Loss of p53 induces cell proliferation via Ras-independent activation of the Raf/Mek/Erk signaling pathway.

The Ras family of small GTPases constitutes a central node in the transmission of mitogenic stimuli to the cell cycle machinery. The ultimate receptor of these mitogenic signals is the retinoblastoma (Rb) family of pocket proteins, whose inactivation is a required step to license cell proliferation. However, little is known regarding the molecular events that connect Ras signaling with the cell cycle. Here, we provide genetic evidence to illustrate that the p53/p21 Cdk-interacting protein 1 (Cip1)/Rb axis is an essential component of the Ras signaling pathway. Indeed, knockdown of p53, p21Cip1, or Rb restores proliferative properties in cells arrested by ablation of the three Ras loci, H-, N- and K-Ras. Ras signaling selectively inactivates p53-mediated induction of p21Cip1 expression by inhibiting acetylation of specific lysine residues in the p53 DNA binding domain. Proliferation of cells lacking both Ras proteins and p53 can be prevented by reexpression of the human p53 ortholog, provided that it retains an active DNA binding domain and an intact lysine residue at position 164. These results unveil a previously unidentified role for p53 in preventing cell proliferation under unfavorable mitogenic conditions. Moreover, we provide evidence that cells lacking Ras and p53 proteins owe their proliferative properties to the unexpected retroactivation of the Raf/Mek/Erk cascade by a Ras-independent mechanism.

Genetic analysis of Ras signalling pathways in cell proliferation, migration and survival.

We have used mouse embryonic fibroblasts (MEFs) devoid of Ras proteins to illustrate that they are essential for proliferation and migration, but not for survival, at least in these cells. These properties are unique to the Ras subfamily of proteins because ectopic expression of other Ras-like small GTPases, even when constitutively active, could not compensate for the absence of Ras proteins. Only constitutive activation of components of the Raf/Mek/Erk pathway was sufficient to sustain normal proliferation and migration of MEFs devoid of Ras proteins. Activation of the phosphatidylinositol 3-kinase (PI3K)/PTEN/Akt and Ral guanine exchange factor (RalGEF)/Ral pathways, either alone or in combination, failed to induce proliferation or migration of Rasless cells, although they cooperated with Raf/Mek/Erk signalling to reproduce the full response mediated by Ras signalling. In contrast to current hypotheses, Ras signalling did not induce proliferation by inducing expression of D-type Cyclins. Rasless MEFs had normal levels of Cyclin D1/Cdk4 and Cyclin E/Cdk2. However, these complexes were inactive. Inactivation of the pocket proteins or knock down of pRb relieved MEFs from their dependence on Ras signalling to proliferate.

Using cells devoid of RAS proteins as tools for drug discovery.

Mutational activation of RAS proteins occurs in nearly 30% of all human tumors. To date direct pharmacological inhibition of RAS oncoproteins has not been possible. As a consequence, current strategies are focusing on the development of inhibitors that target those kinases acting downstream of RAS proteins, including those of the RAF/MEK/ERK and PI3K/AKT pathways. Most of these inhibitors have undesired off-target effects that mask the potential therapeutic effect of blocking their targeted kinases. To facilitate the screening of selective inhibitors, we have generated lines of mouse embryonic fibroblasts that lack endogenous Ras proteins. These cells proliferate due to ectopic expression of either Ras oncoproteins that selectively activate the Raf/Mek/Erk pathway such as H-Ras(G12V/D38E) or constitutively active kinases such as B-Raf and Mek1. These cell lines were exposed to inhibitors against the RAF, MEK, and AKT kinases as well as inhibitors of other kinases known to crosstalk with RAS signaling such as JNK and p38. Amongst all compounds tested, only the MEK inhibitors U0126 and PD0325901, showed the expected specificity pattern. Yet, PD0325901, but not U0126, was able to inhibit a cell line lacking Ras proteins that owed its proliferative properties to loss of p53. Thus, suggesting unexpected off-target activities for this compound. The use of cell lines whose proliferative properties exclusively depend on selective targets provide a novel strategy to analyze the specificity of selective inhibitors designed against molecular targets implicated in human cancer.

Early asymmetry of gene transcription in embryonic human left and right cerebral cortex.

The human left and right cerebral hemispheres are anatomically and functionally asymmetric. To test whether human cortical asymmetry has a molecular basis, we studied gene expression levels between the left and right embryonic hemispheres using serial analysis of gene expression (SAGE). We identified and verified 27 differentially expressed genes, which suggests that human cortical asymmetry is accompanied by early, marked transcriptional asymmetries. LMO4 is consistently more highly expressed in the right perisylvian human cerebral cortex than in the left and is essential for cortical development in mice, suggesting that human left-right specialization reflects asymmetric cortical development at early stages.

Overexpression of LMO4 induces mammary hyperplasia, promotes cell invasion, and is a predictor of poor outcome in breast cancer.

The zinc finger protein LMO4 is overexpressed in a high proportion of breast carcinomas. Here, we report that overexpression of a mouse mammary tumor virus (MMTV)-Lmo4 transgene in the mouse mammary gland elicits hyperplasia and mammary intraepithelial neoplasia or adenosquamous carcinoma in two transgenic strains with a tumor latency of 13-18 months. To investigate cellular processes controlled by LMO4 and those that may be deregulated during oncogenesis, we used RNA interference. Down-regulation of LMO4 expression reduced proliferation of human breast cancer cells and increased differentiation of mouse mammary epithelial cells. Furthermore, small-interfering-RNA-transfected breast cancer cells (MDA-MB-231) had a reduced capacity to migrate and invade an extracellular matrix. Conversely, overexpression of LMO4 in noninvasive, immortalized human MCF10A cells promoted cell motility and invasion. Significantly, in a cohort of 159 primary breast cancers, high nuclear levels of LMO4 were an independent predictor of death from breast cancer. Together, these findings suggest that deregulation of LMO4 in breast epithelium contributes directly to breast neoplasia by altering the rate of cellular proliferation and promoting cell invasion.

Loss of the LIM domain protein Lmo4 in the mammary gland during pregnancy impedes lobuloalveolar development.

LMO4, a member of the LIM-only family of zinc-finger proteins, is overexpressed in a significant proportion of breast carcinomas and acts as a negative regulator of mammary epithelial differentiation. To delineate cell types within the developing mouse mammary gland that express Lmo4, we analysed different stages of mammopoiesis by immunohistochemistry. Lmo4 was found to be highly expressed in the proliferating cap cells of the terminal end bud and in the ductal and alveolar luminal cells of the mature mammary gland but was negligible or low in myoepithelial cells. To assess the physiological role of Lmo4 in the mammary gland, we generated conditionally targeted mice lacking Lmo4 in the mammary epithelium during pregnancy. Acute loss of Lmo4 in late pregnancy impaired lobuloalveolar development, accompanied by a two-fold reduction in the percentage of BrdU-positive cells. In contrast, germline loss of Lmo4 did not alter lobuloalveolar development arising from transplanted mammary anlagen, implying the existence of a compensatory mechanism in these knockout mice. Thus, the use of a conditional targeting strategy has revealed that Lmo4 is required for proper development of the mammary gland during pregnancy and indicated that Lmo4 acts as a positive regulator of alveolar epithelial proliferation.

The LIM domain protein Lmo4 is highly expressed in proliferating mouse epithelial tissues.

LMO4 belongs to the LIM-only family of zinc finger proteins that have been implicated in oncogenesis. The LMO4 gene is overexpressed in breast cancer and oral cavity carcinomas, and high levels of this protein inhibit mammary epithelial differentiation. Targeted deletion of Lmo4 in mice leads to complex phenotypic abnormalities and perinatal lethality. To further understand the role of LMO4, we have characterized Lmo4 expression in adult mouse tissues by immunohistochemical staining using monoclonal anti-Lmo4 antibodies. Lmo4 was highly expressed within specific cell types in diverse tissues. Expression was prevalent in epithelial-derived tissues, including the mammary gland, tongue, skin, small intestine, lung, and brain. High levels of Lmo4 were frequently observed in proliferating cells, such as the crypt cells of the small intestine and the basal cells of the skin and tongue. Lmo4 was highly expressed in the proliferative cap cell layer of the terminal end buds in the peripubertal mammary gland and in the lobuloalveolar units during pregnancy. The expression profile of Lmo4 suggests that this cofactor is an important regulator of epithelial proliferation and has implications for its role in the pathogenicity of cancer.

A classic zinc finger from friend of GATA mediates an interaction with the coiled-coil of transforming acidic coiled-coil 3.

Classic zinc finger domains (cZFs) consist of a beta-hairpin followed by an alpha-helix. They are among the most abundant of all protein domains and are often found in tandem arrays in DNA-binding proteins, with each finger contributing an alpha-helix to effect sequence-specific DNA recognition. Lone cZFs, not found in tandem arrays, have been postulated to function in protein interactions. We have studied the transcriptional co-regulator Friend of GATA (FOG), which contains nine zinc fingers. We have discovered that the third cZF of FOG contacts a coiled-coil domain in the centrosomal protein transforming acidic coiled-coil 3 (TACC3). Although FOG-ZF3 exhibited low solubility, we have used a combination of mutational mapping and protein engineering to generate a derivative that was suitable for in vitro and structural analysis. We report that the alpha-helix of FOG-ZF3 recognizes a C-terminal portion of the TACC3 coiled-coil. Remarkably, the alpha-helical surface utilized by FOG-ZF3 is the same surface responsible for the well established sequence-specific DNA-binding properties of many other cZFs. Our data demonstrate the versatility of cZFs and have implications for the analysis of many as yet uncharacterized cZF proteins.

Dimerization of CtIP, a BRCA1- and CtBP-interacting protein, is mediated by an N-terminal coiled-coil motif.

CtIP is a transcriptional co-regulator that binds a number of proteins involved in cell cycle control and cell development, such as CtBP (C terminus-binding protein), BRCA1 (breast cancer-associated protein-1), and LMO4 (LIM-only protein-4). The only recognizable structural motifs within CtIP are two putative coiled-coil domains located near the N and C termini of the protein. We now show that the N-terminal coiled coil (residues 45-160), but not the C-terminal coiled coil, mediates homodimerization of CtIP in mammalian 293T cells. The N-terminal coiled coil did not facilitate binding to LMO4 and BRCA1 proteins in these cells. A protease-resistant domain (residues 27-168) that minimally encompasses the putative N-terminal coiled coil was identified by matrix-assisted laser desorption ionization time-of-flight mass spectrometry. This region is predicted to contain two smaller coiled-coil regions. The CtIP-(45-160) dimerization domain is helical and dimeric, indicating that the domain does form a coiled coil. The two smaller domains, CtIP-(45-92) and CtIP-(93-160), also formed dimers of lower binding affinity, but with less helical content than the longer peptide. The hydrodynamic radius of CtIP-(45-160) is the same as those of CtIP-(45-92) and CtIP-(93-160), implying that CtIP-(45-160) does not form a single long coiled coil, but a more compact structure involving homodimerization of the two smaller coiled coils, which fold back as a four-helix bundle or other compact structure. These results suggest a specific model for CtIP homodimerization via its N terminus and contribute to an improved understanding of how this protein might assemble other factors required for its role as a transcriptional corepressor.

Defective neural tube closure and anteroposterior patterning in mice lacking the LIM protein LMO4 or its interacting partner Deaf-1.

LMO4 belongs to a family of transcriptional regulators that comprises two zinc-binding LIM domains. LIM-only (LMO) proteins appear to function as docking sites for other factors, leading to the assembly of multiprotein complexes. The transcription factor Deaf-1/NUDR has been identified as one partner protein of LMO4. We have disrupted the Lmo4 and Deaf-1 genes in mice to define their biological function in vivo. All Lmo4 mutants died shortly after birth and showed defects within the presphenoid bone, with 50% of mice also exhibiting exencephaly. Homeotic transformations were observed in Lmo4-null embryos and newborn mice, but with incomplete penetrance. These included skeletal defects in cervical vertebrae and the rib cage. Furthermore, fusions of cranial nerves IX and X and defects in cranial nerve V were apparent in some Lmo4(-/-) and Lmo4(+/-) mice. Remarkably, Deaf-1 mutants displayed phenotypic abnormalities similar to those observed in Lmo4 mutants. These included exencephaly, transformation of cervical segments, and rib cage abnormalities. In contrast to Lmo4 nullizygous mice, nonexencephalic Deaf-1 mutants remained healthy. No defects in the sphenoid bone or cranial nerves were apparent. Thus, Lmo4 and Deaf-1 mutant mice exhibit overlapping as well as distinct phenotypes. Our data indicate an important role for these two transcriptional regulators in pathways affecting neural tube closure and skeletal patterning, most likely reflecting their presence in a functional complex in vivo.

Mutational analysis of the LMO4 gene, encoding a BRCA1-interacting protein, in breast carcinomas.

The LIM domain-only genes LMO1 and LMO2 are translocated in acute T cell leukemia (T-ALL) and have been shown to be oncogenes in T lymphoid cells. LMO4, the fourth member of this family, is overexpressed in more than 50% of sporadic breast cancers, suggesting a role in breast oncogenesis. We recently found that LMO4 interacts with the breast/ovarian tumor suppressor BRCA1 and that LMO4 can repress its transcriptional activity. Since proto-oncogene deregulation can result from activating mutations in their coding or regulatory sequences, we explored whether the LMO4 gene undergoes somatic mutagenesis in breast cancer. Mutation analysis of the coding and 3' untranslated regions of the LMO4 gene was performed on 82 primary breast and 22 tumor cell lines. A somatic mutation was detected in one primary breast cancer, at the 3' end of exon 2, but was not present in normal DNA derived from the same patient. This mutation causes a frame-shift and potentially results in a truncated LMO4 polypeptide, LIM1(mut), lacking the second LIM domain. This mutant protein could still bind Ldb1 but no longer associated with CtIP or BRCA1. Our results show that somatic mutations within the LMO4 gene do occur in breast cancer but at a very low frequency. Thus, the primary mechanism by which LMO4 is deregulated in breast cancers appears to reflect overexpression of the gene rather than the acquisition of activating genetic mutations.

Two promoters within the human LMO4 gene contribute to its overexpression in breast cancer cells.

LMO4, a member of the LIM-only family of zinc-finger proteins, is overexpressed in more than 50% of primary breast cancers and cell lines, implying a role in the pathogenesis of this cancer. Southern blot analysis of these cell lines did not reveal amplification or rearrangement of the LMO4 gene. To investigate further the mechanism underlying LMO4 overexpression and the generation of two patterns of transcripts, we isolated genomic clones spanning the human gene. Similar to the mouse Lmo4 gene, there are two 5' noncoding exons, exon 1a and exon 1b, which we show are differentially expressed in breast epithelial cells. This reflects differential promoter usage in combination with alternative splicing. Two promoter regions were defined, one upstream of exon 1a and the other upstream of exon 1b. Both promoters exhibited strong activity in breast cancer cells, with up to 400-fold activity above basal levels. These promoters were significantly more active in T-47D and MCF-7 cells relative to SKBR3 cells, consistent with RNA levels. Thus, overexpression of the LMO4 gene in breast cancer cells reflects increased promoter activity and appears to involve aberrant activation of the second promoter in a subset of these cells.

Structural basis for the recognition of ldb1 by the N-terminal LIM domains of LMO2 and LMO4.

LMO2 and LMO4 are members of a small family of nuclear transcriptional regulators that are important for both normal development and disease processes. LMO2 is essential for hemopoiesis and angiogenesis, and inappropriate overexpression of this protein leads to T-cell leukemias. LMO4 is developmentally regulated in the mammary gland and has been implicated in breast oncogenesis. Both proteins comprise two tandemly repeated LIM domains. LMO2 and LMO4 interact with the ubiquitous nuclear adaptor protein ldb1/NLI/CLIM2, which associates with the LIM domains of LMO and LIM homeodomain proteins via its LIM interaction domain (ldb1-LID). We report the solution structures of two LMO:ldb1 complexes (PDB: 1M3V and 1J2O) and show that ldb1-LID binds to the N-terminal LIM domain (LIM1) of LMO2 and LMO4 in an extended conformation, contributing a third strand to a beta-hairpin in LIM1 domains. These findings constitute the first molecular definition of LIM-mediated protein-protein interactions and suggest a mechanism by which ldb1 can bind a variety of LIM domains that share low sequence homology.

CPAP is a novel stat5-interacting cofactor that augments stat5-mediated transcriptional activity.

Stat5, a member of the signal transducer and activators of transcription (Stat) protein family, is a primary mediator of prolactin (PRL) signaling in the mammary gland. There are two distinct Stat5 genes, Stat5a and Stat5b. The Stat5a isoform has been demonstrated to have an essential role in mammary epithelial differentiation, whereas Stat5b is required for dimorphic sexual growth. To search for proteins that interact with the C terminus of Stat5a, a highly divergent region amongst Stat family members, we performed a yeast two-hybrid screen of HBL100 and primary breast adenocarcinoma libraries. This led to the identification of a protein that had previously been isolated as a centrosomal P4.1-associated protein (CPAP). CPAP was shown to specifically interact with Stat5a and Stat5b but not with Stat1 or Stat3. Both the tyrosine phosphorylated and unphosphorylated forms of Stat5, as well as Stat5a/Stat5b heterodimers, could associate with CPAP. CPAP was expressed in human breast cancer cell lines and the developing mammary gland as well as in other tissues. Indirect immunofluorescence and cellular fractionation studies revealed that CPAP was predominantly cytoplasmic, with low levels in the nucleus. Nuclear levels of CPAP increased substantially upon activation of the PRL pathway, most likely reflecting cotranslocation of this protein with activated Stat5. Furthermore, CPAP was found to augment Stat5-mediated transcription. Thus, we have identified CPAP as a novel coactivator of Stat5 proteins in the PRL (and probably other) pathways.

The LIM domain protein LMO4 interacts with the cofactor CtIP and the tumor suppressor BRCA1 and inhibits BRCA1 activity.

LMO4 belongs to the LIM-only (LMO) group of transcriptional regulators that appear to function as molecular adaptors for protein-protein interactions. Expression of the LMO4 gene is developmentally regulated in the mammary gland and is up-regulated in primary breast cancers. Using LMO4 in a yeast two-hybrid screen, we have identified the cofactor CtIP as an LMO4-binding protein. Interaction with CtIP appeared to be specific for the LMO subclass of LIM domain proteins and could be mediated by a single LIM motif of LMO4. We further identified the breast tumor suppressor BRCA1 as an LMO4-associated protein. The C-terminal BRCT domains of BRCA1, previously shown to bind CtIP, also mediated interaction with LMO4. Tumor-associated mutations within the BRCT repeats that abolish interaction between BRCA1 and CtIP had no effect on the association of BRCA1 with LMO4. A stable complex comprising LMO4, BRCA1, and CtIP was demonstrated in vivo. The LIM domain binding-protein Ldb1 also participated in this multiprotein complex. In functional assays, LMO4 was shown to repress BRCA1-mediated transcriptional activation in both yeast and mammalian cells. These findings reveal a novel complex between BRCA1, LMO4, and CtIP and indicate a role for LMO4 as a repressor of BRCA1 activity in breast tissue.