Notching up MYC gives a LIC

Abstract

King et al. (Cell, 2013) show that mutations in the ubiquitin ligase subcomponent FBXW7 increase MYC protein levels and expand the number of leukemia initiating cells (LIC) in Notch1-induced T-ALL. Genetic and pharmacological inhibition of Myc abolishes LIC activity with broader therapeutic implications.

PREVIEW

The leukemogenic activities of cancer-specific products, such as translocation-generated fusion proteins or proteins that are constitutively activated or inactivated by amino acid substitutions, are in principle easy to understand even if their ramifications are complicated to enumerate. Less easy to understand is how quantitative changes, sometimes subtle, in the levels of normal proteins provoke or sustain pathologic activities during tumorigenesis and progression. A new study that manipulates Notch1, FBXW7 (the substrate recognition module of an E3-ubiquitin ligase complex that includes both Notch1 and MYC proteins among its clients) and MYC in mice reveals how qualitative changes in protein sequence that inactivate FBXW7, play out to increase MYC levels in tumor cells that in turn increases the number of Leukemia Initiating Cells (LICs).

The transmembrane receptor Notch1 is an important regulator of T cell development (Aster et al. 2011). Following receptor binding by one of its ligands, the intracellular domain of Notch1 is proteolytically processed, released and then trafficked to the nucleus where it interacts with DNA-bound Rbpj and helps to recruit Mastermind-like co-activators to up-regulate targets involved in metabolism, proliferation, cell survival, chromatin remodeling and transcription. In T cells one important downstream effector of Notch1 action is MYC (Palomero et al. 2006). In addition to numerous balanced, non-random chromosomal translocations that occur in T-cell acute lymphoblastic leukemias, ~60% of cases have single nucleotide substitutions or indels that lead to aberrant Notch1 activation. King et al. investigate the underlying mechanism for how mutations in Notch1 and FBXW7 promote leukemia They find that stabilization of MYC levels in a Notch 1 activating background increases the number of LICs thus showing that FBXW7 behaves as a tumor suppressor in T-ALL. The authors show that stabilization of MYC by mutation of FBXW7 leads to impaired T-cell differentiation, a dramatically shortened latency period for leukemogenesis, and the eventual development of a more aggressive T-ALL than occurs through the agency of Notch1 alone. Because removal of the ubiquitin-targeted destabilizing PEST sequence in Notch 1 did not bypass the carcinogenic effects of FBXW7 mutations, the authors conclude that FBXW7 operates more upon MYC than Notch1 to increase the LIC population and provoke disease.

The normal MYC protooncogene may become oncogenic when overexpressed. In T-ALL with FBXW7 mutations, besides verifying the previously described transcriptional upregulation of MYC by activated Notch1, the current study (King et al., 2013) demonstrates MYC protein stabilization due to ineffective FBXW7-mediated degradation. Although in principle, high levels of MYC might drive the pathologic expression of non-physiological targets, more recent studies indicate that MYC amplifies ongoing expression within a cell, but does not directly alter the transcriptional program of the cell (Lin et al. 2012; Nie et al. 2012). In other words, MYC is not a specifier, but an amplifier that controls the flux of materials and information through subcellular networks. For pathways such as proliferation and apoptosis, MYC may push some targets across thresholds (Shachaf et al. 2008) that discriminate between bulk leukemia cells and LICs. Crossing these thresholds may not necessarily be indicative of relentless and irreversible progression toward more aggressive disease, rather it may reveal the existence of a stochastic bistable switch in which cells with fluctuating levels of MYC traffic between the stem-like LIC state and circulating T-lymphoblasts. By tagging GFP-MYC knock-in cells using retroviral insertions, the in vivo fates of individual leukemia clones could be tracked. Mice transplanted with genetically tagged purified high-MYC LICs throw off populations of less aggressive, but similarly tagged, low-MYC cells. Tracking GFP-MYC in vivo allows the visualization of heterogeneity in MYC expression in vivo; such studies demand that a fully functional MYC fusion protein be expressed from its endogenous locus to insure proper regulation and activity(Nie et al. 2012)..

If MYC is an amplifier, then Notch1 is a specifier and a pioneer factor that engages silent genes to reconfigure their chromatin, and turn them on. During T-cell development, Notch1 also directly binds and upregulates the otherwise weakly expressed MYC. The logic of this interaction is simple; by targeting MYC expression, Notch1 and other transcription factors and signaling pathways, increase their own effectiveness as the newly synthesized MYC cooperates with the factor/pathway across the genome. Using ChIP-Seq, the current work shows that Notch1 and MYC co-localize at many if not most sites in T-ALL cells as expected from this sort of coherent feedforward circuit. The statistically high stringencies employed in the Chip-Seq analyses used in this study and many others, rigorously excludes false positive peaks, but almost certainly considerably underestimates the genome-wide extent and degree of Notch1-MYC cooperation.

Based on the demonstration that FBXW7 mutants stabilize MYC, augmenting the number and activity of LICs, King et al, attack the high, oncogenic levels of MYC by targeting its synthesis at the transcriptional level using the BET-Brd4 inhibitor JQ1 and derivatives thereof (Delmore et al. 2011). MYC transcription is exquisitely Brd4 dependent in some tumors, especially those of hematopoietic origin (and most particularly in leukemias and lymphomas where MYC expression is driven by translocation-juxtaposed superhancers”) (Loven et al. 2013), yet in other cancers, especially solid tumors, BET inhibitors are largely without influence. Most likely MYC transcription is driven by different pathways in different tissues and tumors. In the T-ALL studied by King, et al, across the genome MYC, Notch1 and Brd4 extensively co-localize, but especially at enhancers. At pathologic levels, MYC invades enhancers (Loven et al. 2013) and most likely cooperates with Brd4, Notch1, and other enhancer-associated activators. Therefore, attacking MYC via BET inhibitors would not only depress MYC synthesis, but likely also reduce the activity of MYC-associated enhancers by both decreasing the amount of MYC bound to the enhancer and by preventing Brd4 action at these same sites.

The systems degrading or stabilizing MYC may be cell type dependent according to the particular subsets of substrate specificities for the individual E3-complexes. For example, while FBXW7 coordinates Notch1 and MYC degradation, other systems may operate elsewhere and contribute to tissue specific carcinogenesis. MULE is an E3-complex associated with cutaneous tumors (Inoue et al. 2013). MULE targets MYC for degradation in keratinocytes, and to a lesser extent degrades MIZ1, but has not been reported to target Notch1 (Inoue et al. 2013). MIZ1 is a zinc-finger protein that binds to MYC at some of its targets preventing their activation. In skin, MIZ1 prevents MYC amplification of CDK inhibitors (CDKN) p15 and p21 allowing unopposed proliferation. The utilization of different E3-specificty modules (i.e. FBXW7versus Mule) to degrade MYC may provide for its coordinate degradation with different partners in different tissues. Interestingly, activated Notch1 that is pro-proliferative and carcinogenic in T-cells, is a tumor suppressor in skin, suggesting that uncoupling MYC from Notch1 may contribute to the definition of their biological roles.

Besides context dependent degradation, the context specific utilization of transcription factors may exploit MYC to impel developmental(Soufi et al. 2012) or carcinogenic feedforward amplification of gene expression as Notch1 does in T-ALL. For example, androgen receptor activates MYC expression and MYC in turn joins AR at its targets to amplify hormone response in molecular apocrine breast cancer. Exposing the pathways that differentially upregulate or stabilize MYC in distinct pathologic situations may highlight targets beyond Brd4 for pharmacologic intervention. Therefore, it seems that the very complexity of MYC synthesis and degradation that confound its simple understanding, may also provide avenues for safer and more effective tumor specific therapy.

Figure 1. FBXW7 via MYC amplifies Notch1-induced Leukemogenesis by increasing MYC protein levels.

Left—Activated Notch1 after a latent period yields T-ALL. Leukemia initiating cells (LICs) require high level of MYC (saturated red), but throw off (green arrow) bulk leukemia cells expressing lower MYC(pale red). Wildtype FBXW7 depressess MYC levels. We speculate that stochastic elevation of MYC may help promote a bulk leukemia cell to a LIC (pale red arrow). Right—Mutated FBXW7 increases MYC, and increases the frequency and numbers of LICs.