Control of Cell Proliferation and Growth byMyc Proteins

Kaldis: Cell Cycle RegulationDOI 10.1007/004/Published online: 24 February 2006 © Springer-Verlag Berlin Heidelberg 2006 Control of Cell Proliferation and Growth by Myc Proteins Sandra B

P Kaldis: Cell Cycle Regulation

DOI 10.1007/004/Published online: 24 February 2006

© Springer-Verlag Berlin Heidelberg 2006

Control of Cell Proliferation and Growth by Myc Proteins

Sandra Bernard · Martin Eilers (u)

Institute for Molecular Biology and Tumor Research, University of Marburg,

35033 Marburg, Germany

eilers@imt.uni-marburg.de

Abstract Myc proteins act as signal transducers that alter cell proliferation in dependence

on signals from the extracellular environment In normal cells, the expression of MYC

genes is therefore under tight control by growth factor dependent signals The enormous interest in the function of these proteins is motivated by the observation that the close

control of MYC expression is disrupted in a large percentage of human tumors, leading to

deregulated expression of Myc proteins A large body of evidence shows that this dereg-ulation is a major driving force of human tumorigenesis; in cells with deregulated Myc, proliferation often takes place in the complete absence of external stimuli We will dis-cuss current models to understand Myc function and also potential avenues to selectively interfere with the proliferation of Myc-transformed cells.

1

Introduction

MYC genes form a small multigene family; the family has attracted enormous

attention, since the enhanced expression of one of its three members (MYC,

MYCL or MYCN) contributes to multiple human tumors Deregulation of

ex-pression can occur through diverse mechanisms, only some of which involve

the mutation of MYC genes themselves More frequently, mutations occur in human tumors in pathways that control the expression of MYC genes or that

control the function of the encoded proteins The full spectrum of such mu-tations remains to be elucidated, since the regulation of any member of the

MYC gene family is complex As a result, the precise percentage of human

tumors in which MYC genes are activated and/or deregulated remains a

mat-ter of some debate; it is possible that the ‘Myc pathway’, as suggested for the E2F pathway, needs to be deregulated for any human tumor to emerge Alternatively, there is evidence that at least some cells proliferate in a Myc-independent manner and therefore it is possible that tumors derived from

such cells do not have an ‘activated’ MYC gene.

Exogenously introduced MYC genes that are expressed under the con-trol of a strong promoter (to mimic the activation of MYC genes seen in

human tumors) elicit a stereotype response in most cells into which they are introduced: they promote cell proliferation even in the absence of mi-togenic signals, they often promote cell growth and they almost

invari-ably promote apoptosis or at least sensitize cells to apoptotic stimuli Ker-atinocytes are one of only a few examples of cells that respond differently

to Myc: in skin, proliferation is tightly linked to adhesion to the basal lam-ina and Myc disrupts this adhesion; therefore, keratinocytes respond to Myc with premature differentiation and arrest of proliferation (Gandarillas and Watt 1997)

The frequent activation of MYC genes in tumorigenesis has suggested that

Myc proteins might also have an important function in normal prolifera-tion and growth and this suggesprolifera-tion has by now been tested in numerous

experimental systems: Rat1 fibroblasts, in which both alleles of c-myc have

been deleted, show a reduced rate of proliferation and cell growth (Mateyak

et al 1997; Mateyak et al 1999) Similarly, mouse embryo fibroblasts that

carry a floxed allele of c-myc arrest upon cre-mediated excision of c-myc (de Alboran et al 2001) Importantly, deletion of the mnt gene, which encodes

a member of the Mad family of antagonists of Myc, restores proliferation to

c-myc deleted cells, suggesting that the balance of expression of members of

the Myc/Max/Mad network of proteins (see Fig 1) may dictate the response

to loss of Myc (Walker et al 2005)

In mice, the effects of deletion of the c-myc or n-myc genes in vivo are not uniform Deletion of c-myc in mice results in embryonic lethality and

certain cell types do not proliferate (de Alboran et al 2001; Trumpp et al 2001) In the hematopoietic lineage, stem cells continue to proliferate upon genetic ablation of c-myc, whereas differentiated (lineage positive) cells

ar-rest (Wilson et al 2004) While these data show that c-myc is not required for proliferation of hematopoietic stem cells, such cells also express n-myc and therefore it is possible that n-myc provides essential functions of Myc

proteins in these cells Similarly, postnatal proliferation of hepatocytes in the liver does not require c-Myc, but again the potential compensation by other

myc family members is not completely clear (Baena et al 2005) In the small

intestine, deletion of c-myc leads to a transient failure to form normal

num-bers of crypts in the small intestine, but in long-term experiments the mice maintain a normal epithelium in the absence of c-Myc activity and without apparent compensation by N-Myc or L-Myc (Bettess et al 2005) Both findings argue that at least some cell types can proliferate in the absence of functional Myc There are also clear examples for a strict requirement for Myc

func-tion in cell proliferafunc-tion: for example, delefunc-tion of n-myc in neuronal precursor

cells leads to a dramatic loss of proliferative capacity of such cells (Knoepfler

et al 2002) As long as questions of redundancy between different members

of the myc gene family and compensation by loss of antagonists such as mnt

are not fully resolved, it seems to us that no definitive answer is possible as

to whether Myc function is generally required for cell proliferation and cell growth or whether its essential role in proliferation is restricted to specific cell types

Fig 1 Transcriptional regulatory complexes formed by Myc proteins and their co-factors For details, see text Almost certainly, Myc forms more than one repressive complex, so the Myc/Miz1 complex should be viewed as one well-understood example of a group of similar complexes

Mechanisms of Myc Action

Myc proteins act at least in part as transcription factors that activate and re-press large groups of genes (see Fig 2) They activate transcription as part of

a binary complex together with an obligate partner protein, Max; the complex binds to a specific sequence, termed E-Box (CACGTG or related sequences), which is found in all genes that are activated by Myc Interestingly, Myc proteins not only regulate protein-coding genes that are transcribed by poly-merase II

There are at least three exceptions: first, there is evidence that microRNAs can be target genes that are activated by Myc: for example, a microRNA

tar-geting E2F1 is induced by Myc (O’Donnell et al 2005) Since the e2f 1 gene

is at the same time a target for transcriptional upregulation by Myc, the data suggest that there may be a fine tuning of expression of Myc target genes and proteins (Baudino et al 2003) In the case of E2F1, there is evidence both that

it mediates proliferation downstream of Myc (in B-lymphocytes) and that it mediates Myc-dependent apoptosis, so the microRNA-mediated regulation of E2F1 protein levels may ensure the correct balance between the two (Baudino

et al 2003; Leone et al 2001)

The second exception are ribosomal RNA genes Myc activates transcrip-tion through direct binding to canonical binding sites in the rDNA promoter (Arabi et al 2005; Grandori et al 2005; Poortinga et al 2004) rDNA genes are transcribed by polymerase I in the nucleolus, arguing that at least a fraction

of Myc proteins are localized in this compartment Indeed, immunofluores-cence experiments show that Myc can be found in the nucleolus, in particular after proteasome inhibition The latter finding also suggests that Myc can be degraded in the nucleolus and at least one E3 ligase that targets Myc, Fbw7γ,

is specifically localized in this compartment (Welcker et al 2004a,b; Yada et al 2004)

The third exception are several tRNA genes, demonstrating that Myc can also stimulate polymerase III-dependent transcription (Gomez-Roman et al

Fig 2 Genetic targets of Myc in cell proliferation and their proposed functions

2003) There is clear evidence that this effect is direct and not mediated

by activation of protein-coding genes that regulate polymerase III function; however, the mechanism of activation has not been clarified

Myc proteins also act as transcriptional repressors; all current models suggest that this is mediated by protein/protein interactions with other tran-scription factors One such factor is the Myc-interacting zinc finger protein, Miz1: through interaction with Miz1 (and likely additional proteins) Myc is recruited to non-canonical sites in the genome (Schneider et al 1997; Seoane

et al 2002; Seoane et al 2001; Staller et al 2001; Wanzel et al 2003) Free Miz1 acts as a transcriptional activator; in contrast, the Myc/Miz1 complex re-presses transcription from the same binding sites Interestingly, at least one other oncogene, Bcl6, uses Miz1 as a ‘platform’ to repress transcription, sug-gesting that both Bcl6 and Myc may repress a common set of target genes (Phan et al 2005)

Mechanistically, many questions remain open Myc proteins interact with

a number of potential co-activator and co-repressor proteins and, for some

of them, functions in Myc-dependent activation and repression have been demonstrated Some of these are shown in Fig 1 Potentially, the best-understood interaction is that with TRRAP, since Myc recruits two distinct histone acetylases, Gcn5 and Tip60, to Myc/Max target sites in vivo through interaction with TRRAP (Bouchard et al 2001; Frank et al 2003; McMahon

et al 1998; McMahon et al 2000) TRRAP binds to a highly conserved do-main in the amino-terminus of Myc proteins (MycboxII), and Myc controls acetylation of its target genes in a MycboxII-dependent manner Mutations

in this domain impair both activation and repression of many, but not all, genes by Myc, and abolish transformation by Myc (Nikiforov et al 2002) To-gether, the data strongly support the view that TRRAP-dependent stimulation

of local histone acetylation is a key function in transcriptional activation by Myc The precise role of most other interactions is less clear, mainly because genetic analyses are missing or have yielded unexpected results For example, while mutations in Tip48 or Tip49 (which also bind to MycboxII) geneti-cally interact with mutations in Myc in Drosophila, the pattern of genes that are regulated by either mutation show only very little overlap (Bellosta et al 2005) Clearly, more work will be required to achieve a clear picture of which protein interactions of Myc contribute to which aspects of its transcriptional regulatory functions and how those link its biological properties

This is most certainly true for an exciting link that emerges between ubi-quitination of Myc and its transcriptional properties There are two aspects

to this link: first, phosphorylation of threonine 58 (T58) by GSK3 stimulates recognition by Fbw7γ, an SCF-type E3 ligase complex (Welcker et al 2004b;

Yada et al 2004) T58 mutations are frequently found in human lymphomas, suggesting that this is one mechanism selecting for enhanced levels of Myc

in human tumors Surprisingly, however, these mutations also show an al-tered gene-regulatory behavior in that they fail to repress p21Cip1, a target of

the Myc/Miz1 complex and, potentially as an indirect result, fail to induced Bim1, a pro-apoptotic protein that is downstream of p21Cip1 (Collins et al 2005; Hemann et al 2005) Consequently, such mutations allow lymphomas to develop, which retain wild-type p53 and ARF and therefore bypass a muta-tional requirement that exist for wild-type Myc Does that mean that Fbw7γ

is a co-factor of repression? Possibly the situation is very similar to that for Mdm2 and p53, where Mdm2-mediated histone ubiquitination contributes to repression of p53 function by Mdm2 (Minsky and Oren 2004) Similarly, ubi-quitination of Myc by Skp2 has been shown to contribute to degradation of Myc and its gene-regulatory properties (Kim et al 2003; von der Lehr et al 2003)

Secondly, the ARF tumor suppressor inhibits transcriptional activation, but not repression by Myc (Qi et al 2004) The mechanism of this repres-sion is not completely clear and may involve direct interactions of both proteins Alternatively, ARF acts through inhibiting the E3-ligase function

of HectH9/p500ARF-BP1, a major ARF binding protein in human cells (Ad-hikary et al 2005; Chen et al 2005) HectH9 is also an E3-ligase for Myc and assembles a K63-linked poly-ubiquitin chain required for transcriptional ac-tivation by Myc Both findings strongly suggest that ubiquitination of Myc is not solely used to control Myc levels, but to control its biological functions as well

3

Targets

All available evidence suggests that Myc proteins affect both cell growth and cell proliferation through multiple pathways and that there is neither a single critical effector gene nor a small group of such genes In cell culture, Myc can affect cell growth, Cdk2 activity, E2F-dependent transcription and intermedi-ary metabolism and the effects on each pathway can be genetically separated from each other

Myc has been shown to stimulate both polymerase I- and polymerase III-dependent transcription, leading to enhanced transcription of ribosomal RNA and tRNA genes and to stimulate transcription of many genes involved

in all steps of protein translation (Arabi et al 2005; Boon et al 2001; Gomez-Roman et al 2003; Grandori et al 2005) While a single ‘master’ gene down-stream of Myc may not exist, there can be little doubt that stimulation of protein translation and biogenesis is a major effector pathway of Myc Un-doubtedly, the stimulation of cell growth will also feed back and contribute

to the stimulation of cell proliferation by Myc, although there is no evi-dence to suggest that this is the ‘primary’ function of Myc in stimulating cell proliferation in mammalian cells (Beier et al 2000; Trumpp et al 2001) Simi-lar arguments hold for many enzymes of intermediary metabolism that are

activated by Myc, such as lactate dehydrogenase A (LDH-A) and ornithine decarboxylase: induction of these genes contributes to and can be required for Myc-induced proliferation and tumorigenesis, but there is little evidence that the effects of Myc on cell proliferation are in any way mediated by the activation of these genes (Guo et al 2005; Shim et al 1997)

Several direct target genes of Myc contribute to the phosphorylation of the retinoblastoma protein and the regulation of E2F-dependent transcription Cyclins D1 and D2 have been suggested to be downstream targets of the Myc protein (Bouchard et al 1999; Kenney et al 2003; Oliver et al 2003; Oster-hout et al 1999; Perez-Roger et al 1999), and detailed mechanistic studies have been published for cyclin D2 regulation by Myc (Bouchard et al 2001, 2004) Also, Cdk4 is a target for transactivation by Myc (Hermeking et al 2000) and the Cdk4 inhibitor, p15Ink4b, is a direct target for repression by Myc (Seoane et al 2001; Staller et al 2001) In vivo, N-Myc is essential to main-tain expression of cyclin D2 in neuronal precursor cells and the phenotypes

of an N-Myc knockout are similar to those of a cyclin D2 knockout (Knoepfler

et al 2002; Kowalczyk et al 2004) N-Myc is also required to suppress ex-pression of p18Ink4c in vivo, but it is not absolutely clear whether this is

a direct regulation (Knoepfler et al 2002) Cdk4 is essential for Myc-induced tumorigenesis in skin (Miliani de Marval et al 2004), and fibroblasts lack-ing D-type cyclins are resistant to transformation by Myc (Kozar et al 2004), while fibroblasts expressing any single D-type cyclin can be transformed by Myc (Yu et al 2005) Repression of p15Ink4bis critical for stimulation of pro-liferation by Myc in keratinocytes (Gebhardt et al 2005) Taken together, the data strongly suggest that regulation of Cdk4 function is a critical function of Myc in tumorigenesis

Most likely, the regulation of Cdk4 reflects at least in part a requirement for phosphorylation of the retinoblastoma protein or related pocket proteins in Myc-driven proliferation, since Myc leads to an activation of E2F-dependent transcription and distinct activating E2F proteins act downstream of Myc in controlling cell proliferation and tumorigenesis (Baudino et al 2003; Jansen-Dürr et al 1993; Leone et al 2001) In addition, expression of E2F-target genes

is closely correlated with expression of specific MYC genes in human tumors,

arguing that Myc proteins act upstream of E2F-dependent transcription in human tumorigenesis (Hernando et al 2004) [B Samans and M Eilers, un-published observations]

The requirement for Cdk4 function may also reflect a need to sequester p27Kip1 or p21Cip1, leading to activation of cyclin E/Cdk2 kinase activity downstream of Myc (Leone et al 1997; Steiner et al 1995; Vlach et al 1996) Indeed, transcription of both p27Kip1 and p21Cip1is suppressed by enhanced expression of Myc, supporting the notion that Myc may regulate Cdk2 kinase activity (Herold et al 2002; Seoane et al 2002; Yang et al 2001) Tissue-culture experiments suggest that activation of Cdk2 is required for Myc-induced S-phase entry (Rudolph et al 1996) However, the recent findings that Cdk2

is dispensable for cell proliferation in vivo forces one to re-consider this is-sue Rigorous experiments analyzing Myc-induced proliferation and/or trans-formation in cells lacking Cdk2 or cyclins E1 or E2 are currently missing

A related issue is whether Cdk2 activation is required for Myc-induced apop-tosis and there are models suggesting that suppression of p21Cip1 by Myc favors Myc-dependent apoptosis and p53-dependent apoptosis upon DNA damage at the expense of cell-cycle arrest, potentially through upregulation

of Bim1 (Collins et al 2005; Hemann et al 2005)

4

Checkpoints and Apoptosis

One of the most exciting questions in the field addresses the issues of what failsafe processes limit Myc-induced proliferation and transformation and which mutations disable such failsafe mechanisms during tumorigenesis There are multiple indications that activation of Myc causes multiple prob-lems for the cells, the clearest being that Myc can induce apoptosis or dramat-ically sensitize the cells to apoptotic stimuli Several mechanisms contribute

to this effect, including repression of Bcl-xl, the induction of ARF, an activa-tor of p53, and the regulation of Fas ligand Blocking apoptosis by providing exogenous bclxldramatically accelerates Myc-induced tumorigenesis in vivo, demonstrating that apoptosis indeed provides a physiological barrier to Myc-dependent tumorigenesis

There is also clear evidence that deregulation of Myc causes DNA damage: for example, cells in which Myc has been acutely activated stain positive for phosphorylated histone H2Ax and show foci of Mre11, indicative of double-strand breaks (Bartkova et al 2005; Vafa et al 2002); a potential explanation

is provided by the observation that the repair of double-strand breaks occurs

inefficiently in human fibroblasts that express c-myc (Karlsson et al 2003) It

has been argued that these are tissue-culture problems due to unphysiologic-ally high concentrations of oxygen (Soucek and Evan 2002) This is formunphysiologic-ally possible: however, several Myc-induced tumors are genomically unstable and show complex karyotypes and enhanced rates of double-strand breaks, which suggests that deregulated expression of Myc can also promote genomic insta-bility in vivo Whether induction of DNA damage by Myc is a causal event

in Myc-induced tumorigenesis is an open question: clearly, a transgenic situ-ation can be engineered in mice where this is almost certainly not the case (Pelengaris et al 2002) However, such an engineered situation is likely to bypass key steps of early tumorigenesis by activating Myc simultaneously in

a large group of cells or even entire organs

A second important issue is whether both induction of apoptosis and induction of DNA damage by Myc are causally related Mediators of Myc-induced apoptosis include p53 and E2F1 and both have been implicated in

DNA damage responses (Hermeking and Eick 1994; Leone et al 2001) How-ever, activation of p53 by Myc is mediated by induction of ARF, at least in lymphoid cells and in mouse embryo fibroblasts (Zindy et al 1998) While induction of ARF by Myc is not direct and the intermediate steps in ARF activation have not been resolved, little evidence exists to suggest that ARF expression is responsive to DNA damage The situation is different in hu-man fibroblasts, where induction of p53 by Myc occurs independently of p14ARFand can be blocked by pharmacological inhibition of ATM and ATR; these findings suggest that Myc activates key components of the DNA dam-age pathway (Lindstrom and Wiman 2003) Whether this activation occurs

as a response to DNA damage or as a response to alterations to chromatin structure (which activate ATM) is open In vivo studies that measure the im-pact of ATM mutations on Myc-induced tumorigenesis will be required to unequivocally answer these important questions

If indeed induction of DNA damage and apoptosis are hallmarks of Myc-dependent tumors in vivo, then tumors in which Myc has been activated should harbor secondary mutations that are required to suppress the check-point responses to damaged DNA This would be an exciting class of muta-tions to explore, since restoration of checkpoints or targeting specific apop-totic pathways may lead to a synthetic lethal effect, selectively killing Myc-dependent tumor cells while sparing normal cells At present, there are two indications that such mutations may indeed exist: one is the finding that cells lacking the Werner syndrome (WRN) helicase, a protein that has been implicated in DNA repair, undergo strongly accelerated senescence follow-ing expression of Myc (Grandori et al 2003) A second indication is the recent identification of synthetic lethal interactions of deregulated Myc with the TRAIL-receptor pathway, which argue that targeting Myc degradation may lead to selective killing of tumor cells that express high levels of Myc (Rottmann et al 2005; Wang et al 2004) Clearly, the systematic identifica-tion of such synthetic lethal interacidentifica-tions holds a promise for the therapy of Myc-dependent tumors

5

Conclusions

Myc proteins act at least in part as transcription factors that activate and repress large groups of genes; as a result, tumors that express high levels

of Myc proteins have a distinct gene-expression profile Many of the regula-tory mechanisms that control Myc function and the transcriptional mechan-isms through which Myc proteins act have been clarified over the last years

In contrast, questions remain about the identity of the genes that mediate Myc’s various biological functions Similarly, convincing evidence shows that Myc-induced proliferation is balanced by apoptosis, yet the precise

mech-anisms that recognize Myc-induced proliferation as aberrant remain to be elucidated

Acknowledgements Work in the authors’ laboratory is supported by the Deutsche Forschungsgemeinschaft, the European Community through the Framework 6 program, the Thyssen- and the Sander-Stiftung, AICR and the Deutsche Krebshilfe.

References

Adhikary S, Marinoni F, Hock A, Hulleman E, Popov N, Beier R, Bernard S, Quarto M, Capra M, Goettig S, Kogel U, Scheffner M, Helin K, Eilers M (2005) The ubiquitin lig-ase HectH9 regulates transcriptional activation by Myc and is essential for tumor cell proliferation Cell 123:409–421

Arabi A, Wu S, Ridderstrale K, Bierhoff H, Shiue C, Fatyol K, Fahlen S, Hydbring P, Soder-berg O, Grummt I, Larsson LG, Wright AP (2005) c-Myc associates with ribosomal DNA and activates RNA polymerase I transcription Nat Cell Biol 7:303–310

Baena E, Gandarillas A, Vallespinos M, Zanet J, Bachs O, Redondo C, Fabregat I, Mar-tinez AC, de Alboran IM (2005) c-Myc regulates cell size and ploidy but is not essential for postnatal proliferation in liver Proc Natl Acad Sci USA 102:7286–7291

Bartkova J, Horejsi Z, Koed K, Kramer A, Tort F, Zieger K, Guldberg P, Sehested M, Nesland JM, Lukas C, Orntoft T, Lukas J, Bartek J (2005) DNA damage response as

a candidate anti-cancer barrier in early human tumorigenesis Nature 434:864–870 Baudino TA, Maclean KH, Brennan J, Parganas E, Yang C, Aslanian A, Lees JA, Sherr CJ, Roussel MF, Cleveland JL (2003) Myc-mediated proliferation and lymphomagenesis, but not apoptosis, are compromised by E2f1 loss Mol Cell 11:905–914

Beier R, Burgin A, Kiermaier A, Fero M, Karsunky H, Saffrich R, Moroy T, Ansorge W, Roberts J, Eilers M (2000) Induction of cyclin E-cdk2 kinase activity, E2F-dependent transcription and cell growth by myc are genetically separable events EMBO J 19:5813–5823

Bellosta P, Hulf T, Balla Diop S, Usseglio F, Pradel J, Aragnol D, Gallant P (2005) Myc interacts genetically with Tip48/Reptin and Tip49/Pontin to control growth and pro-liferation during Drosophila development Proc Natl Acad Sci USA 102:11 799–11 804 Bettess MD, Dubois N, Murphy MJ, Dubey C, Roger C, Robine S, Trumpp A (2005) c-Myc

is required for the formation of intestinal crypts but dispensable for homeostasis of the adult intestinal epithelium Mol Cell Biol 25:7868–7878

Boon K, Caron HN, van Asperen R, Valentijn L, Hermus MC, van Sluis P, Roobeek I, Weis I, Voute PA, Schwab M, Versteeg R (2001) N-myc enhances the expression of

a large set of genes functioning in ribosome biogenesis and protein synthesis EMBO J 20:1383–1393

Bouchard C, Dittrich O, Kiermaier A, Dohmann K, Menkel A, Eilers M, Luscher B (2001) Regulation of cyclin D2 gene expression by the Myc/Max/Mad network: Myc-dependent TRRAP recruitment and histone acetylation at the cyclin D2 promoter Genes Dev 15:2042–2047

Bouchard C, Marquardt J, Bras A, Medema RH, Eilers M (2004) Myc-induced proliferation and transformation require Akt-mediated phosphorylation of FoxO proteins EMBO J 23:2830–2840

Bouchard C, Thieke K, Maier A, Saffrich R, Hanley-Hyde J, Ansorge W, Reed S, Sicinski P, Bartek J, Eilers M (1999) Direct induction of cyclin D2 by Myc contributes to cell cycle progression and sequestration of p27 EMBO J 18:5321–5333