Subtractive Hybridization Identifies Stem Cell-Associated Genes in an Acute Myeloid Leukemia with Poor Prognosis

ABSTRACT

Introduction: Current prognostic markers have improved survival prediction, however, it has not advanced treatment strategies. Gene expression profiling may identify biological markers suitable as therapeutic targets. Leukaemia stem cell is associated with adverse outcome, however, its biological characteristics are still being investigated. We observed higher in vitro cell viability in acute myeloid leukaemia (AML) samples with poor prognosis, which may be stem cell related. Objective: The objective of this study was to profile highly expressed genes in an AML sample of poor prognosis/high viability and compare with a sample of good prognosis/low viability. Method: Subtractive hybridization was performed on two AML samples with high blast counts (>80%), a poor prognosis, PP (disease free survival, DFS12 months) sample. The PP sample had higher CD34+ counts (73% vs 46%) and higher cell viability than the GP sample. cDNA libraries were subsequently cloned and sequenced. Results: cDNA subtracted from the PP samples was identified as genes active during fetal/embryonic development (LCOR, CNOT1, ORMDL1), HOX- related genes (HOXA3, PBX3, SF3B1), hematopoiesis (SELL, IL-3RA) and aerobic glycolysis/hypoxia (PGK1, HIGD1A) -associated genes. Majority of GP clones isolated contained genes involved in oxidative phosphorylation, OXPHOS (COXs, ATPs, MTND4 and MTRNR2), protein synthesis (including ribosomal proteins, initiating and elongation factors), chromatin remodeling (H2AFZ, PTMA), cell motility (MALAT1, CALM2, TMSB4X), and mitochondria (HSPA9, MPO) genes. Conclusion: Thus, the PP sample exhibited stem cell-like features while the GP sample showed cells at a high level of cell activity. These genes are potential prognostic markers and targets for therapy.