As cancer develops, at least two cell processes are disrupted — cell growth is promoted, and cell death inhibited. It seems that mutated versions of the notorious cancer-promoting protein MYC can accomplish both at once.
The MYC gene is one of the classic cancer-promoting ‘oncogenes’. It is overexpressed in many types of tumour, and the MYC protein it encodes causes a surge in the proliferation of cells. But it has another effect: it enhances programmed cell death, or ‘apoptosis’. So under normal circumstances, the extra cell divisions MYC causes when overexpressed are cancelled out by a rise in cell fatalities. In MYC-associated tumours, however, there is usually a mutation in an ancillary protein that disrupts the apoptosis pathway, releasing the brakes on cell proliferation. In this issue, Hemann et al. (page 807)1 report that MYC does not need to rely on a partner-in-crime to cause tumours — it can itself be mutated in a way that interferes with its apoptotic function.
The authors examined MYC function in Burkitt's lymphoma, a malignancy of immune cells called B cells, which normally produce immunoglobulins in response to an immune attack. The cancer is characterized by gross overexpression of MYC because the position of the MYC gene is swapped with that of the immunoglobulin gene. As this latter gene is often switched on in B cells, and as the ‘translocated’ MYC gene is slotted in next to the regulatory sequences for the immunoglobulin gene (the ‘promoter’), it too is switched on. Translocated MYC genes often harbour specific point mutations that tend to result in altered amino acids in a specific part of the MYC protein (the amino-terminal domain). This clustering of mutations led to the suspicion that they augment the tumorigenic potential of MYC. However, in vitro assays with these mutants failed to show either enhanced cancer-promoting activity or reduced apoptosis2.
To assess the cancer-promoting potential of MYC mutants, the authors used mice that had been irradiated to permit reconstitution of their immune system. They reconstituted these cells from progenitor cells, called haematopoietic stem cells (HSC), that had been manipulated to express either normal (wild-type) or mutated MYC, and then followed the formation of tumours. When HSC expressing wild-type MYC were used, lymphomas developed in a small fraction of the mice only after a long period. By contrast, most of the mice grafted with HSC carrying mutant MYC quickly developed tumours, indicating that the MYC mutants are far more tumorigenic than wild-type MYC.
To identify the mechanism underlying this difference, the authors checked whether the gene mutants had different effects on the apoptotic signalling pathways induced by MYC overexpression. They found that the ARF–p53 pathway, a known target of MYC, was induced equally by wild-type and mutant MYC. However, the apoptosis-promoting protein BIM, which is part of a different pathway, was activated only by wild-type MYC. The mutants' failure to activate BIM seems to contribute to their enhanced tumorigenicity, because both wild-type and mutant MYC were equally oncogenic when HSC lacking BIM were used. This implies that BIM normally constrains the carcinogenic potential of wild-type MYC, consistent with previous observations3.
Activation of BIM by wild-type MYC does not require p53 signalling, as BIM levels were elevated in p53-deficient cells overexpressing wild-type MYC. Therefore, the oncogene activates BIM through an independent route. The authors' model predicts that BIM expression is not induced in Burkitt's lymphoma cells carrying MYC mutations. Indeed, high levels of BIM were found in all seven human Burkitt's lymphoma samples examined that carried a wild-type MYC gene, but in only one of seven Burkitt's lymphoma samples carrying a mutant gene.
Although the MYC mutants seemed to have little effect on p53 itself, they did activate one of its downstream effectors, an inhibitor of cell division called p21. How does this effect on p21 occur, and does it contribute to MYC-driven signals? Wild-type MYC inhibits p21 production by binding to the p21 gene promoter in a complex with the protein Miz-1 and recruiting further repressor proteins4,5. Perhaps the mutations disrupt MYC's interaction with the repressors. Or, as overexpression of mutant MYC also suppresses its own promoter, perhaps there is no wild-type MYC to form complexes with Miz-1, thereby relieving the suppression of p21. Consistent with this notion, overexpression of wild-type MYC reduced p21 levels.
BIM and p21 levels seem to be inversely regulated in this system, but it remains unclear whether wild-type MYC directly activates BIM or does so through p21. In the latter case, an interesting scenario emerges in which p21 acts upstream of BIM6, serving as a switch to determine whether a cell will stop dividing or undergo apoptosis.
Both wild-type and mutant MYC have similar activating effects on three components of the p53-controlled apoptotic pathway — Bax, PUMA and NOXA. However, activation of these apoptosis-promoting factors does not trigger apoptosis if BIM activation is compromised (Fig. 1). Similarly, BIM activation by wild-type MYC does not induce apoptosis if the p53 pathway is disabled. The picture that emerges suggests that impairment of either the p53 or the BIM signalling route is enough to make wild-type MYC as oncogenic as the MYC mutants. The apoptotic signals conveyed by either pathway seem to be similar and additive. In mouse models, loss of one copy of the BIM gene confers strong resistance to apoptosis3, and it will be interesting to learn whether such a loss is sufficient to abolish the difference between wild-type and mutant MYC in inducing lymphomas.
As chemotherapy usually activates the p53 pathway, these observations prompt a comparison of the response to chemotherapy between lymphoma patients carrying a translocated mutant MYC gene and normal p53 and patients carrying a translocated wild-type MYC and mutant p53. For patients with a mutant MYC, one might argue that, despite the reduced apoptosis resulting from suppression of BIM, forceful activation of the p53 pathway might potently induce apoptosis. However, it is possible that p21 would divert the signal towards cell-cycle arrest rather than death6. This comparison would also show whether the resistance to chemotherapy observed in mouse models of Burkitt's lymphoma7 that overexpress MYC and are deficient in p53 faithfully mimics the response of human lymphomas with similar characteristics. If this is the case, it will offer the opportunity to refine the treatment of Burkitt's lymphoma.
Hemann, M. T. et al. Nature 436, 807–811 (2005).
Chang, D. W., Claassen, G. F., Hann, S. R. & Cole, M. D. Mol. Cell. Biol. 20, 4309–4319 (2000).
Egle, A., Harris, A. W., Bouillet, P. & Cory, S. Proc. Natl Acad. Sci. USA 101, 6164–6169 (2004).
Wu, S. et al. Oncogene 22, 351–360 (2003).
Seoane, J., Le, H. V. & Massagué, J. Nature 419, 729–734 (2002).
Collins, N. L. et al. Mol. Cell. Biol. 25, 5282–5291 (2005).
Schmitt, C. A. et al. Cell 109, 335–346 (2002).
About this article
Posttranscriptional deregulation of MYC via PTEN constitutes a major alternative pathway of MYC activation in T-cell acute lymphoblastic leukemia
Trends in Molecular Medicine (2010)
Constitutive Phosphorylation Mutation in Fas-associated Death Domain (FADD) Results in Early Cell Cycle Defects
Journal of Biological Chemistry (2007)