Molecular classification of acute myeloid leukemia (AML) has identified several candidate genes that could potentially define prognosis and response to therapy. One such candidate, identified from microarray studies, is the Class I homeobox gene HOXA9. The HOX gene network encodes master regulators of developmental processes including hemopoiesis. To quantify the contribution of this network of genes in AML, we carried out specific RQ-PCR analysis on twenty-four de novo patient samples using a subset of genes (12 HOX and MEIS1) selected on the basis of their recently reported expression in AML. HOXA6 was ranked, as the most highly expressed gene (range 1 x 103 – 2 x 107 copies per 50 ng RNA), substantially higher than HOXA9 (see Table). Further analysis identified high expression of HOXA6 in both human myeloid cell lines and CD34+ enriched primary progenitors. Parallel studies with murine progenitors (c-Kit+, Lin−) and cell lines also showed a preponderance of Hoxa6 expression over other family members including Hoxa9 and Hoxb4. Several hemopoietic cell lines, namely Ba/F3, EML, FDCP-Mix A4 and 32Dcl3 were subsequently used to investigate Hoxa6 regulation following differentiation or growth factor stimuli. Hoxa6 expression decreased with cell differentiation and growth factor depletion/replenishment studies indicated a cell-cycle component for Hoxa6 regulation. Direct evaluation of cell-cycle status, using Hoechst 33342 staining and cell sorting, identified peak expression of Hoxa6 during S-phase. Gene deletion studies involving Hox tend to result in either a moderate or no phenotype, presumably due to intrinsic compensatory mechanisms. We therefore overexpressed HOXA6 in the Ba/F3 cell line to gain functional insights. Ba/F3-A6 cells were compared to mock-transfected and vector controls on the basis of proliferation, maturation, cell-cycle status, growth factor-dependence and apoptosis. The Ba/F3-A6 cells displayed a growth advantage over normal cells in the presence of IL-3 and maturation was not impaired. Cell-cycle analysis showed a reduction in the number of cells in both G2M and S-phase, associated with accumulation in the pre G1-phase, indicative of increased apoptosis. IL-3 depletion studies of Ba/F3-A6 cells indicated substantial factor-independent growth compared to controls, implying oncogenic potential for HOXA6. In support of this, a recent report (Mamo et al, Blood. 2006 Jul 15;108(2):622–9) indicated Hoxa6 as a potential collaborator in a Meis1-induced model of AML. Taken together these findings identify Hoxa6 as a novel candidate gene in AML with the capacity to alter growth and survival of hemopoietic cells.Gene Expression Ranking of HOX and MEIS1 in AML.GENEEXPRESSION RANGEMEAN RANKS.D.OVERALL RANKHOXA61.2 x 103 – 1.7 x 1072.21.61HOXB39.3 x 101 – 8.4 x 1063.22.52HOXB2*7.9 x 102 – 5.4 x 1063.42.03HOXA94.0 x 101 – 5.3 x 1065.32.44MEIS10.6 x 101 – 8.4 x 1065.42.75HOXA10*2.4 x 102 – 1.7 x 1055.53.26HOXB41.5 x 102 – 7.8 x 1055.53.27HOXA7*5.3 x 103 – 1.8 x 1065.71.78HOXB62.3 x 101 – 8.8 x 1056.62.89HOXA44.1 x 101 – 1.1 x 1057.93.410HOXA5*3.4 x 101 – 4.3 x 1049.32.811HOXC61.0 x 101 – 3.2 x 1039.72.312HOXA11*4.0 x 101 – 6.1 x 10310.62.213Expression values (copies per 50 ng RNA) compiled from primary AML patient samples (n=24) or * (n=12). S.D = standard deviation.