Twins actin’ differently

In this issue of Blood, Stritt et al describe how twinfilin 2a (Twf2a), an actin-binding protein, negatively regulates platelet reactivity and turnover in a manner that is intriguingly different from other cytoskeleton-related proteins.¹ 

Upon activation, platelets undergo dramatic changes in shape and pliability, both of which are intricately dependent on the actin cytoskeleton. In addition, the actin cytoskeleton is critically involved in many other aspects of platelet physiology, including thrombopoiesis, platelet clearance, and granule release. Its involvement is evidenced by the diverse and often severe abnormalities in a number of genetically altered human and murine platelets. Because of its complexity and dynamic nature, it has been difficult to elucidate the precise underlying molecular mechanism that connects the actin cytoskeleton to a platelet function. An effective approach, however, may be through the comparison and analysis of phenotypes between knockout mouse platelets that lack different proteins. For instance, missing filamin A (FlnA) in megakaryocytes (MKs) and platelets results in severe macrothrombocytopenia and impaired proplatelet formation,²  which is similar to that observed in platelets missing the surface receptor glycoprotein GPIbα.^3,4  This similarity can be attributed to the interaction of these 2 proteins and also to domain 17 of FlnA and a stretch of the cytoplasmic domain of GPIbα that mediate the interaction.⁵  Mutations in other domains of FlnA are consistently associated with abnormalities in other tissues. In addition, missing Wiskott-Aldrich syndrome protein (WASP) in MKs results in premature proplatelet formation, which is distinct from abnormalities of FlnA^−/− MKs and platelets, because WASP impacts signaling induced by fibrillar collagen I and its platelet receptors.⁶ 

Twinfilins are evolutionarily conserved actin-sequestering or capping proteins. They bind actin monomer (G-actin), especially adenosine 5′-diphosphate–bound G-actin, and inhibit the assembly and dynamic changes of actin filament.⁷  MKs and platelets express Twf1 and Twf2a. By thorough and rigorous characterization, Stritt et al demonstrated that Twf2a^−/− mice exhibit mild macrothrombocytopenia, shortened platelet lifespan, enhanced integrin activation, accelerated thrombus formation, and other associated alterations in their platelets. The authors attributed these abnormalities to the interactions of Twf2a with actin. Although these effects are largely mild, they are different from most effects in thrombocytopenic mice that have been documented, thus bringing fresh insights on the molecular machineries regulating platelet function and turnover. The work of Stritt et al suggests that the lower platelet count in Twf2a^−/− mice can be attributed to the accelerated clearance of Twf2a^−/− platelets by macrophages in the spleen and that the accelerated clearance is likely due to the heightened reactivity of Twf2a^−/− platelets. It is similar to another case of hyperreactive platelets in which Ras GTPase–activating protein 3 is knocked out or mutated, although the latter presents more severe thrombocytopenia.⁸  This is in line with the expectation that activated platelets are cleared rapidly, although it remains unclear how activated platelets are recognized as such. Curiously, fewer Twf2a^−/− platelets than wild-type platelets expose phosphatidylserine (PS) lipid on the surface in response to agonists and, relatedly, agonist-induced mitochondrial depolarization was also less pronounced in Twf2a^−/− platelets. This seems to contradict the idea that the recognition of exposed PS by scavenger receptors on the macrophage⁹  is used to clear activated platelets.

Stritt et al also reported no detectable abnormalities with Twf1^−/− platelets. This is a bit surprising because both Twf1 and Twf2a bind G-actin, and they share more than 65% sequence identity. It is tempting to speculate that a unique feature of Twf2a that is not shared by Twf1 may be critical to its negative regulation of platelet reactivity. The molecular nature of this unique feature remains unclear. Perhaps a comparison of Twf1 and Twf2a for their differential effects on platelet functions and associated actin dynamics will help to unearth it.