Ras-Erk signaling induces phosphorylation of human TLE1 and downregulates its repressor function.

Signaling mediated by the Ras-extracellular signal-regulated kinase (Erk) pathway often leads to the phosphorylation of transcriptional regulators, thereby modulating their activity and causing concerted changes in gene expression. In Drosophila, the induction of multiple Ras-Erk pathway target genes depends on prior phosphorylation of the general co-repressor Groucho, a modification that downregulates its repressive function. Here, we show that TLE1, one of the four human Groucho orthologs, is similarly phosphorylated in response to Ras-Erk pathway activation, and that this modification attenuates its capacity to repress transcription. Specifically, unphosphorylated TLE1 dominantly suppresses the induction of Ras-Erk pathway target genes in cultured human cells, and the expression of an unphosphorylatable TLE1 derivative causes severe phenotypes in a transgenic Drosophila model system, whereas a phosphomimetic variant of TLE1 exerts only negligible effects. We present data indicating that TLE1 is rapidly excluded from the nucleus following epidermal growth factor receptor pathway activation, an effect that likely accounts for its inability to mediate effective repression under such conditions. Significantly, we find that unphosphorylated TLE1 blocks oncogenic phenotypes induced by mutated H-Ras in human mammary cells, both in vitro and following their implantation in mice. Collectively, our data strongly indicate that phosphorylation of TLE family members and the consequent downregulation of their repressor function is a key conserved step in the transcriptional responses to Ras-Erk signaling, and possibly a critical event in the tumorigenic effects caused by excessive Ras-Erk pathway activity.

Fbw7 regulates the activity of endoreduplication mediators and the p53 pathway to prevent drug-induced polyploidy.

Fbw7 is a tumor suppressor that is mutated in numerous cancers. It encodes an E3 ubiquitin ligase, whose ability to decrease the levels of pivotal regulators of cell growth and proliferation underlies its tumor suppressor function. Here, we explore the consequences of Fbw7 inactivation on the outcome of chemotherapeutic treatments. When exposed to spindle toxins such as vinblastine and taxol, Fbw7-deficient cells undergo extensive mitotic slippage and endoreduplication, rendering them polyploid. A combined deregulation of several Fbw7 target proteins is required for this phenotype. Specifically, elevated expression of cyclin E and Aurora A in Fbw7-deficient cells is required for drug-induced polyploidy. However, overexpression of either cyclin E or Aurora A alone is not sufficient for drug-induced polyploidy. In addition, we demonstrate that Fbw7 deficiency limits the ability of p53 to respond to mitotic toxins but not to DNA damage. Furthermore, Fbw7 expression regulates the p53-dependent induction of genes such as Lats2 and p21 in response to vinblastine. Hence, we suggest that Fbw7 serves as a master regulator of the mitotic and tetraploidy checkpoints.

Depletion of the AP-1 repressor JDP2 induces cell death similar to apoptosis.

JDP2 is a ubiquitously expressed nuclear protein that efficiently represses the activity of the transcription factor AP-1. Thus far, all studies of JDP2 function have relied on the ectopic expression of the protein. In this study, we use a different approach: depletion of JDP2 from cells. Specific depletion of JDP2 resulted in p53-independent cell death that resembles apoptosis and was evident at 72 h. The death mechanism was caspase dependent as the cells could be rescued by treatment with caspase inhibitor zVAD. Our studies suggest that JDP2 functions as a general survival protein, not only following UV-irradiation, as reported earlier, but also under normal culture conditions. Thus, our data support that JDP2 is a cellular survival protein whose presence is necessary for normal cellular function.

AP-1 in cell proliferation and survival.

A plethora of physiological and pathological stimuli induce and activate a group of DNA binding proteins that form AP-1 dimers. These proteins include the Jun, Fos and ATF subgroups of transcription factors. Recent studies using cells and mice deficient in individual AP-1 proteins have begun to shed light on their physiological functions in the control of cell proliferation, neoplastic transformation and apoptosis. Above all such studies have identified some of the target genes that mediate the effects of AP-1 proteins on cell proliferation and death. There is evidence that AP-1 proteins, mostly those that belong to the Jun group, control cell life and death through their ability to regulate the expression and function of cell cycle regulators such as Cyclin D1, p53, p21(cip1/waf1), p19(ARF) and p16. Amongst the Jun proteins, c-Jun is unique in its ability to positively regulate cell proliferation through the repression of tumor suppressor gene expression and function, and induction of cyclin D1 transcription. These actions are antagonized by JunB, which upregulates tumor suppressor genes and represses cyclin D1. An especially important target for AP-1 effects on cell life and death is the tumor suppressor p53, whose expression as well as transcriptional activity, are modulated by AP-1 proteins.

The mammalian UV response: c-Jun induction is required for exit from p53-imposed growth arrest.

The mammalian UV response results in rapid and dramatic induction of c-jun. Induction of a protooncogene, normally involved in mitogenic responses, by a genotoxic agent that causes growth arrest seems paradoxical. We now provide an explanation for the role of c-Jun in the UV response of mouse fibroblasts. c-Jun is necessary for cell-cycle reentry of UV-irradiated cells, but does not participate in the response to ionizing radiation. Cells lacking c-Jun undergo prolonged cell-cycle arrest, but resist apoptosis, whereas cells that express c-Jun constitutively do not arrest and undergo apoptosis. This function of c-Jun is exerted through negative regulation of p53 association with the p21 promoter. Cells lacking c-Jun exhibit prolonged p21 induction, whereas constitutive c-Jun inhibits UV-mediated p21 induction.

Stress-induced JNK activation is independent of Gadd45 induction.

DNA damage and environmental stress activate signaling and induce genes involved in cell cycle and cell death. Expression of the Gadd45 protein is induced following DNA damage and other stress. Gadd45 is believed to play a role in growth arrest and possibly in cell death. The JNK signaling pathway is also activated by some DNA-damaging agents. This activation leads to phosphorylation and activation of transcription factors, such as c-Jun/AP-1 and ATF2, which mediate immediate early gene induction. Recently Gadd45 was suggested to be involved in JNK activation. However, as this suggestion relied on in vitro experiments and ectopic overexpression of Gadd45 protein, we examined whether physiological levels of Gadd45 that are induced following exposure to DNA damaging agents and stress can lead to JNK induction. We found that JNK activation by UV irradiation and anisomycin treatment precedes the induction of gadd45 mRNA by these agents. Gadd45 protein induction by methyl methanesulfonate also lagged behind JNK activation. The use of protein synthesis inhibitors suggested that newly synthesized proteins, including the stress-induced Gadd45, make only a marginal contribution to JNK activation. We also found that stresses such as gamma irradiation induce Gadd45 and do not activate JNK in mouse fibroblasts. Therefore, stress-induced JNK does not depend on Gadd45 induction.

Modulation of Mdm2 expression and p53-induced apoptosis in immortalized human ovarian granulosa cells.

The activity of the tumor suppressor gene p53 is implicated in arrest of the cell cycle and the induction of apoptosis. The mdm2 oncogene is transcriptionally activated by p53, and the protein products of this gene can down-modulate biochemical activities and biological effects of p53 in a cell context-dependent manner. We have established highly steroidogenic human granulosa cell lines expressing the Ha-ras oncogene and a temperature sensitive (ts) mutant of p53 (p53val135) to test the involvement of p53-downstream genes in the modulation of apoptosis in these cells. We find that ras-transformed granulosa cells expressing p53val135 undergo apoptosis following a shift from 37 C to 32 C, a temperature at which p53val135 exerts its wild-type activity. Elevating the cellular content of cAMP at 32 C markedly enhances apoptosis. Basic fibroblast growth factor (bFGF) effectively blocks the p53/cAMP-induced apoptosis, but suppresses steroidogenesis. A naturally produced basement membrane-like extracellular matrix (ECM) containing immobilized bFGF exerts a similar antiapoptotic effect, but unlike soluble bFGF, it enhances steroidogenesis in these cells. While cAMP markedly suppresses the p53-induced Mdm2 expression, bFGF and ECM elevate Mdm2 expression 3-5-fold. These effects on Mdm2 expression are most pronounced 2-4 h after the shift to 32 C, before nuclear fragmentation is detected. Cells grown at 32 C in contact with ECM have a more developed actin cytoskeleton both in the absence and presence of cAMP stimulation, compared with cells grown on plastic dishes. We conclude that bFGF and components of the ECM can cross-talk with p53/cAMP-generated signals for apoptosis. These signals may, at least in part, be coordinated by the modulation of Mdm2 expression, which precedes the biochemical events characteristic of apoptosis. The multicomponent ECM also induced differentiation in these ras-transformed cells, while soluble bFGF inhibited differentiation, suggesting that ECM components other than bFGF stimulate differentiation. Organization of the actin cytoskeleton is likely to play an important role in the cross-talk between p53/cAMP- and bFGF/ECM-generated signals. Because the tumor suppressor gene p53 is implicated with apoptosis of primary granulosa cells and the ECM is involved in the prevention of this process, the newly established cell lines can serve as a useful model for apoptosis in highly luteinized granulosa cells.

Zf9, a Kruppel-like transcription factor up-regulated in vivo during early hepatic fibrosis.

Wound repair in the liver induces altered gene expression in stellate cells (resident mesenchymal cells) in a process known as "activation." A zinc finger transcription factor cDNA, zf9, was cloned from rat stellate cells activated in vivo. Zf9 expression and biosynthesis are increased markedly in activated cells in vivo compared with cells from normal rats ("quiescent" cells). The factor is localized to the nucleus and the perinuclear zone in activated but not quiescent cells. Zf9 mRNA also is expressed widely in nonhepatic adult rat tissues and the fetal liver. The zf9 nucleotide sequence predicts a member of the Kruppel-like family with a unique N-terminal domain rich in serine-proline clusters and leucines. The human zf9 gene maps to chromosome 10P near the telomere. Zf9 binds specifically to a DNA oligonucleotide containing a GC box motif. The N-terminal domain of Zf9 (amino acids 1-201) is transactivating in the chimeric GAL4 hybrid system. In Drosophila schneider cells, full length Zf9 transactivates a reporter construct driven by the SV40 promoter/enhancer, which contains several GC boxes. A physiologic role for Zf9 is suggested by its transactivation of a collagen alpha1(I) promoter reporter. Transactivation of collagen alpha1(I) by Zf9 is context-dependent, occurring strongly in stellate cells, modestly in Hep G2 cells, and not at all in D. schneider cells. Our results suggest that Zf9 may be an important signal in hepatic stellate cell activation after liver injury.

Induction of Mdm2 and enhancement of cell survival by bFGF.

Basic fibroblast growth factor (bFGF) can exert mitogenic and viability-promoting effects in a wide range of biological systems. The biochemical activities mediating the cell survival function of bFGF are largely unknown. We report here that exposure of fibroblasts to bFGF, which confers upon them increased survival, also causes at the same time an increase in cellular levels of the Mdm2 oncoprotein. Cells constitutively exposed to a bFGF autocrine loop are more refractory to killing by cisplatin. This increased chemoresistance coincides with elevated Mdm2 and reduced activation of the endogenous p53, resulting in inefficient transcriptional activation of the bax gene promoter. Importantly, unlike Mdm2 accumulation in fibroblasts exposed to DNA damage, induction of Mdm2 by bFGF does not occur through a p53-mediated pathway. The role of p53 in DNA damage-induced apoptosis and the ability of Mdm2 to block p53-mediated cell death are well established. These findings therefore suggest that induction of Mdm2 and the subsequent inhibition of p53 function may contribute, at least partially, to the anti-apoptotic effects of bFGF and possibly some other survival factors.

Resistance to p53-mediated growth arrest and apoptosis in Hep 3B hepatoma cells.

Mutations in the tumor suppressor p53 are a common event in hepatocellular carcinoma (HCC). Because HCCs typically occur in livers with chronic injury and impaired function, we have explored the role of wild-type p53 in regulating the growth and differentiation of Hep 3B hepatoma cells, a p53-negative line derived from a liver cancer. Stable Hep 3B cell lines were generated in which inducible p53 was introduced using either a temperature-sensitive mutant (p53val135) or a tamoxifen-regulated p53-estrogen receptor chimera (p53-mERtm-pBabepuro). In both cell lines, induction of transcriptionally active p53 was confirmed by assessing several p53 targets: Mdm2 protein, p21waf1 mRNA and protein, and the cyclin G promoter. Despite marked induction of p21waf1, cells with active p53 failed to undergo growth arrest, which is probably due to the presence of a non-functional retinoblastoma protein (pRb) in these cells. Apoptosis also was not observed, even after prolonged (48 h) serum starvation or exposure to cisplatinum. Lack of apoptosis was correlated with unchanged bax mRNA levels following p53 induction. Additionally, albumin mRNA levels remained unchanged, and there was no change in basal transactivation of a reporter containing the promoter of the haptoglobin gene, encoding an acute phase protein. This suggests that growth arrest may be required to promote liver-specific gene expression. Overall, our data demonstrate that introduction of transcriptionally active p53 does not alter the malignant, dedifferentiated phenotype of Hep 3B hepatoma cells. Hence, not all cancer cells are equally responsive to the re-activation of wild-type 53. The ability of a cancer cell to undergo p53-mediated phenotypic alterations may depend on the retention of functional downstream effector pathways.

p53 mediated apoptosis in HeLa cells: transcription dependent and independent mechanisms.

The most frequent target for genetic alterations in human cancers is the p53 tumor suppressor gene. Mutations in p53 abrogate its ability to inhibit cell growth and to suppress tumor progression. The anti-proliferative activity of p53 can be mediated by the induction of growth arrest and/or programmed cell death (apoptosis). Recent in vivo studies support the involvement of apoptosis in tumor suppression by p53. To gain further insight into the mechanisms by which p53 induces apoptosis, the activity of p53 was studied in HeLa cells using a transient transfection assay. To define the functional domains of p53 required for apoptosis a C-terminal deletion mutant of p53 was used. This mutant, p53d1214, lacks the oligomerization domain, the nuclear localization signal and a large part of the core DNA binding domain. This mutant was shown to be deficient in sequence specific transactivation activity. Overexpression of wt p53 induced an efficient apoptosis in transiently transfected HeLa cells. Surprisingly p53d1214, containing only the first 214 N-terminal residues induced extensive apoptosis. The induction of apoptosis by p53d1214 is slower than that induced by wt p53. Furthermore, p53d1214 suppressed the transformation of rat embryo fibroblasts by several oncogene combinations, such as myc plus ras. In view of the fact that p53d1214 lacks measurable transactivation potential, our findings suggest the existence of two p53 dependent-apoptotic pathways--one involves activation of specific target genes, and the other is independent of it. Transactivation independent apoptosis may play a central role in tumor suppression by p53.

Induction of apoptosis in HeLa cells by trans-activation-deficient p53.

The p53 tumor suppressor protein is a transcriptional activator, which can mediate apoptotic cell death in a variety of cell types. To determine whether sequence-specific trans-activation is a prerequisite for the induction of apoptosis by p53, the apoptotic effects of various p53 deletion mutants were monitored in an assay based on the transient transfection of HeLa cells. A truncated protein (p53dl214), containing only the first 214 amino-terminal residues of murine p53, induced extensive apoptosis, albeit at a slower rate than trans-activation-competent wild-type p53. p53dl214 also suppressed the transformation of rat fibroblasts by several oncogene combinations and particularly by myc plus ras and HPV E7 plus ras. p53dl214 lacks a major portion of the DNA-binding domain and cannot activate p53-responsive promoters. Moreover, a human p53 protein carrying mutations in residues 22 and 23 also triggered HeLa cell apoptosis, despite failing to induce significant activation of relevant p53 target promoters. These data suggest the existence of two p53-dependent apoptotic pathways--one requiring activation of specific target genes, and the other independent of sequence-specific trans-activation. The latter pathway may actually be totally uncoupled from the binding of p53 to its consensus DNA sites. The relative contribution of trans-activation-independent apoptosis to tumor suppression by p53 may be dictated by the specific genetic lesions present in the particular tumor.

Transcriptional repression by the C-terminal domain of p53.

We have previously shown that monomeric p53 can transactivate target genes in vivo and that C-terminal fragments of p53 are oncogenic. To further elaborate these findings a series of C-terminal truncations of p53 was generated. The transactivation capacity and the ability of the truncated p53 to suppress oncogene-mediated transformation were studied. We found that p53 truncated at amino acid 303 (p53wtdl303) can still function in both assays, though less efficiently than full length wild type (wt) p53. Transforming C-terminal fragments inhibited transactivation induced by full length wt p53. Surprisingly, they also inhibited transactivation by wtdl303, with which they do not share any overlapping sequences. Furthermore, the C-terminal fragments repressed the transactivation domains of several viral and cellular transcriptional activators. These data raise the possibility that the C-terminal domain of p53 may compete with the p53 transactivation domain for a common basal transcription factor.

A model for tumor suppression using H-1 parvovirus.

A model system is proposed to investigate, at the molecular level, the pathways of tumor suppression. As a tool for the selection of cells with a suppressed phenotype, we used the H-1 parvovirus that preferentially kills various neoplastic cells. From the human K562 leukemia cells, we isolated a clone, KS, that is resistant to the cytopathic effect of the H-1 virus and displays a suppressed malignant phenotype. The suppressed malignancy and the cellular resistance to H-1 killing appear to depend on the activity of wild-type p53. Whereas the KS cells express wild-type p53, the protein is undetectable in the parental K562 cells. Experiments with p53 mutants suggest that wild-type p53, in its functionally intact state, contributes to the resistance against the cytopathic effect of H-1 parvovirus.

Tight DNA binding and oligomerization are dispensable for the ability of p53 to transactivate target genes and suppress transformation.

The p53 tumor suppressor protein can bind tightly to specific sequence elements in the DNA and induce the transactivation of genes harboring such p53 binding sites. Various lines of evidence suggest that p53 binds to its target site as an oligomer. To test whether oligomerization is essential for the biological and biochemical activities of p53, we deleted a major part of the dimerization domain of mouse wild-type p53. The resultant protein, termed p53wt delta SS, was shown to be incapable of forming detectable homo-oligomers in vitro and is, therefore, likely to be predominantly if not exclusively monomeric. In agreement with the accepted model, p53wt delta SS indeed failed to exhibit measurable DNA binding in vitro. Surprisingly, though, it was still capable of suppressing oncogene-mediated transformation and of transactivating in vivo a target gene containing p53 binding sites. These findings indicate that dimerization-defective p53 is biologically active and may engage in productive sequence-specific DNA interactions in vivo. Furthermore, p53 dimerization probably leads to cooperative binding to specific DNA sequences.

Identification of a minimal transforming domain of p53: negative dominance through abrogation of sequence-specific DNA binding.

Mutations in the p53 gene are most frequent in cancer. Many p53 mutants possess transforming activity in vitro. In cells transformed by such mutants, the mutant protein is oligomerized with endogenous cell p53. To determine the relevance of oligomerization for transformation, miniproteins containing C-terminal portions of p53 were generated. These miniproteins, although carrying no point mutation, transformed at least as efficiently as full-length mutant p53. Transforming activity was coupled with the ability to oligomerize with wild-type p53, as well as with the ability to abrogate sequence-specific DNA binding by coexpressed wild-type p53. These findings suggest that p53-mediated transformation may operate through a dominant negative mechanism, involving the generation of DNA binding-incompetent oligomers.

Modulation of the cellular phenotype by integrated adeno-associated virus.

The adeno-associated virus (AAV) rep gene encodes a series of overlapping, multifunctional, nonstructural proteins (Rep proteins) which regulate the viral life cycle and which are also capable of trans-regulating nonviral gene expressions (reviewed in Berns, 1990, Microbiol. Rev. 54, 316-329). To investigate the expression of the AAV rep gene in a cellular chromosomal context, SV40-transformed Chinese hamster embryo (OD4) cells were infected with an AAV/neo hybrid virus and progeny resistant to the antibiotic G418 were selected and amplified. Chromosomal integration and RNA transcription of the AAV and neo DNA inserts were confirmed by Southern and Northern blotting procedures. One of the G418R cell lines stably expressed a protein which reacted specifically with AAV anti-Rep antiserum in Western immunoblots. The stable integration of AAV rep DNA, which did not interfere with cell proliferation under normal growth conditions, was associated with two changes in cellular phenotype: eight of nine lines were markedly more sensitive to UV light (254 nm) than were the parental OD4 cells; and seven of the nine lines had lost the capacity to promote SV40 origin DNA amplification in vitro, in contrast to the parental OD4 cells. OD4 cells transformed to G418R by AAV/neo DNA constructs with a deleted rep gene, or by a neo DNA construct lacking AAV DNA, did not display these phenotypic changes. It is suggested that stable integration of the AAV rep gene interferes with cellular processes connected with DNA repair and gene amplification.

Surgery in patients with circulating lupus anticoagulant.

Twelve surgical procedures were performed in eight patients with circulating lupus anticoagulant. Bleeding complications were noted intra and postoperatively in three patients, either with an associated hypoprothrombinemia and/or thrombocytopenia. Five patients in whom only circulating lupus anticoagulant was found had an uneventful intra and postoperative course. It is concluded that the presence of lupus anticoagulant, when unassociated with other hemostatic defects, is not a contraindication for surgery.

Circulating anticoagulant and serological tests for syphilis.

Circulating anticoagulant (CA) and particularly lupus anticoagulant are commonly associated with biological false-positive tests for syphilis (BFP-STS) in patients with collagen diseases. CA was presently found in the sera of 8 out of 30 subjects with chronic BFP-STS without collagen or autoimmune diseases. It was not found in any of the 21 patients with various stages of syphilis. In 5 out of 21 elderly subjects (age greater than 70 years) in whom a positive BFP-STS was detected, there was no CA. It is concluded that the association between CA and BFP-STS is found only in patients with collagen and autoimmune diseases and in some of the younger chronic BFP-STS reactors. It is not detected in syphilitic patients or elderly subjects in whom a high incidence of BFP-STS can be found. The difference in the incidence of this association is probably due to the differing biologic behaviour of these autoantibodies.