Recent investigations into the relationship between cancer and cellular metabolism have revealed the strong dependency of different cancers on a diverse array of metabolic pathways. For instance, a variety of cancers depend on the PI3K/AKT pathway for a wide range of glucose-related mechanisms, including AKT-mediated membrane translocation of glucose transporters, activation of the glycolytic enzymes hexokinase and phosphofructokinase, and up-regulation of de novo fatty acid synthesis. A major area of therapeutic and biological interest in the study of cancer metabolism is the process by which these metabolic pathways become deregulated in the first place en route to, or as a result of, the development of cancer. The deregulation of metabolic pathways can occur through alterations in the cellular landscape brought about by mutations in metabolic enzymes (e.g., IDH1/IDH2 and SDH), aberrant expression of transcription factors, such as the proto-oncogene MYC, or the loss of tumor suppressors, such as p53.

Through gene expression and metabolic profiling analyses, we found that the transcription factor EVI-1, whose overexpression in acute and chronic myeloid leukemia (AML and CML) is correlated with poor patient outcome, induced key metabolic perturbations in hematopoietic progenitor cells. These result in a decrease of the mitochondrial oxygen consumption rate, a blockade of the de novo purine and pyrimidine synthesis, and an increase in glycolysis. Using a library of pooled shRNAs targeting genes involved in each of these pathways, we then established a direct link between EVI-1 expression and the development of a heretofore undescribed cellular dependency on the overexpression of the ATP-buffering mitochondrial creatine kinase protein CKMT1B. We showed that EVI-1 directly promotes CKMT1B expression through repression of the master regulator of myeloid differentiation RUNX1. Alteration of the CKMT1B-dependent pathway, either with shRNA or with the small molecule cyclocreatine, impairs production of intracellular phospho-creatine, which in turn alters cell viability specifically in EVI-1-positive (n = 8) versus EVI-1-negative AML cell lines (n = 8), and in EVI-1-positive (n = 5) versus EVI-1-negative (n = 17) primary AML blasts. This decrease in cell viability is associated with the activation of an erythroid differentiation program concomitant with a downregulation of the immature lineage marker c-KIT, both induced by GSK3A/B inhibition. Overexpression of a constitutively activated form of GSK3B (S9A) or GSK3A (S21A) impairs the cell differentiation induced by CKMT1B inhibition. Finally, suppression of CKMT1B alleviates leukemic burden in vivo in two AML mouse models: i) an orthotopic model of transplanted human EVI1-positive AML cells and ii) a syngeneic model of transplanted murine NrasG12D + Evi1 AML cells.

This new interplay between EVI-1 and the creatine pathway uncovers CKMT1B as a new target of interest in EVI-1-positive AML, a high-risk subtype of AML for which current treatment regimens remain inadequate.


DeAngelo:Novartis: Consultancy; Ariad: Consultancy; Bristol Myers Squibb: Consultancy; Pfizer: Consultancy; Amgen: Consultancy; Incyte: Consultancy; Agios: Consultancy; Celgene: Consultancy. Stone:Celgene: Consultancy; Agios: Consultancy; Karyopharm: Consultancy; Sunesis: Consultancy, Other: DSMB for clinical trial; Novartis: Research Funding; Celator: Consultancy; Merck: Consultancy; Roche/Genetech: Consultancy; Abbvie: Consultancy; Amgen: Consultancy; Pfizer: Consultancy; AROG: Consultancy; Juno: Consultancy. Stegmaier:Novartis Pharmaceuticals: Consultancy.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution