Cdk Consensus Sites Research Articles

Cdc14 phosphatase counteracts Cdk-dependent Dna2 phosphorylation to inhibit resection during recombinational DNA repair

Cyclin-dependent kinase (Cdk) stimulates resection of DNA double-strand breaks ends to generate single-stranded DNA (ssDNA) needed for recombinational DNA repair. Here we show in Saccharomyces cerevisiae that lack of the Cdk-counteracting phosphatase Cdc14 produces abnormally extended resected tracts at the DNA break ends, involving the phosphatase in the inhibition of resection. Over-resection in the absence of Cdc14 activity is bypassed when the exonuclease Dna2 is inactivated or when its Cdk consensus sites are mutated, indicating that the phosphatase restrains resection by acting through this nuclease. Accordingly, mitotically activated Cdc14 promotes Dna2 dephosphorylation to exclude it from the DNA lesion. Cdc14-dependent resection inhibition is essential to sustain DNA re-synthesis, thus ensuring the appropriate length, frequency, and distribution of the gene conversion tracts. These results establish a role for Cdc14 in controlling the extent of resection through Dna2 regulation and demonstrate that the accumulation of excessively long ssDNA affects the accurate repair of the broken DNA by homologous recombination.

Abstract P5-05-06: Progesterone receptor (PR) isoform-specific expansion of breast cancer stem-like cells occurs through a phospho-PR/FOXO1 driven signaling axis

Abstract Luminal breast cancers account for ˜75% of cases and are positive for progesterone receptor (PR) and estrogen receptor (ER) expression. PR is a classical ER-target gene and is used as a biomarker of ER activity; however, a growing body of evidence supports the role of PR as an important modifier of ER actions and a key driver of luminal breast cancer progression. Progesterone signaling is mediated by two PR isoforms: full-length PR-B and truncated PR-A, which lacks the N-terminal 164 amino acids. Little is known about PR isoform-specific actions in PR-expressing breast tumors, given that total PR expression is measured clinically. Herein, we sought to identify phenotypic differences in luminal breast cancer cells (T47D) overexpressing PR-A (T47D-YA) or PR-B (T47D-YB). We demonstrate that PR-B expression is required for anchorage-independent colony formation, while PR-A expressing cells fail to proliferate in soft agar. PR-B driven proliferation has been mapped to PR phosphorylation events, which include MAPK or CDK consensus sites such as Ser294. We demonstrate that in contrast to previous reports, PR-A is well phosphorylated at Ser294 in response to progestins (e.g. R5020) using our custom phospho-PR (Ser294) polyclonal antibody. Interestingly, Ser294 phosphorylation of PR-A occurs more rapidly and robustly following hormone treatment compared to PR-B expressing cells. Our findings indicate that PR-A is a dominant driver of stem-like expansion in breast cancer cells. PR-A tumorspheres exhibited enriched ALDH+ and CD44+/CD24- populations compared to PR-B and promoted heightened gene expression of stem-like genes (e.g. FOXO1). We demonstrate that the PR target gene and co-activator FOXO1 promotes both PR-A and PR-B phosphorylation at Ser294 and augments tumorsphere formation. Direct inhibition of FOXO1 levels abrogates phospho-PR (Ser294) levels and tumorsphere formation in PR expressing cells. Finally, we show that Ser294 is required for PR-A induced expression of stem-like genes (e.g. FOXO1) and stem-like behavior as measured by ALDH+/CD44+ tumorspheres. Taken together, our data reveal unique functions of PR isoforms as modulators of distinct and opposing pathways (i.e. proliferation versus stem-like expansion) in luminal breast cancer models. A clear understanding of PR isoform-specific actions, phosphorylation events, and the role of co-factors such as FOXO1 may lead to novel biomarkers of advanced tumor behavior and reveal new approaches to pharmacologically target PR isoforms in luminal breast cancers. Citation Format: Truong TH, Dwyer AR, Diep CH, Lange CA. Progesterone receptor (PR) isoform-specific expansion of breast cancer stem-like cells occurs through a phospho-PR/FOXO1 driven signaling axis [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P5-05-06.

Cyclin D1/Cdk4 increases the transcriptional activity of FOXM1c without phosphorylating FOXM1c

Anders et al. (2011) [11] reported that cyclinD1/Cdk4 and cyclinD3/Cdk6 enhance the transcriptional activity of FOXM1c by phosphorylating its TAD. They defined 12 Cdk consensus sites as essential for the activation of FOXM1c by cyclinD1/Cdk4 and cyclinD3/Cdk6 and stated that the 12 Cdk-sites are positioned within the TAD of FOXM1c. In contrast, this study demonstrates that all potential cyclin/Cdk phosphorylation sites S/T-P of FOXM1c are located outside its TAD so that the TAD of FOXM1c contains no potential cyclin/Cdk site, which excludes a phosphorylation of the FOXM1c-TAD by cyclinD1/Cdk4 and cyclinD3/Cdk6. This study shows that the activation of FOXM1c by cyclinD1/Cdk4 is lost without removal of any cyclin/Cdk site and gained without addition of any cyclin/Cdk site because it depends on a FOXM1c domain with no potential cyclin/Cdk site, namely on the interaction domain for the tumor suppressor RB, which binds to and represses FOXM1c. CyclinD1/Cdk4 activates FOXM1c because cyclinD1/Cdk4 releases FOXM1c from its repression by RB through removal of RB from FOXM1c. For this purpose, cyclinD1/Cdk4 phosphorylates only RB, but not FOXM1c, so that cyclinD1/Cdk4 increases the transcriptional activity of FOXM1c without phosphorylating FOXM1c and activates FOXM1c independently of cyclin/Cdk phosphorylation sites in FOXM1c. In summary, this study changes the model of Anders et al. (2011) [11] completely because it disproves their central conclusion that cyclinD1/Cdk4 and cyclinD3/Cdk6 enhance the transcriptional activity of FOXM1c by phosphorylating its TAD at the 12 Cdk-sites.

Cdc14 Phosphatases Preferentially Dephosphorylate a Subset of Cyclin-dependent kinase (Cdk) Sites Containing Phosphoserine

Mitotic cell division is controlled by cyclin-dependent kinases (Cdks), which phosphorylate hundreds of protein substrates responsible for executing the division program. Cdk inactivation and reversal of Cdk-catalyzed phosphorylation are universal requirements for completing and exiting mitosis and resetting the cell cycle machinery. Mechanisms that define the timing and order of Cdk substrate dephosphorylation remain poorly understood. Cdc14 phosphatases have been implicated in Cdk inactivation and are thought to be generally specific for Cdk-type phosphorylation sites. We show that budding yeast Cdc14 possesses a strong and unusual preference for phosphoserine over phosphothreonine at Pro-directed sites in vitro. Using serine to threonine substitutions in the Cdk consensus sites of the Cdc14 substrate Acm1, we demonstrate that phosphoserine specificity exists in vivo. Furthermore, it appears to be a conserved property of all Cdc14 family phosphatases. An invariant active site residue was identified that sterically restricts phosphothreonine binding and is largely responsible for phosphoserine selectivity. Optimal Cdc14 substrates also possessed a basic residue at the +3 position relative to the phosphoserine, whereas substrates lacking this basic residue were not effectively hydrolyzed. The intrinsic selectivity of Cdc14 may help establish the order of Cdk substrate dephosphorylation during mitotic exit and contribute to roles in other cellular processes.

Inactivation and Disassembly of the Anaphase-Promoting Complex during Human Cytomegalovirus Infection Is Associated with Degradation of the APC5 and APC4 Subunits and Does Not Require UL97-Mediated Phosphorylation of Cdh1

Infection of quiescent cells by human cytomegalovirus (HCMV) elicits severe cell cycle deregulation, resulting in a G(1)/S arrest, which can be partly attributed to the inactivation of the anaphase-promoting complex (APC). As we previously reported, the premature phosphorylation of its coactivator Cdh1 and/or the dissociation of the core complex can account for the inactivation. We have expanded on these results and further delineated the key components required for disabling the APC during HCMV infection. The viral protein kinase UL97 was hypothesized to phosphorylate Cdh1, and consistent with this, phosphatase assays utilizing a virus with a UL97 deletion mutation (ΔUL97 virus) indicated that Cdh1 is hypophosphorylated at early times in the infection. Mass spectrometry analysis demonstrated that UL97 can phosphorylate Cdh1 in vitro, and the majority of the sites identified correlated with previously characterized cyclin-dependent kinase (Cdk) consensus sites. Analysis of the APC core complex during ΔUL97 virus infection showed APC dissociation occurring at the same time as during infection with wild-type virus, suggesting that the UL97-mediated phosphorylation of Cdh1 is not required for this to occur. Further investigation of the APC subunits showed a proteasome-dependent loss of the APC5 and APC4 subunits that was temporally associated with the disassembly of the APC. Immediate early viral gene expression was not sufficient for the degradation of APC4 and APC5, indicating that a viral early gene product(s), possibly in association with a de novo-synthesized cellular protein(s), is involved.

Sequential Counteracting Kinases Restrict an Asymmetric Gene Expression Program to early G1

Gene expression is restricted to specific times in cell division and differentiation through close control of both activation and inactivation of transcription. In budding yeast, strict spatiotemporal regulation of the transcription factor Ace2 ensures that it acts only once in a cell's lifetime: at the M-to-G1 transition in newborn daughter cells. The Ndr/LATS family kinase Cbk1, functioning in a system similar to metazoan hippo signaling pathways, activates Ace2 and drives its accumulation in daughter cell nuclei, but the mechanism of this transcription factor's inactivation is unknown. We found that Ace2's nuclear localization is maintained by continuous Cbk1 activity and that inhibition of the kinase leads to immediate loss of phosphorylation and export to the cytoplasm. Once exported, Ace2 cannot re-enter nuclei for the remainder of the cell cycle. Two separate mechanisms enforce Ace2's cytoplasmic sequestration: 1) phosphorylation of CDK consensus sites in Ace2 by the G1 CDKs Pho85 and Cdc28/CDK1 and 2) an unknown mechanism mediated by Pho85 that is independent of its kinase activity. Direct phosphorylation of CDK consensus sites is not necessary for Ace2's cytoplasmic retention, indicating that these mechanisms function redundantly. Overall, these findings show how sequential opposing kinases limit a daughter cell specific transcriptional program to a brief period during the cell cycle and suggest that CDKs may function as cytoplasmic sequestration factors.

Ionizing radiation-dependent and independent phosphorylation of the 32-kDa subunit of replication protein A during mitosis

The human single-stranded DNA-binding protein, replication protein A (RPA), is regulated by the N-terminal phosphorylation of its 32-kDa subunit, RPA2. RPA2 is hyperphosphorylated in response to various DNA-damaging agents and also phosphorylated in a cell-cycle-dependent manner during S- and M-phase, primarily at two CDK consensus sites, S23 and S29. Here we generated two monoclonal phospho-specific antibodies directed against these CDK sites. These phospho-specific RPA2-(P)-S23 and RPA2-(P)-S29 antibodies recognized mitotically phosphorylated RPA2 with high specificity. In addition, the RPA2-(P)-S23 antibody recognized the S-phase-specific phosphorylation of RPA2, suggesting that during S-phase only S23 is phosphorylated, whereas during M-phase both CDK sites, S23 and S29, are phosphorylated. Immunofluorescence microscopy revealed that the mitotic phosphorylation of RPA2 starts at the onset of mitosis, and dephosphorylation occurs during late cytokinesis. In mitotic cells treated with ionizing radiation (IR), we observed a rapid hyperphosphorylation of RPA2 in addition to its mitotic phosphorylation at S23 and S29, associated with a significant change in the subcellular localization of RPA. Our data also indicate that the RPA2 hyperphosphorylation in response to IR is facilitated by the activity of both ATM and DNA-PK, and is associated with activation of the Chk2 pathway.

Colocalization of Sensors Is Sufficient to Activate the DNA Damage Checkpoint in the Absence of Damage

Previous work on the DNA damage checkpoint in Saccharomyces cerevisiae has shown that two complexes independently sense DNA lesions: the kinase Mec1-Ddc2 and the PCNA-like 9-1-1 complex. To test whether colocalization of these components is sufficient for checkpoint activation, we fused these checkpoint proteins to the LacI repressor and artificially colocalized these fusions by expressing them in cells harboring Lac operator arrays. We observed Rad53 and Rad9 phosphorylation, Sml1 degradation, and metaphase delay, demonstrating that colocalization of these sensors is sufficient to activate the checkpoint in the absence of DNA damage. Our tethering system allowed us to establish that CDK functions in the checkpoint pathway downstream of damage processing and checkpoint protein recruitment. This CDK dependence is likely, at least in part, through Rad9, since mutation of CDK consensus sites compromised its checkpoint function.

A Quantitative Results-driven Approach to Analyzing Multisite Protein Phosphorylation

Multisite protein phosphorylation appears to be quite common. Nevertheless our understanding of how multiple phosphorylation events regulate the function of a protein is limited in many cases. The ability to measure temporal changes in the site-specific phosphorylation profile of a protein in response to a given stimulus or cellular activity would provide an immediate indication of the functional significance of any phosphorylation site to a given process. Here we describe a mass spectrometry-based method to identify functionally relevant phosphorylation sites on a protein. It combines stable isotope labeling with a highly selective mass spectrometry analysis to detect and quantitate phosphorylation sites in response to a cellular signal. This approach requires no a priori knowledge of the phosphorylation state of the protein, does not require purification of phosphopeptides, and reliably detects substoichiometric levels of phosphorylation. Following a review of the quantitative results, only those phosphorylation sites that show a change in relative abundance are selected for identification and further study. We used this results-driven approach to study phosphorylation of the budding yeast transcription factor Pho4 in response to phosphate starvation. Phosphorylation of Pho4 on five cyclin-dependent kinase (Cdk) consensus sites has been shown to regulate the transcriptional activity of Pho4 in response to changes in environmental phosphate levels. Here we show that in phosphate-rich medium Pho4 is phosphorylated on at least 15 distinct sites including the five Cdk sites described previously. In excellent agreement with the known mechanism for regulation of Pho4 we found that phosphorylation at all five of the Cdk sites was repressed in phosphate-depleted medium. In addition to these five sites, we identified four novel phosphorylation sites that were also responsive to changes in phosphate availability. Selecting a limited number of Pho4 phosphorylation sites, we performed a more detailed kinetic analysis using an isotope-free strategy. We used LC-MS with selected reaction monitoring to greatly improve the accuracy, sensitivity, and dynamic range of the subsequent experiments. A detailed analysis of the cell-based phosphorylation at the selected Pho4 sites confirmed an apparent site preference for the Pho80-Pho85 cyclin-cyclin-dependent kinase complex.

Identification of Mcm2 Phosphorylation Sites by S-phase-regulating Kinases

Minichromosome maintenance 2-7 proteins play a pivotal role in replication of the genome in eukaryotic organisms. Upon entry into S-phase several subunits of the MCM hexameric complex are phosphorylated. It is thought that phosphorylation activates the intrinsic MCM DNA helicase activity, thus allowing formation of active replication forks. Cdc7, Cdk2, and ataxia telangiectasia and Rad3-related kinases regulate S-phase entry and S-phase progression and are known to phosphorylate the Mcm2 subunit. In this work, by in vitro kinase reactions and mass spectrometry analysis of the products, we have mapped phosphorylation sites in the N terminus of Mcm2 by Cdc7, Cdk2, Cdk1, and CK2. We found that Cdc7 phosphorylates Mcm2 in at least three different sites, one of which corresponds to a site also reported to be phosphorylated by ataxia telangiectasia and Rad3-related. Three serine/proline sites were identified for Cdk2 and Cdk1, and a unique site was phosphorylated by CK2. We raised specific anti-phosphopeptide antibodies and found that all the sites identified in vitro are also phosphorylated in cells. Importantly, although all the Cdc7-dependent Mcm2 phosphosites fluctuate during the cell cycle with kinetics similar to Cdc7 kinase activity and Cdc7 protein levels, phosphorylation of Mcm2 in the putative cyclin-dependent kinase (Cdk) consensus sites is constant during the cell cycle. Furthermore, our analysis indicates that the majority of the Mcm2 isoforms phosphorylated by Cdc7 are not stably associated with chromatin. This study forms the basis for understanding how MCM functions are regulated by multiple kinases within the cell cycle and in response to external perturbations.

Phosphorylation of the Chromosomal Passenger Protein Bir1 Is Required for Localization of Ndc10 to the Spindle during Anaphase and Full Spindle Elongation

The Saccharomyces cerevisiae inhibitor of apoptosis (IAP) repeat protein Bir1 localizes as a chromosomal passenger. A deletion analysis of Bir1 identified two regions important for function. The C-terminal region is essential for growth, binds Sli15, and is necessary and sufficient for the localization of Bir1 as a chromosomal passenger. The middle region is not essential but is required to localize the inner kinetochore protein Ndc10 to the spindle during anaphase and to the midzone at telophase. In contrast, precise deletion of the highly conserved IAP repeats conferred no phenotype and did not alter the cell cycle delay caused by loss of cohesin. Bir1 is phosphorylated in a cell cycle-dependent manner. Mutation of all nine CDK consensus sites in the middle region of Bir1 significantly decreased the level of phosphorylation and blocked localization of Ndc10 to the spindle at anaphase. Moreover, immunoprecipitation of Ndc10 with Bir1 was dependent on phosphorylation. The loss of Ndc10 from the anaphase spindle prevented elongation of the spindle beyond 7 microm. We conclude that phosphorylation of the middle region of Bir1 is required to bring Ndc10 to the spindle at anaphase, which is required for full spindle elongation.

RUNX1/AML1 Is Phosphorylated at Both Its N- and C-Terminus by cdk6/cyclin D3 or cdk1/cyclin B.

RUNX1/AML1 is a key transcriptional mediator of hematopoiesis and leukemogenesis. AML1 regulates myeloid and lymphoid differentiation via activation of lineage-specific genes such as those encoding myeloperoxidase or the T cell receptor δ and participates in apoptotic response pathways via its ability to tranactivate the p14/p19ARF gene. In addition, AML1 accelerates G1 to S cell cycle progression, via activation of the cyclin D3 and potentially the cdk4 genes. CBF oncoproteins such as AML1-ETO or CBFβ-SMMHC interfere with the activities of AML1 and block myeloid differentiation and slow cell cycle progression, and mutations such as loss of p16 which accelerate G1 prevent cell cycle inhibition and cooperate with CBF oncoproteins to induce acute leukemia in mice. In addition to regulation of the cell cycle by AML1, we have been interested in how AML1 activities vary and may be regulated during cell cycle progression. We recently reported that endogenous AML1 levels increase in hematopoietic cell lines as they progress from G1 to S and then diminish again at the end of mitosis (Bernardin-Fried et al J. Biol. Chem. 279:15678, 2004). RNA levels did not vary, but exogenous AML1 mimicked the behaviour of the endogenous protein, suggesting regulation at the level of protein stability. Mutation of two Ras-dependent phosphorylation sites, S276 and S293, to alanine did not prevent cell cycle variation. We have therefore set out to evaluate whether AML1 stability might be regulated by cyclin-dependent kinase (cdk) phosphorylation. AML1 contains 480 amino acids and binds DNA via its N-terminal Runt domain. Both the cdk6/cyclin D3 and the cdk1/cyclin B complex, expressed from baculovirus vectors, phosphorylated GST-AML1(1-290) and GST-AML1(290–480). The Runt domain alone, in GST-AML1(86–217), was not phosphorylated. Interestingly, exogenous DNA-binding domain alone did not vary during the cell cycle. This is the first demonstration that a specific kinase phosphorylates AML1 in vitro. There are three (S/T)PX(K/R) cdk consensus sites in AML1, with serines at residues 48, 303, and 424. Mutation of S424 to alanine did not prevent phosphorylation of GST-AML1(290–480). Additional mutations of these and other serines or threonines adjacent to proline are being generated to further map the cdk phosphorylation sites and to enable in vivo experiments designed to evaluate the effects of these mutations on cell cycle-specific AML1 expression. We propose a model in which accumulated phosphorylation of AML1 during the S and G2/M cell cycle phases leads to ubiquitin-mediated AML1 destabilization at the end of mitosis. The increased stability of AML1 in the presence of proteosome inhibitors supports this model. Phosphorylation-mediated destabilization of AML1 may complement the recent finding that direct interaction of cyclin D3 with AML1 inhibits its activity as a transcriptional activator. Each of these mechanisms may help regulate the proliferation of hematopoietic stem/progenitor cells. Finally, perhaps loss of destabilizing C-terminal phosphorylation sites in the AML1-ETO oncoprotein increases its ability to dominantly repress AML1-target genes during myeloid leukemogenesis.

Cdc14 activates cdc15 to promote mitotic exit in budding yeast.

Inactivation of mitotic cyclin-dependent kinases (Cdks) is required for cells to exit mitosis [1] [2]. In the budding yeast Saccharomyces cerevisiae, Cdk inactivation is triggered by the phosphatase Cdc14, which is activated by a complex network of regulatory proteins that includes the protein kinase Cdc15 [3] [4] [5] [6]. Here we show that the ability of Cdc15 to promote mitotic exit is inhibited by phosphorylation. Cdc15 is phosphorylated in vivo at multiple Cdk-consensus sites during most of the cell cycle, but is transiently dephosphorylated in late mitosis. Although phosphorylation appears to have no effect on Cdc15 kinase activity, a non-phosphorylatable mutant of Cdc15 is a more potent stimulator of mitotic exit than wild-type Cdc15, indicating that phosphorylation inhibits Cdc15 function in vivo. Interestingly, inhibitory phosphorylation of Cdc15 is removed by the phosphatase Cdc14 in vitro, and overproduction of Cdc14 leads to Cdc15 dephosphorylation in vivo. Thus, Cdc15 serves both as an activator and substrate of Cdc14. Although this scheme raises the possibility that positive feedback promotes Cdc14 activation, we present evidence that such feedback is not essential for Cdc14 activation in vivo. Instead, Cdc15 dephosphorylation may promote some additional function of Cdc15 that is independent of its effects on Cdc14 activation.

Sud1(+) targets cyclin-dependent kinase-phosphorylated Cdc18 and Rum1 proteins for degradation and stops unwanted diploidization in fission yeast.

In the fission yeast Schizosaccharomyces pombe, S phase is limited to a single round per cell cycle through cyclin-dependent kinase phosphorylation of critical replication factors, including the Cdc18 replication initiator protein. Because defects in Cdc18 phosphorylation lead to a hyperstable and hyperactive form of Cdc18 that promotes high levels of overreplication in vivo, we wished to identify the components of the Cdc18 proteolysis pathway in fission yeast. In this paper we describe one such component, encoded by the sud1(+) gene. sud1(+) shares homology with the budding yeast CDC4 gene and is required to prevent spontaneous re-replication in fission yeast. Cells lacking sud1(+) accumulate high levels of Cdc18 and the CDK inhibitor Rum1, because they cannot degrade these two key cell cycle regulators. Through genetic analysis we show that hyperaccumulation of Rum1 contributes to re-replication in Deltasud1 cells, but is not the cause of the defect in Cdc18 proteolysis. Rather, Sud1 itself is associated with the ubiquitin pathway in fission yeast and binds to Cdc18 in vivo. Most importantly, Sud1-Cdc18 binding requires prior phosphorylation of the Cdc18 polypeptide at CDK consensus sites. These results provide a biochemical mechanism for the phosphorylation-dependent degradation of Cdc18 and other cell cycle regulators, including Rum1. Evolutionary conservation of the Sud1/CDC4 pathway suggests that phosphorylation-coupled proteolysis may be a general feature of nearly all eukaryotic cell cycles.

Inhibition of poly(A) polymerase requires p34cdc2/cyclin B phosphorylation of multiple consensus and non-consensus sites.

We showed previously that p34(cdc2)/cyclin B (MPF) hyperphosphorylates poly(A) polymerase (PAP) during M-phase of the cell cycle, causing repression of its enzymatic activity. Mutation of three cyclin-dependent kinase (cdk) consensus sites in the PAP C-terminal regulatory domain prevented complete phosphorylation and MPF-mediated repression. Here we show that PAP also contains four nearby non-consensus cdk sites that are phosphorylated by MPF. Remarkably, full phosphorylation of all these cdk sites was required for repression of PAP activity, and partial phosphorylation had no detectable effect. The consensus sites were phosphorylated in vitro at a 10-fold lower concentration of MPF than the non-consensus sites. Consistent with this, during meiotic maturation of Xenopus oocytes, consensus sites were phosphorylated prior to the non-consensus sites at metaphase of meiosis I, and remained so throughout maturation, while the non-consensus sites did not become fully phosphorylated until after 12 h of metaphase II arrest. We propose that PAP's multiple cdk sites, and their differential sensitivity to MPF, provide a mechanism to link repression specifically to late M-phase. We discuss the possibility that this reflects a general means to control the timing of cdk-dependent regulatory events during the cell cycle.