The contribution of the RING finger domain of MDM2 to cell cycle progression.

The MDM2 oncoprotein binds to p53 and abrogates p53-mediated G1 arrest and apoptosis. We show that MDM2 over-expression accelerates cell cycle progression of RPM12650 cells by overcoming the negative effect of endogenous wild type p53 at the G1/S checkpoint. The interaction with p53 and transcription inhibition are necessary but not sufficient. The RING finger domain of MDM2 is also required for the positive effect of MDM2 on the cell cycle. Surprisingly, several point mutants in the zinc binding sites of the RING finger are fully competent for cell cycle stimulation even though they abolish MDM2-directed degradation of p53 and MDM2 E3-ligase activity. In contrast, alterations in and around the cryptic nucleolar localization sequence (KR motif) inhibit MDM2-mediated cell cycle progression as well as p53 degradation and MDM2 E3 ligase activity. We found that all the RING mutants decrease inhibition of both p53 dependent reporters and endogenous p21CIP1/WAF1/SDI1. These results indicate that the RING finger of MDM2 has a role in the regulation of the cell cycle that is independent of p53 degradation and endogenous p21CIP1/WAF1/SDI1 regulation.

The contribution of the acidic domain of MDM2 to p53 and MDM2 stability.

p53 and MDM2 are both degraded by the ubiquitin-proteasome pathway. MDM2 binds p53 and promotes its rapid degradation. MDM2 is an E3 ligase that activates self and p53 ubiquitylation. Moreover, MDM2 nuclear-cytoplasmic shuttling contributes to p53 degradation in the cytoplasm. We have identified a new region of MDM2 which regulates the stability of both p53 and MDM2. The first 50 amino-acids of the MDM2 acidic domain (222-272) contribute to MDM2 and MDM2-mediated p53 degradation by a mechanism which is independent of either MDM2 E3-ligase activity or MDM2 nucleo-cytoplasmic shuttling. The transcriptional coactivator p300 could have been involved, since it binds to the MDM2 acidic domain. However, we found that p300 stabilises MDM2, even in absence of an intact acidic domain, indicating that the MDM2 acidic region contributes to proteolysis independently of p300. We propose that the MDM2 acidic domain is required for unbiquitylated MDM2 and p53 to be degraded by cytoplasmic proteasomes.

Involvement of p21 in mitotic exit after paclitaxel treatment in MCF-7 breast adenocarcinoma cell line.

It has been shown recently that expression of p21 is enhanced by paclitaxel. This cytotoxic compound induces mitotic spindle damage resulting in blockade of the mitotic cell cycle associated or not with apoptotic cell death. In the present study, we showed that, in MCF-7 cells, paclitaxel induced accumulation of p21 in cells with a G2/M DNA content, corresponding to cells either in abnormal mitosis or in an interphase-like state (decondensed chromatin) with multiple nuclei. In MCF-7 cells, the increase in p21 was subsequent to the mitotic arrest and was associated with the exit from abnormal mitosis leading to formation of cells with micronuclei. In this cell line, we noted a relationship between the elevation of p21 expression and the inhibition of p34cdc2 activity. High levels of p21 protein were also found to be associated with inactive p34cdc2/cyclin B protein complex after treatment with paclitaxel. Treatment with p21 antisense oligonucleotide partially blocked induction of p21 expression by paclitaxel and significantly reduced survival of MCF-7 cells exposed to this agent. In NIH-OVCAR-3 cells, which are deficient in basal and paclitaxel-induced p21 expression, paclitaxel led to a prolonged activation of p34cdc2 and a delayed mitotic exit associated with apoptotic cell death. These observations suggest that p21 is not required for the mitotic arrest in response to paclitaxel, but argue in favor of a role for this inhibitor in facilitating the exit from abnormal mitosis. This effectively enhances cell survival after paclitaxel-induced spindle damage.

Early gene responses associated with transforming growth factor-beta 1 growth inhibition and autoinduction in MCF-7 breast adenocarcinoma cells.

In the human breast carcinoma cell line (MCF-7), exogenous TGF-beta 1 induces a dose-dependent inhibition of cell proliferation. In a MCF-7 cell subline [MCF-7(-)], which has an undetectable level of type II TGF-beta receptor, exogenous TGF-beta 1 does not inhibit cell proliferation but is still able to induce its own message. In both cell lines, TGF-beta 1 stimulates expression of c-jun, whereas a rapid, transient and marked increase in c-fos mRNA is only observed in the MCF-7 cells sensitive to the growth inhibitory effect of TGF-beta 1. Depletion of protein kinase C abolishes the c-fos but not the c-jun response to TGF-beta 1. Our results suggest that growth inhibition and autoinduction by TGF-beta 1 are mediated by different signalling pathways. In addition, a PKC-dependent increase in c-fos expression seems to be associated with the growth inhibitory effect of TGF-beta 1.

Involvement of p21 in the PKC-induced regulation of the G2/M cell cycle transition.

Activation of protein kinase C (PKC) inhibits cell cycle progression at the G1/S and G2/M transitions. We found that phorbol 12-myristate 13-acetate (PMA) induced upregulation of p21, not only in MCF-7 cells arrested in the G1 phase as previously shown, but also in cells delayed in the G2 phase. This increase in p21 in cells accumulated in the G1 and G2/M phases of the cell cycle after PMA treatment was inhibited by the PKC inhibitor GF109203X. This indicates that PKC activity is required for PMA-induced p21 upregulation and cell cycle arrest in the G1 and G2/M phases of the cell cycle. To further assess the role of p21 in the PKC-induced G2/M cell cycle arrest independently of its G1 arrest, we used aphidicolin-synchronised MCF-7 cells. Our results show that, in parallel with the inhibition of cdc2 activity, PMA addition enhanced the associations between p21 and either cyclin B or cdc2. Furthermore, we found that after PMA treatment p21 was able to associate with the active Tyr-15 dephosphorylated form of cdc2, but this complex was devoid of kinase activity indicating that p21 may play a role in inhibition of cdc2 induced by PMA. Taken together, these observations provide evidence that p21 is involved in integrating the PKC signaling pathway to the cell cycle machinery at the G2/M cell cycle checkpoint.

Effects of TGF-beta 1 (transforming growth factor-beta 1) on the cell cycle regulation of human breast adenocarcinoma (MCF-7) cells.

The antiproliferative effects of TGF-beta 1 were investigated in a human breast adenocarcinoma cell line (MCF-7). We report that TGF-beta 1 inhibits proliferation through cell cycle arrest in G1. A MCF-7 cell subline (MCF-7(-)), in which the type II TGF-beta receptor is not detected, was shown to be resistant to TGF-beta 1 growth inhibitory effect. Cdk2 kinase activity was inhibited in the MCF-7 sensitive cell subline in parallel with the inhibition of cell cycle progression. In both sensitive and resistant cell lines, TGF-beta 1 treatment did not affect cdk2, cdk4, cyclin E and cyclin D1 mRNA and protein levels. However, in the MCF-7 sensitive cell subline, a time-dependent increase in cells positive for p21WAF1/CIP1 nuclear localization was observed after TGF-beta 1 treatment. These findings suggest that TGF-beta 1 inhibition of MCF-7 cell proliferation is achieved through a type II receptor-dependent down-regulation of Cdk2 kinase activity without modification of Cdk and cyclin expression, but correlated with an increase in p21WAF1/CIP1 nuclear accumulation.

Bcl-2 phosphorylation and proteasome-dependent degradation induced by paclitaxel treatment: consequences on sensitivity of isolated mitochondria to Bid.

Several studies have suggested that Bcl-2 phosphorylation, which occurs during mitotic arrest induced by paclitaxel, inhibits its antiapoptotic function. In the present study, we demonstrated that the level of phosphorylated Bcl-2 was threefold higher in mitochondria than in the nuclear membrane or endoplasmic reticulum. Our results show, in isolated mitochondria, that phosphorylation of Bcl-2 in mitosis does not modify either its integration into the mitochondrial membrane or the ability to release cytochrome c in response to Bid, a cytochrome c releasing agent. In HeLa cells, in which paclitaxel induces apoptosis, the nonphosphorylated form of Bcl-2 is degraded by a proteasome-dependent degradation pathway, whereas the phosphorylated forms of mitochondrial Bcl-2 appear to be resistant to proteasome-induced degradation. We found that low concentrations of recombinant Bid triggered a greater release of cytochrome c from mitochondria isolated from paclitaxel-treated HeLa cells than from mitochondria isolated from control HeLa cells. Taken together, these results show that Bcl-2 phosphorylation does not inhibit its function. On the contrary, Bcl-2 phosphorylation indirectly regulated its antiapoptotic action via protection against degradation. Indeed, in response to paclitaxel treatment, the level of Bcl-2 expression in mitochondria rather than its phosphorylation state could regulate the sensitivity of mitochondria to cytochrome c releasing agents in vitro.

Increased level of p21 in human ovarian tumors is associated with increased expression of cdk2, cyclin A and PCNA.

We have demonstrated over-expression of the cyclin-dependent kinase inhibitor p21 in various ovarian-cancer cell lines as well as in ovarian-tumor biopsies. This increase in p21 expression relative to that observed in normal ovarian epithelial cells is unrelated to proliferation index. In the present study, we found that p21 is functional, since the protein extracted from IGROVI cells is still able to inhibit cdk2-kinase activity. We then investigated how IGROVI cells overcome the growth-inhibitory function of p21. Immunofluorescence assays and subcellular fractionation showed that p21 is located in cytoplasm and nucleus both in normal and in tumoral cells. Compared with normal ovarian epithelial cells in culture, the increase in level of p21 in IGROVI cells was found to be associated with increased expression of cdk2, cyclin-A and PCNA proteins. In IGROVI cells, p21 is associated with inactive cdk2/cyclin-A complex, indicating that it acts as an inhibitory factor rather than an assembly factor. Over-expression of cdk2 and of cyclin A observed in IGROVI cells allows them to escape to p21-inhibitory activity. The fact that cells from ovarian-tumor biopsies exhibited a concomitant increase in p21 and in its partners cdk2 and PCNA suggest that ovarian-tumor cells can tolerate high levels of functional p21 via over-expression of other cell-cycle-regulatory proteins.

Expression of p21WAF1/CIP1 is heterogeneous and unrelated to proliferation index in human ovarian carcinoma.

The growth-inhibitory protein p21WAF1/CIP1 is a potent inhibitor of various cyclin-dependent kinases, the expression of which is regulated at the transcriptional level by p53-dependent and -independent mechanisms. We examined p21WAF1/CIP1 mRNA and protein expression in 5 human ovarian-adenocarcinoma cell lines, 1 primary culture of normal surface epithelium and 17 human ovarian-tumor specimens. In culture cells, the p21WAF1/CIP1 protein was expressed in normal ovarian epithelial cells and at a high level in the adenocarcinoma 2008 and IGROV-1 cell lines. p21 WAF1/CIP1 expression was undetectable at the mRNA and protein levels in the NIH-OVCAR-3 and SKOV-3 ovarian-adenocarcinoma cell lines which are respectively mutated and deleted in the p53 gene. Heterogeneous expression of p21WAF1/CIP1 observed in ovarian-cancer cell lines in culture was also found in vivo on tumor specimens. p21WAF1/CIP1 expression is undetectable in 25% of the ovarian biopsies examined. Since it has been found that the p53 gene is mutated in 79% of ovarian cancer, the absence of p21WAF1/CIP1 expression in 25% of these ovarian cancer could not be correlated with p53 mutation. The proliferation index of the 17 tumors showed great variation from one tumor to another. However, no significant correlation was found between p21WAF1/CIP1 expression and the proliferation rate of the tumors.