Deregulated centrosome duplication can result in genetic instability and contribute to tumorigenesis. Here, we show that centrosome duplication is regulated by the activity of an E3-ubiquitin ligase that employs the F-box protein FBXW5 (ref. 3) as its targeting subunit. Depletion of endogenous FBXW5 or overexpression of an F-box-deleted mutant version results in centrosome overduplication and formation of multipolar spindles. We identify the centriolar protein HsSAS-6 (refs 4,5) as a critical substrate of the SCF-FBXW5 complex. FBXW5 binds HsSAS-6 and promotes its ubiquitylation in vivo. The activity of SCF-FBXW5 is in turn negatively regulated by Polo-like kinase 4 (PLK4), which phosphorylates FBXW5 at Ser 151 to suppress its ability to ubiquitylate HsSAS-6. FBXW5 is a cell-cycle-regulated protein with expression levels peaking at the G1/S transition. We show that FBXW5 levels are controlled by the anaphase-promoting (APC/C) complex, which targets FBXW5 for degradation during mitosis and G1, thereby helping to reset the centrosome duplication machinery. In summary, we show that a cell-cycle-regulated SCF complex is regulated by the kinase PLK4, and that this in turn restricts centrosome re-duplication through degradation of the centriolar protein HsSAS-6.

No abstract

Chinese hamster ovary (CHO) cells are the number one production system for therapeutic proteins. A pre-requirement for their use in industrial production of biopharmaceuticals is to be clonal, thus originating from a single cell in order to be phenotypically and genomically identical. In the present study it was evaluated whether standard procedures, such as the generation of a recombinant cell line in combination with selection for a specific and stable phenotype (expression of the recombinant product) or subcloning have any impact on karyotype stability or homogeneity in CHO cells. Analyses used were the distribution of chromosome counts per cell as well as chromosome painting to identify specific karyotype patterns within a population. Results indicate that subclones both of the host and the recombinant cell line are of comparable heterogeneity and (in)stability as the original pool. In contrast, the rigorous selection for a stably expressing phenotype generated cell lines with fewer variation and more stable karyotypes, both at the level of the sorted pool and derivative subclones. We conclude that the process of subcloning itself does not contribute to an improved karyotypic homogeneity of a population, while the selection for a specific cell property inherently can provide evolutionary pressure that may lead to improved chromosomal stability as well as to a more homogenous population.

No abstract

No abstract