et al. expression of CENP-BINCENP, with mean times of mitotic exit of 30 16 min and 29 12 min, respectively. Thus, 5-ITu overrides the mitotic arrest caused by the expression of the CENP-BINCENP fusion HC-067047 web protein. Collectively, these results indicate that 5-ITu is a rather potent SAC inhibitor. Forcing centromeric localization of Aurora B does not restore a f
