Metabolic enzymes regulated by the Myc oncogene are possible ...

Health Implications of Dietary Amines

Metabolic enzymes regulated by the Myc oncogene are possible targets for chemotherapy or chemoprevention S. Rimpi and J.A. Nilsson1 Department of Molecular Biology, Umeå University, SE-901 87 Umeå, Sweden

Abstract The Myc oncogenes are dysregulated in 70% of human cancers. They encode transcription factors that bind to E-box sequences in DNA, driving the expression of a vast amount of target genes. The biological outcome is enhanced proliferation (which is counteracted by apoptosis), angiogenesis and cancer. Based on the biological effects of Myc overexpression it was originally assumed that the important Myc target genes are those encoding components of the cell cycle machinery. Recent work has challenged this notion and indicates that Myc target genes encoding metabolic enzymes deserve attention, as they may be critical arbiters of Myc in cancer. Thus targeting metabolic enzymes encoded by Myc-target genes may provide a new means to treat cancer that have arisen in response to deregulated Myc oncogenes.

Introduction Myc (c-Myc, N-Myc, L-Myc) oncogenes are regulated directly or indirectly at many levels of control by several of the major signalling pathways including receptor tyrosine kinases, wnt signalling, NF-κB (nuclear factor κB) signalling and the hedgehog pathway. Since many of these pathways are over-activated, and Myc proteins themselves can become deregulated by gene translocation, gene amplification and point mutations altering protein stability, it is safe to assume that a majority of human cancers have a deregulated Myc oncogene [1,2]. Myc is essential for cell proliferation and is sufficient to drive cells into S-phase from a quiescent state. On the other hand, Myc also induces cell death by apoptosis, which is thought to reflect a cellular defence against illegitimate cell proliferation [3,4]. This means that malignant transformation requires disruption of the apoptotic pathway, which can occur through loss of p53 or p19Arf , activation of anti-apoptotic Bcl-2 family members or through reinforcement of oncogenic signalling by for instance Ras [1]. In cells that have evaded the apoptotic machinery, deregulated Myc will, among other things, give the tumours the ability to proliferate uncontrollably, induce angiogenesis, override cell cycle arrest, increase metabolism and down-regulate cell adhesion molecules. These are properties of great value to a malignant cell, explaining the high degree of selection for Myc activation in cancer. The compelling connection between Myc and human cancer has prompted researchers to develop animal models that Key words: glycolysis, lymphomagenesis, Myc, nucleotide biogenesis, polyamine. Abbreviations used: Amd1, S-adenosylmethionine decarboxylase; bHLH-LZ, basic helix–loop– helix leucine-zipper; Cad, carbamoyl phosphate synthetase 2/aspartate transcarbamylase/ dihydro-orotase; DFMO, α-difluoromethylornithine; DLBCL, diffuse large B-cell lymphoma; Eµ, heavy chain enhancer; LDH-A, lactate dehydrogenase A; NF-κB, nuclear factor κB; Odc, ornithine decarboxylase; RT, reverse transcription; Shmt, serine hydroxymethyltransferase; Srm, spermidine synthase. To whom correspondence should be addressed (email [email protected]).

1

mimic Myc-induced cancer. One of the first mouse models of cancer was the Eµ-Myc transgenic mouse developed to model the human disease Burkitt’s lymphoma [5]. These mice overexpress c-Myc in B-cells by virtue of the immunoglobulin heavy chain enhancer (Eµ), similar to the Myc:Ig translocation found in the human disease. The tumours arising in Eµ-Myc mice do not perfectly phenocopy human Burkitt’s lymphoma but they are pre-B- and immature B-cell lymphomas which accumulate in lymph nodes 3– 6 months after birth [5]. To better model human Burkitt’s lymphoma, additional models which develop more mature B-cell lymphomas have been developed by conventional transgenesis (λ-Myc [6]) and knock-in technology (iMycEµ [7]). Still, however, an accurate mouse model of human Burkitt’s lymphoma is lacking. For the work discussed in this mini-review, we have focused on using the Eµ-Myc mouse model so we can compare data directly with the very large body of work present in the literature.

Gene regulation through interactions in the Max network Myc is a member of the Max network of heterodimeric transcription factors interacting through their bHLH-LZ (basic helix–loop–helix leucine-zipper) regions. Myc induces target genes in complex with Max by attracting histone acetylases and/or through displacement of Mad/Mnt–Max repressional complexes from the E-boxes (CACGTG) in the vicinity of the target gene promoters [1,8,9]. Myc can also cause transrepression of genes by interfering with Miz-1 gene activation at Inr elements (Figure 1).

Myc target genes E-boxes are present in thousands of genes, but not all are known to be direct transcriptional targets of Myc. Various studies have focused on finding true Myc targets and those C 2007

Biochemical Society

305

306

Biochemical Society Transactions (2007) Volume 35, part 2

Figure 1 Regulation of target gene expression by members of the Myc and Mad/Mnt families of transcription factors The central molecule in the regulation of E-box (CACGTG)-containing genes is the small bHLH-LZ protein Max. 1. In the classic way of transactivation, Myc dimerizes with Max and binds DNA at E-box sequences [38–41]. Thereafter, histone acetyltransferase activities are

involved in metabolism (Table 1). Our current belief is that these targets are relevant for Myc-induced cancers and we will discuss metabolic pathways undergoing current investigation in our laboratory.

recruited via interaction with TRRAP [42–45]. Studies on the Cad gene have also implicated a post-RNA polymerase II recruitment step that involves the stimulation of elongation mediated by the transcriptional

Myc-induced genes involved in metabolism and cancer

activator P-TEFb [31,46]. 2. Mnt/Mad proteins in complex with Max repress Myc target genes by recruiting SIN3, N-CoR (nuclear receptor co-repressor) and histone deacetylases [47–50]. Mad/Mnt proteins are

Polyamine biosynthesis

induced by differentiation in some cell types and serve to antagonize the growth-promoting effect of Myc [41]. 3. In this revised model, Myc serve as the antagonist of Mnt and activates genes by relief of transrepression. It is expected from this model that loss of Mnt should mimic the effect of Myc overexpression, i.e. enhanced cell proliferation, apoptosis and tumorigenesis, which seems to be the case in certain cellular contexts [9,51–55]. 4. Myc–Max dimers can repress transcription by blocking the activity of the transcription factor Miz-1. By repressing the transcription of selected target genes involved in growth inhibition, such as the cyclin-dependent kinase inhibitors p15Ink4b and p21Cip1 , Myc can override a G1 arrest [56–59].

convincingly supported by experimental evidence are listed at http://www.myccancergene.org [10] along with relevant references. An important question to ask when considering the various reported c-Myc target genes is whether all of these genes are really relevant under normal conditions or whether some of them are more or less relevant only when c-Myc is overexpressed? In addition, is the full Myc transcriptome essential for Myc-induced cancer or are some genes more important than others? The latter question opens up a large field of research which may lead to the identification of new targets for chemotherapy or chemoprevention. We have started to dig into this pile of target genes in a biased manner, fully aware that genomewide screens will be necessary to complete the job. Despite the ability of Myc to promote cell cycle progression, the majority of Myc target genes do not encode proteins involved in cell cycle control, albeit focus has long been on such target genes [11,12]. Recent data have instead warranted a closer look at other genes controlled by Myc, such as those C 2007

Biochemical Society

A comprehensive review of polyamine metabolism is present in other papers of this issue of Biochemical Society Transactions. Odc (ornithine decarboxylase), Amd1 (S-adenosylmethionine decarboxylase) and Srm (spermidine synthase) all have E-boxes in their regulatory regions and are elevated in Myc-expressing cells (Table 1), but only Odc has been studied in great detail with respect to its regulation by Myc [13,14]. Odc was one of the first Myc target genes identified and as such it has received proper attention. Odc is not a strong transcriptional target of Myc, but it is reliable. In fact, very few Myc target genes are induced more than 2–3-fold, perhaps reflecting the limiting amount of TATAbinding protein present in a cell [15]. To circumvent this limitation some genes, e.g. Srm, carry several E-boxes within their regulatory regions which results in a higher degree of transcriptional activation [16]. The idea of blocking polyamine synthesis with chemical inhibitors as a means to treat cancer has been around since the late 1960s/early 1970s, when it was found that cancer cells have elevated Odc activity and overproduce polyamines. Unfortunately, and surprisingly, very few successes with polyamine synthesis inhibitors on established tumours have been reported in clinical trials [17]. The ability of cancer cells to compensate for synthesis inhibition by increasing uptake of dietary polyamines has been put forward as an explanation and, indeed, polyamine-transport-deficient cancer cells are easily cleared by in vivo administration of the Odc inhibitor DFMO (α-difluoromethylornithine) [18]. Thus, when asked to analyse the role of Odc in Myc-induced lymphomagenesis by postdoctoral advisor Dr John L. Cleveland (St Jude Children’s Research Hospital, Memphis, TN, U.S.A.), the second author of the present mini-review accepted the task with some initial scepticism. The experimental setup was to utilize the newly developed Odc heterozygous knockout mice and breed these to the aforementioned Eµ-Myc mouse model. In addition, the Eµ-Myc mice were to receive DFMO in their drinking water from weanling age on. Strikingly, both the genetic and pharmacological experiments showed that reducing polyamine levels by ∼50% is compatible with normal mouse development but not with lymphomagenesis [19]. Findings presented in the study suggest that tumour cells have an elevated requirement for energy metabolism, since deletion of one of the alleles of the Odc gene, or treatment of mice with DFMO reduced the levels of polyamines in tumour cells back to the levels found in normal cells. Treatment with DFMO before the onset of tumorigenesis increased the overall survival of the Eµ-Myc mice; however, treatment of


Table 1 Selected Myc metabolic target genes Adapted from http://www.myccancergene.org Gene Target

Description

Polyamine metabolism AMD1 S-Adenosylmethionine decarboxylase 1 ODC1 SRM

Sugar metabolism ENO1

ENO2 GAPDH

GPI H6PD HK2 LDH-A

LDH-B PFKM PGK1 PKM2 RP1A

CTPS

DHFR MTHFR

SHMT1*

TK1 UMPK

Cell type

Polyamine metabolism

Up

Daudi, U-937 monocyte, [60–62] P493 lymphocyte, fibroblast (rat) Fibroblast, lymphocyte [13] HL60, P493 lymphocyte, [60] fibroblast, U-937 monocyte, lymphocyte, endothelial

Human

Human Human, rat

Enolase 1 (α)

Glycolysis

Up

Mouse, rat, chicken, human

Enolase 2 (γ ) Glyceraldehyde-3phosphate dehydrogenase Glucose phosphate isomerase Hexose-6-phosphate dehydrogenase Hexokinase 2 Lactate dehydrogenase A

Glycolysis Glycolysis

N/A† Up

Glycolysis Pentose phosphate pathway Glycolysis Glycolysis

Glycolysis

Up

Glycolysis

Up

Glycolysis Glycolysis Pentose phosphate pathway, nucleotide metabolism, DNA synthesis

Nucleotide metabolism AMPD3 AMP deaminase CAD

Regulation Organism

Ornithine decarboxylase 1 Polyamine metabolism Up Spermidine synthase Polyamine metabolism, Up RNA synthesis

Lactate dehydrogenase 2, B-chain Phosphofructokinase, muscle Phosphoglycerate kinase 1 Pyruvate kinase, muscle Ribose-5-phosphate isomerase A (ribose-5-phosphate epimerase)

Carbamoyl-phosphate synthetase 2/ aspartate transcarbamylase/ dihydro-orotase CTP synthetase

Dihydrofolate reductase Methylenetetrahydrofolate reductase Serine hydroxymethyltransferase 1, soluble Thymidine kinase 1 UMP kinase

Reference(s) (selected)

Function

[60,63]

Up

P493 lymphocyte, lymphocyte, liver, fibroblast, HL60, U-937 monocyte, endothelial Human Burkitt lymphoma, Daudi Human, rat, mouse HL60, liver, P493 lymphocyte, fibroblast, U-937 monocyte Rat, mouse, human Fibroblast, liver

N/A†

Human

Burkitt lymphoma, Daudi

[61]

Up Up

Rat Rat, human, chicken

[62] [25]

Up Up N/A†

Fibroblast Fibroblast, endothelial, Burkitt lymphoma, Daudi, lymphocyte Human, rat Epithelial, kidney, endothelial Human, rat, mouse Liver, endothelial, fibroblast Mouse, rat Liver, fibroblast Human Burkitt lymphoma, Daudi Human Fibroblast, lymphocyte

Nucleotide metabolism

Up

Human

[60]

DNA repair, nucleotide metabolism, DNA synthesis

Up

Rat, human

DNA synthesis, Up nucleotide metabolism DNA metabolism Up Amino acid metabolism

Human

DNA metabolism

Up

Human, rat

DNA metabolism Nucleotide metabolism

Up Up

Human Human

Human Human

P493 lymphocyte, U-937 monocyte, HL60 P493 lymphocyte, U-937 monocyte, fibroblast, HL60, lymphocyte, endothelial, Burkitt lymphoma, Daudi Lymphocyte

P493 lymphocyte, HL60, U-937 monocyte

[61] [60,63]

[63,64]

[65] [65] [63] [61] [66]

[30,34]

[67]

[29] [65]

[28]

[68] [67]

Lymphocyte

*SHMT2 was also found to be induced by Myc in the same study. †Myc-binding has been shown using chromatin immunoprecipitation, but no expression analysis has been performed. C 2007

Biochemical Society

307

308


mice that already displayed symptoms of tumour development could neither stop disease progress nor delay it. This adds an additional explanation of why previous clinical studies have failed; polyamines are likely to be essential for the development of tumours, but perhaps not for the maintenance of established tumours. These data are in accordance with the accumulating data praising Odc as a target for chemoprevention [17,20], a topic discussed elsewhere in this issue of Biochemical Society Transactions. Although the mechanism of chemoprevention is yet to be completely resolved, data on the Eµ-Myc mouse indicate that reduction of polyamine levels reduces Myc-induced proliferation and blocks secondary mutations of predominantly the tumour suppressor Arf , two mechanisms that are likely to be interconnected [19].

Figure 2 B-cells from Myc-transgenic mice overexpress genes encoding metabolic enzymes Magnetically sorted B-cells from wild-type and Eµ-Myc transgenic littermates were analysed using real-time RT-PCR. The expression levels of the indicated genes were compared with that of Ubiquitin (Ub) which is not regulated by Myc. For abbreviations of the genes, see Table 1. wt, wild-type.

Glycolysis Glycolysis is the process whereby glucose is oxidized to pyruvate yielding four ATP molecules of energy while consuming two. The fate of pyruvate is to either be converted into acetyl-CoA for the TCA (tricarboxylic acid) cycle or to be converted into lactate when oxygen supplies are low. It has been known since the 1920s from pioneering work by Nobel laureate Otto Warburg that tumour cells rely primarily on glycolysis even in an aerobic environment, an observation known as the Warburg effect [21,22]. As virtually all glycolytic enzymes are induced by Myc (Table 1 and [23]), it is conceivable that part of the mechanism behind the Warburg effect is that oncogenes such as Myc induce glycolytic enzymes that drive glucose metabolism in cancer cells. Furthermore, the increased use of glycolysis in tumour cells may represent a survival benefit under hypoxic conditions, as well as selecting for cells that are especially well suited for growing in the acidic environment that increased lactate production creates, meaning that cells with high survival capacity and malignant behaviour are favoured. Desirable features of anticancer treatment is that it be targeted to a known pathway and that it be selective against cancer cells. Here glycolysis inhibition could be a good candidate pathway due to the Warburg effect and that substances inhibiting glycolytic enzymes are generally tolerated at rather high doses (reviewed in [24]). Although targeting of glycolytic enzymes is likely to preferentially kill malignant cells, there is the risk of harming normal tissues that mainly rely on glycolysis, such as the brain, retina and testis. In the case of the brain, where neurotoxicity needs to be avoided, anti-glycolytic drugs that do not cross the blood–brain barrier should be developed [24]. The enzyme involved in glycolysis that has received the most attention is LDH-A (lactate dehydrogenase A). The gene encoding this enzyme is a direct c-Myc target where Myc–Max complexes bind the two E-boxes in the LDH-A promoter to transactivate the gene [23,25]. We have recently confirmed this by showing that the expression of the glycolytic gene Ldh1 (encoding LDH-A in mouse) is elevated in B-cells from Eµ-Myc transgenic mice, as were the other glycolytic genes Pkm2 (pyruvate kinase) and Hk2 (hexokinase) (Figure 2). Overexpression of LDH-A in Myc-trans C 2007

Biochemical Society

formed cells is essential since reducing the levels of LDHA by antisense knockdown severely reduces clonogenicity in soft agar assays [25,26]. Notably, Myc-expressing cells grown in a hypoxic environment were even more sensitive to LDH-A knockdown, suggesting that the altered metabolism caused by LDH-A overexpression gives neoplastic growth advantages in a hypoxic environment [25,26]. More recently, a role for LDH-A as an essential tumour maintenance gene was suggested, further supporting the concept of glycolysis inhibition as an attractive anti-tumour strategy [27]. Down-regulation of LDH-A activity severely affects mitochondrial bioenergetics and decreases the proliferative capacity of tumour cells in culture. In addition, cultured tumour cells with an LDH-A knockdown grow remarkably slowly when transplanted into mice and the mice consequently live considerably longer compared with mice transplanted with ‘normal’ or LDH-A complemented tumour cells [27]. Taken together, the compiled data on glycolysis inhibition as a novel means to treat Myc-induced cancer and other cancers make a strong case, but it needs to be strengthened by using spontaneous mouse models of cancer, which we are in the process of doing.

Nucleotide biogenesis As self-evident as increased energy metabolism, an increase in the pools of deoxynucleotides must accompany the high rate of DNA synthesis and proliferation observed in cancer cells. Many of the biosynthetic enzymes involved in the synthesis of deoxyribonucleotides are regulated by transcription factors induced by growth, such as the Myc and E2f


families. We have recently focused on three of the most extensively studied genes with respect to Myc: Shmt1, Shmt2 and Cad (Table 1 and Figure 2). Both the soluble form of Shmt (serine hydroxymethyltransferase), Shmt1, and the mitochondrial form, Shmt2, are direct Myc transcriptional targets [28]. Shmts are involved in nucleotide and amino acid biosynthesis as well as folate metabolism. They were originally identified as Myc targets in an elegant screen for genes that can partially rescue the slow proliferation of HO.15 myc-null Rat1 fibroblasts [28]. They carry E-boxes in their promoters and they are transcriptionally induced in B-cells from Eµ-Myc transgenic mice (Figure 2). Given their role in several metabolic pathways relevant to cancer, Shmt enzymes have previously been proposed as suitable targets for anticancer treatment, but no in vivo or clinical data have to our knowledge been generated. It is noteworthy, however, that another enzyme involved in folate and nucleotide biogenesis, the E2F and Myc transcriptional target dihydrofolate reductase [29], has long been a target for chemotherapy by methotrexate so targeting Shmt1 may also be fruitful. Another Myc target involved in nucleotide biogenesis is Cad (carbamoyl phosphate synthetase 2/aspartate transcarbamylase/dihydro-orotase), the rate-limiting enzyme of pyrimidine biosynthesis [30]. Cad (as well as Odc) has been used as a prototype gene in determining how Myc selects and transactivates target genes [31–34]. Although these studies have been informative, they have also shed light on the fact that not all Myc target genes are activated in an identical way (Figure 1). The biological importance of Cad in Mycinduced cancers has yet to be determined, but the reverse has been analysed. By using HO.15 myc-null cells, it has been shown that Cad was the only gene, out of the few identified at the time, which was completely dependent on Myc for its induction by serum starvation [35]. Given the strong correlation between Myc expression and Cad expression in rat and mouse cells (Figure 2 and Table 1), it is therefore somewhat surprising that one publication states that Burkitt’s lymphoma cell lines do not overexpress Cad, as determined by an RNase protection assay [36]. We recently repeated this experiment with real-time RT (reverse transcription)– PCR and were able to show that Cad is induced by Myc also in Burkitt’s lymphoma cells (S. Rimpi and J.A. Nilsson, unpublished work). We are currently in the process of analysing the role of Cad in Myc-induced lymphomagenesis.

Future perspectives The strength of using mouse models of cancer is that the importance of a gene can be determined both for the induction of tumorigenesis and for the maintenance of tumours. This gives a clear indication of whether or not a protein should be targeted as a means to prevent or treat cancer. The field of chemoprevention is discussed at length in other minireviews in this issue of Biochemical Society Transactions, but not in the context of lymphomas. Admittedly, it is difficult to know whether or not an individual is at risk of develop-

ing lymphomas as opposed to, for example, colon cancer. However, chemoprevention may actually be utilized in leukaemia and lymphoma, since it may moderate the frequency of relapse in response to DNA-damaging drugs. Relapse is often due to the occurrence of additional mutations which result in the reappearance of malignant clones resistant to the initial treatment. When targeting the metabolic enzyme Odc, the secondary ‘hit’ required for Myc-induced lymphomagenesis was severely hampered. If it is possible to prevent the secondary mutations required for relapse to occur by targeting metabolic enzymes using chemoprevention, an increased use of chemoprevention in the clinic is foreseeable. Another group of individuals at risk of developing lymphoma is HIV/AIDS patients. These patients often develop highgrade B-cell lymphomas such as Burkitt’s lymphoma and DLBCL (diffuse large B-cell lymphoma) [37]. Since both of these tumours often carry Myc translocations, they should be sensitive to the chemopreventative effect of drugs such as DFMO or newer drugs targeting metabolic enzymes that are essential for tumour development. To summarize, we propose that Myc target genes encoding metabolic enzymes can no longer stand in the shadows of cell cycle genes; they need to be analysed. The results of analyses like these are likely to provide the starting point for the development of new drugs that will serve to treat or prevent Myc-induced cancers. We thank the following agencies for financial support: The Swedish Cancer Society, The Collected Charities of the Faculty of Medicine at Umeå University, The Kempe Foundation (Umeå University), The Lions/Norrlands Cancer Foundation and The Faculty of Science and Technology at Umeå University (Biotechnology grant). J.A.N. would like to take the opportunity to thank Dr John L. Cleveland for 31/2 productive postdoctoral years in his laboratory.

References 1 Nilsson, J.A. and Cleveland, J.L. (2003) Oncogene 22, 9007–9021 2 Sears, R.C. (2004) Cell Cycle 3, 1133–1137 3 Askew, D.S., Ashmun, R.A., Simmons, B.C. and Cleveland, J.L. (1991) Oncogene 6, 1915–1922 4 Evan, G.I., Wyllie, A.H., Gilbert, C.S., Littlewood, T.D., Land, H., Brooks, M., Waters, C.M., Penn, L.Z. and Hancock, D.C. (1992) Cell 69, 119–128 5 Adams, J.M., Harris, A.W., Pinkert, C.A., Corcoran, L.M., Alexander, W.S., Cory, S., Palmiter, R.D. and Brinster, R.L. (1985) Nature 318, 533–538 6 Kovalchuk, A.L., Qi, C.F., Torrey, T.A., Taddesse-Heath, L., Feigenbaum, L., Park, S.S., Gerbitz, A., Klobeck, G., Hoertnagel, K., Polack, A. et al. (2000) J. Exp. Med. 192, 1183–1190 7 Zhu, D., Qi, C.F., Morse, III, H.C., Janz, S. and Stevenson, F.K. (2005) Blood 105, 2135–2137 8 Eisenman, R.N. (2000) Harvey Lect. 96, 1–32 9 Nilsson, J.A. and Cleveland, J.L. (2004) Cell Cycle 3, 588–590 10 Zeller, K.I., Jegga, A.G., Aronow, B.J., O’Donnell, K.A. and Dang, C.V. (2003) Genome Biol. 4, R69 11 Dang, C.V., O’Donnell, K.A., Zeller, K.I., Nguyen, T., Osthus, R.C. and Li, F. (2006) Semin. Cancer Biol. 16, 253–264 12 Patel, J.H., Loboda, A.P., Showe, M.K., Showe, L.C. and McMahon, S.B. (2004) Nat. Rev. Cancer 4, 562–568 13 Bello-Fernandez, C., Packham, G. and Cleveland, J.L. (1993) Proc. Natl. Acad. Sci. U.S.A. 90, 7804–7808 14 Wagner, A.J., Meyers, C., Laimins, L.A. and Hay, N. (1993) Cell Growth Differ. 4, 879–883 C 2007

Biochemical Society

309

310


15 Barrett, J.F., Lee, L.A. and Dang, C.V. (2005) BMC Biochem. 6, 7 16 Nilsson, J.A. and Cleveland, J.L. (2006) in Polyamine Cell Signaling: Physiology, Pharmacology and Cancer Research (Wang, J.-Y. and Casero, R.J., eds.), pp. 249–266, Humana Press Inc., Totowa 17 Gerner, E.W. and Meyskens, Jr, F.L. (2004) Nat. Rev. Cancer 4, 781–792 18 Persson, L., Holm, I., Ask, A. and Heby, O. (1988) Cancer Res. 48, 4807–4811 19 Nilsson, J.A., Keller, U., Baudino, T.A., Yang, C., Norton, S., Old, J.A., Nilsson, L.M., Neale, G., Kramer, D.L., Porter, C.W. and Cleveland, J.L. (2005) Cancer Cell 7, 433–444 20 Wallace, H.M. and Fraser, A.V. (2004) Amino Acids 26, 353–365 21 Warburg, O. (1956) Science 123, 309–314 22 Warburg, O., Posener, K. and Negelein, E. (1924) Biochem. Z. 152, 309–344 23 Kim, J.W., Zeller, K.I., Wang, Y., Jegga, A.G., Aronow, B.J., O’Donnell, K.A. and Dang, C.V. (2004) Mol. Cell. Biol. 24, 5923–5936 24 Pelicano, H., Martin, D.S., Xu, R.H. and Huang, P. (2006) Oncogene 25, 4633–4646 25 Shim, H., Dolde, C., Lewis, B.C., Wu, C.S., Dang, G., Jungmann, R.A., Dalla-Favera, R. and Dang, C.V. (1997) Proc. Natl. Acad. Sci. U.S.A. 94, 6658–6663 26 Shim, H., Chun, Y.S., Lewis, B.C. and Dang, C.V. (1998) Proc. Natl. Acad. Sci. U.S.A. 95, 1511–1516 27 Fantin, V.R., St-Pierre, J. and Leder, P. (2006) Cancer Cell 9, 425–434 28 Nikiforov, M.A., Chandriani, S., O’Connell, B., Petrenko, O., Kotenko, I., Beavis, A., Sedivy, J.M. and Cole, M.D. (2002) Mol. Cell. Biol. 22, 5793–5800 29 Mai, S. and Jalava, A. (1994) Nucleic Acids Res. 22, 2264–2273 30 Miltenberger, R.J., Sukow, K.A. and Farnham, P.J. (1995) Mol. Cell. Biol. 15, 2527–2535 31 Eberhardy, S.R. and Farnham, P.J. (2001) J. Biol. Chem. 276, 48562–48571 32 Boyd, K.E. and Farnham, P.J. (1999) Mol. Cell. Biol. 19, 8393–8399 33 Boyd, K.E., Wells, J., Gutman, J., Bartley, S.M. and Farnham, P.J. (1998) Proc. Natl. Acad. Sci. U.S.A. 95, 13887–13892 34 Boyd, K.E. and Farnham, P.J. (1997) Mol. Cell. Biol. 17, 2529–2537 35 Bush, A., Mateyak, M., Dugan, K., Obaya, A., Adachi, S., Sedivy, J. and Cole, M. (1998) Genes Dev. 12, 3797–3802 36 Mac, S.M. and Farnham, P.J. (2000) Mol. Carcinog. 27, 84–96 37 Carbone, A. and Gloghini, A. (2005) Br. J. Haematol. 130, 662–670 38 Blackwell, T.K., Kretzner, L., Blackwood, E.M., Eisenman, R.N. and Weintraub, H. (1990) Science 250, 1149–1151 39 Blackwood, E.M. and Eisenman, R.N. (1991) Science 251, 1211–1217 40 Blackwood, E.M., Luscher, B. and Eisenman, R.N. (1992) Genes Dev. 6, 71–80 41 Eisenman, R.N. (2001) Genes Dev. 15, 2023–2030 42 McMahon, S.B., Van Buskirk, H.A., Dugan, K.A., Copeland, T.D. and Cole, M.D. (1998) Cell 94, 363–374 43 McMahon, S.B., Wood, M.A. and Cole, M.D. (2000) Mol. Cell. Biol. 20, 556–562 44 Park, J., Kunjibettu, S., McMahon, S.B. and Cole, M.D. (2001) Genes Dev. 15, 1619–1624 45 Park, J., Wood, M.A. and Cole, M.D. (2002) Mol. Cell. Biol. 22, 1307–1316

C 2007

Biochemical Society

46 Eberhardy, S.R. and Farnham, P.J. (2002) J. Biol. Chem. 277, 40156–40162 47 Ayer, D.E., Lawrence, Q.A. and Eisenman, R.N. (1995) Cell 80, 767–776 48 Heinzel, T., Lavinsky, R.M., Mullen, T.M., Soderstrom, M., Laherty, C.D., Torchia, J., Yang, W.M., Brard, G., Ngo, S.D., Davie, J.R. et al. (1997) Nature 387, 43–48 49 Hurlin, P.J., Queva, C. and Eisenman, R.N. (1997) Genes Dev. 11, 44–58 50 Koskinen, P.J., Ayer, D.E. and Eisenman, R.N. (1995) Cell Growth Differ. 6, 623–629 51 Walker, W., Zhou, Z.Q., Ota, S., Wynshaw-Boris, A. and Hurlin, P.J. (2005) J. Cell Biol. 169, 405–413 52 Toyo-oka, K., Hirotsune, S., Gambello, M.J., Zhou, Z.Q., Olson, L., Rosenfeld, M.G., Eisenman, R., Hurlin, P. and Wynshaw-Boris, A. (2004) Hum. Mol. Genet. 13, 1057–1067 53 Nilsson, J.A., Maclean, K.H., Keller, U.B., Pendeville, H., Baudino, T.A. and Cleveland, J.L. (2004) Mol. Cell. Biol. 24, 1560–1569 54 Hurlin, P.J., Zhou, Z.Q., Toyo-Oka, K., Ota, S., Walker, W.L., Hirotsune, S. and Wynshaw-Boris, A. (2004) Cell Cycle 3, 97–99 55 Hurlin, P.J., Zhou, Z.Q., Toyo-Oka, K., Ota, S., Walker, W.L., Hirotsune, S. and Wynshaw-Boris, A. (2003) EMBO J. 22, 4584–4596 56 Wu, S., Cetinkaya, C., Munoz-Alonso, M.J., von der Lehr, N., Bahram, F., Beuger, V., Eilers, M., Leon, J. and Larsson, L.G. (2003) Oncogene 22, 351–360 57 Seoane, J., Le, H.V. and Massague, J. (2002) Nature 419, 729–734 58 Staller, P., Peukert, K., Kiermaier, A., Seoane, J., Lukas, J., Karsunky, H., Moroy, T., Bartek, J., Massague, J., Hanel, F. and Eilers, M. (2001) Nat. Cell Biol. 3, 392–399 59 Seoane, J., Pouponnot, C., Staller, P., Schader, M., Eilers, M. and Massague, J. (2001) Nat. Cell Biol. 3, 400–408 60 Fernandez, P.C., Frank, S.R., Wang, L., Schroeder, M., Liu, S., Greene, J., Cocito, A. and Amati, B. (2003) Genes Dev. 17, 1115–1129 61 Li, Z., Van Calcar, S., Qu, C., Cavenee, W.K., Zhang, M.Q. and Ren, B. (2003) Proc. Natl. Acad. Sci. U.S.A. 100, 8164–8169 62 O’Connell, B.C., Cheung, A.F., Simkevich, C.P., Tam, W., Ren, X., Mateyak, M.K. and Sedivy, J.M. (2003) J. Biol. Chem. 278, 12563–12573 63 Osthus, R.C., Shim, H., Kim, S., Li, Q., Reddy, R., Mukherjee, M., Xu, Y., Wonsey, D., Lee, L.A. and Dang, C.V. (2000) J. Biol. Chem. 275, 21797–21800 64 Coller, H.A., Grandori, C., Tamayo, P., Colbert, T., Lander, E.S., Eisenman, R.N. and Golub, T.R. (2000) Proc. Natl. Acad. Sci. U.S.A. 97, 3260–3265 65 Menssen, A. and Hermeking, H. (2002) Proc. Natl. Acad. Sci. U.S.A. 99, 6274–6279 66 Guo, Q.M., Malek, R.L., Kim, S., Chiao, C., He, M., Ruffy, M., Sanka, K., Lee, N.H., Dang, C.V. and Liu, E.T. (2000) Cancer Res. 60, 5922–5928 67 Schuhmacher, M., Kohlhuber, F., Holzel, M., Kaiser, C., Burtscher, H., Jarsch, M., Bornkamm, G.W., Laux, G., Polack, A., Weidle, U.H. and Eick, D. (2001) Nucleic Acids Res. 29, 397–406 68 Pusch, O., Soucek, T., Hengstschlager-Ottnad, E., Bernaschek, G. and Hengstschlager, M. (1997) DNA Cell Biol. 16, 737–747

Received 29 September 2006