Abstract 772

Following thrombin stimulation, platelet PIP5KI synthesizes phosphatidylinositol 4,5-bisphosphate (PIP2), which can be hydrolyzed by phospholipase C to generate second messengers such as IP3. PIP2 also regulates cytoskeletal dynamics by directly interacting with actin-binding proteins such as talin. Three isoforms of PIP5KI (α, β, and γ) are all capable of phosphorylating PI4P to synthesize PIP2. We have generated and characterized murine lines lacking individual PIP5KI isoforms. While mice lacking PIP5KIα and PIP5KIα have absent second messenger formation and partially impaired integrin activation, they are viable. In contrast, mice lacking PIP5KIγ die in utero due to a cardiovascular developmental defect. Megakaryocytes derived from PIP5KIγ-null embryos bleb their membranes due to impaired anchoring of the cell membrane with the underlying cytoskeleton. Since platelets can not be obtained from these embryos, we employed a genetic approach. We used a MLC-2v Cre transgene that targets Cre expression to myocytes, and generated living mice lacking PIP5KIγ by a conditional rescue. PIP5KIγ-/- MLC-2v Cre+ mice expressed PIP5KIγ in the myocardium, but they had absent expression of PIP5KIγ in all other analyzed tissue. These mice had normal appearing hearts, brains, livers, and bone marrow morphology, as well as normal platelet counts. Since mice lacking PIP5KIα and PIP5KIβ have impaired platelet PIP2 production that causes absent IP3 formation, we analyzed platelets lacking PIP5KIγ for second messenger formation. Even though PIP5KIγ an abundant PIP5KI isoform in platelets, loss of PIP5KIγ does not affect IP3 formation or Akt phosphorylation. It has been previously demonstrated that PIP5KIγ can directly bind talin, a protein that regulates the function of integrins. An existing proposed model for integrin activation is that talin-associated PIP5KIγ synthesize PIP2. This newly synthesized PIP2 then binds a FERM domain within talin. The model suggests that this complex of PIP5KIγ-PIP2-talin is critical for the final step that stimulates integrins to bind their ligand. We found three lines of evidence that disprove this model of integrin activation. First, we found that PIP5KIγ-/- platelets had normal integrin-mediated aggregation in response to all analyzed doses of thrombin, ADP, collagen, and a thromboxane analogue (U46619). Second, we observed that PIP5KIγ-null platelets exhibited normal binding of Jon/A, an antibody that recognizes only the activated form of αIIb/β3. Third, we determined that platelets lacking PIP5KIγ spread normally upon adherent fibrinogen. Together, these results disprove the existing model that PIP5KIγ is a critical component of talin-mediated integrin activation. To determine the true function of PIP5KIγ within platelets, we extended our previous studies by analyzing the role PIP5KIγ plays in the regulation of the cytoskeleton. Therefore, we analyzed platelets lacking this enzyme for their ability to anchor the cytoskeleton to the cell membrane. We used optical tweezers to pull the cell membrane apart from the cytoskeleton. Wild type cells had rigid membranes that resisted stretching by trapped fibrinogen-coated beads that were pulled by the optical trap. In contrast, the PIP5KIγ-null platelets had flexible membranes that were easily stretched, and ultimately allowed membrane tethers to form. We further analyzed whether this defect in anchoring the cell membrane to the underlying cytoskeleton causes a defect in vivo using a carotid artery arterial injury model. Mice lacking platelet PIP5KIγ exhibited unstable adhesion in vivo suggesting that impaired cytoskeletal dynamics causes impaired platelet adhesion under flow. Together, our studies demonstrate that the abundant PIP5KI isoform, PIP5KIγ does not contribute to a pool of PIP2 required for second messenger formation or integrin activation. However it does synthesize the pool of PIP2 required to preserve the integrity of the membrane cytoskeleton, and support stable platelet adhesion under conditions of shear.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution