The functional relationship between epigenetics and metabolism in cancer progression has not been carefully examined. Many epigenetic enzymes catalyzing DNA or histone modifications are susceptible to changes in co-substrates of metabolism, but little is known about whether and how altered epigenetics influences cellular metabolism during cancer progression. Mutations of EZH2, the histone H3 lysine 27 methyltransferase, are frequently found in myeloid malignancies and correlate with poor prognosis. We recently developed a new mouse model of myeloid neoplasms by inactivation of EZH2 and activation of oncogenic NRas (G12D). While G12D alone led to an indolent myeloproliferation, EZH2 inactivation markedly accelerated disease progression resulting in myelofibrosis, leukemic transformation and mortality. With this model that faithfully recapitulates leukemia progression, we unexpectedly identified branched-chain amino acid (BCAA) metabolism as the most significantly upregulated metabolic pathway in EZH2-deficient leukemia-initiating cells (LICs). BCAT1, the first enzyme catalyzing BCAA transamination, is repressed by EZH2 in hematopoiesis and aberrantly activated in EZH2-deficient myeloid neoplasms in mice and humans. Increased BCAT1 promotes BCAA production in LICs, resulting in activated mTOR signaling. Genetic and pharmacological inhibition of BCAT1 selectively impairs EZH2-deficient LICs and constitutes a metabolic vulnerability. These findings for the first time connect dysregulation of EZH2 with altered metabolic pathways in cancer progression. The objective of this project is to elucidate the causal mechanisms controlling metabolic liabilities of LICs by focusing on the role of BCAA metabolism in EZH2-deficient myeloid neoplasms. The central hypothesis is that EZH2 deficiency induces metabolic rewiring by activating BCAT1 and BCAA metabolism to create a metabolic dependency for LICs, and that inhibition of BCAT1 will selectively eradicate LICs by disabling the metabolic liability of EZH2-deficient LICs. This hypothesis has been formulated on the basis of our preliminary studies using an in vivo model of leukemia progression and a newly discovered molecular link between altered epigenetics and metabolism. Guided by these preliminary data, this hypothesis will be tested by three specific aims: 1) Define the functional role of BCAT1 in EZH2-deficiency-induced myeloid neoplasms in vivo using BCAT1 knockout and inducible expression mouse models. 2) Determine the mechanisms by which EZH2-deficient LICs exhibit a metabolic dependency on BCAA metabolism. 3) Determine the effects of targeting BCAA metabolism in human AML stem cells using genetic, pharmacological and dietary manipulations. Together these studies will not only validate a selective metabolic liability for EZH2-mutant myeloid neoplasms, but also uncover new pathways that can be exploited to selectively eradicate LICs. Such results are expected to advance our mechanistic understanding of the functional relationship between epigenetics and metabolism in cancer progression, and to guide the design of more effective therapies to target the metabolic liabilities of cancer-initiating cells.
How epigenetic gene regulation and cellular metabolism cooperate to control cancer progression is a fundamentally important question with significant clinical implication. We recently discovered that inactivation of the histone methyltransferase EZH2 promotes the progression of myeloproliferative neoplasms to leukemic transformation by aberrant activation of branched-chain amino acid (BCAA) metabolism in leukemia-initiating cells, establishing a new molecular link between altered epigenetics and metabolism in cancer progression. Elucidating the causal mechanisms that regulate the metabolic liabilities of leukemia-initiating cells will lead to an improved understanding of basic cancer pathophysiology, and guide the design of more effective therapeutics for the treatment of affected cancer patients.