Endothelial Activation by Sickle Mouse Red Cells and Their Endothelial Adhesion In Vivo is Abolished by Catalase Treatment of Sickle Cells, but Not Quiescent Endothelium, Implying a Role of Heme-Mediated Peroxide Generation in Sickle Cell Adhesion.

Abstract 902

Multiple adhesion molecules, expressed on sickle red blood cells (SS RBCs) and activated endothelium, have been implicated in SS RBC adhesion to vascular endothelium. Moreover, intrinsic differences among heterogeneous SS RBC subpopulations, involving differences in red cell adhesion molecules and cell deformability, may contribute to their adhesive and obstructive properties and lead to postcapillary obstruction. However, the role of SS RBCs in endothelium activation and adhesion has not been evaluated despite the insightful studies of Hebbel and coworkers (JCI, 1982) demonstrating that SS RBCs generate excessive amounts of reactive oxygen species due to the presence of unstable hemoglobin S (HbS) and autoxidation of iron in heme. RBCs from transgenic-knockout sickle (BERK) mice similarly show a pronounced increase in heme degradation (Nagababu et. al. Blood Cells Mol Dis, 2008). We hypothesize that hypoxic conditions in venules (oxygen tension,∼30 mm Hg) will accelerate autoxidation of RBC membrane-bound HbS and release H₂O₂ that will be transferred to adjoining endothelium resulting in its activation (i.e., up-regulation of endothelial adhesion molecules) and SS RBC adhesion. To test the hypothesis that HbS-containing red cells from BERK mice will result in activation of quiescent endothelium in normal mice, we infused FITC (fluorescein isothiocynate)-labeled BERK red cells into congenic C57BL mice. BERK mice, expressing exclusively human β^S- and α-globins, have been extensively backcrossed onto C57BL background. Intravital observations were made in the cremaster muscle microcirculatory bed. A single bolus of 150 μl of FITC-labeled BERK RBCs (Hct 30%) was infused into the recipient C57BL mouse via the jugular vein over a period of 5 min to avoid any shear related platelet aggregation. Infusion of FITC-labeled control (C57BL) mouse RBCs into C57BL recipient mice resulted in rare or no RBC adhesion, suggesting that there was no activating effect on endothelium. In contrast, infusion of BERK mouse RBCs into C57BL mice resulted in time-dependent increase in adhesion to venular endothelium. Adhesion became discernable after 3 minutes and showed a 3-5 fold increase after 5-min compared with the number of adherent RBCs at 3 min (P<0.01). Next, we investigated if the infusion of BERK mouse RBCs would induce increased endothelial oxidants. To this end, the cremaster preparation was suffused for 15 min with 123 dihydrorhodamine (DHR), a H₂O₂-sensitive probe (10 μl/L), followed by a bolus infusion of BERK mouse RBCs, and time-dependent changes in DHR fluorescence intensity were monitored in venules, the sites of adhesion. Infusion of BERK mouse RBCs, but not C57BL RBCs, resulted in time-dependent increase in the fluorescence intensity (ΔI) in venular endothelium, with almost 5-fold increase in DHR intensity after 5 min of BERK RBC infusion (P<0.001) compared with ΔI at 1 min. When infusion of catalase (900 U/mouse) into recipient C57BL mice was followed 30 min later by a bolus of FITC-labeled BERK mouse RBCs, BERK RBC adhesion and pronounced DHR fluorescence in endothelium were observed, demonstrating that intravascular infusion of catalase had little effect on oxidant generation by BERK mouse RBCs. In contrast, infusion of BERK RBCs pre-treated with catalase (100 U in 0.2 ml RBC suspension, 9-fold less catalase per mouse) to quench RBC generated H₂O₂ inhibited endothelial DHR fluorescence and BERK RBC adhesion. These results strongly suggest an obligatory role of heme-mediated peroxide generation by SS RBC in endothelial activation and SS RBC adhesion, and support the notion that heme-mediated oxidant generation may play a vital role in endothelial dysfunction in sickle cell disease.