Genome-wide analysis of p53-regulated transcription in Myc-driven lymphomas.

The tumour suppressor p53 is a transcription factor that controls cellular stress responses. Here, we dissected the transcriptional programmes triggered upon restoration of p53 in Myc-driven lymphomas, based on the integrated analysis of p53 genomic occupancy and gene regulation. p53 binding sites were identified at promoters and enhancers, both characterized by the pre-existence of active chromatin marks. Only a small fraction of these sites showed the 20 base-pair p53 consensus motif, suggesting that p53 recruitment to genomic DNA was primarily mediated through protein-protein interactions in a chromatin context. p53 also targeted distal sites devoid of activation marks, at which binding was prevalently driven by sequence recognition. In all instances, the relevant motif was the canonical unsplit consensus element, with no clear evidence for p53 recruitment by split motifs. At promoters, p53 binding to the consensus motif was associated with gene induction, but not repression, indicating that the latter was most likely indirect. Altogether, our data highlight key features of genome recognition by p53 and provide unprecedented insight into the pathways associated with p53 reactivation and tumour regression, paving the way for their therapeutic application.

Dual regulation of Myc by Abl.

The tyrosine kinase c-Abl (or Abl) and the prolyl-isomerase Pin1 cooperatively activate the transcription factor p73 by enhancing recruitment of the acetyltransferase p300. As the transcription factor c-Myc (or Myc) is a known target of Pin1 and p300, we hypothesized that it might be regulated in a similar manner. Consistent with this hypothesis, overexpression of Pin1 augmented the interaction of Myc with p300 and transcriptional activity. The action of Abl, however, was more complex than predicted. On one hand, Abl indirectly enhanced phosphorylation of Myc on Ser 62 and Thr 58, its association with Pin1 and p300 and its acetylation by p300. These effects of Abl were exerted through phosphorylation of substrate(s) other than Myc itself. On the other hand, Abl interacted with the C-terminal domain of Myc and phosphorylated up to five tyrosine residues in its N-terminus, the principal of which was Y74. Indirect immunofluorescence or immunohistochemical staining suggested that the Y74-phosphorylated form of Myc (Myc-pY74) localized to the cytoplasm and coexisted either with active Abl in a subset of mammary carcinomas or with Bcr-Abl in chronic myeloid leukemia. In all instances, Myc-pY74 constituted a minor fraction of the cellular Myc protein. Thus, our data unravel two potential effects of Abl on Myc: first, Abl signaling can indirectly augment acetylation of Myc by p300, and most likely also its transcriptional activity in the nucleus; second, Abl can directly phosphorylate Myc on tyrosine: the resulting form of Myc appears to be cytoplasmic, and its presence correlates with Abl activation in cancer.

Genome-wide mapping of Myc binding and gene regulation in serum-stimulated fibroblasts.

The transition from quiescence to proliferation is a key regulatory step that can be induced by serum stimulation in cultured fibroblasts. The transcription factor Myc is directly induced by serum mitogens and drives a secondary gene expression program that remains largely unknown. Using mRNA profiling, we identify close to 300 Myc-dependent serum response (MDSR) genes, which are induced by serum in a Myc-dependent manner in mouse fibroblasts. Mapping of genomic Myc-binding sites by ChIP-seq technology revealed that most MDSR genes were directly targeted by Myc, but represented a minor fraction (5.5%) of all Myc-bound promoters (which were 22.4% of all promoters). Other target loci were either induced by serum in a Myc-independent manner, were not significantly regulated or were negatively regulated. MDSR gene products were involved in a variety of processes, including nucleotide biosynthesis, ribosome biogenesis, DNA replication and RNA control. Of the 29 MDSR genes targeted by RNA interference, three showed a requirement for cell-cycle entry upon serum stimulation and 11 for long-term proliferation and/or survival. Hence, proper coordination of key regulatory and biosynthetic pathways following mitogenic stimulation relies upon the concerted regulation of multiple Myc-dependent genes.

A positive role for Myc in TGFbeta-induced Snail transcription and epithelial-to-mesenchymal transition.

Myc and transforming growth factor-beta (TGFbeta) signaling are mutually antagonistic, that is Myc suppresses the activation of TGFbeta-induced genes, whereas TGFbeta represses c-myc transcription. Here, we report a positive role for Myc in the TGFbeta response, consisting in the induction of an epithelial-to-mesenchymal transition (EMT) and the activation of the EMT-associated gene Snail. Knockdown of either Myc or the TGFbeta effectors SMAD3/4 in epithelial cells eliminated Snail induction by TGFbeta. Both Myc and SMAD complexes targeted the Snail promoter in vivo, DNA binding occurring in a mutually independent manner. Myc was bound prior to TGFbeta treatment, and was required for rapid Snail activation upon SMAD binding induced by TGFbeta. On the other hand, c-myc downregulation by TGFbeta was a slower event, occurring after Snail induction. The response of Snail to another cytokine, hepatocyte growth factor (HGF), also depended on Myc and SMAD4. Thus, contrary to their antagonistic effects on Cip1 and INK4b, Myc and SMADs cooperate in signal-dependent activation of Snail in epithelial cells. Although Myc also targeted the Snail promoter in serum-stimulated fibroblasts, it was dispensable for its activation in these conditions, further illustrating that the action of Myc in transcriptional regulation is context-dependent. Our findings suggest that Myc and TGFbeta signaling may cooperate in promoting EMT and metastasis in carcinomas.

Expression of cyclins E1 and E2 during mouse development and in neoplasia.

Cyclin E1 (formerly called cyclin E) and the recently described cyclin E2 belong to the family of E-type cyclins that operate during the G(1)/S phase progression in mammalian cells. The two E-cyclins share a catalytic partner, cyclin-dependent kinase 2 (CDK2), and activate their associated kinase activities at similar times during cell cycle progression. Despite these similarities, it is unknown whether the two proteins perform distinct functions, or, alternatively, they control S-phase entry of different cell types in a tissue-specific fashion. To start addressing in vivo functions of E-cyclins, we determined the expression pattern of cyclins E1 and E2 during normal mouse development. We found that the two E-cyclins showed very similar patterns of expression; both were expressed within the proliferating compartment during embryo development. Analyses of cells and tissues lacking members of the retinoblastoma (pRB) family of proteins revealed that the expression of both cyclins is controlled in a pRB-dependent, but p107- and p130-independent fashion, likely through the pRB-dependent E2F transcription factors. We also found that cyclins E1 and E2 are expressed at high levels in mouse breast tumors driven by the Myc oncogene. Last, we found that cyclin E2 is overexpressed in approximately 24% of analyzed human mammary carcinomas. Collectively these findings suggest that the expression of cyclins E1 and E2 is governed by similar molecular circuitry.

Suppression of bile acid synthesis, but not of hepatic cholesterol 7alpha-hydroxylase expression, by obstructive cholestasis in humans.

Regulation of bile acid synthesis, a key determinant of cholesterol homeostasis, is still incompletely understood. To elucidate the feedback control exerted on bile acid biosynthesis in humans with obstructive cholestasis, 16 patients with bile duct obstruction were studied. In vivo 7alpha-hydroxylation, reflecting bile acid synthesis, was assayed in 13 of them by tritium release analysis. Serum 27-hydroxycholesterol was determined by gas chromatography-mass spectrometry. In a subgroup, hepatic cholesterol 7alpha-hydroxylase mRNA was assayed by real-time polymerase chain reaction (PCR), enzyme activity was determined by isotope incorporation, and microsomal cholesterol content was assayed by gas chromatography-mass spectrometry. Age-matched control subjects were studied in parallel. Hydroxylation rates were lower in cholestatic patients (108 +/- 33 mg of cholesterol per day, mean +/- SEM; controls: 297 +/- 40 mg/d; P <.01). The reduction was proportional to the severity of cholestasis, and synthetic rates were normalized in 4 subjects restudied after resolution of biliary obstruction. Consistent findings were obtained by analysis of serum 7alpha-hydroxycholesterol levels. On the other hand, hepatic cholesterol 7alpha-hydroxylase mRNA, microsomal enzyme activity, and cholesterol content tended to be increased in cholestasis. Finally, serum 27-hydroxycholesterol levels were slightly reduced in cholestatic subjects and were not related with the severity of the disease. Suppression of in vivo bile acid synthesis with no corresponding reduction in tissue 7alpha-hydroxylase expression and activity is consistent with nontranscriptional, posttranslational levels of regulation; these may play a role in the feedback control of bile acid synthesis in particular conditions. Alteration of the alternate biosynthetic pathway seems unlikely according to the present data.

Binding of c-Myc to chromatin mediates mitogen-induced acetylation of histone H4 and gene activation.

The Myc protein binds DNA and activates transcription by mechanisms that are still unclear. We used chromatin immunoprecipitation (ChIP) to evaluate Myc-dependent changes in histone acetylation at seven target loci. Upon serum stimulation of Rat1 fibroblasts, Myc associated with chromatin, histone H4 became locally hyperacetylated, and gene expression was induced. These responses were lost or severely impaired in Myc-deficient cells, but were restored by adenoviral delivery of Myc simultaneous with mitogenic stimulation. When targeted to chromatin in the absence of mitogens, Myc directly induced H4 acetylation. In addition, Myc recruited TRRAP to chromatin, consistent with a role for this cofactor in histone acetylation. Finally, unlike serum, Myc alone was very inefficient in inducing expression of most target genes. Myc therefore governs a step, most likely H4 acetylation, that is required but not sufficient for transcriptional activation. We propose that Myc acts as a permissive factor, allowing additional signals to activate target promoters.

Function of the c-Myc oncoprotein in chromatin remodeling and transcription.

Deregulated expression of the c-myc proto-oncogene contributes to malignant progression of a variety of tumors. The c-Myc protein (or Myc) is a transcription factor that positively or negatively regulates expression of distinct sets of target genes. Transcriptional activation by Myc is mediated through dimerization with Max and binding to the DNA consensus sequence CA(C/T)GTG (the E-box). Transcriptional inhibition is mediated through distinct DNA elements, and may be due to functional interference with factors that transactivate via these sequences. We review here our current knowledge on these transcriptional activities of Myc and their relationship to its biological function. The findings that Myc interacts with subunits of histone acetyl-transferase (HAT) complexes and of the ATP-dependent chromatin remodeling complex, SWI/SNF, suggest that localized changes in chromatin structure may mediate Myc function. We present a working hypothesis for the concerted action of HAT and SWI/SNF complexes in Myc-activated transcription and argue that this model should prompt re-thinking of the experimental strategies and criteria used to identify Myc target genes.

Cholesterol metabolism in primary biliary cirrhosis during simvastatin and UDCA administration.

Little is known about the effects of cholesterol-lowering agents in hypercholesterolemic patients with primary biliary cirrhosis (PBC). The aim of this study was to compare the changes induced by simvastatin and ursodeoxycholic acid (UDCA) on cholesterol metabolism in patients with PBC and preserved liver function. Six patients with PBC were administered simvastatin (40 mg/day) for 30 days and, after a washout period of 30 days, ursodeoxycholic acid (600 mg/day) for 30 days. Serum levels of lathosterol, campesterol, 7 alpha-hydroxycholesterol, and 27-hydroxycholesterol were measured by gas chromatography-mass spectrometry. During simvastatin administration, reduction of cholesterol levels (34% in 30 days) was paralleled by the decrease of lathosterol (55%), whereas concentrations of campesterol and of the two hydroxysterols were not substantially modified. During ursodeoxycholic acid administration, a trend toward a decrease of serum cholesterol concentrations was observed after only one year of treatment, and these changes were paralleled by the decrease of campesterol serum levels. Both simvastatin and UDCA were well tolerated, and a reduction of serum liver enzyme levels occurred with the latter. Simvastatin proved to be safe and effective in reducing serum cholesterol levels in patients with PBC by an inhibitory effect on cholesterol synthesis occurring within 24 h. --Del Puppo, M., M. Galli Kienle, A. Crosignani, M. L. Petroni, B. Amati, M. Zuin, and M. Podda. Cholesterol metabolism in primary biliary cirrhosis during simvastatin and UDCA administration. J. Lipid Res. 2001. 42: 437--441.

Recruitment of TRRAP required for oncogenic transformation by E1A.

TRRAP links Myc with histone acetylases and appears to be an important mediator of its oncogenic function. Here we show that interaction with TRRAP is required for cellular transformation not only by Myc, but also by the adenovirus E1A protein. Substitution of the 262 N-terminal residues of Myc with a small domain of E1A (residues 12-54) restores Myc transforming function. E1A(12-54) contains a TRRAP-interaction domain, that recruits TRRAP to either E1A-Myc chimeras, or the native 12S E1A protein. Overexpression of a competing TRRAP fragment in vivo blocks interaction of cellular TRRAP with either E1A-Myc or E1A, and suppresses cellular transformation by both oncoproteins. Moreover, E1A(Delta26-35) that fails to bind TRRAP but is capable of binding the Retinoblastoma (Rb)-family and p300/CBP proteins is defective in cellular immortalization, transformation and cell cycle deregulation. Thus in addition to disrupting Rb and p300/CBP functions, E1A must recruit TRRAP to transform cells.

Conserved region 2 of adenovirus E1A has a function distinct from pRb binding required to prevent cell cycle arrest by p16INK4a or p27Kip1.

Ectopic expression of the CDK inhibitors (CKIs) p16INK4a and p27Kip1 in Rat1 fibroblasts induces dephosphorylation and activation of Retinoblastoma-family proteins (pRb, p107 and p130), their association with E2F proteins, and cell cycle arrest in G1. The growth-inhibitory action of p16, in particular, is believed to be mediated essentially via pRb activation. The 12S E1A protein of human Adenovirus 5 associates with pRb-family proteins via residues in its Conserved Regions (CR) 1 and 2, in particular through the motif LXCXE in CR2. These interactions are required for E1A to prevent G1 arrest upon co-expression of CKIs. We show here that mutating either of two conserved motifs adjacent to LXCXE in CR2, GFP and SDDEDEE, also impairs the ability of E1A to overcome G1 arrest by p16 or p27. Strikingly, however, these mutations affect neither the association of E1A with pRb, p07 and p130, nor its ability to derepress E2F-1 transcriptional activity in transient transfection assays. One of the EIA mutants, however, is defective in derepressing several endogenous E2F target genes in the presence of p16 or p27. Thus, CR2 possesses an essential function besides pRb-binding. We speculate that this function might be required for the full derepression of E2F-regulated genes in their natural chromatin context.

Myc induces the nucleolin and BN51 genes: possible implications in ribosome biogenesis.

The c-Myc oncoprotein and its dimerization partner Max bind the DNA core consensus sequence CACGTG (E-box) and activate gene transcription. However, the low levels of induction have hindered the identification of novel Myc target genes by differential screening techniques. Here, we describe a computer-based pre-selection of candidate Myc/Max target genes, based on two restrictive criteria: an extended E-box consensus sequence for Myc/Max binding and the occurrence of this sequence within a potential genomic CpG island. Candidate genes selected by these criteria were evaluated experimentally for their response to Myc. Two Myc target genes are characterized here in detail. These encode nucleolin, an abundant nucleolar protein, and BN51, a co-factor of RNA polymerase III. Myc activates transcription of both genes via E-boxes located in their first introns, as seen for several well-characterized Myc targets. For both genes, mutation of the E-boxes abolishes transcriptional activation by Myc as well as repression by Mad1. In addition, the BN51 promoter is selectively activated by Myc and not by USF, another E-box-binding factor. Both nucleolin and BN51 are implicated in the maturation of ribosomal RNAs, albeit in different ways. We propose that Myc, via regulation of these and probably many other transcriptional targets, may be an important regulator of ribosome biogenesis.

Cyclin E2: a novel CDK2 partner in the late G1 and S phases of the mammalian cell cycle.

We report here the cloning and characterization of human and mouse cyclin E2, which define a new subfamily within the vertebrate E-type cyclins, while all previously identified family-members belong to the cyclin El subfamily. Cyclin E2/CKD2 and cyclin E/CDK2 complexes phosphorylate histone H1 in vitro with similar specific activities and both are inhibited by p27Kip1. Cyclin E2 mRNA levels in human cells oscillate throughout the cell cycle and peak at the G1/S boundary, in parallel with the cyclin E mRNA. In cells, cyclin E2 is complexed with CDK2, p27 and p21. Like cyclin E, cyclin E2 is an unstable protein in vivo and is stabilized by proteasome inhibitors. Cyclin E2-associated kinase activity rises in late G1 and peaks very close to cyclin E activity. In two malignantly transformed cell lines, cyclin E2 activity is sustained throughout S phase, while cyclin E activity has already declined and cyclin A activity is only beginning to rise. We speculate that cyclin E2 is not simply redundant with cyclin E, but may regulate distinct rate-limiting pathway(s) in G1-S control.

A novel function of adenovirus E1A is required to overcome growth arrest by the CDK2 inhibitor p27(Kip1).

We show here that the adenovirus E1A oncoprotein prevents growth arrest by the CDK2 inhibitor p27(Kip1) (p27) in rodent fibroblasts. However, E1A neither binds p27 nor prevents inhibition of CDK2 complexes in vivo. In contrast, the amount of free p27 available to inhibit cyclin E/CDK2 is increased in E1A-expressing cells, owing to reduced expression of cyclins D1 and D3. Moreover, E1A allows cell proliferation in the presence of supraphysiological p27 levels, while c-Myc, known to induce a cellular p27-inhibitory activity, is only effective against physiological p27 concentrations. E1A also bypasses G1 arrest by roscovitine, a chemical inhibitor of CDK2. Altogether, these findings imply that E1A can act downstream of p27 and CDK2. Retinoblastoma (pRb)-family proteins are known CDK substrates; as expected, association of E1A with these proteins (but not with p300/CBP) is required for E1A to prevent growth arrest by either p27 or the CDK4/6 inhibitor p16(INK4a). Bypassing CDK2 inhibition requires an additional function of E1A: the mutant E1A Delta26-35 does not overcome p27-induced arrest, while it binds pRb-family proteins, prevents p16-induced arrest, and alleviates pRb-mediated repression of E2F-1 transcriptional activity (although E1A Delta26-35 fails to restore expression of E2F-regulated genes in p27-arrested cells). We propose that besides the pRb family, E1A targets specific effector(s) of CDK2 in G1-S control.

Myc and the cell cycle.

Ectopic expression of the c-Myc oncoprotein prevents cell cycle arrest in response to growth-inhibitory signals, differentiation stimuli, or mitogen withdrawal. Moreover, Myc activation in quiescent cells is sufficient to induce cell cycle entry in the absence of growth factors. Thus, Myc transduces a potent mitogenic stimulus but, concomitantly, induces apoptosis in the absence of survival factors. We review here recent progress in our understanding of the molecular mechanisms linking Myc activity to cell cycle control. Myc is a positive regulator of G1-specific cyclin-dependent kinases (CDKs) and, in particular, of cyclin E/CDK2 complexes. Cyclin D/CDK4 and CDK6 may conceivably also be activated by Myc, but the circumstances in which this occurs remain to be explored. Myc acts via at least three distinct pathways which can enhance CDK function: (1) functional inactivation of the CDK inhibitor p27Kip1 and probably also of p21Cip1 and p57Kip2, (2) induction of the CDK-activating phosphatase Cdc25A and (3) - in an ill understood and most likely indirect way - deregulation of cyclin E expression. Constitutive expression of either Myc or cyclin E can prevent growth arrest by p16INK4a (an inhibitor of cyclin D/CDK4, but not of cyclin E/CDK2). In cells, p16INK4a inhibits phosphorylation, and thus induces activation of the Retinoblastoma-family proteins (pRb, p107 and p130). Surprisingly, this effect of p16 is not altered in the presence of Myc or cyclin E. Thus, Myc and cyclin E/CDK2 activity unlink activation of p16 and pRb from growth arrest. Finally, Myc may itself be a functional target of cyclin D/CDK4 through its direct interaction with p107. We discuss how the effects of Myc on cell cycle control may relate to its oncogenic activity, and in particular to its ability to cooperate with activated Ras oncoproteins.

Cyclin E and c-Myc promote cell proliferation in the presence of p16INK4a and of hypophosphorylated retinoblastoma family proteins.

Retroviral expression of the cyclin-dependent kinase (CDK) inhibitor p16(INK4a) in rodent fibroblasts induces dephosphorylation of pRb, p107 and p130 and leads to G1 arrest. Prior expression of cyclin E allows S-phase entry and long-term proliferation in the presence of p16. Cyclin E prevents neither the dephosphorylation of pRb family proteins, nor their association with E2F proteins in response to p16. Thus, cyclin E can bypass the p16/pRb growth-inhibitory pathway downstream of pRb activation. Retroviruses expressing E2F-1, -2 or -3 also prevent p16-induced growth arrest but are ineffective against the cyclin E-CDK2 inhibitor p27(Kip1), suggesting that E2F cannot substitute for cyclin E activity. Thus, cyclin E possesses an E2F-independent function required to enter S-phase. However, cyclin E may not simply bypass E2F function in the presence of p16, since it restores expression of E2F-regulated genes such as cyclin A or CDC2. Finally, c-Myc bypasses the p16/pRb pathway with effects indistinguishable from those of cyclin E. We suggest that this effect of Myc is mediated by its action upstream of cyclin E-CDK2, and occurs via the neutralization of p27(Kip1) family proteins, rather than induction of Cdc25A. Our data imply that oncogenic activation of c-Myc, and possibly also of cyclin E, mimics loss of the p16/pRb pathway during oncogenesis.

Phosphorylation-dependent degradation of the cyclin-dependent kinase inhibitor p27.

The p27(Kip1) protein associates with G1-specific cyclin-CDK complexes and inhibits their catalytic activity. p27(Kip1) is regulated at various levels, including translation, degradation by the ubiquitin/proteasome pathway and non-covalent sequestration. Here, we describe point mutants of p27 deficient in their interaction with either cyclins (p27(c-)), CDKs (p27(k-)) or both (p27(ck-)), and demonstrate that each contact is critical for kinase inhibition and induction of G1 arrest. Through its intact cyclin contact, p27(k-) associated with active cyclin E-CDK2 and, unlike wild type p27, p27(c-) or p27(ck-), was efficiently phosphorylated by CDK2 on a conserved C-terminal CDK target site (TPKK). Retrovirally expressed p27(k-) was rapidly degraded through the proteasome in Rat1 cells, but was stabilized by secondary mutation of the TPKK site to VPKK. In this experimental setting, exogenous wild-type p27 formed inactive ternary complexes with cellular cyclin E-CDK2, was not degraded through the proteasome, and was not further stabilized by the VPKK mutation. p27(ck-), which was not recruited to cyclin E-CDK2, also remained stable in vivo. Thus, selective degradation of p27(k-) depended upon association with active cyclin E-CDK2 and subsequent phosphorylation. Altogether, these data show that p27 must be phosphorylated by CDK2 on the TPKK site in order to be degraded by the proteasome. We propose that cellular p27 must also exist transiently in a cyclin-bound non-inhibitory conformation in vivo.

Growth arrest by the cyclin-dependent kinase inhibitor p27Kip1 is abrogated by c-Myc.

We show here that c-Myc antagonizes the cyclin-dependent kinase (CDK) inhibitor p27Kip1. p27 expressed from recombinant retroviruses in Rat1 cells associated with and inhibited cyclin E/CDK2 complexes, induced accumulation of the pRb and p130 proteins in their hypophosphorylated forms, and arrested cells in G1. Prior expression of c-Myc prevented inactivation of cyclin E/CDK2 as well as dephosphorylation of pRb and p130, and allowed continuous cell proliferation in the presence of p27. This effect did not require ubiquitin-mediated degradation of p27. Myc altered neither the susceptibility of cyclin E/CDK2 to inhibition by p27, nor the intrinsic CDK-inhibitory activity of p27, but induced sequestration of p27 in a form unable to bind cyclin E/CDK2. Neither Myc itself nor other G1-cyclin/CDK complexes were directly responsible for p27 sequestration. Retroviral expression of G1 cyclins (D1-3, E or A) or of the Cdc25A phosphatase did not overcome p27-induced arrest. Growth rescue by Myc required dimerization with Max, DNA binding and an intact transcriptional activation domain, as previously shown for cellular transformation. We propose that this activity is mediated by the product of an as yet unknown Myc-Max target gene(s) and represents an essential aspect of Myc's mitogenic and oncogenic functions.