Abstract 818

Hydroxycarbamide (HC, or Hydroxyurea) is the only drug with a proven efficiency to treat patients affected by sickle cell disease (SCD), notably by decreasing vasoocclusive crisis (VOC) frequency, the hallmark of the disease. It is now admitted that abnormal red blood cells (RBC) carrying hemoglobin S exhibit increased adhesion to the vascular endothelium (VE) and that this inflammatory-strengthened adhesion plays a major role in VOC onset. Beyond its inconstant action on fetal hemoglobin stimulation, HC decreases adhesion between RBC and VE. Our laboratory and others have shown that HC modulates adhesion molecules expression on circulating RBC. We were the first to show that this action was reproducible in vitro on erythroid precursors and that HC also targets vascular endothelial cells (VEC) by modulating some adhesion molecules, pro-inflammatory cytokines and endothelin-1 (ET-1) expression. However, HC is a cytostatic agent that was firstly used to treat myeloproliferative disorders. Its mode of action is poorly understood and the potential deleterious effects of a life-long treatment are unknown. Our objective is then to decipher the sequence of events leading to target genes modulation to identify potential safer therapeutic targets and with an increased specificity of action. By conducting a network analysis of our microarray data obtained on VECs from the bone marrow microcirculation (TrHBMEC) treated or untreated with HC under basal or inflammatory conditions, we identified two candidate genes, namely transcription factors GATA-2 and GATA-6, as potential relevant intermediates in the HC-induced decreased endothelial expression of adhesion proteins and ET-1. GATA-2 and 6 expression was measured at the mRNA level by RQ-PCR and at the protein level by quantifying western blot signals in HC-treated TrHBMEC from 5 to 48h in basal conditions and in the presence of pro-inflammatory cytokines. In TrHBMEC, HC decreases mRNA levels of GATA-2 and GATA-6 time-dependently in basal and inflammatory states. Indeed, GATA-2 is significantly decreased from 16h-basal and 24h-inflammatory HC treatment with a maximal decrease at 48h (50%, p-value<0.0001 and 35%, p-value<0.0001 respectively). GATA-6 is significantly decreased from 16h-basal and 48h-inflammatory HC treatment, with a maximal decrease at 48h (42%, p-value = 0.0008 and 22%, p-value = 0.0004 respectively). Protein analyses performed by signals quantification following western blot did not show any GATA-6 protein modulation at 24h. However, at 48h, HC significantly decreases GATA-6 protein level in basal condition (44%, p-value = 0.0018) but in a less and non-significative extent in inflammatory condition. Studies of GATA-2 modulation at the protein level are still in process. Microarray experiments followed by RQ-PCR allowed us to point out and to confirm the mRNA downregulation of two GATA factors: GATA-2 and GATA-6 involved in the expression of previously-identified HC-modulated-adhesion genes. These results were validated at the protein level for GATA-6, for which a significant decrease at 48h in basal condition treatment was observed, in agreement with the observed HC effect on the GATA-6-dependent gene vWF (von Willebrand Factor). Functional HC effect on GATA2/6 binding to GATA-responsive elements is being currently tested by gel shift assays. These data will complete this study by the analysis of GATA activity modulation in response to HC treatment. Our results are consistent with our previous results concerning the HC effect on the endothelial expression of genes pertinent to SCD pathophysiology and known to contain GATA sites in their promoter: ET-1 and PECAM-1 (Platelet Endothelial Cell adhesion Molecule-1) for GATA-2, vWF for GATA-6 and VCAM-1 (Vascular Cell Adhesion Molecule-1) for both GATA-2 and GATA-6. It was proposed that HC acts through the NO-sGC-cGMP pathway; but mediators, notably transcription factors, between cGMP and final target genes are unknown. Our results suggest that GATA-2 and GATA-6 are mediators in the VEC HC response.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution