Determinants of Sickle Cell Density Distribution: A Study of Benin Twins.

Abstract Red cell density distribution affects both hemolysis and vaso-occlusion; however, currently recognized factors cannot account for all of the variation seen. We hypothesized that a range of genetically controlled factors contributes to red cell density distribution and hemolysis, which has recently received a great deal of attention from Gladwin et al for its role in sickle cell anemia (SCA) and its impact on NO metabolism. Our previous studies have demonstrated that although RBC density distribution varies significantly from patient to patient, the pattern for individual patients is stable in the absence of disease. Some genetically determined factors that affect red cell density distribution have been defined, such as alpha-thalassemia and % HbF (Fabry et al, Blood, 1982); however, neither of these factors completely predicts density distribution. The study of identical twins offers the unique opportunity to minimize some of the genetic variability between individuals that may not be relevant to RBC density while allowing the remaining differences to be detected. Because SCA patients from the US have a complex mixture of Caucasian, other ethnicities, and genes from all parts of Africa including all of the sickle haplotypes, we have chosen to recruit our population from Benin that has a single beta-globin haplotype, thus further minimizing differences arising from admixture from inside and outside of Africa. We have collected samples from six sets of monozygous twins from Benin and validated their monozygosity by DNA analysis. Of the 6 twin sets analyzed to date, 4 have alpha-thalassemia. We compared density gradients on two separate occasions, approximately one year apart, for these twins and found that density gradients for both members of all twin sets without medical complications (malaria, painful crisis) were indistinguishable. This is not true for pairs of randomly chosen individuals even after their % HbF and alpha-thalassemia status has been determined and matched. After elimination of WBCs, we were able to isolate sufficient RNA to obtain microarray data without amplification. We compared subjects with a low % dense cells vs a high % dense cells. In the combinations that were analyzed, we found a consistent pattern of up- and down-regulated genes. Down-regulated genes included the Gardos channel (KCNN4), K-Cl cotransporter (KCC1), NOS3, CAIV, and PKC. Up-regulated genes included ferritin heavy chain (probably in mitochondria that are present in reticulocytes) and 2,3-bisphosphomutase that was elevated in twins with higher MCHC (density). The latter is consistent with our previous observations and those of Poillon et al that DPG can affect polymer formation and RBC density. We conclude that the study of twins demonstrates that there is a strong genetic component in the control of sickle cell density distribution and that a better understanding of factors controlling density distribution may lead to new forms of treatment.