CD36-POSITIVE STRESS RETICULOCYTOSIS IN SICKLE-CELL-ANEMIA

Vasoocclusive episodes in sickle cell anemia may be initiated by adher ence of erythrocytes to endothelium, one mechanism of which involves t hrombospondin binding to CD36 on the red blood cell (RBC). We compared CD36 expression and its relationship to stress reticulocytosis in pat ients with sickle cell anemia and other chronic hemolytic disorders, i ncluding some after splenectomy. Adults with sickle cell anemia had si gnificantly more CD36-positive cells (4.1% +/- 3.4%, mean +/-: SD, n = 12) in unfractionated blood than did normal adults, who had almost no ne (0.13% +/- 0.15%, n = 8, p < 0.05). In density-fractionated blood, sickle samples contained significantly more CD36-positive cells (39.8% +/- 21.9%, n = 10) in the lowest density layers than did splenectomiz ed high-reticulocyte controls (8.5% +/- 6.4%, n = 4, p < 0.05). ''Stre ss'' reticulocytes (identified by their unique morphology) were signif icantly more frequent in low-density layers from sickle blood (44.3% /- 23.9%, n = 10) than from nonsplenectomized high-reticulocyte contro ls (10.5% +/- 14.8%, n = 5, p < 0.05). There was a strong correlation (r = 0.92) between stress reticulocyte count and number of CD36-positi ve cells for all patients except thalassemics, in whom CD36-positive c ells were more frequent than stress reticulocytes. Flow cytometry conf irmed that the maximal CD36 signal was found on immature reticulocytes . We conclude that CD36-positive stress reticulocytes occur more frequ ently in sickle cell anemia than in other chronic hemolytic states, ev en after surgical splenectomy, suggesting that this enhanced CD36-posi tive stress reticulocytosis does not simply reflect absent splenic fun ction. These results explain why RBCs from high-reticulocyte control p atients fail to show the significant CD36-dependent thrombospondin-med iated adhesion to endothelium that is exhibited by sickle red cells.