Class I Phosphatidylinositol Transfer Proteins (PITPs) Promote Platelet Activation by Functioning As Essential Cofactors for Phospholipid Kinases and Hydrolases

Abstract

Abstract 91In contrast to the more abundant phospholipids within the platelet membrane bilayer, phosphatidylinositol (PI) can be phosphorylated by PI kinases to generate seven distinct phosphoinositides that function as signaling molecules in series of cellular events including platelet activation. The synthesis of individual phosphoinositides in different cellular compartments is tightly regulated both in time and in space, and the relative amount of various phosphoinositides change within a few seconds after agonist stimulation of platelets. Thus, demonstrating how the synthesis of specific phosphoinositides is regulated in platelet activation would be critical to understanding their role in platelet biology. Class I PhosphatidylInositol Transfer Proteins (PITPs) are a small family of proteins that bind and transfer PI monomers from one cellular compartment to another in vitro. Studies in yeast cells suggested that PITP proteins are essential for the biosynthesis of phosphoinositides. Mammalian cells class I PITP has two members, PITPα and PITPβ. These two isoforms are 77% identical in primary sequence and are 94% homologous. It is notable that PITPα is approximately 7-fold more abundant than PITPβ in murine platelets. To characterize the role of each PITP isoform in platelet activation, we generated mice containing conditional null mutations within the gene of each isoform specifically in their megakaryocytes, and consequently to knock out these proteins in their platelets. Mice lacking individual platelet PITP isoforms have approximately 25% lower platelet counts than their littermate controls. Mice lacking both PITP isoforms have platelet counts that are 45% lower than wild type littermates, but otherwise have normal blood counts and appear phenotypically normal. Although the loss of either PITP isoform caused only mild ex vivo platelet function defects, loss of both isoforms led to significant impairment of cell spreading, aggregation, and secretion. To distinguish the role of both PITP isoforms in platelet phosphoinositide production, we 32P-labeled platelets ex vivo, and then analyzed by thin layer chromatography the concentration of individual phosphoinositides. Despite the fact that PITPα is far more abundant than PITPβ, we found that the loss of either isoform impaired the synthesis of PI(4)P and PI(4,5)P2 by 40–50% in either resting or thrombin stimulated platelets. To determine whether PITPs mediate their effect on phosphoinositide synthesis via PI transfer activity, we analyzed in vitro phospholipid kinase activity in lysates of knockout platelets using either PI or PI(4)P as the exogenous substrate. We reasoned that providing abundant quantities of exogenous substrate should eliminate the need for any transfer activity, and any effect of PITP on phosphoinositide synthesis in this circumstance would instead be due to an effect of PITPs on phospholipid kinase activity. We observed that the loss of either PITPα or PITPβ resulted in decreased synthesis of PIP and PI(4,5)P2 in vitro. Interestingly, even though PITPβ is far less abundant than PITPα, PITPβ is required for the majority (approximately 70%) of thrombin induced PI(4,5)P2 synthesis in vitro. In contrast, PITPα is required for 60% of thrombin induced PI(4)P synthesis. As expected, we could reverse the phosphoinositide synthesis defect by adding back recombinant PITPα to the PITPα-null platelet lysates or by adding back recombinant PITPβ to the PITPβ-null platelet lysates. Finally, we analyzed whether the loss of either PITP isoform affected the ability of the production of the second messenger IP3 (a product of PLC-mediated hydrolysis of PI(4,5)P2). We observed that the PITPα-null mutation caused an 80% decrease of thrombin-induced IP3 formation, and the PITPβ-null mutation caused a 56% loss in IP3 production. Together, the data demonstrate that although both PITPα and PITPβ are required for phosphoinositides synthesis and IP3 formation, they appear to have non-redundant functions. PITPα plays a larger role in maintaining PI(4)P levels and PLC signaling, and PITPβ contributes more to PI(4,5)P2 synthesis. Both isoforms cooperate together to promote platelet activation. Most importantly, our work demonstrates that despite their name, PhosphatidylInositol Transfer Proteins (PITPs) do more than just transfer phospholipids. They also possess critical cofactor activity for the synthesis of phosphoinositides. Disclosures:No relevant conflicts of interest to declare.