The Spliceosomal Component Sf3b1 Is Essential for Hematopoietic Differentiation

Myelodysplastic syndrome (MDS) is a disorder arising from hematopoietic stem and progenitor cell (HSPC) dysfunction resulting in ineffective hematopoiesis. A multitude of recurrent somatic mutations in spliceosomal components were recently identified in MDS that likely contribute to the pathogenesis of the disease. The lack of in vivo models to study cell-type specific effects of spliceosomal mutations limits our understanding of why such mutations lead to hematopoietic abnormalities. Using a zebrafish with a loss-of-function mutation in sf3b1 (sf3b1^hi3394), an essential member of the spliceosome, we demonstrate hematopoietic cell differentiation and hematopoietic stem and progenitor cell (HSPC) specification are processes sensitive to spliceosomal malfunction. Primitive erythropoiesis initiates normally in sf3b1 mutants as evidenced by expression of scl in the posterior lateral plate mesoderm at 14 hours post fertilization (hpf) as well as gata1 and beta-globin at 24 hpf. Flow cytometry quantification of gata1:gfp positive erythrocytes showed sf3b1 mutants have 25% more cells at 24 hpf, but greater than 3-fold fewer cells at 36 and 48 hpf, time points when wild type erythroblasts are expanding and differentiating. At 48 hpf, we also observed decreased levels of o-dianisidine positive erythrocytes, low numbers of morphologically mature erythroblasts, and higher levels of immature erythroblasts in sf3b1 mutants. Similarly, we observed normal initiation of primitive myelopoiesis marked by stem cell leukemia (scl) expression in the anterior lateral plate mesoderm at 14 hpf, but diminished expression of more differentiated markers, l-plastin and myeloperoxidase at 24 and 28 hpf in sf3b1 mutants. Quantification of lysozyme C:dsred positive myeloid cells using flow cytometry also showed 24-fold fewer mature myeloid cells in sf3b1 mutants at 36 hpf. Our data on primitive erythropoiesis and myelopoiesis indicate sf3b1 is required for hematopoietic cell differentiation. Additionally, sf3b1 mutants have diminished expression of the definitive HSPC marker runx1 within the aorta at 28 hpf. In contrast, we observed normal expression of the pan-endothelial marker kinase insert domain receptor-like (kdrl) and aorta-specific markers notch1b and notch3 at 24 hpf. Flow cytometry quantification of kdrl:gfp endothelial cells at 24 hpf shows no difference in the frequency of endothelial cells in sf3b1 mutants. Moreover, we observed fewer cmyb:gfp; kdrl:dsred double positive HSPCs along the dorsal aorta. The data suggest that decreased HSPC formation in sf3b1 mutants is due to a failure in hemogenic induction. From these studies, we show sf3b1 is required at specific stages of hematopoietic cell development. These results provide novel insight into the role of splicing in blood cell development, and can afford a deeper understanding of the mechanism of splicing regulation on the origins of MDS.