Phosphatidlyserine (PS) exposure on the erythrocyte surface endows the cell with the propensity of adhering to vascular endothelium. Because individuals with sickle cell disease (SCD) manifest loss of erythrocyte membrane asymmetry with PS exposure, we have assessed the contribution of this marker to the process of sickle erythrocyte-microendothelial adhesion. Assays for plasma-induced adhesion were conducted on unactivated endothelium, in the absence of immobilized ligands, such that PS was compared to the erythrocyte adhesion receptor CD36. Blocking studies with erythrocytes pretreated with annexin V (to cloak PS) or anti-CD36 or both revealed an inhibitory effect on adhesion of 36% ؎ 10% and 23% ؎ 8% with blocking of both sites suggestive of an additive effect. We next evaluated 87 blood samples from patients with SCD and grouped them into 4 categories based on adhesion marker (CD36 and PS) levels. Results revealed a striking correlation between erythrocyte PS positivity and adhesion. Analyses of the individual patient data demonstrated a positive correlation between PS and adhesion (R ‫؍‬ 0.52, P < .000 001), whereas none was noted between adhesion and CD36 (R ‫؍‬ 0.2, P > .07). The effect of PS on adhesion appears to be related to the quantitative differences in erythrocyte markers in SCD, with PS the predominant marker when compared to CD36 both in the total erythrocyte population, and when the adherence-prone erythrocyte, the CD71 ؉ stress reticulocyte, was evaluated. Our study signals the entrance of an important new contributor to the field of sickle erythrocyte-endothelial adhesion. The implications of erythrocyte PS exposure in relation to the vascular pathology of SCD need to be assessed. (Blood. 2002;99:1564-1571

To assess whether fetal hemoglobin (HbF) modulates the adhesion of sickle erythrocytes to endothelium, children with homozygous sickle cell anemia (SS disease) were studied, using this physiologically crucial period to evaluate the relationships between HbF and the major erythrocyte adhesion markers. The mean level of CD36 ؉ erythrocytes was 2.59% ؎ 2.15% (؎ SD, n ‫؍‬ 40) with an inverse relationship between CD36 positivity and F cells (R ‫؍‬ ؊0.76, P < .000 00 002). In univariate analyses, significant correlations with various hematologic parameters and age were noted. Multiple regression analyses, however, revealed a relationship solely with F cells. Minimal levels of very late activation antigen-4 ؉ (VLA4 ؉ ) erythrocytes (0.31% ؎ 0.45%, n ‫؍‬ 40) with relationships similar to those noted for CD36 ؉ cells were also observed. The subpopulation of strongly adhesive stress reticulocytes was further assessed, using CD71 as their marker. The mean level of CD71 ؉ erythrocytes was 5.81% ؎ 4.21%, with statistical correlates in univariate and multivariate analyses similar to those discussed above. When adhesion ratios were evaluated, inverse correlations were noted between basal and plasma-induced adhesion and F-cell numbers (R ‫؍‬ ؊0.54, P < .0005; R ‫؍‬ ؊0.53, P < .0006, n ‫؍‬ 39). In addition, in analyses where basal or plasma-

In sickle cell disease (SCD), loss of erythrocyte membrane phospholipid asymmetry occurs with the exposure of phosphatidylserine (PS), which provides a docking site for coagulation proteins. In vivo sickling/desickling, with resulting red cell membrane changes and microvesicle formation, appears to be one of the factors responsible for PS exposure. We evaluated children with SCD homozygous for sickle hemoglobin (SS disease) and controls (n = 65) and demonstrate that high levels of fetal hemoglobin (assessed as F cells) are associated with decreased microvesicle formation, PS exposure, and thrombin generation. F cells correlated inversely with both microvesicles and PS positivity (P < .000001) in SS disease. Multiple regression analyses using various hematologic parameters as independent variables, and either microvesicles or PS positivity as the dependent variable, showed a strong relationship only with F cells. Additionally, plasma prothrombin fragment F1.2 levels (a marker for thrombin generation) correlated with both PS positivity (P < .001) and F cells (P < .01). An F-cell level of approximately 70% was associated with normal levels of prothrombin fragment F1.2 and with microvesicle formation indistinguishable from control values. We suggest that the use of such surrogate biologic markers in conjunction with F-cell numbers may provide valuable insights into the biology and consequences of in vivo sickling.

In sickle cell disease (SCD), loss of erythrocyte membrane phospholipid asymmetry occurs with the exposure of phosphatidylserine (PS), which provides a docking site for coagulation proteins. In vivo sickling/desickling, with resulting red cell membrane changes and microvesicle formation, appears to be one of the factors responsible for PS exposure. We evaluated children with SCD homozygous for sickle hemoglobin (SS disease) and controls (n = 65) and demonstrate that high levels of fetal hemoglobin (assessed as F cells) are associated with decreased microvesicle formation, PS exposure, and thrombin generation. F cells correlated inversely with both microvesicles and PS positivity (P &lt; .000001) in SS disease. Multiple regression analyses using various hematologic parameters as independent variables, and either microvesicles or PS positivity as the dependent variable, showed a strong relationship only with F cells. Additionally, plasma prothrombin fragment F1.2 levels (a marker for thrombin generation) correlated with both PS positivity (P &lt; .001) and F cells (P &lt; .01). An F-cell level of approximately 70% was associated with normal levels of prothrombin fragment F1.2 and with microvesicle formation indistinguishable from control values. We suggest that the use of such surrogate biologic markers in conjunction with F-cell numbers may provide valuable insights into the biology and consequences of in vivo sickling.