The beta1-Tubulin Binding Protein RanBP10 Plays a Critical Role for Platelet Discoid Shape and Hemostasis

Blood platelets in the circulation are released from megakaryocytes (MKs), their precursor cells in the bone marrow. Mature MKs are located at the blood vessels and release platelets into the blood stream by intermediate structures referred to as proplatelets. Every MK can release up to a thousand virtually identical platelets. During this fragmentation process every single platelet is equipped with a rather constant number of assorted granules and a peripheral microtubule ring consisting of several coiled filaments. 8–12 microtubule filaments are found on average in the coil, which is essential for maintaining the platelet discoid shape. Generating these dynamic microtubule filaments at the very cell periphery in the absence of classical microtubule nucleating factors is strictly dependent on de novo nucleation of filaments. Beta1-tubulin is a MK/platelet-specific and the predominant isoform present in proplatelets and platelets. Mice lacking beta1-tubulin are thrombocytopenic and reveal platelet spherocytosis. In humans, a Q43P mutation in the beta1-tubulin gene Tubb1 also leads to large and spheric platelets. We have recently identified RanBP10 as a cytoplasmic beta1-tubulin binding protein that is selectively expressed in hematopoietic tissues like bone marrow and MKs. RNA interference (RNAi)- mediated RanBP10 ablation in primary MKs resulted in the collapse of long microtubule filaments whereas retrovirally forced overexpression of RanBP10 led to a marked increase in microtubule filaments, most likely by number or augmented bundling. RanBP10 harbors guanine nucleotide exchange factor activity toward Ran by exchanging GDP with GTP. As a local increase in Ran-GTP precedes nucleation of non-centrosomal microtubules in other cell types this mechanism might explain how positional information about new microtubule filaments is provided in the cell periphery of mature and proplatelet elaborating MKs. We generated a gene trap-mediated mouse model for RanBP10 deficiency. MKs derived from nullizygous animals phenocopied the collapse of microtubule filaments found by RNAi. Although RanBP10 was dispensable for overall platelet biogenesis with counts about 90–95% of normal, the ultrastructural analysis revealed that loss of RanBP10 protein resulted in a significant decline in the elliptic coefficient: Mutant platelets were more spherical and electron micrographs showed a wider variety in both microtubule filament number and coil localization including additional and incomplete coiling. Most intriguingly, RanBP10- deficient mice demonstrated a markedly prolonged bleeding time compared to littermate controls in a standardized tail bleeding assay. Platelets from nullizygous animals failed to respond normally to ADP by CD62P expression in flow cytometry. Our data thus suggest that RanBP10 plays a critical role in maintaining the tightly controlled platelet shape and function and that RanBP10 is essential for hemostasis.