Clinical and Genetic Variability of Red Blood Cell Hemolysis in Sickle Cell Anemia

Abstract 1077

Intravascular hemolysis is an important pathological mechanism underlying some complications of sickle cell disease and other hemolytic anemias. Hemolysis contributes to endothelial dysfunction, pulmonary and systemic vasculopathy, and platelet and hemostatic activation via nitric oxide catabolism by plasma hemoglobin and arginine catabolism by red blood cell arginase. Little is known about the molecular mechanisms of hemolysis and how the propensity of erythrocytes to hemolyze is modulated. Hemoglobin F concentration and the presence of ∝ thalassemia affect the level of hemolysis but it is likely that other genes and their products are also important. We hypothesize that genetic variation, much of which is outside the β-globin gene-like cluster, underlies the susceptibility of erythrocytes to hemolyze in response to diverse disease stressors.

We first characterized hemolysis by creating a principal component analysis (PCA) of age-adjusted values for LDH, AST, reticulocyte count and total bilirubin, but not hemoglobin concentration, to develop a hemolytic component that reflects shared variability among markers. The development of such a component helps to resolve the problem of dealing with correlated predictors in multivariate analyses and confounding variables such as site, and it permits for adjustment for the degree of anemia. To validate the PCA, we measured the plasma hemoglobin levels and red cell microparticle levels in the first and fourth quartile of PCA intensity of hemolysis in 118 HbS-only patients without detectable HbA, from the Walk-PHASST cohort We observed a highly significant increase in plasma hemoglobin (p<0.0001) and red cell microparticles (p=0.0004) based on PCA quartile. Despite the small sample size of this validating cohort we reproduced significant associations between high hemolytic rate and the subphentypes of low arterial oxygen saturation, high pulse pressure, leg ulcers, TRV, high NT-proBNP levels, and low 6-minute walk test distance. More patients with ∝ thalassemia and more females were present in the lower hemolytic index quartile (p=0.006).

The hemolytic index and its individual components were then used as phenotypes in genome-wide association studies (GWAS) in the CSSCD (Cooperative Study of Sickle Cell Disease) and walk-PHaSST cohorts to discover novel genes that might be associated with hemolysis. As further validation of our approach using PCA stratification, patients in the quartile with the lowest hemolytic index from the CSSCD also had a much higher prevalence of ∝ thalassemia than patients within the highest quartile of hemolytic index (p=2.2E-16). We first examined 1117 cases from the CSSCD and found 303 SNPs, 265 with a MAF >0.05, that reached a threshold of significance of p<5E-4. For replication, we examined these SNPs in the Walk-PHASST cohort. Eight SNPs replicated with the same effects in a GWAS in 449 subjects from Walk-PHAAST and p-value<0.01. Of the 8 SNPs that replicated, 4 SNPs were in olfactory receptor (OR) genes on chromosome (chr) 11p; OR51L2 (rs7948471, rs7938426. rs1391617), and OR51L1 (rs2445284). Several of these SNPs were also associated with HbF in previous GWAS analyses. Polymorphisms in the OR gene cluster upstream of HBG might modulate HbF levels by altering chromatin structure within the HBB globin gene-like cluster. One SNP in an intron of NPRL3 (rs7203566) on chr16p is ∼34 kb upstream from a SNP causing ∝ thalassemia. In CSSCD cases there was an association of SNPs in NPRL3 with reticulocytes (p=5.1E-0006) and LDH (p=0.0003). In silico analysis did not predict any function for this SNP.

Genetic studies to discover new biologic modifiers of hemolysis will help to identify critical molecular determinants of hemolysis for functional studies, to develop new disease severity biomarkers, and to suggest candidate therapies for some common human diseases with intravascular hemolysis. We anticipate that our studies will identify genetic variants enriched in the African-American population primarily determined by the evolved human response to endemic malaria infection. These studies are expected to broadly impact many human diseases and blood banking by providing genomic markers of susceptibility to hemolytic anemia, red cell storage stability and transfusion risk, and insights into novel strategies to reduce anemia and to enhance red blood cell storage and post-transfusion erythrocyte recovery.