The proto-oncogene c-myc and apoptosis - Nature

Oncogene (1998) 17, 3351 ± 3357 ã 1998 Stockton Press All rights reserved 0950 ± 9232/98 $12.00 http://www.stockton-press.co.uk/onc

The proto-oncogene c-myc and apoptosis Barbara Homan*,1,2 and Dan A Liebermann1,2 1

Fels Institute for Cancer Research and Molecular Biology, Temple University School of Medicine, 3307 North Broad Street, Philadelphia, Pennsylvania 19140, USA and 2Department of Biochemistry, Temple University School of Medicine, 3307 North Broad Street, Philadelphia, Pennsylvania 19140, USA

Deregulated expression of c-Myc not only promotes proliferation, but also can either induce or sensitize cells to apoptosis. Inappropriate expression of c-Myc under conditions which inhibit growth and down-regulate endogenous c-Myc expression, including serum deprivation and exposure to cytotoxic agents including the anticancer agents vinblastine, etoposide, Ara-C, and nocodazole, usually results in programmed cell death in many dierent cell types. Also, inappropriate Myc expression is associated with an apoptotic response elicited by induction of dierentiation. The proapoptotic property of c-Myc requires an intact N-terminal transactivation domain and bHLHZip domain, as well as interaction with Max, thereby implicating c-Myc target genes in this apoptotic process. Although some target genes, namely cdc25A and ODC, have been shown to participate in Myc-mediated apoptosis, no target gene has yet been identi®ed which is essential for this apoptotic response. It is possible that the response of cells inappropriately expressing c-Myc is due not only to the growth arrest signals per se, but also to signals elicited by speci®c growth inhibitors in the context of a particular biological setting. Also regulating the response of the cells is expression of other oncogenes and tumor suppressor genes, as well as paracrine and autocrine survival factors. Apoptosis associated with inappropriate Myc expression limits the tumorigenic eect of the cmyc proto-oncogene. Mechanisms which inhibit apoptosis should enhance or promote tumorigenesis. Keywords: c-myc; proliferation; apoptosis; tumorigenesis

Introduction The proto-oncogene c-myc has been shown to play a pivotal role in growth control, dierentiation and apoptosis, and its abnormal expression has been associated with many naturally occurring neoplasms (Askew et al., 1991; Evan and Littlewood, 1993; Evan et al., 1992; Shi et al., 1992). c-Myc is expressed in almost all proliferating normal cells where its expression is strictly dependent on mitogenic stimuli, and is downregulated in many kinds of cells when they are induced to terminally dierentiate (Evan and Littlewood, 1993; Ryan and Birnie, 1997). Forced expression of c-Myc was shown to inhibit differentiation and its associated growth arrest in several cell

*Correspondence: B Homan

types, including myeloid, erythroid, myogenic, preadipocyte, and nerve cells (Freytag, 1988; HomanLiebermann and Liebermann, 1991; Packham and Cleveland, 1995; Selvakumaran et al., 1993, 1996). The proto-oncogene c-myc plays a role in both positive and negative growth via having an in¯uence on proliferation, dierentiation and apoptosis. The eects of c-Myc are revealed both by appropriate and inappropriate expression. As expected for a gene product which participates in these dierent processes, alterations in c-Myc expression is associated with tumor growth. Therefore, it is important to understand how c-Myc carries out its multiple functions, both with regard to normal development, tumor formation and maintenance, and responsiveness to dierent cancer therapies. Myc is not the only oncogene to promote the antagonistic processes of cell proliferation and apoptosis. Other genes associated with cell cycle progression, including E1A, E2F and c-fos, are also implicated in the control or execution of apoptosis (Harrington et al., 1994b). It is important to understand the phenomenon of Myc-mediated apoptosis because of the implications for controlling cell numbers and modulating the eects of oncogenesis. With this in mind we will discuss the functional domains of the Myc protein, the role(s) of genes believed to be directly regulated by Myc, the eect of other genes on the apoptotic response which either suppress or are necessary for apoptosis, and a possible mechanistic pathway. We will conclude by assessing the role of the apoptotic function of Myc in tumorigenesis and the ability of Myc-associated tumors to respond to dierent therapies. The Myc network The gene product encoded by the c-myc oncogene is a transcription factor. The carboxy terminus of c-Myc contains three structural domains that are homologous to domains found in characterized transcription factors, including a leucine zipper (Zip), a helix ± loop ± helix motif (HLH), and an adjacent domain rich in basic amino acids (b) (Blackwell et al., 1990; Landshulz et al., 1988; Murre et al., 1989). The HLH and Zip motifs promote protein-protein interaction, and the basic region mediates sequence-speci®c DNA binding (Blackwell et al., 1990; Landshulz et al., 1988; Murre et al., 1989). These motifs are contiguous, thus c-Myc is a bHLHZip protein (Henriksson and Luscher, 1996). The transcriptional activation domain (TAD) has been mapped to the amino terminus. Both the bHLHZip and TAD domains are essential for

c-myc and apoptosis B Hoffman and DA Liebermann

3352

transcriptional transactivation, proliferation and transformation (Henriksson and Luscher, 1996). Consistent with the predominant localization of c-Myc in the nucleus, there are two nuclear localization signals in the carboxy terminus (Henriksson and Luscher, 1996). Recent experimental evidence suggests that c-Myc functions as a transcriptional regulator as part of a network of interacting factors (Ayer et al., 1995; Evan and Littlewood, 1993). Under physiological conditions Myc cannot form homodimers (Henriksson and Luscher, 1996). Max, a ubiquitous bHLHZip protein appears to be an obligate heterodimeric partner for cMyc in mediating its functions (Amati and Land, 1994; Ayer and Eisenman, 1993). Initially two partners for Max had been identi®ed, Mxi1 and Mad (Amati and Land, 1994). Analysis of mouse Mxi1 has led to the identi®cation of two transcript forms, Mxi-SR and Mxi-WR (Schreiber-Agus et al., 1995). Two additional Max-interacting proteins, Mad3 and Mad4, which behave similarly to Mad (Mad1), have been reported (Henriksson and Luscher, 1996; Hurlin et al., 1995). It has been proposed that transcription activation is mediated exclusively by Myc : Max complexes, whereas Max : Max, Max:Mad and Max : Mxi complexes mediate transcription repression through identical binding sites (Ayer and Eisenman, 1993; Ayer et al., 1993; Zervos et al., 1993). Max expression is not highly regulated and its protein is very stable; in contrast, Mad protein appears to have a short half life and to be highly regulated (Amati and Land, 1994). Mxi protein is also regulated; however, substantial amounts are already present in many undierentiated cell types (Amati and Land, 1994; Zervos et al., 1993). In addition, mSin3A and mSin3B, possessing sequence homology with the yeast transcription repressor Sin3, bind speci®cally to Mad and Mxi. Evidence is consistent with mSin3 forming a ternary complex with Mad-Max and Mxi1-Max that recognizes E box binding sites and participates in transcriptional repression (Ayer et al., 1995; Harper et al., 1996; Rao et al., 1996; Schreiber-Agus et al., 1995); this repression is a consequence of the association of Sin3 with multiprotein complexes including histone deacetylases which alter chromatin structure and block transcription (Alland et al., 1997; Hassig et al., 1997; Laherty et al., 1997). Very recently another Maxinteracting partner, MNT, has been identi®ed which is coexpressed with Myc yet mediates repression of expression (Hurlin et al., 1997). It has been established that Max is esssential for cMyc-mediated gene transactivation, transformation, cell cycle progression and apoptosis (Amati and Land, 1994; Packham and Cleveland, 1995). Overexpression of Max in transgenic mice (Eu-myc/Eumax) also overexpressing c-Myc reduced the rate of lymphoma onset compared to Eu-myc transgenic mice, suggesting that elevation of Max expression in vivo inhibits the function of c-Myc (Lindeman et al., 1995). Overexpression of Mxi1 (or Mxi-SR) and Mad can antagonize c-Myc activity in cellular transformation assays (Koshinen et al., 1995; Lahox et al., 1994) and proliferation (Roussel et al., 1996; Wu et al., 1996), and can diminish the malignant phenotype of tumor cells (Chen, 1995). In addition, overexpression of Mad1 reversed the dierentiation blocking eect of c-Myc in erythroleukemia cells (Cultraro et al., 1997). These

®ndings are consistent with the notion that Mad and Mxi1 suppress c-Myc function (Packham and Cleveland, 1995). Myc promotes apoptosis Deregulated expression of c-Myc not only promotes proliferation, but also can either induce or sensitize cells to apoptosis. This was ®rst demonstrated in interleukin-3 (IL-3) dependent 32Dcl3 myeloid progenitor cells. Following IL-3 withdrawal, c-Myc is down-regulated and cells become growth arrested, accumulating in the G0/G1 phase of the cell cycle and rapidly undergo apoptosis. Enforced c-Myc expression accelerates the apoptotic response in the absence of IL-3 (Askew et al., 1991). When 32Dcl3 cells are cultured to high density in the presence of IL3, the cells undergo growth arrest, maintain viability and c-Myc is down-regulated; however, when c-Myc expression is deregulated, the cells die by apoptosis. This indicates that IL-3 cannot protect the cells (Askew et al., 1993). Induction of apoptosis by deregulated cMyc was observed both in the Rat-1 ®broblast cell line and primary rat embryo ®broblasts (REFs) deprived of serum, as well as in Rat-1 ®broblasts growth arrested by either a thymidine block or isoleucine starvation (Evan et al., 1992). The extent of the apoptotic response is correlated with the level of Myc expression, and the level of Myc expression similar to the level in untransformed cells is sucient to induce apoptosis (Evan et al., 1992). Analagous observations were made in various mouse cell lines and mouse embryo ®broblasts (MEFs) (Packham and Cleveland, 1995). Inappropriate expression of c-Myc under conditions which inhibit growth and down-regulate endogenous c-Myc expression, including serum deprivation and exposure to cytotoxic agents including the anticancer agents vinblastine, etoposide, Ara-C, and nocodazole, usually results in programmed cell death in many dierent cell types. Recently it was demonstrated that activation of apoptotic cell death in quiescent renal epithelial cells by nephrotoxicants requires c-Myc induction (Zhan et al., 1997). In addition to myeloid cells also ®broblasts, hepatocytes, osteosarcomas, epithelial cells, and lymphoid cells can be subjected to Myc-mediated apoptotic responses (Packham and Cleveland, 1995; Thompson, 1998). However, deregulated Myc expression does not always cause or raise susceptibility to apoptosis when cells are unable to grow; in Chinese hamster ovary (CHO) cells overexpression of c-Myc protein promotes cell death under some but not all conditions that block cell division (Gibson et al., 1995). Although inappropriate expression of c-Myc under restricted growth conditions results in apoptosis, many instances of apoptosis do not require expression of cMyc. This includes apoptosis following withdrawal of IL-3 from 32Dcl3 cells, as well as observations made in this laboratory where apoptosis was observed in myeloid leukemic M1 cells following treatment with either TGFb or activation of the p53ts transgene, although endogenous c-Myc was down-regulated (Guillouf et al., 1995; Selvakumaran et al., 1994). It can be asked if Myc expression is relevant to physiological apoptosis, by either mediating the


response or being a requirement. One instance of the latter is the requirement for c-Myc in apoptosis which takes place in anti-TCR antibody stimulated T-cell hybridomas (Shi et al., 1992). Previous work from this laboratory has shown that deregulated expression of c-Myc in M1 myeloid leukemic cells blocked IL-6-induced dierentiation and its associated growth arrest; however the cells proliferated at a signi®cantly reduced rate compared to untreated cells (Homan-Liebermann and Liebermann, 1991). Recently it was shown that the increased doubling time for IL-6-treated M1myc cells can be accounted for by the induction of an apoptotic pathway. Therefore, both proliferation and apoptosis were continuously ongoing with steady state dynamics (Amanullah et al., submitted for publication). Thus, inappropriate Myc expression is also associated with an apoptotic response elicited by induction of dierentiation. This eect blunts the tumorigenic eect of the Myc-mediated dierentiation block. Myc target genes The proapoptotic property of c-Myc requires an intact N-terminal transactivation domain and bHLHZip domain, as well as interaction with Max (Amati et al., 1994; Evan et al., 1992; Grandori et al., 1996; Henriksson and Luscher, 1996). Therefore, it is expected that c-Myc target genes and downstream eectors of the target genes participate in c-Mycmediated apoptosis. Although there is evidence that some of the known target genes play a role in the Mycmediated apoptotic response, no target gene has yet been identi®ed which is essential. Included among the target genes induced by c-Myc are cdc25A, cyclin A, cyclin E, a-prothymosin, lactate dehydrogenase A, MrDb, e1F-2a and e1F-4E, ornithine decarboxylase (ODC) and rcl, in which the evidence is con¯icting for some of the genes being bona®de Myc target genes (Grandori and Eisenman, 1997; Packham and Cleveland, 1995). Cdc25 represents a family of CDK-activating phosphatases, in which cdc25A and cdc25B each cooperate with Ha-ras in the oncogenic transformation of primary rodent ®broblasts (Galaktionov et al., 1995). Cdc25A has been shown to be a c-Myc target gene via several independent criteria (Galaktionov et al., 1996). Fibroblast cell lines depleted of growth factor can be induced to undergo apoptosis by constitutive expression of cdc25A (Galaktionov et al., 1996). C-Myc-induced apoptosis requires cdc25A; however, blocking cdc25A expression does not obliterate the Myc-mediated apoptotic response (Galaktionov et al., 1996). These data suggest that either the block in cdc25A expression is not complete or other genes, presumably c-Myc target genes, can compensate for the absence of cdc25A. There is evidence that cdc25A acts on CDK2 complexes, and its activity is required for the G1/S transition; cyclinD-CDK4 complexes may also be cdc25 targets, although the physiological substrates for cdc25A have not been con®rmed (Galaktionov et al., 1996). Expression of c-Myc leads to activation of cyclinECDK2 and cyclinD-CDK4 complexes in quiescent cells (Hunter, 1997; Steiner et al., 1995). This activation

seems to be due to loss or inhibition of p27 rather than via direct activation of the CDKs (Grandori and Eisenman, 1997; Vlach et al., 1996). In addition, cyclin A and cyclin E are induced by c-Myc (Jansen-Durr et al., 1993), and cyclin D1 is repressed by c-Myc (Philipp et al., 1994). How regulation of the cell cycle machinery by c-Myc mediates an apoptotic response is still not clear. It has been demonstrated that overexpression of cyclin A is sucient to induce apoptosis; however, it has not been shown that cyclin A expression is necessary for Myc-mediated apoptosis (Hoang et al., 1994). a-prothymosin, another Myc target gene whose expression is correlated with c-Myc expression and may participate in chromatin remodelling (Diaz-Julien et al., 1996), has, thus far, not been demonstrated to play a role in any c-Myc functions including apoptosis (Evan and Littlewood, 1998 ). Expression of the Myc responsive gene lactate dehydrogenase A (LDH-A) is often increased in human cancers and, when ectopically expressed, induces apoptosis under conditions of glucose deprivation (Shim et al., 1998). Constitutive expression of the recently identi®ed Myc-responsive gene rcl in Rat 1A ®broblasts results in anchorageindependent growth, although to a lesser degree compared to ectopic expression of c-Myc (Lewis et al., 1997). One myc target gene MrDb, encodes for an evolutionarily conserved RNA helicase of the DEAD box family, which may participate in RNA processing and translational control (Grandori et al., 1996; Grandori and Eisenman, 1997). Two other putative Myc target genes are translation initiation factors e1F2a and e1F-4E, the mRNA cap binding protein (Rosenwald et al., 1993). It has been shown that overexpression of e1F-4E can inhibit Myc-mediated apoptosis in serum starved ®broblasts. This ®nding suggests that Myc not only activates an apoptotic pathway, but also provides negative feedback to the apoptotic pathway by regulating e1F-4E (Polunovsky et al., 1996). Which of these target genes participates in Myc-mediated apoptosis, and how the putative target gene e1F-4E inhibits apoptosis remain to be determined. Ornithine decarboxylase (ODC), a rate-limiting enzyme of polyamine biosynthesis, has been shown to be required for entry into and progression through the cell cycle and to be a transcriptional target of the proto-oncogene c-myc (Bello-Fernandez et al., 1993; Packham and Cleveland, 1995; Wagner et al., 1993). Like cdc25A, ODC can cooperate with c-H-ras in cell transformation (Hibshoosh et al., 1991). Several lines of investigation have shown that ODC activity is critical for cell transformation (Auvinen et al., 1992; Hibshoosh et al., 1991; Moshier et al., 1993). It also has been shown that ODC expression is sucient to induce accelerated apoptosis following Interleukin-3 (IL-3) withdrawal in IL-3 dependent 32Dcl3 myeloid cells, although the level of apoptosis is less than that induced by c-Myc (Packham and Cleveland, 1994). Using a-di¯uoromethylornithine (DFMO), a speci®c, irreversible inhibitor of ODC enzyme activity, it was possible to assess the eect of blocking ODC activity on the rates of apoptosis following IL-3 withdrawal of 32Dcl3 and 32Dcl3myc clones. It was found that DFMO inhibited the death of 32Dcl3myc clones;

3353


3354

however, the rate of apoptosis was still greater than in similarly treated parental cells. These observations demonstrate that ODC is a mediator of c-Mycinduced accelerated apoptosis of 32Dcl3 cells, but there is a pathway which does not require ODC for apoptosis (Packham and Cleveland, 1994). Although ODC is critical for cell transformation and mediates part of the eect of c-Myc in apoptosis of 32Dcl3 cells, it appears to play no role in the Myc-mediated block in myeloid dierentiation (Homan-Liebermann et al., 1996; Selvakumaran et al., 1996). In addition to Myc being an activator of transcription, expression of Myc is also correlated with downregulation of genes (Grandori and Eisenman, 1997; Packham and Cleveland, 1995; Thompson, 1998). Some of the genes are repressed via inhibition of initiator-dependent transcription, whereas repression of other genes may be indirect (Packham and Cleveland, 1995). The transcription factor C/EBP-a, which regulates dierentiation associated genes, is downregulated by c-Myc, probably through its initiator element (Li et al., 1994). This may account, at least partially, for the Myc-mediated block in terminal dierentiation observed for several dierent cell types. Repression of cyclin D1 by c-Myc does not require association with Max in vivo nor hormone activation of Mycer (Philipp et al., 1994), suggesting that Myc regulates gene expression in multiple ways. In addition to Myc repressing C/EBP, it has been shown to repress the growth arrest genes gadd45 and gadd153. C/EBP also has been shown to repress gadd45 gene expression. Whereas c-Myc down-regulates gadd45, p53 induces gadd45, these being noncompetitive co-regulatory events (Marhin et al., 1997). However, Myc also activates p53, although p53 is predominantly regulated post-transcriptionally (Packham and Cleveland, 1995). If this intricate regulatory loop participates in Myc mediated apoptosis remains to be determined. Modulation of apoptosis Cytokines, such as insulin-like growth factors and PDGF, the endothelium-derived vasoactive peptide endothelin-1 protect ®broblasts against Myc-induced apoptosis under conditions of low serum, and this protection from apoptosis is not linked to promotion of cell proliferation (Harrington et al., 1994a; Shichiri et al., 1998). The survival signaling pathway activated by IGF-1 in ®broblasts is directed by Ras activation of phosphatidylinositol-3-kinase [PI(3)K] through activation of the serine-threonine kinase PKB/Akt (Kauffman et al., 1997); Ras also can promote apoptosis through the Raf pathway (Kauman et al., 1997). Deregulated Myc expression drastically reduced clonogenicity of ®broblasts, where overexpression of Bcl-2 restored the plating eciency to normal levels, strongly suggesting that lowered clonogenicity was due to apoptosis. In addition, the plating eciency was also increased if cells were plated either on a feeder layer of lethally irradiated ®broblasts or in conditioned media from Rat-1 cells. These results indicate that a paracrine factor from Rat-1 ®broblasts can suppress c-Mycinduced apoptosis (Rupnow et al., 1998). Overexpression of CDK2 or CDK3 in Myc-overexpressing cells is accompanied by upregulaton of the

speci®c kinase activites. It was reported that whereas neither CDK2 nor CDK3 alone can induce apoptosis, there is a signi®cant enhancement of Myc-mediated apoptosis; in addition, expression of either CDK2 or CDK3 overrides cytokine-mediated protection from Myc induced apoptosis (Braun et al., 1998). Because of previous observations that Bcl-2 can promote cell survival and block apoptosis, the eect of Bcl-2 expression on Myc-mediated apoptosis was investigated. Bcl-2 protected glucocorticoid and cyclic AMP-treated T and B cells and factor dependent cells in the absence of survival factors from apopotosis (Alnemri et al., 1992; Vaux et al., 1988; Wagner et al., 1994). Bcl-2 was originally identi®ed due to its translocation to the immunoglobulin heavy chain enhancer in the t(14;18) translocation, associated with more than 80% of human follicular lymphomas (Cleary et al., 1986). Eu-Bcl2 transgenic mice develop follicular lymphomas which are at risk to develop into malignant lymphomas (McDonnell and Korsmeyer, 1991). Bcl-2 has been shown to inhibit Myc-mediated apoptosis in some but not all cell systems (Bissonnette et al., 1992; Fanidi et al., 1992; Green et al., 1994; Packham and Cleveland, 1995; Thompson, 1998; Wagner et al., 1994), while having no eect on the proliferative function of c-Myc (Fanidi et al., 1992; Packham and Cleveland, 1995). This interaction between c-Myc and Bcl-2 provides an explanation for the cooperation of Myc and Bcl-2 in cell transformation (Bissonette et al., 1992; Vaux et al., 1988). BHRF1, an Epstein-Barr virus protein and a member of the Bcl-2 family, inhibits c-Myc-induced apoptosis which occurs in the absence of survival factors, and like Bcl-2 has no eect on growth promotion (Fanidi et al., 1998). Similarly, Mcl-1, another Bcl-2 family gene, provides protection to Myc-mediated apoptosis (Reynolds et al., 1994). Bax, a pro-apoptotic member of the Bcl-2 family has been observed to be elevated when Myc is overexpressed (Sakamuro et al., 1995). Studies on the functional interaction between p53 and c-Myc have shown that c-Myc-mediated apoptosis is dependent on p53 in some cell types, but not in others (Hermeking and Eick, 1994; Hsu et al., 1995; Sakamuro et al., 1995; Wagner et al., 1994). Ectopic expression of c-Myc does not induce apoptosis in serum starved mouse MEFs from p53 null mice, whereas apoptosis occurs in similarly treated MEFs from wild type mice. Enforced expression of Myc in 10.1Val5 cells, which contain a temperature-sensitive p53 transgene, revealed that only when p53 is active do the cells undergo Myc-mediated apoptosis and this p53 activity did not require induction of p53 target proteins (Wagner et al., 1994). There is evidence that Myc activates p19(ARF) and p53 gene expression in primary MEFs to induce apoptosis, where p53 and ARF are tumor suppressors which act in a common biochemical pathway (Zindy et al., 1998). In lymphoid and myeloid cells Myc-driven apoptosis is p53independent, at least for the speci®c cells and treatments investigated (Hsu et al., 1995; Selvakumaran et al.,1994). Both p53-dependent and -independent mechanisms for c-Myc-induced apoptosis have been shown to exist in epithelial cells (Sakamuro et al., 1995). The implications of the relationship between p53 and c-Myc for tumor growth and therapy will be discussed below.


Taken together, it is apparent that apoptosis in response to inappropriate c-Myc expression is dependent on a particular biological setting. Also regulating the response of the cells is expression of other oncogenes and tumor suppressor genes, as well as paracrine and autocrine survival factors. Mechanism of action Ascertaining the speci®c apoptotic pathways mediated by c-Myc in dierent cell types and under dierent conditions is one approach towards understanding how the Myc-mediated apoptotic response is regulated. It can be asked if all Myc-mediated apoptotic responses follow the same pathway or distinct paths, which may or may not ultimately converge. It has been demonstrated that caspase 3 is involved in the apoptotic response of the Rat 1A mycer ®broblast cell line under conditions of low serum (Kangas et al., 1998). Fibroblasts, including Swiss 3T3, mouse primary embryo ®broblasts (MEFs) and Rat-1 cells have been shown to require the CD95 (FAS or APO-1) receptorligand pathway for c-Myc-induced apoptosis under conditions of low serum (Hueber et al., 1997). The apoptosis triggered by anti-CD95 antibodies indicated that an intact CD95 apoptotic signaling pathway is present in MEFs and Swiss 3T3 cells (Hueber et al., 1997). A direct role for the CD95-CD95 ligand apoptotic pathway in Myc-mediated apoptosis in serum starved ®broblasts was demonstrated by the ability of antibodies to CD95L to block the apoptotic response. Furthermore, expression of a dominant negative Fadd mutant, which can interact with CD95 but not caspase 8, eectively preventing progression of the apoptotic caspase cascade, also blocked the Mycmediated apoptotic response (Hueber et al., 1997). How inappropriate expression of Myc in ®broblasts activates the intact CD95-CD95L apoptotic pathway is still not known. The role of the CD95-CD95L pathway in Myc-mediated apoptosis in dierent cell types and under dierent conditions is an additional area open to investigation. Models for myc-mediated apoptosis Two models have been put forth to explain the ability of c-Myc to promote both proliferation and apoptosis. In the `clash' or `con¯ict' model the proliferative signals induced by inappropriate expression of c-Myc in the presence of growth arrest signals causes a con¯ict in growth signals which the cell senses, and apoptosis is induced. In the second model, the `dual signal' model (Grandori et al., 1996; Evan and Littlewood, 1998), Myc induces both proliferation and apoptosis signals, and the presence or absence of survival factors determines the outcome. There is data which supports and refutes each model. The fact that there is not a complete correlation between growth arrest and apoptosis in cells which express c-Myc inappropriately (Gibson et al., 1995), as well as the observation that survival factors such as the cytokines IGF-1 and PDGF protect ®broblasts against Myc-induced apoptosis without promoting cell proliferation (Harrington et al., 1994a;

Shichiri et al., 1998) argue against the con¯ict model, thereby supporting the dual signal model. However, there are instances in which even in the presence of survival factors cells continue to undergo programmed cell death (Harrington et al., 1994a), arguing against the dual signal model. It is possible that the response of cells inappropriately expressing c-Myc is due to signals elicited by speci®c growth inhibitors in the context of a particular biological setting. This predicts that for a given cell at a given stage of development, the same apoptotic pathway should be activated by deregulated c-Myc regardless of the trigger. Myc and tumorigenesis Alterations in the level of Myc expression or protein structure are associated with many malignancies in humans and animals (Askew et al., 1991; Evan and Littlewood, 1993; Evan et al., 1992). Myc genes have been shown to be activated by a variety of genetic alterations including chromosomal translocation, proviral insertion, retroviral transduction and gene ampli®cation, although most often the basis for altered Myc expression is not yet understood (Ryan and Birnie, 1997). Deregulated expression of c-Myc causes cellular immortalization, but is not sucient to induce transformation of primary cells; however, Myc and Ras have been shown to cooperate in both transformation and tumorigenesis in vivo (Henricksson and Luscher, 1996; Ryan and Birnie, 1997). Apoptosis associated with inappropriate Myc expression, as with other oncogenes such as E1A, limits the tumorigenic eect of the oncogene. Mechanisms which inhibit apoptosis should enhance or promote tumorigenesis. The observed oncogenic synergy between Myc and Bcl-2 demonstrates that this is indeed the case. Bcl-2 has been shown to inhibit Mycmediated apoptosis in some but not all cell systems, while having no eect on the proliferative function of c-Myc (Adams and Cory, 1998; Fanidi et al., 1992; Bissonnette et al., 1992). Bcl-2 cooperated with c-Myc to promote proliferation of B-cell precursors, and some of these cells became tumorigenic. This cooperation was exhibited in vivo as well, where the doubly transgenic Eu-bcl-2/myc mice developed tumors much faster than Eu-myc mice (Strasser et al., 1990). Tumors arising in Eu-bcl-2/myc and Eu-myc mice are clonal and appear only after a latency period, indicative of the requirement for further somatic mutations for tumor formation. The synergy for oncogenesis was also demonstrated for breast cancer in bi-transgenic mice (Jager et al., 1997). Other genes besides Bcl-2 cooperate with c-Myc to accelerate tumorigenesis, including pim1, bmi-1/bla-1, pal-1, raf-1, ras and abl (Evan and Littlewood, 1993; Packham and Cleveland, 1995), where some of these genes, namely, pim1, raf and abl, exhibit anti-apoptotic activity. As indicated earlier, c-Myc-mediated apoptosis is dependent on p53 in some cell types, but not in others (Hermeking and Eick, 1994; Hsu et al., 1995; Sakamuro et al., 1995; Wagner et al., 1994). Deregulated Myc expression results in more tumors when expressed in mice with a p53 null background (Evan and Littlewood, 1998; McCormack et al., 1998). Since many cells do not require p53 for the Myc-

3355


3356

mediated apoptotic response, and the absence of p53 results in increased mutation rates, it is possible that enhanced tumor formation in the p53 null background is due to the generation of additional somatic mutations. As noted previously the CD95/CD95L apoptotic pathway is activated in Myc-mediated apoptosis in serum starved ®broblasts. Any of the genes coding for proteins which carry out essential functions in this pathway would be expected to behave as tumor suppressor genes which cooperate with c-Myc in tumor formation (Hueber et al., 1997). The apoptotic capability of oncogene-expressing cells accounts for the sensitivity of many primary tumors to anticancer agents relative to their normal counterparts. This is the basis for the success of many cancer therapies. However, further mutations often eliminate the apoptotic response, rendering the treatment ineective. Since many Myc-expressing cells require p53 to undergo apoptosis, the p53 status of tumor cells will determine if the cell responds to a particular therapeutic treatment. Similarly, activation of genes encoding for proteins such as Bcl-2, which inhibit the apoptotic response would also render cells resistant to certain treatments.

Concluding remarks C-Myc in¯uences multiple biological activities, some in the context of appropriate expression and some detected by deregulated expression. These include cell cycle progression, transformation, bypass of cell cycle arrest, blocking terminal dierentiation and apoptosis. All tested biological activities of c-Myc have been shown to require the Myc transcription function to remain intact (Henriksson and Luscher, 1996). Therefore, Myc target genes are essential to mediate its dierent biological eects, yet the biological functions of c-Myc cannot be completely explained by the known Myc target genes. The biological setting appears to determine how a cell responds to altered Myc expression. It is possible that the apoptotic response of cells inappropriately expressing c-Myc is due not only to the growth arrest signals per se, but also to signals elicited by speci®c growth inhibitors in a particular context. Also regulating the response of the cells is expression of other oncogenes and tumor supppressor genes, as well as paracrine and autocrine survival factors.

References Adams JM and Cory S. (1998). Science, 281, 1322 ± 1326. Alland L, Muhle R, Hou Jr H, Potes J, Chin L, SchreiberAgur N and DePinho RA. (1997). Nature, 387, 49 ± 55. Alnemri ES, Fernandes TF, Haldar S, Croce CM and Litwack G. (1992). Cancer Res., 52, 491 ± 495. Amati B and Land H. (1994). Curr. Opin. Gen. Dev., 4, 102 ± 108. Amati B, Littlewood T, Evan G and Land H. (1994). EMBO J., 12, 5083 ± 5087. Askew DS and Ihle JN. (1993). Blood, 82, 2079 ± 2087. Askew DS, Ashmun RA, Simmons BC and Cleveland JL. (1991). Oncogene, 6, 1915 ± 1922. Askew DS, Ihle JN and Cleveland JL. (1993). Blood, 82, 2079 ± 2087. Auvinen M, Paasinen A, Andersson LC and Holtta E. (1992). Nature, 360, 355 ± 358. Ayer DE and Eisenman N. (1993). Genes Devel., 7, 2110 ± 2119. Ayer DE, Kretzner L and Eisenman RN. (1993). Cell, 72, 211 ± 222. Ayer DE, Lawrence QA and Eisenman RN. (1995). Cell, 80, 767 ± 776. Bello-Fernandez C, Packham G and Cleveland JL. (1993). Proc. Natl. Acad. Sci. USA, 90, 7804 ± 7808. Bissonnette RP, Echeverri F, Mahboubi A and Green DR. (1992). Nature, 359, 552 ± 554. Blackwell TK, Kretzner L, Backwood EM, Eisenman RN and Weintraub H. (1990). Science, 250, 1149 ± 1151. Braun K, Holzl G, Pusch O and Hengstschlager M. (1998). DNA Cell Biol., 17, 789 ± 798. Cleary M, Smith S and Sklar J. (1986). Cell, 47, 19 ± 28. Chen J, Willingham T, Margraf LR, Schreiber-Agus N, DePinho RA and Nisen PD. (1995). Nature Med., 1, 638 ± 643. Cultraro CM, Bino T and Segal S. (1997). Mol. Cell. Biol., 17, 2353 ± 2359. Diaz-Julian C, Perez-Estevez A, Covelo G and Freire M. (1996). Biochim. Biophs. Acta, 1296, 219 ± 227. Evan GI and Littlewood TD. (1993). Curr. Opin. Genet. Dev., 3, 44 ± 49. Evan G and Littlewood T. (1998). Science, 281, 1317 ± 1322.

Evan GI, Wyllie AH, Gilbert CS, Littlewood TD, Land H, Brooks M, Waters CM, Penn LZ and Hancock DC. (1992). Cell, 69, 119 ± 128. Fanidi A, Harrington EA and Evan GI. (1992). Nature, 359, 554 ± 556. Fanidi A, Hancock DC and Littlewood TD. (1998). J. Virol., 72, 8392 ± 8395. Freytag SO. (1988). Mol. Cell. Biol., 8, 1614 ± 1624. Galaktionov K, Lee AK, Eckstein J, Draetta G, Meckler J, Loda M and Beach D. (1995). Science, 269, 1575 ± 1577. Galaktionov K, Chen X and Beach D. (1996). Nature, 382, 511 ± 517. Gibson AW, Cheng T and Johnston RN. (1995). Exp. Cell. Res., 218, 351 ± 358. Grandori C and Eisenman RN. (1997). Trends Biochem. Sci., 22, 177 ± 181. Grandori C, Mac J, Siebelt F, Ayer DE and Eisenman RN. (1996). EMBO J., 15, 4344 ± 4357. Green DR, Mahboubi A, Nishioka W, Oja S, Echeverri F, Shi Y, Glynn J, Yang Y, Ashwell J and Bissonnette R. (1994). Immunol. Rev., 142, 321 ± 342. Guillouf C, Grana X, Selvakumaran M, Giordano A, Homan B and Liebermann DA. (1995). Blood, 85, 2691 ± 2698. Harper SE, Qiu Y and Sharp PA. (1996). Proc. Natl. Acad. Sci. USA, 93, 8536 ± 8540. Harrington EA, Bennett MR, Fanidi A. and Evan GI. (1994a). EMBO J., 14, 3286 ± 3295. Harrington EA, Fanidi A and Evan GI. (1994b). Curr. Opin. Genet. Dev., 4, 120 ± 129. Hassig CA, Fleischer TC, Billin AN, Schreiber SL and Ayer DE. (1997). Cell, 89, 341 ± 347. Henriksson M and Luscher G. (1996). Adv. Cancer Res., 68, 110 ± 182. Hermeking H and Eick D. (1994). Science, 265, 2091 ± 2093. Hibshoosh H, Jonson M and Weinstein BI. (1991). Oncogene, 6, 739 ± 743. Hoang AT, Cohen KJ, Barrett JF, Berstrom DA and Dang CV. (1994). Proc. Natl. Acad. Sci. USA, 91, 6875 ± 6879. Homan-Liebermann B and Lieberman D. (1991). Mol. Cell. Biol., 11, 2375 ± 2381.


Homan B, Liebermann DA, Selvakumaran M and Nguyen HQ. (1996). Curr. Topics Micro Immun., 211, 17 ± 27. Hsu B, Marin MC, El-Naggar AK, Stephens LC, Brisbay S and McDonnell TJ. (1995). Oncogene, 11, 175 ± 179. Hueber AO, Zornig M, Lyon D, Suda T, Nagata S and Evan GI. (1997). Science, 278, 1305 ± 1309. Hunter T. (1997). Cell, 88, 333 ± 346. Hurlin PJ, Queva C, Koshinen PJ, Steingrimsson E, Ayer DE, Copeland NG, Jenkins NA and Eisenman RN. (1995). EMBO J., 14, 5646 ± 5659. Hurlin PJ, Queva C and Eisenman RN. (1997). Genes Devel., 11, 44 ± 58. Jager R, Herzer U, Schenkel J and Weiher H. (1997). Oncogene, 15, 1787 ± 1795. Jansen-Durr P, Meichle A, Steiner P, Pagano M, Finke K, Botz J, Wessbecher J, Draetta G and Eilers M. (1993). Proc. Natl. Acad. Sci. USA, 90, 3685 ± 3689. Kangas A, Nicholson DW and Holtta E. (1998). Oncogene, 16, 387 ± 398. Kaumann-Zeh A, Rodriguez-Vicianna P, Ulrich E, Gilbert C, Coer P, Downward J and Evan G. (1997). Nature, 385, 544 ± 548. Koshinen PJ, Ayer DE and Eisenman RN. (1995). Cell Growth Di., 6, 623 ± 629. Laherty D, Yang W, Sun J, Davie JR, Seto E and Eisenman RN. (1997). Cell, 89, 349 ± 356. Lahox EG, Xu L, Schreiber-Agus N and DePinho RA. (1994). Proc. Natl. Acad. Sci. USA, 91, 5503 ± 5507. Landshulz WH, Johnson PF and McKnight SL. (1988). Science, 240, 1759 ± 1764. Lewis BC, Shim H, Li Q, Wu CS, Lee LA, Maity A and Dang CV. (1997). Mol. Cell. Biol., 17, 4967 ± 4978. Li LH, Nerlov C, Prendergast G, MacGregor D and Zi EB. (1994). EMBO J., 13, 4070 ± 4079. Lindeman GJ, Harris AW, Bath ML, Eisenman RN and Adam JM. (1995). Oncogene, 10, 1013 ± 1017. McDonnell TJ and Korsmeyer SJ. (1991). Nature, 349, 254 ± 256. McCormack SJ, Weaver Z, Deming S, Natarajan G, Torri J, Johnson MD, Liyanage M, Ried T and Dickson RB. (1998). Oncogene, 16, 2755 ± 2766. Marhin WW, Chen S, Facchini LM, Fornace AJ and Penn LZ. (1997). Oncogene, 14, 2825 ± 2834. Moshier JA, Dosecu J, Skunca M and Luk GD. (1993). Cancer Res., 53, 2618 ± 2622. Murre C, Schonleber-McCaw P and Baltimore D. (1989). Cell, 56, 777 ± 783. Packham G and Cleveland JL. (1994). Mol. Cell. Biol., 14, 5741 ± 5747. Packham G and Cleveland JL. (1995). Biochim. Biophysi. Acta, 1242, 11 ± 28. Philipp A, Schneide A, Vasrik I, Finke K, Xiong Y, Beach D, Alitalo K and Eilers M. (1994). Mol. Cell. Biol., 14, 4032 ± 4043. Polunovsky VA, Rosenwald IB, Tan AT, White J, Chiang L, Sonenberg N and Bitterman PB. (1996). Mol. Cell. Biol., 16, 6573 ± 6581.

Rao G, Alland L, Guida P, Schreiber-Agus N, Chen K, Chin L, Rochelle JM, Seldin MF, Skoultchi AI and DePinho RA. (1996). Oncogene, 12, 1165 ± 1172. Reynolds JE, Yang T, Qian L, Jenkinson JD, Zhou P, Eastman A and Craig RW. (1994). Cancer Res., 54, 6348 ± 6352. Rosenwald ID, Broads DB, Callahan LD, Isselbacker KJ and Schmidt EM. (1993). Proc. Natl. Acad. Sci., 90, 6175 ± 6178. Roussel MF, Ashmun RA, Sherr CJ, Eisenman RN and Ayer DE. (1996). Mol. Cell. Biol., 16, 2796 ± 2801. Rupnow BA, Murtha AD, Chen E and Knox SJ. (1998). Cancer Lett., 127, 211 ± 219. Ryan KM and Birnie GD. (1997). Oncogene, 14, 2835 ± 2843. Sakamuro D, Eviner V, Elliot KJ, Showe L, White E and Prendergast GC. (1995). Oncogene, 11, 2411 ± 2418. Schreiber-Agus N, Chin L, Chen K, Torres R, Rao G, Guida P, Skoultchi A and DePinho RA. (1995). Cell, 80, 777 ± 786. Selvakumaran M, Liebermann DA and Homan-Liebermann B. (1993). Blood, 81, 2257 ± 2262. Selvakumaran M, Lin HK, TjinThamSjin R, Reed JC, Liebermann DA and Homan B. (1994). Mol. Cell. Biol., 14, 2352 ± 2360. Selvakumaran M, Liebermann DA and Homan B. (1996). Blood, 88, 1248 ± 1255. Shi Y, Glynn JM, Guilbert LJ, Cotter TG, Bissonnette RP and Green DR. (1992). Science, 257, 212 ± 214. Shichiri M, Sedivy JM, Marumo F, and Hirata Y. (1998). Mol. Endocrinol., 12, 172 ± 180. Shim H, Chun YS, Lewis BC and Dang CV. (1998). Proc. Natl. Acad. Sci., 95, 1511 ± 1516. Steiner P, Philipp A, Lukas J, Godden-Kent D, Pagano M, Milttnacht S, Bartek J and Eilers M. (1995). EMBO J., 14, 4814 ± 4826. Strasser A, Harris AW, Bath ML and Cory S. (1990). Nature, 348, 331 ± 333. Thompson EB. (1998). Annu. Rev. Physiol., 60, 575 ± 600. Vaux DL, Cory S and Adams JM. (1988). Nature, 335, 440 ± 442. Vlach J, Hennecke S, Alevizopoulos K, Conti D and Amati B. (1996). EMBO J., 15, 6595 ± 6604. Wagner AJ, Meyers C, Laimins LA and Hay N. (1993). Cell Growth Di., 4, 879 ± 883. Wagner AJ, Kokontis JM and Hay N. (1994). Genes Dev., 8, 2817 ± 2830. Wu SJ, Pena A, Korcz A, Soprano DR and Soprano KJ. (1996). Oncogene, 12, 621 ± 629. Zervos AS, Gyuris J and Brent R. (1993). Cell, 72, 223 ± 232. Zhan Y, Cleveland JL and Stevens JL. (1997). Mol. Cell. Biol., 12, 6755 ± 6764. Zindy E, Eischen CM, Randle DH, Kamijo T, Cleveland JL, Sherr CJ and Roussel MF. (1998). Genes Dev., 12, 2424 ± 2433.

3357