Amino acid dependence in acute myeloid leukaemia

Abstract

Acute myeloid leukaemia (AML) is a heterogeneous disease which remains curable in only a minority of patients. Outcome is dismal in the subgroup with adverse genetic features. Amino acid (AA) depletion is a therapeutic strategy that is used to deprive malignant cells of nutrients. This thesis explores the dependence of AML on specific AAs. I show that primary AML cells need valine, methionine and a combination of serine and glycine to survive in culture and that this varies across AML risk groups. Depletion of these causes significantly increased apoptosis when compared to normal bone marrow cells although the combination of cytarabine chemotherapy was no more effective. This dependence was confirmed was confirmed using mass spectrometry which demonstrated increased consumption of these AAs in media following co-culture. Intracellular AA levels were difficult to measure consistently due to low cell numbers. Further analysis suggests that adverse risk AML consumes AAs more avidly than intermediate risk AML. The enzymes of the methionine salvage pathway are strongly expressed in AML consistent with the need for methionine. Further work is needed to elucidate the role of the GCN2/eIF2a/ATF4 integrated stress response pathway. Analysis of the AML proteome in response to valine and methionine depletion reveals that only a small number of proteins change significantly suggesting that the effect is generally at a metabolomic level. A significant increase in PHGDN, a downstream enzyme in the ISR pathway, was however seen in response to AA depletion. These data show that AA deprivation is effective in killing primary AML cells in vitro. This needs to be tested in combination with other chemotherapy drugs. Potential clinical applications include combining this with conditioning chemotherapy to reduce the AML burden prior to allogeneic stem cell transplantation.