The histone- and PRMT5-associated protein COPR5 is required for myogenic differentiation.

Myogenic differentiation requires the coordination between permanent cell cycle withdrawal, mediated by members of the cyclin-dependent kinase inhibitor (CKI) family, and activation of a cascade of myogenic transcription factors, particularly MYOGENIN (MYOG). Recently, it has been reported that the Protein aRginine Methyl Transferase PRMT5 modulates the early phase of induction of MYOG expression. Here, we show that the histone- and PRMT5-associated protein COPR5 (cooperator of PRMT5) is required for myogenic differentiation. C2C12 cells, in which COPR5 had been silenced, could not irreversibly exit the cell cycle and differentiate into muscle cells. This phenotype might be explained by the finding that, in cells in which COPR5 was downregulated, p21 and MYOG induction was strongly reduced and PRMT5 recruitment to the promoters of these genes was also altered. Moreover, we suggest that COPR5 interaction with the Runt-related transcription factor 1 (RUNX1)-core binding factor-ß (CBFß) complex contributes to targeting the COPR5-PRMT5 complex to these promoters. Finally, we present evidence that COPR5 depletion delayed the in vivo regeneration of cardiotoxin-injured mouse skeletal muscles. Altogether, these data extend the role of COPR5 from an adaptor protein required for nuclear functions of PRMT5 to an essential coordinator of myogenic differentiation.

The periodic down regulation of Cyclin E gene expression from exit of mitosis to end of G(1) is controlled by a deacetylase- and E2F-associated bipartite repressor element.

The expression of cyclin E and that of a few other bona fide cell cycle regulatory genes periodically oscillates every cycle in proliferating cells. Although numerous experiments have documented the role of E2F sites and E2F activities in the control of these genes as cells exit from G(0) to move through the initial G(1)/S phase transition, almost nothing is known on the role of E2Fs during the subsequent cell cycles. Here we show that a variant E2F-site that is part of the Cyclin E Repressor Module (CERM) (Le Cam et al., 1999b) accounts for the periodic down regulation of the cyclin E promoter observed between the exit from mitosis until the mid/late G(1) phase in exponentially cycling cells. This cell cycle-dependent repression correlates with the periodic binding of an atypical G(1)-specific high molecular weight p107-E2F complex (Cyclin E Repressor Complex: CERC2) that differs in both size and DNA binding behaviors from known p107-E2F complexes. Notably, affinity purified CERC2 displays a TSA-sensitive histone deacetylase activity and, consistent with this, derepression of the cyclin E promoter by trichostatin A depends on the CERM element. Altogether, this shows that the cell cycle-dependent control of cyclin E promoter in cycling cells is embroiled in acetylation pathways via the CERM-like E2F element.

pRB binds to and modulates the transrepressing activity of the E1A-regulated transcription factor p120E4F.

The retinoblastoma protein pRB is involved in the transcriptional control of genes essential for cell cycle progression and differentiation. pRB interacts with different transcription factors and thereby modulates their activity by sequestration, corepression, or activation. We report that pRB, but not p107 and p130, binds to and facilitates repression by p120(E4F), a ubiquitously expressed GLI-Kruppel-related protein identified as a cellular target of E1A. The interaction involves two distinct regions of p120(E4F) and the C-terminal part of pRB. In vivo pRB-p120(E4F) complexes can only be detected in growth-arrested cells, and accordingly contain the hypophosphorylated form of pRB. Repression of an E4F-responsive promoter is strongly increased by combined expression of p120(E4F) and pRB, which correlates with pRB-dependent enhancement of p120(E4F) binding activity. Elevated levels of p120(E4F) have been shown to block growth of mouse fibroblasts in G(1). We find this requires pRB, because RB(-/-) fibroblasts are significantly less sensitive to excess p120(E4F).

A CDE/CHR-like element mediates repression of transcription of the mouse RB2 (p130) gene.

The bipartite repressor elements, termed cell cycle-dependent element (CDE)/cell cycle regulatory element (CCRE)-cell cycle homology region (CHR) control the growth-dependent transcription of the cyclin A, cdc25C, cdc2 genes. Here, we have identified a functional element displaying the signature of the CDE-CHR in the promoter of the mouse RB2 (p130) gene, encoding the retinoblastoma protein family (pRB)-related protein p130. This element locates close to the major transcription start site where it makes major groove contacts with proteins that can be detected in a cellular context using in vivo genomic footprinting techniques. Inactivation of either the CDE or CHR sequence strongly up-regulates the p130 promoter activity in exponentially growing cells, a situation where endogenous p130 gene expression is almost undetectable. Electrophoretic mobility shift assays suggest that two different protein complexes bind independently to the p130 CDE and CHR elements, and that the protein(s) bound to the CDE might be related to those bound on cyclin A and cdc2 promoters.

Human E2F5 gene is oncogenic in primary rodent cells and is amplified in human breast tumors.

E2F transcription factors (E2F1 to 6) are central players in the control of animal cell proliferation as regulators of genes involved in cell cycle progression and in transformation. In this report, we have investigated the potential involvement of the E2F5 gene in tumorigenesis. We show that E2F5 can promote the formation of morphologically transformed foci in primary baby rat kidney cells (BRK) when it is overexpressed in the presence of its heterodimeric partner DP1 and activated RAS. This suggests that E2F5 behaves like a MYC-type cooperating oncogene in functional assays, prompting us to monitor potential amplifications of the E2F5 gene in primary human tumors. We mapped the human E2F5 gene to 8q21.1-21.3 equidistant from the MOS (8q12) and MYC (8q24) oncogenes. Since the long arm of chromosome 8 is frequently the site of increased gene copy number (ICN) in breast cancer, we screened 442 breast tumor DNAs for gains of E2F5, MOS, and MYC genes. The three genes showed ICN, albeit at variable incidence and levels of amplification, with the ICN of E2F5 occurring concomitantly with those of MOS and/or MYC in almost half of the cases. Moreover, a marked increase of the 2. 5-kb E2F5 transcript was also detected in some tumors and tumor cell lines. In conclusion, the evidence that sustained unregulated expression of E2F5 can cooperate with other oncogenes to promote cell transformation in functional assays, together with the detection of chromosomal amplifications and overexpressions of the E2F5 gene in breast tumors, provides the first indications that E2F5 deregulation may have a role in human tumor development.

Inhibition of mammalian cell proliferation by genetically selected peptide aptamers that functionally antagonize E2F activity.

The p16-cyclin D-pRB-E2F pathway is frequently deregulated in human tumors. This critical regulatory pathway controls the G1/S transition of the mammalian cell cycle by positive and negative regulation of E2F-responsive genes required for DNA replication. To assess the value of the transcription factors E2Fs as targets for antiproliferative strategies, we have initiated a program aiming to develop inhibitors targeting specifically these proteins in vitro and in vivo. The cellular activity of E2F is the result of the heterodimeric association of two families of proteins, E2Fs and DPs, which then bind DNA. Here, we use a two hybrid approach to isolate from combinatorial libraries peptide aptamers that specifically interact with E2Fs DNA binding and dimerization domains. One of these is a potent inhibitor of E2F binding activity in vitro and in mammalian fibroblasts, blocks cells in G1, and the free variable region from this aptamer has the same effect. Our experiments argue that the variable region of this aptamer is structured, and that it functions by binding E2F with a motif that resembles a DP heterodimerization region, and blocking E2F's association with DP. These results show that cell proliferation can be inhibited using genetically-selected synthetic peptides that specifically target protein-protein interaction motifs within cell cycle regulators. These results also emphasize the critical role of the E2F pathway for cell proliferation and might allow the design of novel antiproliferative agents targeting the cyclin/CDK-pRB-E2F pathway.

Timing of cyclin E gene expression depends on the regulated association of a bipartite repressor element with a novel E2F complex.

Transient induction of the cyclin E gene in late G1 gates progression into S. We show that this event is controlled via a cyclin E repressor module (CERM), a novel bipartite repressor element located near the cyclin E transcription start site. CERM consists of a variant E2F-binding site and a contiguous upstream AT-rich sequence which cooperate during G0/G1 to delay cyclin E expression until late G1. CERM binds the protein complex CERC, which disappears upon progression through G0-G1 and reappears upon entry into the following G1. CERC disappearance correlates kinetically with the liberation of the CERM module in vivo and cyclin E transcriptional induction. CERC contains E2F4/DP1 and a pocket protein, and sediments faster than classical E2F complexes in a glycerol gradient, suggesting the presence of additional components in a novel high molecular weight complex. Affinity purified CERC binds to CERM but not to canonical E2F sites, thus displaying behavior different from known E2F complexes. In cells nullizygous for members of the Rb family, CERC is still detectable and CERM-dependent repression is functional. Thus p130, p107 and pRb function interchangeably in CERC. Notably, the CERC-CERM complex dissociates prematurely in pRb-/- cells in correspondence with the premature expression of cyclin E. Thus, we identify a new regulatory module that controls repression of G1-specific genes in G0/G1.

Novel retinoid-related molecules as apoptosis inducers and effective inhibitors of human lung cancer cells in vivo.

Lung cancer causes more than 140,000 deaths annually in the United States alone, and the prognosis for non-small cell lung cancer (NSCLC) is particularly poor. Therapies using small molecules that preferentially kill lung tumor cells by inducing cellular suicide (apoptosis) would therefore be highly desirable. Retinoids have shown promise as cancer preventive and cancer therapeutic agents. Retinoid signals are mediated by two classes of nuclear receptors: the retinoic acid receptors (RAR alpha, beta, and gamma) and the retinoid X receptors (RXR alpha, beta and gamma). These receptors usually bind as heterodimers to specific DNA sequences and/or interact with other transcriptional regulators, such as AP-1 (ref. 10) to regulate gene transcription. Synthetic retinoids can be made that activate only specific portions of the complex retinoid response network and activate selective biological programs. To identify retinoids with novel biological activities, we used a high-throughput "biological activity fingerprint" screen on a large library of retinoids and retinoid-related molecules (RRMs). We identified new structures that are highly effective against lung cancer cells in vitro, inducing apoptosis. We show here for one of these compounds that it is very effective against a human NSCLC in vivo in an animal model. These new molecules show a distinct pattern of receptor signaling.

Re-evaluation of the distributions of dystrophin and utrophin in sciatic nerve.

Differential expression of proteins belonging to the dystrophin family was analysed in peripheral nerves. In agreement with previous reports, no full-size dystrophin was detectable, only Dp116, one of the short dystrophin products of the Duchenne muscular dystrophy (DMD) gene. We used specific monoclonal antibodies to fully investigate the presence of utrophin, a dystrophin homologue encoded by a gene located on chromosome 6q24. Evidence is presented here of the presence of two potential isoforms of full-length utrophin in different nerve structures, which may differ by alternative splicing of the 3'-terminal part of the utrophin gene according to the specificities of the monoclonal antiobodies used. One full-length utrophin was co-localized with Dp116 in the sheath around each separate Schwann cell-axon unit, but the other utrophin isoform was found to be perineurium-specific. We also highlighted a potential 80 kDa utrophin-related protein. The utrophin distribution in peripheral nerves was re-evaluated and utrophin isoforms were detected at the protein level. This preliminary indication will require more concrete molecular evidence to confirm the presence of these two utrophin isoforms as well as the potential 80 kDa utrophin isoform, but the results strongly suggest that each isoform must have a specialized role and function within each specific nervous structure.

Dystrophin, the protein that promotes membrane resistance.

Deficiency of dystrophin, a 427-kDa subsarcolemma membrane protein, is responsible for Duchenne muscular dystrophy. The function of this protein is not clear but its subcellular distribution suggests that it is an important link between the cytoskeleton and the extracellular matrix, thus maintaining membrane integrity. The N-terminus of dystrophin was shown to bind actin in vivo and in vitro via two major actin binding sites. The role of dystrophin/actin interactions has been investigated and the results presented here demonstrate for the first time that the N-terminal part of dystrophin is able (i) to interact with G-actin monomers, and (ii) to slowly promote G->F actin transformation. This conversion was shown to be stimulated the presence of calmodulin in a calcium dependent manner. This is evidence that dystrophin is an anchor protein for actin involved in the control of membrane cell shape deformation and developing a calmodulin-calcium induced F-actin network, thus stiffening the myotube membrane cytoskeleton.

Parallel expression level of dystrophin and contractile performances of rat aortic smooth muscle.

We have recently shown that the expression of dystrophin was in line with the conservation of contractility in primary and subcultured vascular smooth muscle cells (VSMC). In this report, we provide experimental evidence that shows the existence of a gradient in dystrophin expression from the aortic arch to the region just above the diaphragm, in line with a parallel gradient of contractility. On the contrary, there were no differences in caldesmon and smooth muscle myosin expression, as previously shown for vimentin and alpha-smooth muscle actin [Osborn, M., Caselitz, J., and Weber, K. (1981) Differentiation 20, 196-202]. Overall, this is the first demonstration, in a normal adult smooth muscle, of a differential expression of dystrophin, in line with its contractile performances. In addition, these results suggest that the experimental modulation of dystrophin expression in cultured VSMC in line with their contractility was the manifestation in vitro of an actual physiological property of those cells in vivo.

N-terminal domain of dystrophin.

Contro-versial experiments have been published on calmodulin binding of dystrophin. In this study, we used recombinant proteins and the techniques of affinity chromatography and ELISA to show that the N-terminal part of dystrophin binds calmodulin specifically in a calcium-dependent manner. The calcium-dependent interaction of calmodulin and dystrophin does not directly regulate binding of actin to dystrophin, but may regulate dystrophin interactions with other associated proteins.

Dystrophin does not influence regular cytoskeletal architecture but is required for contractile performance in smooth muscle aortic cells.

Cultured vascular smooth muscle cells express distinct histological phenotypes due to a contractile to synthetic stage transition. In this study, we compared the behaviour of cultured aortic smooth muscle cells from young normal and mdx mice. Morphological, immunobiochemical, immunocytochemical analyses and contraction studies of these cells demonstrated that (i) the cell cytoskeleton in mdx mice is not affected by the absence of dystrophin since proteins such as caldesmon, a-actin, and vinculin are expressed similarly in normal mice, (ii) utrophin (or dystrophin-related protein) overexpression does not compensate for the physiological and functional role of the lacking dystrophin. These data suggested that dystrophin and utrophin cannot substitute one another and may play different or complementary roles within smooth muscle cells.

Utrophin localization in normal and dystrophin-deficient heart.

BACKGROUND: The localization of dystrophin at the sarcolemma of cardiac skeletal fibers and cardiac Purkinje fibers has been described. Dystrophin deficiency produces clinical manifestations of disease in skeletal muscles and hearts of patients with Duchenne and Becker muscular dystrophy. Utrophin (or dystrophin-related protein), a dystrophin homologous protein, was found to be expressed in fetal muscles and reexpressed in dystrophin-deficient skeletal muscle fibers. We therefore examined utrophin expression in normal and in dystrophin-deficient hearts. METHODS AND RESULTS: The expression and subcellular distribution of utrophin was examined in cardiac muscle by immunoblot and immunofluorescence analysis in normal bovine heart compared with dystrophin. Utrophin expression was also examined in normal and dystrophin-deficient hearts of MDX mice. Three monoclonal antibodies reacting with dystrophin and utrophin solely or reacting with both proteins along with two polyclonal antibodies reacting with either utrophin or dystrophin and utrophin were tested. In normal bovine heart, utrophin was not expressed at the periphery of fibers but was strongly expressed in intercalated disks and in the cytoplasm of cardiac Purkinje fibers. In cardiocytes, utrophin was colocalized along transverse T tubules with dystrophin. Dystrophin was present at the periphery of cardiocytes and cardiac Purkinje fibers as well as in transverse T tubules but was absent or faintly expressed in intercalated disks. The results with monoclonal and polyclonal antibodies were identical. Western blot analysis revealed that the detected molecules corresponded only to a 400-kD protein band and not to possible shorter transcripts of utrophin or dystrophin (apo-utrophin or apo-dystrophin). In dystrophin-deficient hearts of MDX mice, utrophin alone was abundant but not organized in the same networklike distribution. CONCLUSIONS: This first localization of utrophin in normal heart (in Purkinje fibers, transverse tubules, and intercalated disks) showed a distinct subcellular localization of this protein with dystrophin, suggesting an important function of this protein in intercellular communication. In dystrophin-deficient hearts of MDX mice, utrophin alone is overexpressed as in skeletal muscle sarcolemma, an area normally occupied by dystrophin but not organized in the same networklike distribution.

Characterization and localization of a 77 kDa protein related to the dystrophin gene family.

The Duchenne muscular dystrophy gene gives rise to transcripts of several lengths. These mRNAs differ in their coding content and tissue distribution. The 14 kb mRNA encodes dystrophin, a 427 kDa protein found in muscle and brain, and the short transcripts described encode DP71, a 77 kDa protein found in various organs. These short transcripts have many features common to the deduced primary structure of dystrophin, especially in the cysteine-rich specific C-terminal domains. The dystrophin C-terminal domain could be involved in membrane anchorage via the glycoprotein complex, but such a functional role for these short transcript products has yet to be demonstrated. Here we report the first isolation of a short transcript product from saponin-solubilized cardiac muscle membranes using alkaline buffer and affinity chromatography procedures. This molecule was found to be glycosylated and could be easily dissociated from cardiac muscle and other non-muscle tissues such as brain and liver. DP71-specific monoclonal antibody helped to identify this molecule as being related to the dystrophin gene family. Immunofluorescence analysis of bovine or chicken cardiac muscle showed a periodic distribution of DP71 in transverse T tubules and this protein was co-localized with the dystrophin glycoprotein complex in the Z-disk area.

Dystrophin (Xp21), a new phenotype marker of cultured rat aortic myocytes.

Dystrophin is a low-abundance cytoskeletal protein involved in the maintenance of membrane integrity in striated muscle. Very little is known about its role in smooth muscle. Utrophin (a dystrophin-related protein) is an ubiquitous protein whose role is still unclear. Changes in the expression of both proteins (if any) during phenotypic modulation of smooth muscle have not yet been reported. In contrast, modulated expression of heavy-molecular-weight caldesmon (h-CaD), a well-known specific regulatory protein of the contractile apparatus in smooth muscle, is well documented, along with its nonmuscle isoform, low-molecular-weight caldesmon (1-CaD), and other cytoskeletal proteins. We investigated three properties of cultured rat aortic smooth muscle cells: morphology, contractile ability, and expression of dystrophin, utrophin, h-CaD, and 1-CaD. Cells were grown either in serum substitute supplemented medium (U-medium), where they reexpressed contractility, or in fetal calf serum-supplemented medium (F-medium), where they did not. It was found that only cultures grown in U-medium continued expressing dystrophin, even during the proliferation phase, contrary to cells grown in F-medium. However, when F-medium was changed for U-medium the cells recovered their contractility and reexpressed dystrophin. Expression of utrophin, h-CaD, and 1-CaD was similar in both culture types. Dystrophin was demonstrated to be a true phenotype marker of cultured rat aortic smooth muscle cells, particularly with respect to their actual contractility.

Actin-dystrophin interface.

Dystrophin, an elongated cytoskeletal molecule which is deficient in Duchenne muscular disease, contains an actin-binding domain in its N-terminal portion. We show that this part interacted with actin in the native molecule. By molecular biology techniques, four recombinant proteins were expressed in Escherichia coli using the pMAL vector which allowed us to obtain soluble proteins directly after purification. These constructions were tested for their ability to bind actin under various conditions, and their apparent dissociation constants were determined. The effects of other actin-binding proteins such as caldesmon and tropomyosin were analyzed in comparison to the actin-binding properties of these constructions. These results support the potential concept of a multiple actin-binding contact in the N-terminal region of dystrophin. Differences in the functional domains are discussed relative to similar alpha-actinin-actin-binding sites.