Notching Up MYC Gives a LIC

The  leukemogenic  activities   of  cancer- specific products, such  as translocation- generated   fusion   proteins  or   proteins that  are constitutively activated or inacti- vated 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 increase the number of leukemia- initiating cells (LICs) (Figure 1).

The transmembrane receptor Notch1 is an important regulator of T cell develop- ment (Aster et al., 2011). Following recep- tor  binding  by  one   of  its  ligands, the intracellular  domain  of Notch1  is proteo- lytically  processed,  released,  and   traf- ficked  to  the  nucleus, where  it interacts with DNA-bound Rbpj and helps to recruit Mastermind-like  coactivators  to  upregu- late    targets    involved    in    metabolism, proliferation,  cell  survival,  chromatin re- modeling,  and   transcription.  In  T  cells one   important  downstream  effector   of Notch1  action  is  MYC  (Palomero  et  al.,2006). In addition  to numerous balanced, nonrandom  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  mecha- nism  for  how  mutations in Notch1   and FBXW7   promote   leukemia.  They   find that   stabilization  of  MYC   protein   in  a Notch1-activating  background increases the  number  of  LICs,  thus  showing that FBXW7 behaves as  a  tumor  suppressor in T-ALL. The authors show that stabiliza- tion   of  MYC   by   mutation  of   FBXW7 leads to impaired T cell differentiation, a dramatically shortened latency period  for leukemogenesis, and  the eventual devel- opment of a more aggressive T-ALL than that  which occurs through the agency of Notch1   alone.   Because  removal  of  the ubiquitin-targeted              destabilizing             PEST sequence  in Notch1  did  not  bypass the carcinogenic effects of FBXW7   muta- tions,  the  authors conclude  that  FBXW7 operates more  upon  MYC than  Notch1 to  increase the  LIC population and  pro- voke disease.

The     normal     MYC    protooncogene may  become  oncogenic  when   overex- pressed. Besides verifying the previously described transcriptional upregulation of MYC by  activated  Notch1,  the  current study  (King et al., 2013) demonstrates, in T-ALL with FBXW7 mutations, MYC pro- tein    stabilization  due    to    ineffective FBXW7-mediated  degradation. Although in principle, high levels of MYC might drive the pathologic expression  of nonphysio- logical    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 spec- ifier, but an amplifier that controls  the flux of materials and information through sub-  cellular 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 fluctu- ating  levels  of  MYC traffic  between the stem-like LIC state and circulating  T-lym- phoblasts. By tagging  GFP-MYC knockin cells using retroviral insertions, the in vivo fates  of individual leukemia  clones could be tracked. Mice transplanted with genet- ically   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  en- gages  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  transcrip- tion    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 colocalize at many if   not   most  sites  in T-ALL cells,  as expected from this sort of coherent feed- forward    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

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 levels of MYC (saturated red), but throw off (green arrow) bulk leukemia  cells expressing lower levels of MYC (pale  red).  Wild-type  FBXW7 depresses 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.  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 biolog- ical roles. Besides context-dependent  degrada- tion,   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 (AR) 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   con- founds    its   simple   understanding  may also  provide avenues for safer  and  more effective  tumor-specific therapy. underestimates the  genome-wide  extent and  degree of Notch1-MYC cooperation. Based   on    the    demonstration   that FBXW7 mutants stabilize MYC, augment- ing the number and  activity of LICs, King et al. attack the high, oncogenic levels of MYC by targeting its synthesis at the tran- scriptional level using the BET-Brd4 inhib- itor JQ1 and  derivatives thereof  (Delmore et al., 2011). MYC transcription is exqui- sitely  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  influ- ence.  Most  likely  MYC transcription  is driven  by  different  pathways in different tissues and  tumors. In the T-ALL studied by   King   et    al.,   MYC,    Notch1,   and Brd4  extensively  colocalize across  the genome, but especially at enhancers. At pathologic   levels,    MYC    invades    en- hancers (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  syn- thesis, but  likely also  reduce  the  activity of  MYC-associated  enhancers by  both decreasing the amount of MYC bound to the enhancer and  preventing Brd4 action  at these same sites.

The  systems  degrading or  stabilizing MYC may be cell type dependent accord- ing to the particular subsets of substrate specificities  for  the   individual  E3-com- plexes. For example, while FBXW7 coor-  dinates  Notch1  and   MYC degradation, other   systems  may   operate  elsewhere and  contribute to tissue-specific carcino- genesis. MULE is an E3-complex associ- ated 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  re- ported  to  target  Notch1   (Inoue  et  al., 2013).  MIZ1 is a  zinc-finger  protein  that binds  to MYC at some of its targets, pre- venting their activation. In skin, MIZ1 pre- vents MYC amplification of CDK inhibitors (CDKN) p15 and p21, allowing unopposed proliferation.   The  utilization  of  different E3-specificity  modules  (i.e.,    FBXW7 versus MULE) to degrade MYC may pro- vide   its   coordinate   degradation    with Aster,  J.C.,  Blacklow, S.C., and Pear, W.S. (2011). J. Pathol. 223, 263–274.

Delmore, J.E., Issa, G.C., Lemieux,  M.E.,  Rahl, P.B., Shi, J., Jacobs, H.M., Kastritis, E., Gilpatrick, T.,  Paranal, R.M.,  Qi, J.,  et  al.  (2011).  Cell 146,

904–917.

Inoue, S., Hao, Z., Elia, A.J., Cescon, D., Zhou, L., Silvester,  J., Snow,  B., Harris, I.S., Sasaki, M., Li, W.Y., et al. (2013). Genes Dev. 27, 1101–1114.

King,   B.,  Trimarchi,  T.,  Reavie,  L.,  Xu, L., Mul- lenders, J., Ntziachristos, P., Aranda-Orgilles, B., Perez-Garcia, A., Shi, J., Vakoc, C., et al. (2013). Cell 153, 1552–1566.

Lin, C.Y., Loven, J., Rahl, P.B.,  Paranal, R.M., Burge,  C.B.,  Bradner, J.E.,  Lee,  T.I., and  Young, R.A. (2012). Cell 151, 56–67.

Loven,  J., Hoke, H.A., Lin, C.Y., Lau, A., Orlando, D.A., Vakoc,  C.R.,  Bradner, J.E.,  Lee,  T.I., and Young, R.A. (2013). Cell 153, 320–334.

Nie, Z., Hu, G., Wei, G., Cui, K., Yamane, A., Resch, W., Wang, R., Green, D.R., Tessarollo, L., Casellas, R., et al. (2012). Cell 151, 68–79.

Palomero,  T., Lim, W.K., Odom,  D.T., Sulis, M.L., Real,  P.J., Margolin,  A.,  Barnes, K.C., O’Neil, J., Neuberg, D., Weng,  A.P., et al. (2006). Proc.  Natl. Acad. Sci. USA 103, 18261–18266.

Shachaf,  C.M.,  Gentles, A.J.,  Elchuri,  S.,  Sahoo, D., Soen, Y., Sharpe, O., Perez, O.D., Chang, M., Mitchel, D., Robinson, W.H., et al. (2008). Cancer Res.  68, 5132–5142.

Soufi, A., Donahue, G., and Zaret, K.S. (2012). Cell