Specific interaction between 14-3-3 isoforms and the human CDC25B phosphatase.

CDC25 dual-specificity phosphatases are essential regulators that activate cyclin-dependent kinases (CDKs) at critical stages of the cell cycle. In human cells, CDC25A and C are involved in the control of G1/S and G2/M respectively, whereas CDC25B is proposed to act both in S phase and G2/M. Evidence for an interaction between CDC25 phosphatases and members of the 14-3-3 protein family has been obtained in vitro and in vivo in several organisms. On the basis of the work performed with CDC25C, it has been proposed that phosphorylation is required to mediate the interaction with 14-3-3. Here we have examined the molecular basis of the interaction between CDC25B phosphatases and 14-3-3 proteins. We show that in the two-hybrid assay all three splice variants of CDC25B interact similarly and strongly with 14-3-3eta, beta and zeta proteins, but poorly with epsilon and Theta. In vitro, CDC25B interacts at a low level with 14-3-3beta, epsilon, zeta, eta, and Theta isoforms. This interaction is not increased upon phosphorylation of CDC25B by CHK1 and is not abolished by dephosphorylation. In contrast, a specific, strong interaction between CDC25B and 14-3-3zeta and eta isoforms is revealed by a deletion of 288 residues in the amino-terminal region of CDC25B. This interaction requires the integrity of Ser 323, although it is independent of phosphorylation. Thus, interaction between 14-3-3 proteins and CDC25B is regulated in a manner that is different from that with CDC25C. We propose that, in addition to a low affinity binding site that is available for all 14-3-3 isoforms, post-translational modification of CDC25B in vivo exposes a high-affinity binding site that is specific for the zeta and eta14-3-3 isoforms.

Human Cdc25 A inactivation in response to S phase inhibition and its role in preventing premature mitosis.

The Cdc25 A phosphatase is required for the G1-S transition of the cell cycle and is overexpressed in human cancers. We found that it is ubiquitylated and rapidly degraded by the proteasome and that its levels increase from G1 until mitosis. By treating cells with the DNA synthesis inhibitor hydroxyurea, Cdc25 A rapidly decreased in abundance, and this was accompanied by an increase in Cdk2 phosphotyrosine content and a decrease in Cdk2 kinase activity. Cdc25 A overexpression altered the ability of cells to arrest in the presence of hydroxyurea, and caused them to undergo premature chromosome condensation. Cdc25 A overexpression could render tumor cells less sensitive to DNA replication checkpoints, thereby contributing to their genomic instability.

Interaction studies between the p21Cip1/Waf1 cyclin-dependent kinase inhibitor and proliferating cell nuclear antigen (PCNA) by surface plasmon resonance.

The cyclin-dependent kinase (CDK) inhibitor p21Cip1 consists of two domains that interact with CDKs and proliferating cell nuclear antigen (PCNA), respectively. We have investigated the interaction between p21Cip1 and PCNA using surface plasmon resonance (SPR) technology and compared the results with those obtained from other sources such as the yeast two-hybrid system. Whilst other methods are only semi-quantitative, the SPR technique allowed us to determine the kinetic parameters of the interaction. The apparent equilibrium constant KD calculated for these kinetic parameters was 3.2 x 10(-7) M. We further demonstrate the use of SPR to study the interaction between mutant proteins and to determine their actual KD. The interaction between p21Cip1/PCNA is shown to be dependent upon the trimeric conformation of PCNA since a point mutant that abolishes PCNA-PCNA interaction also abolishes PCNA's interaction with p21Cip1. Finally, we demonstrate that SPR can be used to characterise the interaction of p21Cip1 and PCNA in the presence of short competitive peptides.

Heterologous expression of the human cyclin-dependent kinase inhibitor p21Cip1 in the fission yeast, Schizosaccharomyces pombe reveals a role for PCNA in the chk1+ cell cycle checkpoint pathway.

Fission yeast cells expressing the human gene encoding the cyclin-dependent kinase inhibitor protein p21Cip1 were severely compromised for cell cycle progress. The degree of cell cycle inhibition was related to the level of p21Cip1 expression. Inhibited cells had a 2C DNA content and were judged by cytology and pulsed field gel electrophoresis to be in the G2 phase of the cell cycle. p21Cip1 accumulated in the nucleus and was associated with p34cdc2 and PCNA. Thus, p21Cip1 interacts with the same targets in fission yeast as in mammalian cells. Elimination of p34cdc2 binding by mutation within the cyclin-dependent kinase binding domain of p21Cip1 exaggerated the cell cycle delay phenotype. By contrast, elimination of PCNA binding by mutation within the PCNA-binding domain completely abolished the cell cycle inhibitory effects. Yeast cells expressing wild-type p21Cip1 and the mutant form that is unable to bind p34cdc2 showed enhanced sensitivity to UV. Cell cycle inhibition by p21Cip1 was largely abolished by deletion of the chk1+ gene that monitors radiation damage and was considerably enhanced in cells deleted for the rad3+ gene that monitors both DNA damage and the completion of DNA synthesis. Overexpression of PCNA also resulted in cell cycle arrest in G2 and this phenotype was also abolished by deletion of chk1+ and enhanced in cells deleted for rad3+. These results formally establish a link between PCNA and the products of the rad3+ and chk1+ checkpoint genes.

Mechanism of inhibition of proliferating cell nuclear antigen-dependent DNA synthesis by the cyclin-dependent kinase inhibitor p21.

It is known that the direct binding of the cyclin-dependent kinase (Cdk) inhibitor p21, also called Cdk-interacting protein 1 (p21), to proliferating cell nuclear antigen (PCNA) results in the inhibition of PCNA-dependent DNA synthesis. We provide evidence that p21 first inhibits the replication factor C-catalyzed loading of PCNA onto DNA and second prevents the binding of DNA polymerase delta core to the PCNA clamp assembled on DNA. The second effect contributes most to the inhibition of pol delta holoenzyme activity. p21 primarily inhibited the DNA synthesis resulting from multiple reassembly of DNA polymerase delta holoenzyme. On the other hand, an ability of the PCNA clamp to translocate along double-stranded DNA was not affected by p21. These data were confirmed with a mutant of p21 that is unable to bind PCNA and therefore neither inhibited clamp assembly nor prevented the loading of DNA polymerase delta core onto DNA. Our data suggest that p21 does not discriminate in vitro "repair" and "replication" DNA synthesis based on template length but does act preferentially on polymerization which encounters obstacles to progress.

Identification of binding domains on the p21Cip1 cyclin-dependent kinase inhibitor.

Members of the recently discovered family of cyclin-dependent kinases inhibitors (CKIs) appear to play an essential regulatory role in the control of cell proliferation. To investigate the molecular basis of the interaction between these proteins and the cyclin-dependent kinases (CDKs), we performed a systematic mutagenesis of the CKI family member p21Cip1 using the alanine-scanning strategy. We have examined the interaction between in vitro translated human cdk2, cyclins A and D1, purified proliferating cell nuclear antigen (PCNA) and a set of human p21Cip1 mutants fused to glutathione S-transferase. Independent domains that are required for the interaction with cdk2 and with PCNA have been identified. The cdk2 binding domain is located in the N-terminal part of the protein, between residues 45 and 60, a region that is fully conserved in the p27Kip1 inhibitor. A PCNA binding region was localised to the C-terminus of the protein, between residues 142 and 163. These findings define protein motifs that are highly conserved between members of the CKI family and that are likely to play an essential function in the regulation of the G1/S transition.

Evidence for transcriptional control of human mdr1 gene expression by verapamil in multidrug-resistant leukemic cells.

We investigated the mechanism of verapamil (VRP) effects on mdr1 gene expression in two leukemic multidrug-resistant (MDR) cell lines, K562/ADR and CEM VLB100. Exposure to VRP for 24 hr resulted in a decrease in mdr1 mRNA levels that was dose related at concentrations between 15 and 50 microM. The maximal decrease of mdr1 mRNA levels was found to be 6-fold in the K562/ADR cells and 3-fold in the CEM VLB100 cells. The effect of VRP on mdr1 mRNA levels was, however, biphasic. At 100 microM VRP, which strongly inhibited cell proliferation, a 2-fold increase of mdr1 mRNA levels was observed in the K562/ADR cells. To determine whether the decrease of mRNA levels resulted from post-transcriptional mechanisms, mRNA stability was studied after blocking of transcription with actinomycin D in VRP-treated cells and in control cells. This study revealed that mdr1 mRNA was stable in both cell lines and no increase in mdr1 mRNA degradation was observed in the 30 microM VRP-treated cells versus control cells (half-lives of 23 hr versus 14 hr for the K562/ADR cells and 15.5 hr versus 10.0 hr for the CEM VLB100 cells). The suggestion of a transcriptional mechanism was confirmed by nuclear run-on assays. A 4-fold decrease in the mdr1 gene transcription rate was observed in the 30 microM VRP-treated CEM VLB100 cells. The decreased transcription rate could be due to the decrease in mdr1 proximal promoter activity observed in CEM VLB100 cells transiently transfected with the mdr1 promoter fused to the chloramphenicol acetyltransferase gene. Indeed, after exposure to 30 microM VRP, chloramphenicol acetyltransferase activity was decreased by 2-fold. This study reports for the first time a down-regulation of mdr1 gene transcription by a pharmacological agent. These results provide further identification of the regulatory mechanisms involved in the overexpression of mdr1 in MDR cells and may help in the development of new strategies for MDR reversal.

[Pharmacological control of P-glycoprotein expression]. / Modulation pharmacologique de l'expression de la P-glycoprotéine.

Calcium channel inhibitors, such as verapamil, have been identified as having the ability to modulate the multidrug-resistant (MDR) phenotype due to overexpression of P-glycoprotein (Pgp). We have studied the effect of verapamil on Pgp expression levels in a cell line originating from acute myeloblastic leukemia and resistant to adriamycin, K562/ADR. In this line, the addition of 15 microM verapamil in the culture medium gives a 3-fold decrease of Pgp expression after 72 hours of treatment. Similar results have been obtained for two other MDR cell lines, which suggest that this phenomenon is not specific of a single model. The level of mdr1 mRNAs is decreased in the presence of verapamil (with a maximum effect obtained at the 24th hour), which suggests that the mechanism of action of verapamil is transcriptional and/or post-transcriptional. We have also studied the effect of verapamil on the level of expression of mdr1 mRNAs in non-drug selected cells such as the HEL line (human acute myeloblastic leukemia) and the parental K562 line, which present a very low level of expression of Pgp, detectable only by PCR. In these lines, verapamil treatment has no effect on the level of expression of mdr1 mRNAs. The effect of verapamil is therefore restricted to drug-selected lines presenting high levels of Pgp expression. The impact of the negative regulation of Pgp expression on the MDR phenotype has been studied in the K562/ADR line. When the cells are treated for 72 h by verapamil, there is a decrease of resistance and an increase of intracellular accumulation of anticancer agents such as daunorubicin or vinblastine. Negative regulation of Pgp expression appears therefore as a possible strategy for MDR phenotype reversal. The effect of verapamil, whose molecular mechanism of action is being studied, could constitute a basis for this strategy.

Verapamil decreases P-glycoprotein expression in multidrug-resistant human leukemic cell lines.

We studied the effect of verapamil on Pgp expression (Pgp) in MDR human leukemia cell lines, K562/ADR and CEM VLB100. In the K562/ADR cell line, addition of verapamil to the culture medium (15 microM concentration) resulted in a 3-fold decrease in Pgp expression after 72 hr exposure. The effect of verapamil was reversible, and Pgp expression reached the level of untreated controls 24 hr after discontinuation of verapamil. Similar results were obtained with the human vinblastine-resistant cell line, CEM VLB100. On the contrary, no effect on Pgp expression was observed when the cells were treated with nifedipine or diltiazem (2 other calcium-channel blockers), even at doses that inhibited cell proliferation. The level of Pgp mRNA in the presence of verapamil was measured by Northern blot and was also decreased 2-fold (with the maximum reached within 24 hr), suggesting a transcriptional or post-transcriptional mechanism for verapamil. We further established that the effect of verapamil on Pgp expression led to an increase in DNR and VLB accumulation and cytotoxicity. These results suggest that verapamil acts specifically on Pgp expression in these drug-selected leukemic cells. The identification of a potentially novel mechanism of action may provide new insights as to how chemosensitization may be more effectively applied in vivo.