ALBERT

Description: Recent studies by our laboratory have demonstrated unequivocally that factor V is endocytosed by megakaryocytes from plasma to form the platelet-derived factor V pool. In the current study, we have determined the time-dependent acquisition of factor V by platelets in a 67 year old factor V-deficient patient, previously shown to be completely devoid of plasma- and platelet-derived factor V. The patient now receives weekly tranfusions of two units of FFP to prevent gastrointestinal bleeding. Plasma and platelet lysate samples were prepared from fresh, whole blood drawn prior to (0 hrs), and following (2, 6, 24, 96, 168 hrs) patient transfusion. A factor V radioimmunoassay (RIA), which can detect factor V concentrations as low as 0.075 μg/mL, a standard factor V clotting-based activity assay, and/or western blotting analyses, which utilize a mixture of anti-human factor V light chain and heavy chain mAbs, were used to evaluate the appearance, and subsequent disappearance, of factor V in these two blood compartments. Prior to transfusion (t = 0 hr), the patient’s plasma-derived factor V could not be detected by any of the three factor V assays. The patient’s plasma-derived factor V level peaked immediately following transfusion (t = 2 hr) reaching a concentration of 1.3 +/− 0.08 μg/mL as measured by RIA, and declined until it reached an undetectable level at 96 hrs post transfusion. These data were confirmed by the results of both the clotting-based assays and western blotting analyses. As the patient’s platelet-derived factor V levels were below the sensitivity of both the RIA and clotting assay, they were analyzed by western blotting. Such analyses confirmed that subsequent to its endocytosis, the patient’s platelet-derived factor V is stored in a partially protelytically activated form similar to that of a control individual. Due to the partially proteolytically activated state of the platelet-derived cofactor, platelet lysates treated with thrombin to activate the factor V to factor Va were used for quantative western blotting. Interestingly, and perhaps consistent with the patient’s receipt of weekly FFP transfusions, factor V/Va could still be detected in platelets immediately prior to transfusion. Subsequent to transfusion and in marked contrast to changes in the plasma-derived cofactor pool, a significant increase in the patient’s platelet-derived factor V/Va level was not observed until 24 hrs post transfusion. These data are consistent with previous studies demonstrating that platelets do not endocytose factor V. Although the concentration of the platelet-derived factor V/Va pool decreased over the subsequent 6 days, antigen remained detectable even though the plasma-derived pool had been depleted 3 days earlier. These combined observations indicate that, subsequent to FFP administration, the patient’s megakaryocytes acquire and proteolytically process plasma-derived factor V normally. Furthermore, the consistent presence of factor V/Va within the patient’s platelets is in all likelihood preventing gastrointestinal bleeding in this individual, which supports the concept that platelet-derived factor V represents the hemostatically relevant factor V pool.

Description: Key Points Platelet VAMP-3 mediates receptor-mediated endocytosis and endocytic trafficking of cargo. Platelet VAMP-3 regulates spreading, clot retraction, and TPOR/Janus kinase 2 signaling.

Description: Abstract 4316 We recently demonstrated that CADM1, interacting with protein C deficiency, is a novel risk factor for venous thrombosis (VT) Blood 114:3084-3091, 2009. CADM1 likely plays a role in maintaining endothelial barrier function. Prior to this study CADM1 had not been identified in endothelial cells. In this study we have determined the distribution of CADM1 in human vasculature. Methods: Human tissue samples representing all organs as well as large vessels were accessioned from archived paraffin blocks from the surgical pathology division of FAHC. Tissue sections were processed for immunoflourescence using as primary antibodies a chicken monoclonal anti-CADM1 IgY antibody (anti-SynCAM/TSLC1 clone 3E1, MBL Inc, Woburn, MA) and anti-smooth muscle actin (SMA) clone 1A4 (Sigma, St. Louis, MO), anti-von Willebrand factor (vWF) rabbit polyclonal antibody (DAKO Inc., Glostrup, DK). vWF staining was used to identify endothelial cells. The secondary antibodies for CADM1, SMA, and vWF were donkey anti chicken alexa 488, donkey anti mouse alexa 647, and donkey anti rabbit alexa 647, respectively. Appropriate negative and positive controls were run. Intensity of staining was rated as absent, trace or present by three observers (TK, DJT, and EGB). Immunoelectron microscopy (IEM) was performed on cultured human umbilical vein endothelial cells (Allcells LLC, Emeryville, CA) using a post embedding procedure with protein-A gold. Peripheral blood leukocytes were evaluated for the presence of CADM1 using the same antibody following cytospin preparation. Results CADM1 was found ubiquitously in the macro- and micro-vasculature of all organs as well as the aorta and saphenous vein with intensity of staining showing modest variability from organ to organ. CADM1 staining of SMA was observed in both arterial and venous vessels but was considerably stronger on the arterial side. IEM demonstrated cytoplasmic immunogold staining associated with elements of rough endoplasmic reticulum and actin filaments as well as at the membranes of filopodia. In contrast, peripheral blood monocytes, lymphocytes and granulocytes were negative for CADM1expression. Conclusions: CADM1 is expressed ubiquitously in endothelial and smooth muscle cells of the macro-and micro-vasculature with IEM evidence for its presence in filopodial cytoplasm and membranes. Peripheral blood leukocytes did not show evidence of CADM1 expression. This is the first report of CADM1 positivity in vascular smooth muscle cells. The biological role of CADM1 remains to be determined but its ubiquitous expression in the endothelium of the macro- and micro-circulation, together with an apparent role in endothelial cell motility, suggests an important role in endothelial homeostasis. Disclosures: Bovill: Haemotologic Technologies, Essex Jct., VT: Equity Ownership.

Description: Human platelet-derived factor V originates from megakaryocyte endocytosis of its plasma counterpart. Circulating platelets do not endocytose plasma-derived factor V and little, if any, of this hemostatically relevant, and unique, cofactor protein is synthesized by megakaryocytes. Fibrinogen, another α-granule protein, is also endocytosed from plasma, whereas other similarly stored proteins are synthesized by megakaryocytes, e.g. vWF and P-selectin. Using CD34+ex vivo-derived megakaryocytes as a model, mechanisms defining megakaryocyte endocytosis of plasma-derived factor V and its intracellular trafficking to α-granules are being studied. Expression of the well-known megakaryocyte proteins αIIb, β3, vWF, and P-selectin increased in parallel concomitant with the loss of CD34 and increased cellular differentiation. Expression of these proteins was apparent within 4 days of cell culture and was maximal by day 10. In contrast, endocytosis of factor V by cells at earlier stages of megakaryocyte differentiation (day 7) occurred in the absence of αIIb and vWF expression. By day 10 in culture, all cells endocytosing factor V also expressed αIIb and vWF. Fibrinogen endocytosis occurred with the concomitant expression of αIIb/β3. The colocalization of fluorescently-labeled factor V and fibrinogen with various organelle-specific, fluorescent antibodies was determined by confocal microscopy using day 10 or 11 ex vivo-derived megakaryocytes. After a 1 hr endocytosis period, 87.8 ± 7.2% of the factor V colocalized with an antibody to Rab5, an early endosomal marker. Colocalization decreased over time such that only 5% of the Rab5 colocalized with factor V at 4 hrs. Endocytosed factor V also colocalized in a time-dependent manner with an antibody to GM130, a cis-Golgi element, consistent with the hypothesis that endocytosed, plasma-derived factor V is structurally modified to generate the unique platelet-derived molecule. After a 2 hr endocytosis period, 53.2 ± 23.0% of the GM130 specific antibody colocalized with factor V. Colocalization decreased 5-fold by 4 hrs. In contrast, little if any of the GM130 antibody colocalized with endocytosed fibrinogen, while 〉80% of the cis-Golgi element colocalized with vWF. After 19 hrs, substantial colocalization was also observed between endocytosed factor V and vWF (84.2 ± 3.7%) or P-selectin (54.8 ± 3.5%), which is consistent with their storage in α-granules and confirms flow cytometric analyses. The colocalization of endocytosed factor V and fibrinogen was also determined over time. These proteins colocalized quickly (1 hr) likely due to their uptake into early endosomes as suggested by previous studies in a megakaryocyte-like cell line. Colocalization reached a maximum and plateaued after endocytosis periods ≥ 8 hrs again consistent with their ultimate storage in α-granules. At endocytosis periods 〈 8 hrs, less factor V colocalized with fibrinogen consistent with the localization of factor V in the cis-Golgi network at these times. In conclusion, these combined observations suggest that following its endocytosis by megakaryocytes, factor V is taken up into early endosomes and trafficked through the Golgi complex prior to its storage in α-granules. Its transport through the Golgi is consistent with previous observations that platelet-derived factor V contains an O-linked glycosylation not found in its plasma counterpart.

Description: Abstract 4022 Poster Board III-958 Different clinical phenotypes are often observed among individuals with hemorrhagic or thrombotic disorders despite the presence of the same mutation and/or factor levels suggesting that additional genetic and/or environmental factors influence clinical presentation. As thrombin plays a central role in hemostasis, factors that affect an individual's ability to generate thrombin may lead to an increased risk of hemorrhage or thrombosis. The goal of this study was to assess procoagulant platelet formation in individuals with hemophilia A using a rapid, whole blood flow cytometric assay of prothrombinase complex assembly on platelets described previously. In this assay, Ca2+-dependent factor Xa binding to activated platelets is used as a marker of thrombin generating potential as it has been shown previously to correlate with platelet prothrombinase activity in a washed platelet system. Procoagulant platelet subpopulation formation in 15 men with varying degrees of factor VIII-deficiency was evaluated and compared to two independent measures of hemostatic competence: a 5-year mean bleeding score and whole blood clot time. In these individuals, the % activated platelets binding factor Xa in whole blood varied from 2.35 – 9.0% (n = 30), which is consistent with what is observed in unaffected individuals (1.5 – 41.5%, n = 136). Bleeding was scored independently by two experienced hemophilia nurses and one hematologist at Centre Hospitalier Universitaire Sainte Justine based on each individual's bleeding history, including hemarthroses, soft tissue hematoma, and annual factor VIII usage, and averaged. The 5-year mean bleeding scores in these individuals ranged from 0 – 20.4. Linear regression analyses indicated that in this population the % activated platelets binding factor Xa correlated indirectly with the 5-year mean bleeding score, where individuals with lower bleeding scores (i.e. less bleeding episodes and/or factor VIII usage) generated larger procoagulant platelet subpopulations. The time to clot formation in a tissue factor-dependent, contact pathway-suppressed whole blood clotting assay described previously, was also determined. The clot time, which was determined visually, and ranged from ∼3 – 7 min, also correlated indirectly, though less well, with the % activated platelets binding factor Xa in whole blood (r = -0.23). Thus, consistent with what was observed with the bleeding score, those individuals who clotted more quickly (i.e. exhibited a greater degree of hemostatic competence) generated more proacoagulant platelets. Platelet procoagulant subpopulation formation was also compared to other hematological measures. An inverse correlation was observed between the % activated platelets binding factor Xa and mean platelet volume (r = -0.347), suggesting that in this population, individuals with smaller platelets (i.e. platelets with fewer prothrombinase binding sites) may generate a greater number of platelets capable of assembling prothrombinase. In contrast, no correlation was observed between whole blood platelet number and % activated platelets binding factor Xa. Interestingly, factor Xa binding was also negatively correlated with plasma levels of factor IX (r = -0.56) and factor V (r = -0.62). Platelet subpopulation formation was also weakly and negatively associated with the aPTT (r = - 0.28). In this small population of individuals with hemophilia A, whole blood platelet factor Xa binding correlates with bleeding phenotype. These observations support the notion that this measurement can be used to predict an individuals propensity to bleed or clot. Disclosures: No relevant conflicts of interest to declare.

Description: Tissue factor (TF) is an integral membrane protein, which is the key initiator of blood coagulation in vivo. Due to the limited availability of natural TF, recombinant proteins of various lengths and origins have been extensively used in research and clinical laboratories worldwide. Experimental results acquired with recombinant TF proteins are frequently used for the understanding of the coagulation processes occurring in vivo, although there is a lack of data confirming the structural and functional identity of natural TF proteins from various sources and recombinant ones. In the current study, human TF from cultured monocytes and purified from placenta were compared with three different species of recombinant TF: 1–218 (extracellular domain only), 1–242 (lacking cytoplasmic domain) and 1–263 (full-length). Anti-TF mAbs gave 93–98% inhibition of TF activity for all TF species tested, in both natural and relipidated preparations. It was established that purified placental TF has a higher affinity for factor VIIa (Kd 0.13 nM) than recombinant counterparts 1–242 and 1–263 (Kd 0.50–0.80 nM). Similarly, placental TF is more efficient in factor X activation by the extrinsic Xase than recombinant TF 1–242 (the second order rate constants are 3.0x107 and 0.7x107 M−1s−1, respectively). We explored the use of these TF species as well as monocyte TF (purified/relipidated and present on LPS-stimulated monocytes) for the initiation of thrombin generation in two in vitro models of blood coagulation. At equimolar concentrations (5 pM; determined by immunoassay), when evaluated in synthetic plasma reconstituted with 2x108/ml platelets, recombinant TF 1-263 provided an initiation phase of ~4 min. Placental TF and relipidated monocyte TF had similar profiles of thrombin generation with an initiation phase of ~3 min. In contrast, 0.5 pM TF on LPS-stimulated monocytes gave an initiation phase of ~1 min. Even at 0.05 pM concentration, monocyte TF was as active as any relipidated protein at 5.0 pM. A similar pattern of relative TF activity was observed in whole blood and plasma PT clotting assays. TF on stimulated monocytes gave the highest activity, exceeding that of any relipidated protein by 100–200-fold. Recombinant TF 1–242 was more active than recombinant TF 1–263 and placental TF in the PT assay but less active in synthetic plasma and whole blood. The lowest overall activity was observed for relipidated monocyte TF. Our data suggest that TF proteins from different sources are different with respect to their functional properties.