Identifying Direct Targets of the Hoxb4 Transcription Factor Involved in Hematopoietic Cell Self Renewal.

Hoxb4 is a member of the homeobox protein family which when overexpressed, promotes expansion of hematopoietic stem cells (HSCs) both in vivo and ex vivo. The molecular mechanisms underlying the capacity of Hoxb4 to induce self-renewal of HSCs are poorly understood. In order to identify the direct transcriptional targets of Hoxb4 in primary hematopoietic progenitor cells, we created a Hoxb4-ERT2 fusion gene to allow for temporally regulated Hoxb4 activity. This gene was introduced into mouse lineage negative bone marrow (Lin− BM) cells using retroviral-mediated gene transfer. When these cells were treated with Tamoxifen (TAM), translocation of the fusion protein into the nuclei was demonstrated using confocal microscopy. Transduced cells were treated with 300nM of TAM for varying time in suspension culture and then plated into semisolid medium for myeloid progenitor assays. After 7 days, colonies from these primary CFU-C cultures were dispersed and replated for secondary CFU-C as an assay for myeloid progenitor self renewal. Treatment with TAM for 12hr in Hoxb4-ERT2 introduced cells resulted in an increase in the number (5.5 fold) and size (10 fold) of secondary colonies comparing to mock vector transduced controls, verifying that TAM treatment resulted in self-renewal of myeloid progenitors. In order to identify the target genes for this response, RNA from 12hr and 24hr TAM treated cells was analyzed for differential gene expression using Affymetrix 430v2 chips. Analysis of three independent experiments showed good reproducibility and allowed identification of candidate genes. Local pooled error test was used to analyze the data and a false discovery rate threshold at <25% was applied to identify the probe sets significantly differing in samples from TAM-treated Hoxb4-ERT2 samples versus TAM-treated control vector samples. This analysis identified 78 probe sets from cells treated for 12 hours, 103 probe sets from 24hr treated samples, and 20 common probe sets from both time points. A computer analysis of 64 candidate promoters identified in our screen and available in the NCBI database found that 50 % of them contained more than one high or moderate affinity consensus Hoxb4 binding motif. Furthermore, 22 of our candidate genes contained more than two consensus Hoxb4 binding motifs in the promoter regions. In this selected candidate gene group, genes were identified that are involved in various cellular functions: four in signal transduction, two in cell cycle regulation/apoptosis, four in enzyme activity regulation, six in cellular metabolism, two in transcriptional regulation; and eight of them associated with signaling transduction pathways for hematopoietic cells proliferation/self renewal. Therefore, our current results have identified a relatively small number of candidate genes (20) with significant potential to be direct targets of the Hoxb4 transcription factor in primary, self-renewing hematopoietic progenitor cells. We are currently testing several of these genes for direct Hoxb4 binding activity and for functional activity in hematopoietic cells.