There are three related projects in this program: Project 1: Genome sequencing to identify DBA mutations. Project 2: In vitro culture and epigenetic profiling of DBA patient cells. Project 3: Novel mouse models of red blood cell functions. DBA is an inherited bone marrow failure syndrome in which ribosome defects specifically block erythropoiesis while cells of the other hematopoietic lineages are normal in both number and function. In addition to anemia, DBA shares several clinical features with other bone marrow failure syndromes, e.g. Fanconi anemia and Schwachman Diamond syndrome, including: 1) a high frequency of birth defects, 2) a predisposition to cancer (both solid tumors and AML) and 3) variable penetrance and clinical course, ranging from asymptomatic to severe. DBA is inherited as an autosomal dominant with incomplete penetrance and 60% of DBA mutations arise de novo. Despite the phenotypic variation observed among patients with DBA, only a few confirmed genotype/phenotype correlations have been established. Mutations in at least 15 different ribosomal protein (RP) genes have been identified in 65% of DBA patients. These mutations lead to aberrant rRNA processing, impaired production of either the small or large ribosomal subunit and reduced numbers of mature 80S ribosomes. Targeted sequencing of the 80 ribosomal protein genes in >75 DBA patients without a molecular diagnosis failed to identify causative mutations in 35% of DBA patients. Recent exome sequencing efforts have identified three DBA families (4 patients) with GATA1 splicing mutations, another DBA family (1 patient) with mutations in the Ribosome Maturation Factor gene TSR2114 and families with mutations in CECR1 (ADA2 deficiency). Our exome sequencing of 12 DBA patients with no identifiable RP mutations from the resequencing cohort, also did not identify causative mutations in any other genes. Identifying and characterizing the remaining 35% of DBA mutations is the goal of Project 1. An unresolved question regarding the pathogenesis of DBA is the precise stage at which erythroid differentiation is blocked. DBA patient CD34+ cells are deficient in the earliest described erythroid progenitor, the Burst Forming Unit-Erythroid (BFU-E)116,117. The hierarchical models of hematopoiesis would predict that the defect would be at the BFU-E stage or its immediate precursor, the MEP. However, we have shown that the MEP population contains 45% erythroid primed cells34, suggesting that the failure of erythropoiesis in DBA may involve a more primitive Pre-MEP or HSC population. We hypothesize that CD34+ progenitor cells from patients with DBA will be deficient in erythroid primed progenitor cells. In the first part of Project 2 we will test this hypothesis using single cell RNA profiling to identify the cell populations affected in DBA. DBA patients are treated with red cell transfusions, usually followed by a trial of corticosteroid therapy around the age of three years. Durable steroid-induced erythropoiesis is seen in approximately 30-40% of DBA patients. In addition, 15% of DBA patients undergo spontaneous remission and begin relatively normal erythropoiesis for periods of several months to many years. One possible explanation for the variable response to steroid treatment and spontaneous remission is that epigenetic correction in progenitor cells initiates normal hematopoiesis. We hypothesize that the induced or spontaneous initiation of erythropoiesis is accompanied by specific epigenetic changes that will identify critical targets for future therapeutic interventions. In the second part of Project 2 we will investigate the epigenetic profile of erythroid cells cultured from DBA patients. We have preliminary data demonstrating specific and reproducible differences in DNA methylation between cells from transfusion dependent DBA patients and controls. We will extend these studies to include analyses of chromatin accessibility (ATAC-Seq). We will recruit steroid-responsive DBA patients and DBA patients in remission to identify specific epigenetic changes associated with the restoration of erythropoiesis. Our collaboration with Jeff Lipton and Adrianna Vlachos of the North American DBA Registry (DBAR) has been very fruitful, providing us access to DNA collected from DBA patients and family members, who have provided informed consent for genetic testing. The DBAR has over 700 patients registered119 of which approximately 250 lack a molecular diagnosis. The DBAR team have also provided us with DBA patient peripheral blood samples. Experimentally, peripheral blood CD34+ progenitor cells have several advantages over bone marrow derived CD34+ cells. Bone marrow is collected infrequently and only to monitor a change in the patients medical condition. Peripheral blood can be obtained easily and allows for serial sampling of the same patient over time including before and during steroid treatment and after remission. Project 3 will investigate two novel hypotheses about erythrocyte production and function and red cell metabolism. Mammalian red cell deformability and metabolism changes between low oxygen (tissues) and high oxygen (lung) environments. In hypoxic conditions red cells show increased deformability, increased ATP release and pentose pathway activity, while in normoxic conditions red cells revert to normal deformability and activate the glycolysis pathway to produce ATP. A candidate for the oxygen sensor is band 3, the protein encoded by the Slc4a1 gene. Band 3 is the most abundant protein in the mammalian erythrocyte membrane and functions as an anion exchange channel. In vitro, the N-terminus of the cytoplasmic domain of band 3 reversibly binds glycolytic enzymes in normoxic conditions and deoxyhemoglobin in hypoxic conditions but the in vivo function of this reversible binding has not been studied. In sickle cell disease (SCD), polymerization of deoxyhemoglobin S is the fundamental lesion underlying the multi-organ damage that characterizes the disease. DeoxyHbS forms long insoluble polymers that distort the red cell into the characteristic sickle shape. A paradox in the study of SCD is that the polymerization of HbS within red blood cells occurs at 10-fold higher concentrations of oxygen than the polymerization of HbS in vitro. We will test the hypothesis that the binding of deoxyhemoglobin to band 3 is the switch controlling the oxygen dependent changes in deformability, glycolysis and HbS polymerization. The red blood cells of an average adult human male store collectively about 2.5 grams of iron, representing about 65 percent of the total body iron. The continuous production of new red blood cells requires iron, some supplied by the diet and the remainder by macrophages which engulf senescent red blood cells and recycle iron to maturing red cells. Iron metabolism models predict that the iron required for heme synthesis (and therefore hemoglobin assembly) is imported into erythroid cells as transferrin and subsequently incorporated into heme by a series of enzymes associated with the mitochondrial membrane. To keep the levels of heme and globin chains balanced (excess heme is toxic), immature red cells express a heme exporter, FlvcR, which is expressed at its highest levels at the CFU-E stage before declining during terminal erythroid maturation (basophilic erythroblast to reticulocyte). However, the translation of globin mRNA peaks during the most terminal stages of erythroid maturation (orthochromatic erythroblast and reticulocyte) between which the nucleus is extruded and most of the heme producing mitochondria degrade. How then do reticulocytes generate enough heme to complete hemoglobin assembly? We have knocked out a novel heme transporter, HRG1 to answer this question.
Showing the most recent 10 out of 24 publications
