The actin binding proteins cortactin and HS1 are dispensable for platelet actin nodule and megakaryocyte podosome formation

Abstract

A dynamic, properly organised actin cytoskeleton is critical for the production and haemostatic function of platelets. The Wiskott Aldrich Syndrome protein (WASp) and Actin-Related Proteins 2 & 3 Complex (Arp2/3 complex) are critical mediators of actin polymerisation and organisation in many cell types. In platelets and megakaryocytes, these proteins have been shown to be important for proper platelet production and function. The cortactin family of proteins (Cttn & HS1) are known to regulate WASp-Arp2/3-mediated actin polymerisation in other cell types and so here we address the role of these proteins in platelets using knockout mouse models. We generated mice lacking Cttn and HS1 in the megakaryocyte/platelet lineage. These mice had normal platelet production, with platelet number, size and surface receptor profile comparable to controls. Platelet function was also unaffected by loss of Cttn/HS1 with no differences observed in a range of platelet function assays including aggregation, secretion, spreading, clot retraction or tyrosine phosphorylation. No effect on tail bleeding time or in thrombosis models was observed. In addition, platelet actin nodules, and megakaryocyte podosomes, actin-based structures known to be dependent on WASp and the Arp2/3 complex, formed normally. We conclude that despite the importance of WASp and the Arp2/3 complex in regulating F-actin dynamics in many cells types, the role of cortactin in their regulation appears to be fulfilled by other proteins in platelets.

Keywords: Actin nodules; Cortactin; HS1; platelets; podosomes.

Figure 1.

Characterisation of platelets from Cttn KO and DKO mice. (A) Western blots of platelet lysates confirming phosphorylation of Cttn downstream of G-protein coupled and tyrosine kinase linked receptors. Cortactin was immuno-precipitated from stimulated mouse platelet lysates and western blotted with anti-phosphotyrosine (4G10) and anti-cortactin (4F11) antibodies. (B) Whole blood platelet counts and mean platelet volumes were measured from WT, Cttn and DKO mice. No significant difference was observed between the genotypes in either platelet number or volume. Data is mean ± SD (WT, n = 37; Cttn KO, n = 24; DKO, n = 18). (C) Platelet surface receptor levels were analysed by flow cytometry. No significant differences were observed between either genotype and WT controls. Data are % of WT control (Mean ± SEM; n = 3) and values were corrected for IgG background staining.

Figure 2.

Loss of Cttn and HS1 does not affect platelet spreading or F-actin organisation. (A) The loss of Cttn only or Cttn and HS1 from platelets did not affect their ability to adhere and spread on fibrinogen (±0.1 U/ml thrombin) or collagen related peptide (CRP) coated coverslips. (B) Quantitation of the surface area of spread platelets from either Cttn KO or DKO platelets showed no significant differences in spreading. (C) Staining of spread platelets for F-actin with fluorescent phalloidin showed normal actin organisation with filopodia, actin nodules and platelet stress fibres being observed in both WT and DKO platelets. Scale bars in (A) = 10 µm and in (C) = 5 µm.

Figure 3.

Loss of Cttn and HS1 does not affect platelet function. The loss of Cttn alone or both Cttn and HS1 had no effect on the aggregation (A & B) or α-granule secretion (C) of platelets to G-protein coupled or tyrosine kinase linked receptor agonists. Aggregation data are expressed as % final aggregation 6 mins after agonist addition. Alpha granule secretion data are % of WT values measured by flow cytometry 2 mins after addition of agonist. (D) The loss of Cttn and HS1 had no effect on pro-coagulant surface generation as phosphatidylserine exposure was not affected by loss of both cortactin and HS1. (E) Clot retraction in PRP following stimulation by thrombin was unaffected by loss of Cttn or Cttn and HS1. All data are presented as mean ± SEM, n = 3.

Figure 4.

In vivo thrombosis assays and in vitro flow studies. (A) An increase in tail bleeding (mg blood loss/g body weight) was observed in Cttn KO mice following removal of the terminal 3 mm of the tail. However, this increase was not significant and was not observed in DKO mice. Symbols (● = WT, Δ = Cttn KO, × = DKO) represent individual data points, horizontal bars the mean and vertical bars the SEM (n = 18 for WT, 12 for Cttn KOs and 7 for DKOs). (B) In vitro flow assays performed over collagen showed no significant decrease in aggregate formation at shear rates of either 1000 s⁻¹ or 1500 s⁻¹ for either Cttn KO or DKO mice. (C & D) In vivo thrombosis, as determined by the cremaster laser injury model, showed no effect of loss of Cttn on either thrombus size (C) or time to peak intensity (D).