Abstract

In response to agonist stimulation, platelets undergo a rapid reorganization of their actin cytoskeleton. This process involves simultaneous disassembly and assembly of filamentous actin, and is one of the earliest events of platelet activation. Ex vivo flow models support the hypothesis that actin assembly is essential for stable in vivo platelet adhesion and thrombus formation under the shear conditions found within the arterial system. On the cellular level, the reorganization of actin that occurs during platelet activation involves the formation of an actin network similar to that found in lamellipodia at the leading edge of the locomoting fibroblasts. During lamellipod formation in fibroblasts, the predominant actin isoform of the cell leading edge is beta-actin. In these cells, beta-actin undergoes posttranslational arginylation (i.e., addition of arginine onto the N-terminus). We estimate that up to 40% of actin at the leading edge of fibroblasts is arginylated, and that the absence of this modification results in actin filament aggregation in vitro. Arginylation of actin is mediated by the arginyltransferase, ATE1. Embryos lacking ATE1 die between E12 and E17 due to severe cardiovascular defects and massive hemorrhage. Fibroblasts derived from knockout embryos collapse their extending lamella, potentially because their non-arginylated actin aggregates at the extending edge. The beta-isoform of actin is over 85% of the total actin in platelets. To gain insight into whether actin could undergo arginylation in platelets, we analyzed actin from resting and thrombin- or PMA-stimulated platelets fractionated on 2D gels. Our analysis indicates that platelet activation results in a prominent shift of a significant fraction of actin protein into the basic pI range, consistent with the addition of arginine onto the actin molecule. This result suggests that actin might be arginylated in platelets, and that this arginylation is activation-dependent. We hypothesized that if platelet actin reorganization indeed occurs by the mechanisms similar to those seen in fibroblasts, it should also be dependent on the arginylation. To test this hypothesis, we reconstituted the hematopoietic system of lethally irradiated mice with hematopoietic precursor cells collected from the livers of ATE1 null or wildtype embryos at day 13.5 of development. Platelets derived from ATE1 knockout radiation chimeric mice exhibited normal binding to fluorescently labeled fibrinogen in response to stimulation of the thrombin receptor or by PMA. However, in contrast to platelets derived from radiation chimeric mice rescued with normal hematopoietic cells, platelets derived from chimeric mice with ATE1 null hematopoietic cells spread poorly to immobilized fibrinogen (decreased 35.6%). When blood harvested from radiation chimeric mice was allowed to adhere to immobilized collagen under flow, wild type platelets adhered firmly, and rarely formed unstable adhesions (15 ± 10%). In contrast, ATE1 null platelets adhered less well, and frequently failed to form stable adhesions (50% ±15%). Together, these results suggest that beta-actin is rapidly arginylated in agonist-stimulated platelets, and that arginylation is required for actin reorganization during platelet adhesion and lamellipodia formation. We hypothesize that ATE1-mediated arginylation of platelet beta-actin is also essential for in vivo platelet adhesion and thrombosis.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author