HoxA10 Represses Transcription of Multiple Genes Encoding Respiratory Burst Oxidase Proteins in Undifferentiated Myeloid Cells.

HoxA10 is a homeodomain (HD) transcription factor which is maximally expressed in committed myeloid progenitors and is overexpressed in acute myeloid leukemia (AML). Consistent with this, forced overexpression of HoxA10 in murine bone marrow leads to AML. Our studies identify a mechanism by which HoxA10-overexpression blocks differentiation, contributing to leukemogenesis. Previously, we found that HoxA10 represses transcription of the CYBB gene. This gene encodes the gp91phox component of the phagocyte respiratory burst oxidase and is actively transcribed in myeloid cells differentiated beyond the promyelocyte stage. CYBB transcription continues until cell death and is further increased in mature phagocytes by inflammatory mediators such as IFNg and LPS. In undifferentiated myeloid cells, HoxA10 represses CYBB transcription via a proximal promoter cis element with homology to the consensus sequence for HoxA10 DNA-binding as a heterodimer with Pbx1 (also a HD protein). However, we found that HoxA10 repression of CYBB transcription is Pbx1-independent and HDAC2-dependant. During cytokine induced differentiation, phosphorylation of two HoxA10 HD tyrosine residues decreases DNA-binding affinity and transcriptional repression. In the current studies, we investigate the impact of HoxA10 on transcription of the NCF2 gene. This gene, which encodes the p67phox phagocyte oxidase component, is transcribed concurrently with the CYBB gene during myelopoiesis and the inflammatory response. The NCF2 promoter also includes a sequence homologous to the HoxA10/Pbx1 DNA-binding consensus. In the current studies, we determine that HoxA10 functionally represses NCF2 transcription in undifferentiated myeloid cells in a Pbx1-independent, HDAC2-dependant manner. Additionally, we determine that NCF2-transcriptional repression by HoxA10 is abolished by phosphorylation of HD tyrosine residues in response to differentiating cytokines. In these studies, we also investigate the impact of IFNg-treatment of monocytes on HoxA10 phosphorylation and gp91phox and p67phox expression. We find that treatment of monocytes with this inflammatory mediator results in hyper tyrosine phosphorylation of HoxA10, which is necessary for IFNg-induced gp91phox and p67phox expression in these cells. These experiments suggest that HoxA10 contributes to differentiation block in AML by repressing transcription of genes that confer the mature myeloid phenotype. HoxA10 tyrosine phosphorylation abrogates transcriptional repression and may provide a molecular target for differentiation induction in AML. Additionally, HoxA10 hyper phosphorylation during the inflammatory response further increases oxidase gene expression. This represents the first demonstration that functions of HoxA10 contribute to gene regulation in mature phagocytes, contributing to the system of host defense.