The Molecular Mechanism Of Cell Cycle Regulation In Budd

Diabetes, Digestive, Kidney Diseases

Investigators

Linked publications & trials

Abstract

During cell division daughter cells must receive one and only one copy of each and every chromosome. To achieve this all chromosomes must be duplicated, and the duplicated products must be recognized as being equivalent, i.e. sister chromatids, so that they can be segregated to the two daughter cells. The identity of the sister chromatids is maintained by the presence of sister chromatid cohesion, which is dissolved only after all duplicated chromosomes have formed bipolar spindle attachments. A key regulator of this process in budding yeast is the protein Pds1. Functional homologues of Pds1, known as securins, are found in many organisms including human, where securin over expression is linked to genomic instability. Pds1 inhibits anaphase initiation by inactivating a conserved protease known as separase (Esp1 in budding yeast) that dissolves cohesion between sister chromatids. Esp1 becomes active only after Pds1 is degraded in a process that involves a ubiquitin ligase called the anaphase promoting complex/cyclosome (APC/C). During Pds1 ubiquitination the APC/C acts in conjunction with an associated subunit, the Cdc20 protein. Prior to our work (see section 1 below) the precise role of Cdc20 was not known although it was suspected that it provided specificity to the APC/C. Pds1 function becomes crucial in the presence of DNA or spindle damage, when it inhibits both anaphase initiation and mitotic exit until the damage is repaired. In the presence of DNA damage Pds1 is stabilized by Chk1-dependent phosphorylation, and we have recently determined the molecular mechanism for this stabilization (section 1). In addition to its role as an inhibitor of Esp1, Pds1 is also required for the efficient nuclear localization of Esp1. How Pds1 promotes Esp1's localization is still not known but in the past year we uncovered an important mechanism that regulates this process (section 2). Finally, we are in the process of carrying out two genetic screens that are expected to identify proteins involved in spindle function, DNA integrity and Esp1 localization (section 3). 1. The Pds1-Cdc20 interaction under normal conditions and in the presence of DNA damage. To identify Pds1 interacting proteins we carried out a yeast two-hybrid screen using full length Pds1 as bait. We identified Cdc20 as an interacting protein and demonstrated that the Pds1-Cdc20 interaction occurs both in vitro and in vivo. Cdc20 was previously shown to be an APC/C associated protein required for Pds1 ubiquitination but its exact role was unknown. We showed that the Pds1-Cdc20 interaction was dependent on Pds1's destruction box, thus identifying the element needed for substrate recognition by APC/C-Cdc20. We also showed that Pds1 does not interact with another APC/C associated subunit, Cdh1, further supporting a role for Cdc20 and Cdh1 in conferring APC/C's substrate specificity. Finally we showed that in the presence of spindle damage, when Pds1 ubiquitination is inhibited, Cdc20 still binds Pds1, suggesting that the spindle damage checkpoint inhibits APC/C activity rather than substrate recognition. This study was carried out in collaboration with Dr. Chris Hardy (Washington University, St. Louis MO) and was published in Current Biology (Hilioti et al, Curr. Biol., 11, 1347-1352, 2001). In the presence of DNA damage Pds1 is phosphorylated by the Chk1 kinase. To investigate the role of this phosphorylation in mediating the checkpoint arrest we examined whether it is sufficient to block ubiquitination in vitro by the APC/C. In collaboration with Dr. Hongtao Yu (University of Texas Southwestern Medical Center, Dallas TX) we found that to be the case, suggesting that Chk1 phosphorylation is stabilizing Pds1 by preventing its ubiquitination. To further understand the mechanism behind this inhibition we determined that the ubiquitination reaction is inhibited because the DNA damage induced phosphorylation of Pds1 prevents its association with Cdc20, which, as we showed earlier, is a necessary step in the ubiquitination process. These results allowed us to examine the molecular mechanism by which cells recover from the DNA damage induced arrest and led us to the discovery that a phosphatase, the identity of which is currently being sought, is an essential component of this process. A manuscript summarizing these results is currently in preparation. 2. Cdc28-mediated phosphorylation of Pds1 is required to promote nuclear localization, and hence activation, of Esp1. We recently discovered that Pds1 is also phosphorylated in a DNA damage-independent manner. The presence of consensus phosphorylation sites in Pds1 for the cyclin-dependent kinase Cdc28 prompted us to examine whether Pds1 is a substrate for Cdc28. We found that Pds1 is indeed a Cdc28 substrate, both in vivo and in vitro. The Cdc28 consensus phosphorylation sites were mutated and the resultant pds1 mutants were found to confer an esp1 mutant-like phenotype. Our analysis revealed that these pds1 mutants were defective in localizing Esp1 to the nucleus, thus failing to separate sister chromatids, and that the Cdc28-mediated phosphorylation is necessary for an efficient Pds1-Esp1 interaction. These findings uncover a role for Cdc28 in anaphase initiation, they reveal a novel regulatory pathway that controls the Pds1-Esp1 interaction and they shed light on the mechanism of Esp1's nuclear localization. This study was recently published in Genes and Development (Agarwal R. and Cohen-Fix, O. Genes Dev 16, 1371-1382 (2002)). 3. The role of Pds1 as a mitotic regulator. Pds1 is not essential for viability but there are several conditions under which cells cannot survive in the absence of Pds1. These include conditions that lead to spindle defect or induce DNA damage. In addition, we hypothesize that Pds1-independent defects in Esp1's activation or nuclear localization will also render Pds1 essential because under these conditions cell will have to depend on the ability of Pds1 to promote Esp1 activation. To identify proteins required for spindle function, DNA integrity and Esp1 activation we carried out two genetic screens for mutants whose viability depends on Pds1 (known as a synthetic lethality screen). The first screen is based on a colony-sectoring assay in which cells that are mutated at the PDS1 locus and at a second, randomly induced site, cannot survive in the absence of a plasmid carrying the PDS1 gene. The presence or absence of this plasmid is reflected in colony color (plasmid loss results in the appearance of a sector), and cells that cannot lose the PDS1 plasmid will give rise to colonies that don't sector. So far we isolated 14 complementation groups that show synthetic lethality with a pds1 deletion. Of these, several mutants have been identified and were shown to require Pds1 for viability due to spindle defects. Of particular interest was our finding that a gene encoding for protein required for mitotic cyclin degradation, CDH1, is also synthetically lethal with the deletion of PDS1. We found that a cdh1 mutant requires the spindle checkpoint, and Pds1 in particular, to prevent chromosome mis-segregation. These and additional results suggest that an untimely increase in cyclin-Cdk activity can lead to genomic instability. A manuscript describing these results is currently under preparation. We are conducting a second genetic screen in collaboration with Dr. Mike Tyers (Samuel Lunenfeld Research Institute, Toronto Canada), in which a deletion of Pds1 is combined, individually, with a collection of strains each carrying a deletion of one out of the 4000 non-essential genes in budding yeast. So far this screen has yielded several dozen mutants that require Pds1 for viability. We are in the process of determining whether the defects that render Pds1 essential have to do with spind