Oncogene (2001) 20, 5595 ± 5610 ã 2001 Nature Publishing Group All rights reserved 0950 ± 9232/01 $15.00 www.nature.com/onc

Translocations involving c-myc and c-myc function

Linda M Boxer1 and Chi V Dang*,2

1Division of Hematology, Department of Medicine, Stanford University School of Medicine, Stanford, California CA 94305, USA; 2Division of Hematology, Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland MD 21205, USA

c-MYC is the prototype for oncogene activation by sites in the c-myc promoter (Arcinas and Boxer, 1994). chromosomal translocation. In contrast to the tightly NF-kB is an important regulatory factor for both the regulated expression of c-myc in normal cells, c-myc is murine (Kessler et al., 1992; La Rosa et al., 1994) and frequently deregulated in human cancers. Herein, aspects human (Ji et al., 1994) c-myc promoters in B cells. The of c-myc gene activation and the function of the c-Myc c-myc gene is a target of the APC/b-catenin/Tcf protein are reviewed. The c-myc gene produces an pathway in colon carcinoma cells (He et al., 1998), oncogenic transcription factor that a€ects diverse but the signi®cance of this regulatory pathway in cellular processes involved in cell growth, cell prolifera- lymphocytes is not clear. The role of APC and catenins tion, apoptosis and cellular metabolism. Complete in colon carcinogenesis will be discussed below. removal of c-myc results in slowed cell growth and In addition to transcription initiation, transcription proliferation, suggesting that while c-myc is not required elongation plays an important role in the regulation of for cell proliferation, it acts as an integrator and c-myc expression. The down-regulation of c-myc accelerator of cellular metabolism and proliferation. mRNA during di€erentiation of di€erent cell types Oncogene (2001) 20, 5595 ± 5610. occurs by a block to transcription elongation (Bentley and Groudine, 1986; Eick and Bornkamm, 1986; Keywords: c-myc; gene expression; lymphoma; apoptosis Nepveu and Marcu, 1986). Kinetic nuclear run-on experiments and mapping of single stranded DNA in vivo have revealed that the polymerase II complex is Regulation of c-myc expression paused at +30 relative to the major transcription initiation site (Krumm et al., 1992; Strobl and Eick, The c-myc gene is regulated by complex mechanisms at 1992). both the transcriptional and post-transcriptional levels (Bornkamm et al., 1988; Spencer and Groudine, 1991). Four transcriptional promoters have been identi®ed, Translocations of c-myc but RNA initiated at the P2 promoter usually contributes to 80 ± 90% of total c-myc steady-state The c-myc gene is translocated to one of the RNA in normal cells (Taub et al., 1984). The c-myc immunoglobulin loci in virtually all Burkitt's lympho- gene comprises three exons. Exon 1 contains two mas. The typical translocation of c-myc into the promoters and is noncoding. Exons 2 and 3 encode the immunoglobulin heavy chain locus is observed in Myc protein with translation initiation at nucleotide 16 about 80% of Burkitt's lymphomas. The variant of exon 2. translocations of c-myc into either the k or l light The regulation of c-myc expression has been the chain locus each occur at a frequency of about 10%. subject of many studies. Numerous transcription Burkitt's lymphomas are divided into endemic (Afri- factors have been reported to interact with regions of can-derived) and sporadic. The vast majority of the c-myc gene, but results of di€erent studies have endemic Burkitt's cases are associated with Epstein ± yielded con¯icting data (Marcu et al., 1992; Spencer Barr virus (EBV) infection, and while it is likely that and Groudine, 1991). This is due at least in part to the EBV plays a causal role, the mechanisms involved have fact that di€erent cell types and promoter constructs not been completely de®ned. As discussed below, the have been used. Several DNase I hypersensitive sites breakpoints are often di€erent in these two groups of have been identi®ed 5' and internal to the c-myc gene lymphomas. Burkitt's lymphoma accounts for approxi- (Bentley and Groudine, 1986; Dyson et al., 1985; mately 30% of all HIV-associated lymphomas. Trans- Dyson and Rabbitts, 1985). We have shown that locations of c-myc into one of the immunoglobulin transcription factors bind in vivo to a number of these gene loci are also observed in di€use large cell lymphoma. Some of these lymphomas also show a t(14;18) translocation of the bcl-2 gene into the immunoglobulin heavy chain gene and may represent *Correspondence: C Dang, Ross Research Building, Room 1025 720 Rutland Avenue, Baltimore, MD 21205, USA; lymphomas that have transformed from low-grade E-mail: [email protected] follicular lymphoma. MYC activation and function LM Boxer and CV Dang 5596 The c-myc gene is translocated into one of the T cell The breakpoint on the immunoglobulin gene usually receptor loci in some cases of T-cell acute leukemia. occurs at the VDJ region in endemic Burkitt's More recently, translocations of c-myc into one of the lymphoma while a breakpoint at the class switch immunoglobulin genes have been found in multiple region is most common in sporadic Burkitt's lympho- myeloma (Drach et al., 2000; Shou et al., 2000). These ma (Magrath, 1990; Shiramizu et al., 1991). In HIV abnormalities include complex translocations and associated Burkitt's lymphomas, the translocation insertions and are unlikely to be caused by errors in breakpoint also occurs at the class switch region. VDJ or class switch recombination or somatic Examples of some translocation breakpoints are shown hypermutation (Shou et al., 2000). While the transloca- in Figure 1. The intron enhancer in the immunoglo- tion of c-myc into the immunoglobulin or T cell bulin heavy chain gene is located on the same receptor locus is believed to be one of the primary chromosome as the c-myc gene in endemic Burkitt's events in malignant transformation in Burkitt's lymphoma, but it is located on the reciprocal lymphoma and in T-ALL, it is most likely a secondary translocation product in sporadic Burkitt's lymphoma event associated with progression in multiple myeloma. and thus is not available to deregulate c-myc A summary of the diseases frequently associated with expression. As discussed in the next section, there are c-myc translocations is shown in Table 1. several enhancers located 3' of the immunoglobulin heavy chain gene, which are always located on the same chromosome as the translocated c-myc gene. Translocation breakpoints in Burkitt's lymphoma

Translocation breakpoints in Burkitt's lymphomas fall Deregulation of c-myc expression in Burkitt's lymphoma into three classes de®ned by their location with respect to c-myc (Cory, 1986): within the c-myc gene (class I), Translocation of the c-myc gene to the heavy or light immediately 5' to the c-myc gene (class II), and distant chain immunoglobulin locus is seen in virtually every (class III). Class III breakpoints can be more than case of Burkitt's lymphoma and in many cases of 100 kb from the c-myc gene. Thus, in order to detect transformed lymphomas. Transcription of the translo- the translocations by physical linkage to the immu- cated c-myc is greater than that seen in resting B cells; noglobulin locus, pulsed ®eld rather than conventional however, in some cases the magnitude of the increase is electrophoresis must be used. The class I breakpoints cluster within exon 1 and the ®rst intron. These translocations may disrupt regula- tory signals for c-myc. Many of the class II breakpoints are within a few kilobases 5' to exon I. Several protein binding sites have been detected in the 5' ¯anking region by in vitro assays and have been proposed to be positive or negative regulators of transcription. In endemic Burkitt's lymphoma, the translocation breakpoint usually occurs 5' of the c-myc gene, and the distance can be quite large. In these cases, the c-myc gene is intact. In sporadic Burkitt's lymphoma, the breakpoint is often in the ®rst exon or the ®rst intron (Magrath, 1990; Shiramizu et al., 1991). The c-myc coding region consisting of the second and third exons is always intact, however. It is likely that the pathogenesis of endemic and sporadic Burkitt's lymphoma di€er, and the state of di€erentiation of the target cell for transformation is most likely di€erent in the two types of lymphoma.

Table 1 c-Myc translocations and associated diseases Translocation Genes Disease Figure 1 Translocation breakpoints in Burkitt's lymphoma. (a) An example of the typical t(8;14) translocation of c-myc into the t(8;14)(q24;q32) c-myc/IgH Burkitt's lymphoma immunoglobulin heavy chain locus is shown with the chromo- t(2;8)(p12;q24) Igk/c-myc Burkitt's lymphoma some 8 breakpoint 5' of the c-myc gene and the chromosome 14 t(8;22)(q24;q11) c-myc/Igl Burkitt's lymphoma breakpoint in the VDJ region. (b) One of the variant t(8;14)(q24;q32) c-myc/IgH Diffuse large cell lymphoma translocations with the k light chain gene on chromosome 2 is t(8;14)(q24;q11) c-myc/TCRa,b T-cell acute lymphoblastic leukemia shown. The breakpoint on chromosome 8 is located 3' of the c- t(8;14)(q24;q32) c-myc/IgH Multiple myeloma myc gene. (c) The other variant translocation with the l light t(2;8)(p12;q24) Igk/c-myc Multiple myeloma chain gene on chromosome 22 is shown. The chromosome 8 t(8;22)(q24;q11) c-myc/Igl Multiple myeloma breakpoint is 3' of the c-myc gene

Oncogene MYC activation and function LM Boxer and CV Dang 5597 no more than that observed in actively dividing but elements. Although transgenic mice with the c-myc non-malignant B cells, such as those infected with the gene linked to the immunoglobulin heavy chain intron EBV. The normal c-myc allele is usually transcription- enhancer develop B cell malignancies, these studies ally silent in Burkitt's lymphomas (ar-Rushdi et al., shed no light on the biological role of the intron 1983; Cory, 1986; Hayday et al., 1984; Nishikura et al., enhancer in the t(8;14) translocation. They do demon- 1983), and thus the only Myc protein in most Burkitt's strate that high level expression of c-myc in B cells by a cells is derived from the translocated c-myc allele. In regulatory element, which is active in B cells, can lead Burkitt's cell lines, transcription of the translocated c- to development of a B cell malignancy. In fact, the myc is preferentially initiated further upstream at malignancy that develops in these mice does not promoter P1 instead of at P2 (Strobl and Eick, 1992; resemble Burkitt's lymphoma, particularly because the Strobl et al., 1993; Taub et al., 1984). The cause of this malignant cells in these mice are at the pre-B cell stage promoter shift is not known, but the immunoglobulin of di€erentiation, as opposed to the mature B cells in heavy chain gene enhancers are required for induction Burkitt's lymphoma. These mice served to demonstrate of the promoter shift. All of these changes have been that c-myc was an oncogene in B cells, but they do not interpreted as evidence for the activation or deregula- provide a useful model for Burkitt's lymphoma. Mice tion of the translocated c-myc gene (Bernard et al., carrying a yeast arti®cial chromosome (YAC) contain- 1983; Erikson et al., 1983; Leder et al., 1983; Taub et ing the c-myc gene linked to a portion of the al., 1984). immunoglobulin heavy chain gene locus also developed It does not appear that the interruption of c-myc B cell malignancies (Butzler et al., 1997). However, regulatory sequences by class I and II breakpoints is only the intron enhancer was present, and the 3' sucient for malignant transformation. Class III enhancers were not included in this model. In fact, the breakpoints are distant from the c-myc gene and do intron enhancer was deleted, and the mice still not a€ect the 5' ¯anking sequence. It is likely that developed B cell malignancies (Palomo et al., 1999). regulatory elements from the immunoglobulin locus are It is likely that the YAC has integrated into a responsible for the deregulation of c-myc expression. It transcriptionally active region of the chromosome, is possible that the immunoglobulin elements require and additional regulatory elements are not required certain c-myc regulators for the interaction to occur. for expression of c-myc. Depending on the class of translocation, di€erent c- In a number of Burkitt's lymphomas, the immuno- myc elements may be used. For instance, in the Ramos globulin intron enhancer is not available to the c-myc cell line, the 5' ¯anking sequence is located on a gene because it is located on the other chromosome in di€erent chromosome from the c-myc gene, and we the translocation. The best candidate for deregulation have observed in vivo occupancy of an NF-kB site in of the c-myc gene is the 3' enhancer, which has been the c-myc ®rst exon on the translocated allele. In the found in the kappa and heavy chain immunoglobulin Raji cell line this region of the ®rst exon is deleted loci (Judde and Max, 1992; Muller et al., 1990). Matrix from the translocated c-myc gene, and we found in vivo association regions (MARs) are located near the occupancy of an NF-kB site in the 5' ¯anking region of immunoglobulin intron enhancer in the murine and c-myc (Ji et al., 1994). We have also found an in vivo human kappa loci (Whitehurst et al., 1992) and in the footprint over a potential Nm23 binding site in the c- murine heavy chain locus (Cockerill et al., 1987). myc promoter (Ji et al., 1995). Two sites in the P1 MARs are frequently located near enhancers and some promoter region that are required for the k immu- appear to have enhancer activity. They often contain noglobulin gene activation of the P1 promoter have topoisomerase II sites and may therefore be involved in been identi®ed. In transfection studies the TATA box both transcriptional enhancement and swiveling at the and an Sp1 site are both essential and sucient for this base of chromosome loops. activation (Geltinger et al., 1996). Several enhancers have been shown to be important The deregulation of c-myc gene expression is thought for expression of the immunoglobulin heavy chain to play a causal role in malignant transformation. gene. Four B-cell-speci®c and cell-stage-dependent Transgenic mice, which carried c-myc linked to the DNase I hypersensitive sites, HS1 to HS4, are located immunoglobulin heavy chain gene intron enhancer, 10 ± 35 kb 3' of the Ca gene (Madisen and Groudine, developed clonal B cell malignancies (Adams et al., 1994; Michaelson et al., 1995; Pettersson et al., 1990; 1985; Nussenzweig et al., 1988). Furthermore, when Saleque et al., 1997). These regions have been shown to lymphoblastoid cells (immortalized by EBV) were function as a locus control region in B cells (Madisen transfected with a constitutively expressed c-myc gene, and Groudine, 1994). Enhancers have been located the cells became tumorigenic in nude mice (Lombardi downstream of two human Ca genes, and these regions et al., 1987). share some homology with the murine HS1-4, but only limited functional studies have been performed on the human enhancers. The 3' region of the immunoglobu- Mechanisms of deregulation of c-myc in lin heavy chain locus is linked to the translocated c- Burkitt's lymphoma myc gene in all t(8;14) translocations in Burkitt's lymphoma, and it is likely that this region plays a role It is likely that the translocated c-myc gene falls under in the deregulated expression of the translocated c-myc the in¯uence of immunoglobulin gene regulatory gene.

Oncogene MYC activation and function LM Boxer and CV Dang 5598 In addition to a shift in promoter usage from P2 to These ®ndings support a model in which sequences P1, the block to transcription elongation is absent on present in the immunoglobulin gene locus deregulate the translocated c-myc gene. Nuclear run-on studies expression from the cis-linked c-myc allele by promot- have shown that the initiation rate at the c-myc P1 ing interactions between c-myc and immunoglobulin promoter is not increased but that the number of gene regulatory elements that a€ect c-myc initiation paused polymerase complexes near the P2 promoter is and elongation. As proposed in Figure 2, interactions markedly decreased (Strobl et al., 1993). Constructs between multiprotein complexes assembled on both the consisting of the c-myc gene linked to enhancer regions c-myc promoter and immunoglobulin enhancers may from the di€erent immunoglobulin genes have been occur with looping out of the intervening DNA used to study the deregulation of c-myc transcription. sequence. It should be noted, however, that the Linkage of c-myc to the MAR, intron enhancer, translocation breakpoint in many sporadic Burkitt's constant k gene and the 3' enhancer resulted in lymphomas separates the c-myc promoter from the increased c-myc expression and a shift in promoter coding region (Pelicci et al., 1986; Shiramizu et al., usage from P2 to P1 (Polack et al., 1993). Linkage to 1991). In these cases, the regulatory elements of the either the intron or the 3' enhancer alone resulted in immunoglobulin gene enhancers apparently activate c- less activation of c-myc, and there was no promoter myc transcription without interaction with the c-myc shift. All t(2;8) translocations between c-myc and the k promoter elements. Transcription often initiates in the light chain gene result in colocalization of the intron ®rst intron of c-myc in these sporadic Burkitt's and 3' enhancers with c-myc on the translocated lymphomas. chromosome. As discussed above, the intron enhancer on the immunoglobulin heavy chain locus is usually located on a di€erent chromosome from the translo- Sequence mutations in the translocated c-myc gene cated c-myc gene. Linkage of c-myc to the 3' enhancers (HS1234) from the immunoglobulin heavy chain region Most translocated c-myc alleles from Burkitt's lym- resulted in high level expression of c-myc, the promoter phoma cell lines contain sequence alterations, often shift, and loss of the block to transcription elongation point mutations or deletions (Cesarman et al., 1987; at the P2 promoter (Madisen and Groudine, 1994). Rabbits et al., 1984; Siebenlist et al., 1984; Taub et al., The intron enhancer was not required for these e€ects. 1984; Yu et al., 1993). The translocated allele is located Similar results have been obtained with enhancer in the hypermutable immunoglobulin locus and thus regions from the l light chain gene (Gerbitz et al., may be subject to somatic mutations at an increased 1999). More recently, it has been shown that the 3' frequency (Bemark and Neuberger, 2000). Many enhancers from the immunoglobulin heavy chain locus mutations are located near the two promoters in exon increase expression from the c-myc P2 promoter by an I or near the exon I ± intron I boundary. These regions increase in histone acetylation. However, this increase are important regulatory sites, which demonstrate in acetylation does not explain the HS1 ± 4 activation protein binding in vivo. Mutations in the exon I/intron of transcription from the P1 promoter (Madisen et al., 1998). Investigations into the speci®c sites in the immu- noglobulin gene enhancers that are required for the deregulation of c-myc expression have been per- formed. Our studies showed that an NF-kB site in HS4 from the immunoglobulin heavy chain locus was required for the enhancer activity of this region when it was linked to the c-myc promoter. The NF-kB site was also required for the shift in transcription initiation from P2 to P1. The promoter shift due to HS4 was lost when the NF-kB site was mutated, although it is clear that regions of HS1, 2 and 3 also contributed to the full promoter shift (Kanda et al., 2000). Most of these studies have been performed with the murine immunoglobulin heavy chain gene. We identi®ed an NF-kB site in the human HS4 enhancer region and showed that it was active with the c-myc Figure 2 A model for deregulation of c-myc expression in the promoter (Kanda et al., 2000). An NF-kB site in the k t(8;14) translocation in Burkitt's lymphoma. Interactions occur between regulatory sites on the three immunoglobulin heavy chain light chain gene intron enhancer was shown to be gene enhancers (HS3, HS12, and HS4) and several transcription important for the deregulation of c-myc expression in factors that bind to the c-myc promoter, including the TATA the t(2;8) translocation. In addition, a binding site for sequences. By mechanisms that are not completely understood, PU.1/Spi-1 in the k 3' enhancer was also required. the P1 promoter is activated, and the block to transcription Mutation of both the NF-kB and PU sites resulted in elongation at the P2 promoter is relieved. Note that the human immunoglobulin locus is shown and only one of the two sets of the loss of activation of the c-myc P1 promoter enhancers is shown for simplicity. The second set is located 3' of (Wittekindt et al., 2000). the Ca2 gene

Oncogene MYC activation and function LM Boxer and CV Dang 5599 I boundary region of the translocated c-myc gene have cooperates with c-Myc in transformation, inhibits the been found in di€use large cell lymphomas with either phosphorylation of this region and causes c-Myc the t(8;14) or t(8;22) translocations (Bradley et al., protein stabilization (Sears et al., 1999). In several 1993). Although mutations are found in potential c- studies, mutations at or around Thr-58 appear to myc regulatory regions in many cases of Burkitt's augment c-Myc mediated transformation. c-Myc lymphoma, there is no evidence that they are the mediated apoptosis does not generally appear to be critical events in the activation of c-myc or even that a€ected or may be diminished (Chang et al., 2000; the same site is mutated in the majority of lymphomas. Conzen et al., 2000; Hoang et al., 1995). While the role of O-linked glycosylation of c-Myc remains to be established, it appears that hotspot mutations at or Hotspot mutations and activation of c-myc in around Thr-58 result in protein stabilization that may human neoplasias further augment c-Myc levels and presumably Myc- mediated transcriptional regulation. In addition to chromosomal translocations that In addition to activation of the c-myc gene in activate c-myc in lymphoid neoplasms and regulatory lymphoid tumors through chromosomal translocation sequence mutatations, the coding sequence also bears and point mutations, c-myc is frequently deregulated in hotspot non-conserved mutations (Figure 3) (Bhatia et a variety of human tumors (Nesbit et al., 1999). C-myc al., 1993; Yano et al., 1993). The c-Myc hotspot gene amplication accounts for a portion of human mutations are most likely the result of somatic breast, lung, ovarian and prostate carcinoma as well as hypermutation, a hallmark of B-cell maturation in medulloblastoma. C-myc expression is more frequently the germinal center. The consequences of the c-Myc deregulated, however, in a large fraction of colon, mutations are now better appreciated. There is gynecological tumors and melanoma. evidence supporting the inhibition of c-Myc function The mechanism underlying deregulated c-myc ex- by the pocket protein p107 (Beijersbergen et al., 1994; pression in human cancers has not been well- Gu et al., 1994; Hoang et al., 1995). Two studies characterized; however, the Wnt-Adenomatous Poly- suggest that mutations in the N-terminal region of c- posis Coli (APC) pathway has been implicated in colon Myc result in the inability of p107 to inhibit c-Myc carcinogenesis. In particular, one study suggests that function, although this concept remains controversial. loss of APC, which mediates the proteosomal degrada- One of the c-Myc Box I hotspot is Thr-58, which tion of b-catenin, could result in an increase in b- may be phosphorylated or O-linked glycosylated (Chou catenin that can in turn activate the c-myc gene in et al., 1995; Lutterbach and Hann, 1994). Phosphor- cooperation with the transcription factor TCF4/LEF ylation of Thr-58 appears to be required for proteo- (He et al., 1998). While this pathway appears attractive some-mediated degradation of c-Myc, such that as an explanation for augmented c-myc expression in mutation of Thr-58 to Ala results in protein stabiliza- colon cancer, a more recent study indicates that tion (Bahram et al., 2000; Flinn et al., 1998; Gregory elevated c-Myc expression is unnecessary for b-catenin and Hann, 2000). Furthermore, activated RAS, which mediated cell transformation (Kolligs et al., 1999). In addition, g-catenin (plakoglobin) appears to be more consistently connected with increased c-myc expression and cellular transformation (Kolligs et al., 2000). Despite the identi®cation of c-myc as a downstream target of TCF4/LEF, b-catenin is not as robust as plakoglobin in activating c-myc gene expression. Hence, both plakoglobin and b-catenin may be down- stream of the APC pathway and are both required for deregulated c-myc expression in colon carcinogenesis.

Functional domains of c-myc and c-myc transcriptional properties

The c-Myc protein belongs to the larger family of helix ± loop ± helix leucine zipper (HLHzip) transcrip- tion factors and is also a member of the Myc family of proteins (B-Myc, L-Myc and N-Myc) that are Figure 3 c-Myc hotspot mutations in Burkitt's lymphoma. The diagram depicts the t(8;14) translocation, as an example, and the characterized by two conserved N-terminal regions, associated c-Myc polypeptide shown as a bar with numbers termed Myc Box I and Box II (Grandori et al., 2000; indicating amino acid residues. The ®lled diamonds represent Henriksson and Luscher, 1996). The c-myc gene instances of mutations reported in the literature. Note that encodes several polypeptides. A 62 kDa protein is residues 39 and 58 are frequently targeted. The Myc partner partner protein, Max, is also depicted. TAD, transactivation produced from a canonical AUG translational start domain; NTS, nuclear targeting signal; bHLHZip, basic helix ± site and a 64 kDa protein from an upstream CUG loop ± helix leucine zipper (Hann et al., 1992). Internal translational initiation

Oncogene MYC activation and function LM Boxer and CV Dang 5600 results in a shortened 45 kDa form, termed MycS the canonical 5'CACGTG-3' central dinucleotide CG (Spotts et al., 1997). All three polypeptides hetero- (Dang et al., 1992). A conserved Arg residue that dimerize with Max, an HLHzip protein, to bind target directly contacts the central G is the signature of a DNA sequences called E-boxes (Prendergast and Zi€, subclass of Myc-like HLHzip proteins (Ferre-D'Amare 1992). Together with Max, Myc regulates the tran- et al., 1993). The HLHzip region represents a scription of target genes and induces cell growth, dimerization motif that allows for the formation of proliferation, or apoptosis and to inhibit cellular crisscrossing coiled-coil structures between c-Myc and di€erentiation. its partner Max in a fashion that aligns the basic The N-terminal 143 amino acids and the C-terminal regions into the major groove of target DNA 140 amino acids of c-Myc are required for neoplastic sequences. The loop region has been implicated in the transformation and inhibition of cellular di€erentiation contact of the protein to sequences ¯anking the core (Dang, 1999). These two regions correspond to the N- 5'CACGTG-3' hexamer. Crystal structures of the terminal transactivation domain and the C-terminal HLHzip transcriptional factor USF suggests that DNA-binding and HLHzip dimerization domain. The certain HLHzip proteins could form higher order c-Myc transactivation domain was identi®ed through tetrameric complexes through the formation of a studies with GAL4 fusion proteins (Kato et al., 1990). four-helical bundle via the leucine zippers (Ferre- The ®rst 143 amino acids, which comprises of Myc Box D'Amare et al., 1994). It is possible that such I (residues 45 ± 63) and Box II (residues 128 ± 143), are complexes could account for the cooperative activity sucient for gene activation. Deletions of Box II do of c-Myc/Max on promoters that contain tandem E- not a€ect transcriptional activation in transient boxes, although direct evidence for the role of reporter assays; however, Myc-mediated suppression tetramers in transcriptional regulation is lacking. of initiator-driven transcription is abrogated (Lee et al., While most studies explore c-Myc transcriptional 1996; Li et al., 1994). Hence, it has been proposed that regulation in isolation, further understanding on how Box II participates in transcriptional repression that c-Myc cooperates with other transcriptional factors will appears to be required for neoplastic transformation. be required to appreciate c-Myc's full activity in vivo. The function of Box II remains unclear, in particular For example, the AP2 transcriptional factor negatively since a recent study demonstrates that mutation of Box a€ects c-Myc (Gaubatz et al., 1995), whereas one II resulted not only in loss of transcriptional repression report suggests that Sp1 cooperates with c-Myc (Kyo but also in the inability to increase the expression of et al., 2000). C-Myc has also been implicated in endogenous ornithine decarboxylase (ODC), a Myc altering C/EBP function through protein ± protein target gene (Conzen et al., 2000). It is very likely that interactions and not through directly contacting DNA transient reporter assays do not fully re¯ect the binding sites ¯anking those of C/EBP (Mink et al., transcriptional activities of c-Myc on genomic targets. 1996). Hence, c-Myc function is likely to be signi®- Nevertheless, it stands to reason that Box II may cantly a€ected by nearby regulatory sequences as well mediate transcriptional repression and activates a as by protein ± protein contacts that may be cell and speci®c subset of genes such as ODC. tissue type speci®c. In addition, c-Myc transactivates The binding of the cofactor TRRAP, which other transcriptional regulators such as Coup, DP1, associates with histone acetyl transferase activity, to Fra1, HMG(Y)I, and Id2 (Coller et al., 2000; Lasorella Myc Box II region further suggests a role for Box II in et al., 2000; O'Hagan et al., 2000b; Schuhmacher et al., transcriptional activation (McMahon et al., 1998, 2001; Wood et al., 2000). In this manner, c-Myc itself 2000). However, it is notable that mutations in Box may hypothetically regulate promoters that also II only result in the loss of activation of some but not requires transcriptional factors produced from Myc all Myc target genes. As such, it is tempting to targets, such as HMG(Y). In this case, an indirect Myc speculate that c-Myc might regulate di€erent subsets target gene could well be a direct target gene whose of genes through di€erent Myc regions and by di€erent promoter requires both Myc and another transcription mechanisms. Perhaps Box II region is required for the factor that is induced by Myc. The precedence for this activation of gene expression through histone acetyla- has been reported for the expression of the myeloper- tion, whereas Box II-independent transcriptional oxidase gene that requires the transcription factors C/ activation primarily depends on the recruitment of EBP and PU.1, whose expression is induced by C/EBP the transcriptional complex to genomic regions where (Wang et al., 1999). With the emergence of DNA histones are already hyper-acetylated. The use of microarray technology to study c-Myc function, these speci®c c-Myc mutants in combination with arrayed lessons should be kept in mind if we aim to understand gene expression analysis will hopefully provide insight the full spectrum of Myc function in a variety of into the mystery of Myc Box II. di€erent tissues. The C-terminal region of c-Myc, which is required for transformation, contains a unique basic region that represents a distinct class of nuclear targeting Max and Max-associated proteins sequences (Dang and Lee, 1988; Saphire et al., 1998). A second basic region, which weakly localizes proteins While truncated c-Myc, removed of its transcriptional to the nucleus, turns out to be the DNA binding region regulatory domain, is capable of weakly homodimer- of c-Myc that has a characteristic signature for binding izing in vitro and binding DNA, the full-length c-Myc

Oncogene MYC activation and function LM Boxer and CV Dang 5601 polypeptide exists in a conformation that lacks DNA Cellular functions and target genes of Myc binding activity (Kato et al., 1992). DNA binding by c-Myc requires heterodimerization with Max, con- Despite the complexity of Max network of proteins, a verting c-Myc from an inactive form to a hetero- common theme of tumorigenesis is deregulated c-Myc dimer that is capable of binding E-boxes (Blackwood activity. Thus, to understand c-Myc physiologic and and Eisenman, 1991). Max, in contrast to c-Myc tumorigenic activity, we need to understand the gene however, does not contain transregulatory regions, expression programs that are induced by physiologi- suggesting that its role is to switch on c-Myc's DNA cally regulated c-Myc activity versus genes that are binding capability. With this ability, it stands to regulated by ectopic c-Myc expression. To understand reason that Max might be at the center of a complex these nuances, experimentally tractable models are network of protein ± protein interactions where Max required. Unfortunately, the study of c-Myc target could switch on other HLHzip proteins. Indeed, Max genes has not been standardized as to the cell types is able to homodimerize as well as heterodimerize used or the expression system employed (Cole and with Mad family proteins, Mga and Mnt (Grandori McMahon, 1999; Dang, 1999). Perhaps this is re¯ected et al., 2000). Hence, the Max network of proteins by the fact that the vast majority of c-Myc responsive further adds to the complexity of gene regulation by genes do not overlap between di€erent studies. c-Myc. Whether these di€erences are due to di€erent technical In contrast to c-Myc, the Mad proteins contain the approaches or whether they re¯ect cell type di€erences SID (Sin3-interacting domain) motif that is capable of remains to be established. Despite these shortcomings, recruiting Sin3A transrepression complexes and his- a role for c-Myc in cellular functions has emerged tone deacetylase activity (Ayer et al., 1993, 1996). The (Figure 4). For the following discussion, speci®c Myc Mad/Max complexes, which bind E-boxes, could target genes veri®ed by independent studies from therefore antagonize Myc function by competing for several laboratories are used as examples. target sites and render the associated genes silent through histone deacetylation. While it is attractive to hypothesize that Myc and Mad represents the yin and Cell growth and cellular metabolism yang of the Max network, in which Myc activates the same genes that Mad silences, recent studies suggest C-Myc has been implicated in the transition of cells that Mad/Max complexes may regulate a subset of from the resting phase of Go (those deprived of growth genes that are not shared with c-Myc/Max (O'Hagan factors in the experimental setting) into G1 and the et al., 2000b). Furthermore, Mad family members traversal of cells through G1, in which cells accumulate appear to display di€erentiation speci®c expression mass in preparation for DNA synthesis (Roussel, such that only certain Mad proteins are expressed at 1998). Studies of hypofunctioning Myc in the Droso- speci®c stages of cellular di€erentiation (Queva et al., phila provide a slightly di€erent understanding of c- 1998). Myc in cell growth and proliferation (Gallant et al., The fact that Mad could antagonize Myc transcrip- 1996; Schreiber-Agus et al., 1997). Hypo-functioning of tional activity suggests that it may inhibit Myc Drosophila Myc (dmyc) results in small body size, a transforming activity (Ayer et al., 1993; Roussel et phenotype collectively termed Minutes. Intriguingly, al., 1996). Indeed, in several model systems, Mad most molecularly characterized Minutes are defective inhibits Myc-mediated neoplastic transformation (Kos- in speci®c ribosomal protein genes. In addition, it has kinen et al., 1995; Lahoz et al., 1994; Schreiber-Agus et been demonstrated that dmyc hypomorphs and al., 1995). This further suggests the possibility that Minutes, in general, are small not because there are certain MAD genes might behave as tumor suppressor fewer cells, but that the cells are smaller in size genes, the loss of which would allow Myc to function (Johnston et al., 1999). This observation suggests a role in an un-antagonized manner. Among the di€erent for c-Myc in cell mass accumulation and metabolism. Mad family members, only Mxi-1 (Mad2) to date has In mammalian systems, overexpression of c-Myc been implicated as a potential tumor suppressor. results in larger lymphocytes in vivo as well as in vitro Homozygous deletion of Mxi-1 predisposes the (Iritani and Eisenman, 1999; Schuhmacher et al., 1999). knock-out mice to skin tumors and lymphoma In particular, induction of c-Myc expression in a (Schreiber-Agus et al., 1998). In addition, the human human lymphocyte cell line under growth factor chromosomal region 10q24 bearing the Mxi-1 gene deprivation resulted in an increase in cell mass without frequently display loss of heterozygosity in di€erent cell proliferation (Schuhmacher et al., 1999). Likewise, human cancers including prostate (Prochownik et al., Myc-induced cell mass accumulation has been demon- 1998) and the brain tumor, glioblastoma multiforme strated in primary murine ®broblasts (Beier et al., (Wechsler et al., 1997). However, the vast majority of 2000). This induction is separable from Myc induction the remaining allele in these tumors is not mutated. As of E2F2 and cell cycle progression through the use of such, if Mxi-1 functions as a suppressor, it must p27 null ®broblasts. Without a€ecting cyclin E-cdk2 contribute to tumorigenesis through haploinsuciency. activity, c-Myc induces E2F, S-phase entry and It is also possible, although as yet not established, that accumulation of cell mass in p27 null ®broblasts. In the remaining Mxi-1 allele may be silenced through contrast, E2F2 was able to trigger S-phase entry methylation. without cell mass accumulation. An adenoviral gene

Oncogene MYC activation and function LM Boxer and CV Dang 5602

Figure 4 The c-Myc ± Max heterodimer bound to DNA is depicted to a€ect a variety of cellular functions

delivery model of in vivo c-Myc expression also demonstrates the ability of c-Myc, but not E2F1, to cause liver cell hypertrophy without proliferation (Kim et al., 2000). In this case, within 4 days of intravenous delivery of Myc-expressing viruses, the murine livers were grossly enlarged with exuberantly large hepato- cytes that express high levels of ribosomal protein genes. These studies all support the notion that c-Myc regulates the accumulation of cell mass as cells traverse from early to late G1 (Figure 5). Figure 5 The model depicts c-Myc e€ects on cell growth arrest, A number of c-Myc responsive genes appear to be cell cycle regulators and metabolism. When c-Myc is over- involved in protein synthesis and hence might con- expressed the response of primary cells is cell cycle arrest or tribute to the cell growth phenotype (Schmidt, 1999). apoptosis through increases in ARF/p19 or p53 function. c-Myc Since ribosomes account for a signi®cant fraction of also induces cell cycle regulators which are inhibited by the dominant e€ect of ARF or p53. A Myc response distinctly cell mass, regulation of their assembly is likely to a€ect di€erent from cell cycle regulation is its role in cell mass cell mass accumulation. Myc responsive genes involved accumulation that may result from activation of protein synthesis in ribosome biogenesis include nucleolin, nucleophos- min, BN51, ®brillarin, RNA helicase DDX18, and a variety of ribosomal protein genes (Coller et al., 2000; Greasley et al., 2000; Guo et al., 2000; Schuhmacher et aminoacyl-tRNA synthetases have appeared as Myc al., 2001). Whether all of these genes are directly responsive genes (Coller et al., 2000; Guo et al., 2000; regulated by c-Myc remains uncertain, although in the Rosenwald et al., 1993; Schuhmacher et al., 2001; case of nucleolin sucient evidence have accumulated Schuldiner et al., 1996). Taken together, di€erent to support its direct regulation by c-Myc. In fact, studies have implicated the protein synthetic machin- nucleolin has appeared as a Myc target gene in several ery as a target for c-Myc regulation. Which of the independent studies of di€erent systems. What also rate-limiting steps for protein synthesis are a€ected by remains uncertain is whether c-Myc a€ects rate-limiting c-Myc remains unknown. Hence, additional studies steps of ribosomal assembly and whether c-Myc a€ects are necessary to delineate the causal connection ribosomal RNA availability. One study suggests that c- between c-Myc, its target genes involved in protein Myc augments rRNA level post-transcriptionally synthesis and the phenotype of cell mass accumula- (Gibson et al., 1992). tion. Several genes involved in translation have also been For cells to accumulate mass from early to late G1, implicated as Myc target genes. In particular the their biosynthetic capability must be coupled with translational initiation factors eIF2a, eIF3b, eIF4E, appropriate energy metabolism. In several models of eIF4G, eIF5A, the translational elongation factor EF- cell proliferation, it has been demonstrated that there is 2, amino acid transporter ECA39, as well as the a signi®cant increase in glycolytic metabolism before

Oncogene MYC activation and function LM Boxer and CV Dang 5603 cells enter S phase. It is hypothesized that glycolytic Cell proliferation metabolism, which has diminished production of reactive oxygen species as compared to oxidative Although Myc function has not been clari®ed from phosphorylation, would protect DNA from oxidative murine knock-out experiments, in which c-myc null damage (Dang and Semenza, 1999). Intriguingly, many embryos die at post-coital day 10.5, it is clear that cancers are also known to have a high rate of glucose other Myc family members, such as N-myc, are uptake and lactate production, a phenomenon reported insucient for normal development (Davis et al., over seven decades ago and known as the Warburg 1993). In contrast to c-myc, homozygous deletion of e€ect. Also notable is that human tumors tend to have L-myc does not a€ect development (Hatton et al., increased lactate dehydrogenase (LDH) A, and the 1996). Similar to c-myc, however, homozygous deletion serum LDH level has remained a major independent of N-myc also results of embryonic lethality (Stanton et predictor of poor outcome for diverse types of human al., 1992). The embryonic lethality seen in c-myc null cancers. C-Myc not only direct regulates LDHA, but it embryos suggests that either N-myc is unable to also directly activates the expression of the glucose substitute for c-myc or that temporal and spatial transporter 1 (GLUT1), phosphofructokinase, and regulation of c-myc expression is required in the enolase A (Osthus et al., 2000; Schuhmacher et al., embryo. To test this hypothesis, N-myc coding 1999, 2001; Shim et al., 1997). Furthermore, studies sequence has been knocked into the c-myc locus under have provided functional evidence that c-Myc increases the regulation of c-myc regulatory sequences (Malynn glucose uptake and lactate production in rat ®bro- et al., 2000). Intriguingly, it appears that N-Myc is blasts. fully capable of substituting for c-Myc. This experi- An E-box bound by c-Myc in the LDHA promoter ment speci®cally indicates that N-Myc protein could is also recognized by the hypoxia inducible transcrip- substitute for c-Myc during embryogenesis. tion factor HIF1. HIF1 activity increases as the Homozygous deletion of c-myc in an immortalized physiological response of cells to hypoxia, in which Rat1 ®broblast cell line and the use of conditional myc essentially all of the glycolytic enzyme genes are knock-out animals have provide signi®cant insights activated by HIF1 through cis-elements that resemble into c-Myc's role in cellular proliferation (de Alboran E-boxes (Dang and Semenza, 1999). Normal cells et al., 2001; Mateyak et al., 1997). Removal of c-Myc cease to proliferate in response to hypoxia, and the results in a marked prolonged doubling time. In the mechanism by which hypoxic cells cease to proliferate case of Rat1 ®broblasts, the doubling time increases involves activation of the cyclin-dependent kinase from about 16 h to about 50 h without c-Myc. The inhibitor p27 (Gardner et al., 2001). While HIF1 myc null ®broblasts, which do not express any other mediates the metabolic response of cells to hypoxia, c- Myc family proteins, display prolonged G1 and G2 Myc may allow certain tumor cells to adapt to phases with a conserved S phase length. In primary hypoxia through direct activation of glycolysis. murine ®broblasts rendered de®cient of c-Myc through Although induction of glycolysis may be a dominant Cre recombinase removal of ¯oxed myc alleles, the metabolic e€ect of c-Myc, c-Myc responsive genes, doubling time increases from about 20 h to over 200 h cyclophilin, VDAC, cytochrome c, and heme oxyge- (de Alboran et al., 2001). It is notable that viral nase 1, involved in mitochondrial function and transduction of Cre recombinase itself reduced growth presumably mitochondrial respiration have also been rate. However, the extended doubling time resulting identi®ed (Coller et al., 2000; Guo et al., 2000). from the removal of c-myc was associated with an Hence, it may speculated that c-Myc regulates both apparent G1 delay. These studies provide the genetic glycolysis and mitochondrial respiration, although the evidence that while c-Myc is not required for cell balance between oxygen-dependent and oxygen-inde- proliferation per se, it is required for the speed at pendent metabolism may be dictated by speci®c cell which cells progress through the cell cycle. types. The e€ect of ectopic c-Myc on the cell cycle Other Myc responsive genes, such as Ferritin H, machinery is context dependent. In particular, studies IRP-1, and IRP-2, re¯ect the cellular demand in iron performed with immortalized cell lines have reached metabolism, and other genes re¯ect requirements for conclusions that are likely to be confounded by the fact the synthesis of DNA, RNA and other macromolecules that immortalized cell lines tend to have defects in cell (Wu et al., 1999b). For example, the induction of cycle checkpoints. In contrast to immortalized cell carbamoyl phosphate synthase (CAD), dihydrofolate lines, overexpression of c-Myc in primary cells or in reductase, inosine-5'-monphosphate dehydrogenase, vivo results in either cell cycle arrest or apoptosis, ODC, and thymidine kinase by c-Myc suggests a which is thought to be one mechanism by which coupling between energy metabolism and the increase organisms defend against neoplastic cells arising from in DNA synthesis (Bello-Fernandez et al., 1993; Mai deregulated oncogenes. Ectopic expression of c-Myc in and Jalava, 1994; Miltenberger et al., 1995; Pusch et murine ®broblasts result in the activation of p53 and al., 1997b; Schuhmacher et al., 2001). The connection p19/ARF, although the e€ect on ARF is likely to be between c-myc and the potential regulation of a large indirect (Reisman et al., 1993; Zindy et al., 1998). variety of metabolic genes suggests that c-Myc is a key Increased ARF sequesters MDM2, which in turn integrator of cell mass accumulation, energy metabo- releases p53 to induce its growth inhibitory or lism and cell proliferation. apoptosis promoting activities. Likewise, it is shown

Oncogene MYC activation and function LM Boxer and CV Dang 5604 that induction of ectopic c-Myc in an in vivo model is Rb control as well (Alevizopoulos et al., 1997). c-Myc insucient for cell cycle progression unless a G2 appears to activate cyclin E and E2F2 directly and checkpoint is eliminated (Felsher et al., 2000). Hence, perhaps provide a means for bypassing Rb (Beier et al., the dominant e€ect of ectopic c-Myc in the primary, 2000). E2F2 could in turn activate cyclin E in a non-immortalized cell is the activation of cell cycle positive feed-forward loop that pushes cells from G1 arrest. into S phase. Presumably, cyclin E and E2F2 (or other The regulation of normal cell proliferation by c-Myc E2Fs) activates cyclin A, which is essential for S phase. involves several key components of the G1-S regula- Several genes involved in DNA replication such as tory molecules (Figure 6). Evidence to date suggest MCM4, ORC1, and replication factor 4C may also be several general features that may re¯ect the normal role targeted by c-Myc (Coller et al., 2000; Schuhmacher et of c-Myc in physiological cell proliferation. The cyclin- al., 2001). The ability of c-Myc to suppress growth dependent kinase inhibitor, p27, appears to be a critical arresting genes, such as gas1, GADD34, GADD45, target of c-Myc (Amati et al., 1998). In Rat1 cells and GADD153 is also likely to contribute to its growth nullizygous for c-myc, p27 levels are elevated (Mateyak promoting activities in cells that lack normal check- et al., 1997). Likewise, p27 levels are greatly increased points (Amundson et al., 1998; Lee et al., 1997). in primary ®broblasts or lymphocytes that have been Although myc nullizygous rat ®broblasts display a rendered nullizygous for c-myc (de Alboran et al., prolonged G2 phase, the role of c-Myc in G2 is less 2001). However, the inhibitor, p21, is not increased. C- clear and has not been directly studied (Mateyak et al., Myc reduces p27 levels or activity through several 1997, 1999). It may be speculated, however, that the mechanisms. Induction of cyclin D2 and CDK4 by c- lack of Myc's repression of p27 could well account for Myc may sequester p27 from cyclin E/CDK2 into the a prolonged G2 in cells lacking c-Myc. The prolonged cyclin D2/CDK4 complex although studies have also G2 is not apparent in murine embryo ®broblasts suggested that phosphorylation of p27 causes its release lacking c-Myc, and hence whether the G2 delay in from cyclin E/CDK2 (Bouchard et al., 1999; Muller et the absence of c-Myc is cell type dependent is not al., 1997; Perez-Roger et al., 1999; Vlach et al., 1996). known. Additionally, induction of a component of the Ectopic expression of c-Myc results in deregulated ubiquitin conjugating complex, CUL1, is proposed to entry into S phase as well as sensitizing cells to various mediate p27 degradation (O'Hagan et al., 2000a). apoptotic stimuli (Askew et al., 1991). Ectopic Myc Either or both of these mechanisms may contribute expression is sucient to cause S phase entry in growth to the decrease in a critical cyclin kinase inhibitor. c- factor deprived immortalized ®broblasts (Eilers et al., Myc may also cause a decrease in p21 expression, 1991). Ectopic c-Myc induces cyclin E expression in although this e€ect may be context dependent early G1, in contrast the expression of endogenous (Claassen and Hann, 2000; Coller et al., 2000). That Myc that is followed by cyclin E expression in late G1 is, depending on whether c-Myc is ectopically expressed (Perez-Roger et al., 1997; Pusch et al., 1997a). Despite in non-immortalized or immortalized human breast this early expression of cyclin E, one study demon- epithelial cells, p21 can either be induced or repressed strates that cells must attain a critical cell mass for (E Emison and C Dang, unpublished observation). entry into S phase (Pusch et al., 1997a). CDK2 activity The activation of CDK4 by Myc could result in the is activated by ectopic c-Myc and it appears to drive phosphorylation of Rb, which results in the activation cells prematurely into S phase, speci®cally when cells of E2F (Hermeking et al., 2000). E2F, in turn, is are treated with microtubule inhibitors or X-irradiated capable of activating cyclin E and other genes required (Li and Dang, 1999; Yin et al., 1999). This ability may for S phase entry. Myc also directly induces Id2; Id2 is underlie the observations that c-Myc is able to induce able to inactivate Rb through direct binding (Lasorella genomic instability. Whether this instability also et al., 2000). Thus, Myc may inactivate Rb through re¯ects the suppression of the mitotic checkpoint by several mechanisms, but in addition Myc circumvents c-Myc is not yet established.

Apoptosis

When cells are deprived of growth factors (serum), they withdraw from the cell cycle and have diminished c- myc expression. In contrast, cells that express ectopic c- Myc are triggered to release cytochrome c from the mitochondria with serum-deprivation, culminating in the activation of caspases and apoptosis (Juin et al., 1999). What are the target genes of c-Myc that sensitize cells to apoptotic triggers? Both ornithine decarbox- Figure 6 c-Myc is depicted to activate CDK4 and CDK2 ylase and LDHA have been shown to sensitize cells to complexes and activities through the induction of CDK4, cyclin D, E and reduction of p27. In cells that have lost the ARF serum-deprivation and glucose-deprivation induced response to c-Myc, c-Myc drives these cells into cycle in a fashion apoptosis, respectively (Packham and Cleveland, that requires other target genes involved in metabolism 1994; Shim et al., 1998). Serum and glucose deprivation

Oncogene MYC activation and function LM Boxer and CV Dang 5605 represent distinct stimuli since overexpression of cells rendered non-adherent stop proliferating. Normal LDHA sensitizes cells to glucose, but not serum, cells also recognize external cues and undergo contact deprivation. Although these genes, when ectopically inhibited growth when cells are con¯uent. This is in expressed, predispose cells to apoptosis; the molecular contrast to transformed cells that tend to proliferate underpinnings that trigger a central death pathway independently of external cellular cues. C-myc expres- remains poorly understood. sion in lymphocytes is associated with the decreased Myc target genes that regulate the cell cycle, such as expression of LFA-1 that participates in homotypic p53 and cdc25A, have been implicated in Myc-induced interaction between lymphocytes (Inghirami et al., apoptosis (Galaktionov et al., 1996; Reisman et al., 1990). Decreased LFA-1 expression occurs in many 1993). However, the roles of p53 and cdc25A in Myc- Burkitt's lymphomas and hence suggests that deregu- induced apoptosis may be cell line or context lated growth could in part be coupled to loss of LFA- dependent. Myc responsive genes involved in mito- 1. chondrial function such as cyclophilin, HSP70A and Other cellular adhesive molecules are down-regulated Mer5 (thioredoxin-dependent peroxidase) may also by c-Myc either through transcriptional mechanisms or a€ect apoptosis; however, their roles have not been through the induction of metalloproteinases. Collagens reported (Coller et al., 2000; Guo et al., 2000; Lewis et have been repeatedly reported to be down-regulated by al., 1997). Bax, the antagonist of Bcl2 or Bcl-xL, has c-Myc (Yang et al., 1991). In one study, it is proposed also been reported as a Myc target gene, although the that c-Myc interferes with the the regulation of results have not been con®rmed by independent studies collagen genes by nuclear factor 1 (CCAAT/NF1) (Mitchell et al., 2000). Evidence for c-Myc activation of (Yang et al., 1993). This is similar to the proposed upstream events in the apoptotic cascade, such as mechanism for the down-regulation of the PDGFb activation of FasL and Fas, have been provided by receptor gene by c-Myc (Oster et al., 2000). To date, several independent studies. In one study c-Myc however, mechanistic studies to determine whether appears to activate the Fas pathway in ®broblasts collagens could inhibit Myc-mediated tumorigenesis overexpressing Myc (Hueber et al., 1997). Although a have not been performed. Other potential repressed separate study with FADD (a mediator of Fas targets such as ®bronectin have also been reported apoptosis) knockout cells suggests that c-Myc induce (Coller et al., 2000). One study demonstrated that Myc apoptosis independently of the Fas pathway, a recent reduces a3 mRNA expression in a small cell lung study with lymphocytes containing conditional myc cancer cells line, leads to a decrease in a3b1 integrin knockout demonstrate that myc null lymphocytes are receptor and enhances anchorage independent growth unable to express Fas or FasL (de Alboran et al., (Barr et al., 1998). Restoration of a3 expression neither 2001). Furthermore, the FasL gene appears to be reduces the growth rate nor c-Myc expression, but it directly activated by c-Myc through a non-canonical dramatically blocks anchorage-independent growth. Myc binding site (Kasibhatla et al., 2000). DNA This study provides evidence that reduction of a microarray experiments have identi®ed other potential speci®c adhesion receptor by c-Myc enhances ancho- Myc target genes involved in apoptosis (TRAP1, ANT, rage-independent growth. It is further supported by an API2), although de®nitive studies are lacking (Guo et independent study demonstrating that restoration of b1 al., 2000; O'Hagan et al., 2000b; Schuhmacher et al., expression in ®broblasts lacking b1 causes a signi®cant 2001). reduction in the size of myc and ras induced tumors In aggregate, several studies support the role for Fas (Brakebusch et al., 1999). These studies together and FasL in Myc-induced apoptosis. It is also notable suggest that down-regulation of adhesion molecules that Fas-mediated apoptosis is coupled to glucose by c-Myc may be required for neoplastic transforma- transport (Berridge et al., 1996). That is, upon FasL tion. activation glucose transport dramatically decreases and Thrombospondin is a matrix-associated protein with conversely, enhancement of glucose uptake abrogates a negative regulatory role in angiogenesis. It is IL3-deprivation induced apoptosis (Kan et al., 1994). repressed by c-myc, probably through speci®c post- Although the molecular details of how Myc triggers transcriptional mechanisms (Janz et al., 2000). While it apoptosis remain to be delineated, the understanding is clear that c-myc mediates tumorigenesis, whether c- of the links between Myc overexpression, mitochon- Myc induces neovascularization is not yet clear (Ngo et drial function, and glycolysis will be necessary for al., 2000). In vivo models suggest that c-Myc induces better appreciation of the mechanisms underlying c- vascular endothelial growth factor VEGF (Pelengaris Myc induced apoptosis. et al., 1999). However, in several cell culture models, ectopic Myc expression caused a decrease in VEGF mRNA, although whether cell type contributes to the Cell adhesion and di€erentiation di€erences is unknown (Barr et al., 2000). It is likely that the regulation of angiogenesis by c-Myc di€ers A hallmark of cellular di€erentiation is the expression depending on the process, whether it is embryogenesis, of speci®c cell surface molecules that mediate cell ± cell wound repair or tumorigenesis. It is possible that and cell ± extracellular matrix interactions. Most ad- during embryogenesis or perhaps wound repair, herent and even non-adherent cells rely on cell ± cell or physiological expression of Myc directly couples with cell ± matrix signals for mitogenesis, such that normal angiogenesis; in contrast, ectopic overexpression of c-

Oncogene MYC activation and function LM Boxer and CV Dang 5606 Myc could result in the suppression of angiogenesis as to suppress apoptosis through AKT and PI3K, which an additional mechanism to protect organisms against is likely to suppress the apoptotic e€ects of c-Myc tumorigenesis. (Rohn et al., 1998). These recent observations o€er a c-Myc has long been shown to antagonize di€er- new view, which suggests that complementation entiation programs of several cell culture models between oncogenes includes both complementary including myocyte and adipocyte di€erentiation. The growth promoting e€ects as well as cross-suppression failure of cells expressing ectopic Myc to withdraw of normal safeguard mechanisms elicited by overactive from the cell cycle is thought to prevent them from oncogenes (Figure 7). expressing their di€erentiated phenotype. The concept that withdrawal from the cell cycle is required for cellular di€erentiation is now contested by the The spectrum of normal and abnormal myc function observations that continued Myc expression is required for di€erentiation of keratinocytes (Gandarillas et al., Regulated c-myc expression is required for normal 2000). Furthermore, the observation that B-cells with embryogenesis and cellular proliferation, as is re¯ected ectopic c-myc expression in vivo show limited di€er- by the phenotypes of embryos that are homozygously entiation suggests that c-myc expression and di€er- depleted of c-myc or primary cells rendered de®cient of entiation are not necessarily mutually exclusive. c-myc. To reach an understanding of normal c-myc function, reagents derived from these animals will have to be judiciously studied to identify normal c-myc Senescence and immortality target genes. The use of cell lines with inactivated c- myc alleles have yielded novel insights; however, it is Some of the landmark conceptual shifts stemmed from likely that some compensatory mechanisms must be at the study of single versus multiple oncogenes in play to account for a large di€erence between the primary cells. The ability of Myc and Ras to transform doubling time of primary cells deprived of c-myc versus primary rodent cells provided an early conceptual that of an immortalized rodent cell line. The di€erences framework for our understanding of multistep tumor- in growth rates between wild-type and myc null cells igenesis which was later modi®ed by the recognition of will pose a challenge in the interpretation of any tumor suppressors. The oncogenes were thought to studies that rely on growing cells. The experimental cooperate through their respective complementary manipulation of cells through serum deprivation and ability to drive cell proliferation (Land et al., 1986). re-stimulation to trigger mitogenesis remains a feasible That is, in the case of Myc and Ras, it was thought model to study target genes of c-Myc. Perhaps with the that Myc is able to immortalize primary cells and that use of DNA microarrays and chromatin immunopre- Ras provides additional oncogenic cues to drive cells cipitation assays, a better understanding of physiologi- into full tumorigenic potential. This simple view cal Myc target genes will emerge. overlooked the early events of ectopic c-myc expression Deregulated or abnormal function of c-Myc as in primary cells (Evan et al., 1992; Zindy et al., 1998). depicted by chromosomal translocations in B-cell As it is now known and discussed above, c-Myc over- malignancies is likely to elicit a distinctly di€erent expression in primary cells tends to result in cell cycle subset of Myc target genes. To date, one study arrest or apoptosis. Only cells that survive this crisis attempted to address this issue using isogenic wild- and massive exodus by death become immortalized by type, myc null, and wild-type cells plus ectopic myc and c-Myc (Zindy et al., 1998). The recognition that c-Myc activates telomerase activity and directly activates the gene encoding the reverse transcriptase subunit, TERT, suggests a possible mechanism for c-Myc mediated immortalization (Greenberg et al., 1999; Wang et al., 1998; Wu et al., 1999a). Yet Myc and Ras together are unable to transform human primary cells (Hahn et al., 1999). In fact, human primary cells require ectopic expression of hTERT for neoplastic transformation by activated Ras and SV40T antigen. The di€erence in the susceptibility of rodent versus human cells to Myc and Ras transformation may re¯ect the propensity for rodent cells to maintain telomerase activity and have much longer telomeres. Thus, the role of c-Myc in regulating hTERT physiologically or in tumorigenesis Figure 7 The cooperation of c-Myc and Ras is depicted for the remains to be established for human tissues. rat embryo cell transformation assay with a hypothetical model In contrast to c-Myc, activated Ras expression alone for their cooperativity. C-Myc alone induces apoptosis in the in primary cells results in senescence rather than cell absence of Ras. Apoptosis is inhibited by Ras through AKT and cycle arrest per se (Serrano et al., 1997). The potential PI3K, leaving c-Myc to force entry into the cell cycle without the penalty of death. C-Myc activation of telomerase (TERT) may in ability of c-Myc to maintain telomerase activity could turn inhibit the ability of Ras alone to induce senescence, counter the senescent e€ect of Ras. Ras is now known allowing Ras to manifest its oncogenic potential

Oncogene MYC activation and function LM Boxer and CV Dang 5607 DNA microarray technology (Guo et al., 2000). There the abnormal expression of non-physiological target appears to be a distinct subset of c-Myc responsive genes that are not regulated by endogenous c-myc. genes that are more readily detected between ectopic With advances in technologies that enable investiga- myc expressing wild-type cells and control wild-type tions to incorporate spatial and temporal regulation of cells versus myc null cells and control wild-type cells. gene expression, understanding of normal and abnor- In summary, while c-myc is not required for cell mal c-Myc function should be rapidly forthcoming. proliferation as is evident from the proliferation of myc null cells, the pleiotropic transcriptional e€ect of c-Myc suggests that it is a central integrator and acceleration Acknowledgments of physiological cell growth, proliferation and cellular Supported by National Institutes of Health Grant metabolism. Ectopic expression of c-Myc most likely CA69322 (LM Boxer), CA51497 (CV Dang) and contributes to tumorigenesis through both temporally CA57431 (CV Dang). We thank E Emison, L Gardner, disordered expression of physiological target genes and LA Lee and D Wechsler for comments.

References

AdamsJM,HarrisAW,PinkertCA,CorcoranLM, Brakebusch C, Wennerberg K, Krell HW, Weidle UH, Alexander WS, Cory S, Palmiter RD and Brinster RL. Sallmyr A, Johansson S and Fassler R. (1999). Oncogene, (1985). Nature, 318, 533 ± 538. 18, 3852 ± 3861. Alevizopoulos K, Vlach J, Hennecke S and Amati B. (1997). Bradley JF, Rothberg PG, Ladanyi M and Chaganti RSK. EMBO J., 16, 5322 ± 5333. (1993). Genes Chrom. Cancer, 7, 128 ± 130. Amati B, Alevizopoulos K and Vlach J. (1998). Front Biosci., Butzler C, Zou X, Popov AV and Bruggemann M. (1997). 3, D250 ± D268. Oncogene, 14, 1383 ± 1388. Amundson SA, Zhan Q, Penn LZ and Fornace Jr AJ. (1998). Cesarman E, Dalla-Favera R, Bentley D and Groudine M. Oncogene, 17, 2149 ± 2154. (1987). Science, 238, 1272 ± 1275. Arcinas M and Boxer LM. (1994). Oncogene, 9, 2699 ± 2706. Chang DW, Claassen GF, Hann SR and Cole MD. (2000). ar-Rushdi A, Nishikura K, Erikson J, Watt R, Rovera G and Mol. Cell. Biol., 20, 4309 ± 4319. Croce C. (1983). Science, 222, 390 ± 393. Chou TY, Hart GW and Dang CV. (1995). J. Biol. Chem., Askew DS, Ashmun RA, Simmons BC and Cleveland JL. 270, 18961 ± 18965. (1991). Oncogene, 6, 1915 ± 1922. Claassen GF and Hann SR. (2000). Proc. Natl. Acad. Sci. Ayer DE, Kretzner L and Eisenman RN. (1993). Cell, 72, USA, 97, 9498 ± 9503. 211 ± 222. Cockerill PN, Yuen M-H and Garrard WT. (1987). J. Biol. Ayer DE, Laherty CD, Lawrence QA, Armstrong AP and Chem., 262, 5394 ± 5397. Eisenman RN. (1996). Mol. Cell. Biol., 16, 5772 ± 5781. Cole MD and McMahon SB. (1999). Oncogene, 18, 2916 ± Bahram F, von der Lehr N, Cetinkaya C and Larsson LG. 2924. (2000). Blood, 95, 2104 ± 2110. Coller HA, Grandori C, Tamayo P, Colbert T, Lander ES, Barr LF, Campbell SE, Bochner BS and Dang CV. (1998). Eisenman RN and Golub TR. (2000). Proc. Natl. Acad. Cancer Res., 58, 5537 ± 5545. Sci. USA, 97, 3260 ± 3265. Barr LF, Campbell SE, Diette GB, Gabrielson EW, Kim S, Conzen SD, Gottlob K, Kandel ES, Khanduri P, Wagner AJ, Shim H and Dang CV. (2000). Cancer Res., 60, 143 ± 149. O'Leary M and Hay N. (2000). Mol. Cell. Biol., 20, 6008 ± Bernard O, Cory S, Gerondakis S, Webb E and Adams JM. 6018. (1983). EMBO J., 2, 2375 ± 2383. Cory S. (1986). Adv. Cancer Res., 47, 189 ± 234. Beier R, Burgin A, Kiermaier A, Fero M, Karsunky H, Dang CV. (1999). Mol. Cell. Biol., 19, 1 ± 11. Sa€rich R, Moroy T, Ansorge W, Roberts J and Eilers M. Dang CV, Dolde C, Gillison ML and Kato GJ. (1992). Proc. (2000). EMBO J., 19, 5813 ± 5823. Natl. Acad. Sci. USA, 89, 599 ± 602. Beijersbergen RL, Hijmans EM, Zhu L and Bernards R. Dang CV and Lee WM. (1988). Mol. Cell. Biol., 8, 4048 ± (1994). EMBO J., 13, 4080 ± 4086. 4054. Bello-Fernandez C, Packham G and Cleveland JL. (1993). Dang CV and Semenza GL. (1999). Trends Biochem. Sci., 24, Proc. Natl. Acad. Sci. USA, 90, 7804 ± 7808. 68 ± 72. Bemark M and Neuberger MS. (2000). Oncogene, 19, 3404 ± Davis AC, Wims M, Spotts GD, Hann SR and Bradley A. 3410. (1993). Genes Dev., 7, 671 ± 682. Bentley DL and Groudine M. (1986). Nature, 321, 702 ± 706. de Alboran IM, O'Hagan RC, Gartner F, Malynn B, Berridge MV, Tan AS, McCoy KD, Kansara M and Rudert DavidsonL,RickertR,RajewskyK,DePinhoRAand F. (1996). J. Immunol., 156, 4092 ± 4099. Alt FW. (2001). Immunity, 14, 45 ± 55. Bhatia K, Huppi K, Spangler G, Siwarski D, Iyer R and Drach J, Kaufmann H, Urbauer E, Schreiber S, Ackermann Magrath I. (1993). Nat. Genet., 5, 56 ± 61. J and Huber H. (2000). J. Cancer Res. Clin. Onc., 126, Blackwood EM and Eisenman RN. (1991). Science, 251, 441 ± 447. 1211 ± 1217. Dyson PJ, Littlewood TD, Forster A and Rabbitts TH. Bornkamm GW, Polack A and Eick D. (1988). c-myc (1985). EMBO J., 4, 2885 ± 2891. Deregulation by Chromosomal Translocation in Burkitt's Dyson PJ and Rabbitts TH. (1985). Proc. Natl. Acad. Sci. Lymphoma New York: M. Dekker, Inc. USA, 82, 1984 ± 1988. Bouchard C, Thieke K, Maier A, Sa€rich R, Hanley-Hyde J, Eick D and Bornkamm GW. (1986). Nucl. Acids Res., 14, AnsorgeW,ReedS,SicinskiP,BartekJandEilersM. 8331 ± 8345. (1999). EMBO J., 18, 5321 ± 5333.

Oncogene MYC activation and function LM Boxer and CV Dang 5608 Eilers M, Schirm S and Bishop JM. (1991). EMBO J., 10, Hoang AT, Lutterbach B, Lewis BC, Yano T, Chou TY, 133 ± 141. Barrett JF, Ra€eld M, Hann SR and Dang CV. (1995). Erikson J, ar-Rushdi A, Drwinga HL, Nowell PC and Croce Mol. Cell. Biol., 15, 4031 ± 4042. CM. (1983). Proc. Natl. Acad. Sci. USA, 80, 820 ± 824. Hueber AO, Zornig M, Lyon D, Suda T, Nagata S and Evan Evan GI, Wyllie AH, Gilbert CS, Littlewood TD, Land H, GI. (1997). Science, 278, 1305 ± 1309. Brooks M, Waters CM, Penn LZ and Hancock DC. Inghirami G, Grignani F, Sternas L, Lombardi L, Knowles (1992). Cell, 69, 119 ± 128. DM and Dalla-Favera R. (1990). Science, 250, 682 ± 686. Felsher DW, Zetterberg A, Zhu J, Tlsty T and Bishop JM. Iritani BM and Eisenman RN. (1999). Proc. Natl. Acad. Sci. (2000). Proc. Natl. Acad. Sci. USA, 97, 10544 ± 10548. USA, 96, 13180 ± 13185. Ferre-D'Amare AR, Pognonec P, Roeder RG and Burley Janz A, Sevignani C, Kenyon K, Ngo CV and Thomas- SK. (1994). EMBO J., 13, 180 ± 189. Tikhonenko A. (2000). Nucleic Acids Res., 28, 2268 ± 2275. Ferre-D'Amare AR, Prendergast GC, Zi€ EB and Burley Ji L, Arcinas M and Boxer LM. (1994). Mol. Cell. Biol., 14, SK. (1993). Nature, 363, 38 ± 45. 7967 ± 7974. Flinn EM, Busch CM and Wright AP. (1998). Mol. Cell. Ji L, Arcinas M and Boxer LM. (1995). J. Biol. Chem., 270, Biol., 18, 5961 ± 5969. 13392 ± 13398. Galaktionov K, Chen X and Beach D. (1996). Nature, 382, Johnston LA, Prober DA, Edgar BA, Eisenman RN and 511 ± 517. Gallant P. (1999). Cell, 98, 779 ± 790. Gallant P, Shiio Y, Cheng PF, Parkhurst SM and Eisenman Judde J-G and Max EE. (1992). Mol. Cell. Biol., 12, 5206 ± RN. (1996). Science, 274, 1523 ± 1527. 5216. Gandarillas A, Davies D and Blanchard JM. (2000). Juin P, Hueber AO, Littlewood T and Evan G. (1999). Genes Oncogene, 19, 3278 ± 3289. Dev., 13, 1367 ± 1381. Gardner LB, Li Q, Park MS, Flanagan WM, Semenza GL Kan O, Baldwin SA and Whetton AD. (1994). J. Exp. Med., and Dang CV. (2001). J. Biol. Chem., 276, 7919 ± 7926. 180, 917 ± 923. GaubatzS,ImhofA,DoschR,WernerO,MitchellP, Kanda K, Hu H-M, Zhang L, Grandchamps J and Boxer Buettner R and Eilers M. (1995). EMBO J., 14, 1508 ± LM. (2000). J. Biol. Chem., 275, 32338 ± 32346. 1519. Kasibhatla S, Beere HM, Brunner T, Echeverri F and Green Geltinger C, Hortnagel K and Polack A. (1996). Gene Exp., DR. (2000). Curr. Biol., 10, 1205 ± 1208. 6, 113 ± 127. Kato GJ, Barrett J, Villa-Garcia M and Dang CV. (1990). Gerbitz A, Mautner J, Geltinger C, Hortnagel K, Christoph Mol. Cell. Biol., 10, 5914 ± 5920. B, Asenbauer H, Klobeck G, Polack A and Bornkamm Kato GJ, Lee WM, Chen LL and Dang CV. (1992). Genes GW. (1999). Oncogene, 18, 1745 ± 1753. Dev., 6, 81 ± 92. Gibson AW, Ye R, Johnston RN and Browder LW. (1992). Kessler DJ, Spicer DB, La Rosa FA and Sonenshein G. Oncogene, 7, 2363 ± 2367. (1992). Oncogene, 7, 2447 ± 2453. Grandori C, Cowley SM, James LP and Eisenman RN. Kim S, Li Q, Dang CV and Lee LA. (2000). Proc. Natl. Acad. (2000). Ann. Rev. Cell. Dev. Biol., 16, 653 ± 699. Sci. USA, 97, 11198 ± 11202. Greasley P, Bonnard C and Amati B. (2000). Nucleic Acids Kolligs FT, Hu G, Dang CV and Fearon ER. (1999). Mol. Res., 28, 446 ± 453. Cell. Biol., 19, 5696 ± 5706. Greenberg RA, O'Hagan RC, Deng H, Xiao Q, Hann SR, Kolligs FT, Kolligs B, Hajra KM, Hu G, Tani M, Cho KR Adams RR, Lichtsteiner S, Chin L, Morin GB and and Fearon ER. (2000). Genes Dev., 14, 1319 ± 1331. DePinho RA. (1999). Oncogene, 18, 1219 ± 1226. Koskinen PJ, Ayer DE and Eisenman RN. (1995). Cell Gregory MA and Hann SR. (2000). Mol. Cell. Biol., 20, Growth Di€er., 6, 623 ± 629. 2423 ± 2435. Krumm A, Meulia T, Brunvand M and Groudine M. (1992). Gu W, Bhatia K, Magrath IT, Dang CV and Dalla-Favera Genes Dev., 6, 2201 ± 2213. R. (1994). Science, 264, 251 ± 254. Kyo S, Takakura M, Taira T, Kanaya T, Itoh H, Yutsudo Guo QM, Malek RL, Kim S, Chiao C, He M, Ru€y M, M, Ariga H and Inoue M. (2000). Nucleic Acids Res., 28, Sanka K, Lee NH, Dang CV and Liu ET. (2000). Cancer 669 ± 677. Res., 60, 5922 ± 5928. Lahoz EG, Xu L, Schreiber-Agus N and DePinho RA. Hahn WC, Counter CM, Lundberg AS, Beijersbergen RL, (1994). Proc. Natl. Acad. Sci. USA, 91, 5503 ± 5507. Brooks MW and Weinberg RA. (1999). Nature, 400, 464 ± Land H, Chen AC, Morgenstern JP, Parada LF and 468. Weinberg RA. (1986). Mol. Cell. Biol., 6, 1917 ± 1925. Hann SR, Sloan-Brown K and Spotts GD. (1992). Genes La Rosa FA, Pierce JW and Sonenshein GE. (1994). Mol. Dev., 6, 1229 ± 1240. Cell. Biol., 14, 1039 ± 1044. HattonKS,MahonK,ChinL,ChiuFC,LeeHW,PengD, Lasorella A, Noseda M, Beyna M and Iavarone A. (2000). Morgenbesser SD, Horner J and DePinho RA. (1996). Nature, 407, 592 ± 598. Mol. Cell. Biol., 16, 1794 ± 1804. Leder P, Battey J and Lenoir G. (1983). Science, 222, 765 ± Hayday AC, Gillies SD, Saito H, Wood C, Wiman D, 771. Hayward WS and Tonegawa S. (1984). Nature, 307, 334 ± Lee LA, Dolde C, Barrett J, Wu CS and Dang CV. (1996). J. 340. Clin. Invest., 97, 1687 ± 1695. He TC, Sparks AB, Rago C, Hermeking H, Zawel L, da Lee TC, Li L, Philipson L and Zi€ EB. (1997). Proc. Natl. Costa LT, Morin PJ, Vogelstein B and Kinzler KW. Acad.Sci.USA,94, 12886 ± 12891. (1998). Science, 281, 1509 ± 1512. Lewis BC, Shim H, Li Q, Wu CS, Lee LA, Maity A and Dang Henriksson M and Luscher B. (1996). Adv. Cancer Res., 68, CV. (1997). Mol. Cell. Biol., 17, 4967 ± 4978. 109 ± 182. LiLH,NerlovC,PrendergastG,MacGregorDandZi€EB. Hermeking H, Rago C, Schuhmacher M, Li Q, Barrett JF, (1994). EMBO J., 13, 4070 ± 4079. Obaya AJ, O'Connell BC, Mateyak MK, Tam W, Li Q and Dang CV. (1999). Mol. Cell. Biol., 19, 5339 ± 5351. Kohlhuber F, et al. (2000). Proc. Natl. Acad. Sci. USA, Lombardi L, Newcomb EW and Dalla-Favera R. (1987). 97, 2229 ± 2234. Cell, 49, 161 ± 170.

Oncogene MYC activation and function LM Boxer and CV Dang 5609 Lutterbach B and Hann SR. (1994). Mol. Cell. Biol., 14, Pelicci PG, Knowles DM, Magrath I and Dalla-Favera R. 5510 ± 5522. (1986). Proc. Natl. Acad. Sci. USA, 83, 2984 ± 2988. Madisen L and Groudine M. (1994). Genes Dev., 8, 2212 ± Perez-Roger I, Kim SH, Griths B, Sewing A and Land H. 2226. (1999). EMBO J., 18, 5310 ± 5320. Madisen L, Krumm A, Hebber TR and Groudine M. (1998). Perez-Roger I, Solomon DL, Sewing A and Land H. (1997). Mol. Cell. Biol., 18, 6281 ± 6292. Oncogene, 14, 2373 ± 2381. Magrath I. (1990). Adv. Cancer Res., 55, 133 ± 270. Pettersson S, Cook GP, Bruggemann M, Williams GT and Marcu KB, Bossone SA and Patel AJ. (1992). Ann. Rev. Neuberger MS. (1990). Nature, 344, 165 ± 167. Biochem., 61, 809 ± 860. Polack A, Feederle R, Klobeck G and Hortnagel K. (1993). Mai S and Jalava A. (1994). Nucleic Acids Res., 22, 2264 ± EMBO J., 12, 3913 ± 3920. 2273. Prendergast GC and Zi€ EB. (1992). Trends Genet., 8, 91 ± Malynn BA, de Alboran IM, O'Hagan RC, Bronson R, 96. Davidson L, DePinho RA and Alt FW. (2000). Genes Prochownik EV, Eagle Grove L, Deubler D, Zhu XL, Dev., 14, 1390 ± 1399. Stephenson RA, Rohr LR, Yin X and Brothman AR. Mateyak MK, Obaya AJ, Adachi S and Sedivy JM. (1997). (1998). Genes Chromosomes Cancer, 22, 295 ± 304. Cell Growth Di€er., 8, 1039 ± 1048. Pusch O, Bernaschek G, Eilers M and Hengstschlager M. Mateyak MK, Obaya AJ and Sedivy JM. (1999). Mol. Cell. (1997a). Oncogene, 15, 649 ± 656. Biol., 19, 4672 ± 4683. Pusch O, Soucek T, Hengstschlager-Ottnad E, Bernaschek G McMahon SB, Van Buskirk HA, Dugan KA, Copeland TD and Hengstschlager M. (1997b). DNA Cell. Biol., 16, 737 ± and Cole MD. (1998). Cell, 94, 363 ± 374. 747. McMahon SB, Wood MA and Cole MD. (2000). Mol. Cell. Queva C, Hurlin PJ, Foley KP and Eisenman RN. (1998). Biol., 20, 556 ± 562. Oncogene, 16, 967 ± 977. Michaelson JS, Giannini SL and Birshtein BK. (1995). Nucl. Rabbits TH, Forster A, Hamlyn P and Baer R. (1984). Acids Res., 23, 975 ± 981. Nature, 309, 592 ± 597. Miltenberger RJ, Sukow KA and Farnham PJ. (1995). Mol. ReismanD,ElkindNB,RoyB,BeamonJandRotterV. Cell. Biol., 15, 2527 ± 2535. (1993). Cell Growth Di€er., 4, 57 ± 65. Mink S, Mutschler B, Weiskirchen R, Bister K and Rohn JL, Hueber AO, McCarthy NJ, Lyon D, Navarro P, Klempnauer KH. (1996). Proc. Natl. Acad. Sci. USA, 93, Burgering BM and Evan GI. (1998). Oncogene, 17, 2811 ± 6635 ± 6640. 2818. Mitchell KO, Ricci MS, Miyashita T, Dicker DT, Jin Z, Rosenwald IB, Rhoads DB, Callanan LD, Isselbacher KJ Reed JC and El-Deiry WS. (2000). Cancer Res., 60, 6318 ± and Schmidt EV. (1993). Proc. Natl. Acad. Sci. USA, 90, 6325. 6175 ± 6178. Muller B, Stappert H and Reth M. (1990). Eur. J. Immunol., Roussel MF. (1998). Adv. Cancer Res., 74, 1 ± 24. 20, 1409 ± 1411. Roussel MF, Ashmun RA, Sherr CJ, Eisenman RN and Ayer Muller D, Bouchard C, Rudolph B, Steiner P, Stuckmann I, DE. (1996). Mol. Cell. Biol., 16, 2796 ± 2801. Sa€rich R, Ansorge W, Huttner W and Eilers M. (1997). Saleque S, Singh M, Little RD, Giannini SL, Michaelson JS Oncogene, 15, 2561 ± 2576. and Birshtein BK. (1997). J. Immunol., 158, 4780 ± 4787. Nepveu A and Marcu KB. (1986). EMBO J., 5, 2859 ± 2865. Saphire AC, Bark SJ and Gerace L. (1998). J. Biol. Chem., Nesbit CE, Tersak JM and Prochownik EV. (1999). 273, 29764 ± 29769. Oncogene, 18, 3004 ± 3016. Schmidt EV. (1999). Oncogene, 18, 2988 ± 2996. NgoCV,GeeM,AkhtarN,YuD,VolpertO,AuerbachR Schreiber-Agus N, Chin L, Chen K, Torres R, Rao G, Guida and Thomas-Tikhonenko A. (2000). Cell Growth Di€er., P, Skoultchi AI and DePinho RA. (1995). Cell, 80, 777 ± 11, 201 ± 210. 786. Nishikura K, ar-Rushdi A, Erikson J, Watt R, Rovera G and Schreiber-Agus N, Meng Y, Hoang T, Hou Jr. H, Chen K, Croce CM. (1983). Proc. Natl. Acad. Sci. USA, 80, 4822 ± Greenberg R, Cordon-Cardo C, Lee HW and DePinho 4826. RA. (1998). Nature, 393, 483 ± 487. Nussenzweig MD, Schmidt E, Shaw AC, Sinn E, Campos- Schreiber-Agus N, Stein D, Chen K, Goltz JS, Stevens L and Torres J, Mathey-Prevot B, Pattengale PK and Leder P. DePinho RA. (1997). Proc. Natl. Acad. Sci. USA, 94, (1988). Nature, 336, 446 ± 450. 1235 ± 1240. O'Hagan RC, Ohh M, David G, de Alboran IM, Alt FW, Schuhmacher M, Kohlhuber F, Holzel M, Kaiser C, Kaelin WG and DePinho RA. (2000a). Genes Dev., 14, Burtscher H, Jarsch M, Bornkamm GW, Laux G, Polack 2185 ± 2191. A, Weidle UH and Eick D. (2001). Nucleic Acids Res., 29, O'Hagan RC, Schreiber-Agus N, Chen K, David G, Engel- 397 ± 406. man JA, Schwab R, Alland L, Thomson C, Ronning DR, Schuhmacher M, Staege MS, Pajic A, Polack A, Weidle UH, Sacchettini JC. et al. (2000b). Nat. Genet., 24, 113 ± 119. Bornkamm GW, Eick D and Kohlhuber F. (1999). Curr. Oster SK, Marhin WW, Asker C, Facchini LM, Dion PA, Biol., 9, 1255 ± 1258. Funa K, Post M, Sedivy JM and Penn LZ. (2000). Mol. Schuldiner O, Eden A, Ben-Yosef T, Yanuka O, Simchen G Cell. Biol., 20, 6768 ± 6778. and Benvenisty N. (1996). Proc. Natl. Acad. Sci. USA, 93, Osthus RC, Shim H, Kim S, Li Q, Reddy R, Mukherjee M, 7143 ± 7148. Xu Y, Wonsey D, Lee LA and Dang CV. (2000). J. Biol. Sears R, Leone G, DeGregori J and Nevins JR. (1999). Mol. Chem., 275, 21797 ± 21800. Cell., 3, 169 ± 179. Packham G and Cleveland JL. (1994). Mol. Cell. Biol., 14, Serrano M, Lin AW, McCurrach ME, Beach D and Lowe 5741 ± 5747. SW. (1997). Cell, 88, 593 ± 602. Palomo C, Zou X, Nicholson IC, Butzler C and Bruggemann Shim H, Chun YS, Lewis BC and Dang CV. (1998). Proc. M. (1999). Cancer Res., 59, 5625 ± 5628. Natl. Acad. Sci. USA, 95, 1511 ± 1516. Pelengaris S, Littlewood T, Khan M, Elia G and Evan G. (1999). Mol. Cell., 3, 565 ± 577.

Oncogene MYC activation and function LM Boxer and CV Dang 5610 Shim H, Dolde C, Lewis BC, Wu CS, Dang G, Jungmann Wechsler DS, Shelly CA, Petro€ CA and Dang CV. (1997). RA, Dalla-Favera R and Dang CV. (1997). Proc. Natl. Cancer Res., 57, 4905 ± 4912. Acad. Sci. USA, 94, 6658 ± 6663. Whitehurst C, Henney HR, Max EE, Schroeder HWJ, Shiramizu B, Barriga F, Neequaye J, Jafri A, Dalla-Favera Stuber F, Siminovitch KA and Garrard WT. (1992). Nucl. R, Neri A, Guttierez M, Levine P and Magrath I. (1991). Acids Res., 20, 4929 ± 4930. Blood, 77, 1516 ± 1526. Wittekindt NE, Hortnagel K, Geltinger C and Polack A. ShouY,MartelliML,GabreaA,QiY,BrentsLA,Rosckle (2000). Nucl. Acids Res., 28, 800 ± 808. A, Deward G, Kirsch IR, Bergsagel PL and Kuehl WM. Wood LJ, Mukherjee M, Dolde CE, Xu Y, Maher JF, (2000). Proc. Natl. Acad. Sci. USA, 97, 228 ± 233. Bunton TE, Williams JB and Resar LM. (2000). Mol. Cell. Siebenlist U, Henninghausen L, Battey J and Leder P. Biol., 20, 5490 ± 5502. (1984). Cell, 37, 381 ± 391. Wu K, Grandori C, Amacker M, Simon-Vermot N, Polack Spencer CA and Groudine M. (1991). Adv. Cancer Res., 56, A, Lingner J and Dalla-Favera R. (1999a). Nat. Genet., 21, 1±47. 220 ± 224. Spotts GD, Patel SV, Xiao Q and Hann SR. (1997). Mol. Wu KJ, Polack A and Dalla-Favera R. (1999b). Science, 283, Cell. Biol., 17, 1459 ± 1468. 676 ± 679. Stanton BR, Perkins AS, Tessarollo L, Sassoon DA and Yang BS, Geddes TJ, Pogulis RJ, de Crombrugghe B and Parada LF. (1992). Genes Dev., 6, 2235 ± 2247. Freytag SO. (1991). Mol. Cell. Biol., 11, 2291 ± 2295. Strobl LJ and Eick D. (1992). EMBO J., 11, 3307 ± 3314. Yang BS, Gilbert JD and Freytag SO. (1993). Mol. Cell. Strobl LJ, Kohlhuber F, Mautner J, Polack A and Eick D. Biol., 13, 3093 ± 3102. (1993). Oncogene, 8, 1437 ± 1447. Yano T, Sander CA, Clark HM, Dolezal MV, Ja€e ES and Taub R, Moulding C, Battey J, Murphy W, Vasicek T, Ra€eld M. (1993). Oncogene, 8, 2741 ± 2748. Lenoir GM and Leder P. (1984). Cell, 36, 339 ± 348. Yin XY, Grove L, Datta NS, Long MW and Prochownik Vlach J, Hennecke S, Alevizopoulos K, Conti D and Amati EV. (1999). Oncogene, 18, 1177 ± 1184. B. (1996). EMBO J., 15, 6595 ± 6604. Yu BW, Ichinose I, Bonham MA and Zajac-Kaye M. (1993). Wang J, Xie LY, Allan S, Beach D and Hannon GJ. (1998). J. Biol. Chem., 268, 19586 ± 19592. Genes Dev., 12, 1769 ± 1774. Zindy F, Eischen CM, Randle DH, Kamijo T, Cleveland JL, Wang X, Scott E, Sawyers CL and Friedman AD. (1999). Sherr CJ and Roussel MF. (1998). Genes Dev., 12, 2424 ± Blood, 94, 560 ± 571. 2433.

Oncogene