Hydroxyurea (HU) has been shown to augment the production of fetal hemoglobin (HbF) and for this reason is being used in the treatment of sickle cell anemia. While it has been assumed that HU promote HbF production indirectly by perturbing the maturation of erythroid precursors, the molecular mechanism(s) of how hydroxyurea regulates g-globin expression is not known. To illuminate potential genetic pathways responsible for the induction of HbF after HU treatment, we identified genes induced by HU using a two-phase liquid erythroid culture system in conjunction with mRNA differential display. Cells were treated with/without 100uM HU on days 6 and 7 during phase II, and then collected on day 10. RT-PCR-amplified DNA fragments were analyzed by gel electrophoresis, and a comparison between gene expression profiles of HU-treated or untreated RNA samples was performed. 383 unique DNA bands were identified in HU-treated samples using 56 sets of arbitrary primers. After confirmation by reverse dot blots, 8 positive clones exhibiting more than a 5-fold higher gene induction pattern were sequenced, and searched for homology in Genebank. One of 8 clones was found to be 91% homologous to a mouse small GTP binding protein (sGBP). We further cloned the full-length 2.9 Kb cDNA sequence from a human bone marrow cDNA library using 3' and 5' RACE. In vitro transcription/translation revealed a translated peptide of 22 kDa, corresponding to an open reading frame of 594bp. This small GTP-binding protein was localized to chromosome 6 (band 18.27) and is expressed in most tissues, especially in thyroid, spinal cord and adrenal gland. In order to define its functional localization, we transfected this gene tagged with a GFP fusion protein at the 3' end into K562 and COS7 cells; we found this gene expressed primarily in the Golgi-ER complex. In order to determine whether this gene was involved in the induction of the g-globin gene expression by HU, K562 cells stably transfected with this gene (or vector control) were grown and the expression of g-globin gene was determined by Northern Blots. sGBP-transfectants had a 2.4 fold increase in g-globin gene expression, whereas, vector-transfectants only demonstrated a 40% g-globin induction; both compared with parental K562 cells (P<0.05). These preliminary data suggest that this novel sGBP may play a role in the induction of g-globin gene by HU, and an elucidation of possible signal transduction pathways is currently under investigation.
