Thrombin-induced platelet aggregation is mediated by a platelet plasma membrane-bound lectin.

Throbin-activated human platelets cause agglutination of trypsinized, formalinized bovine erythrocytes. This lectin activity of stimulated platelets was blocked by galactosamine, glucosamine, mannosamine, lysine, and arginine, but not by N-acetylated sugars, other neutral sugars, or other amino acids. Inhibitors of the thrombin-induced lectin activity also blocked thrombin-induced platelet aggregation. It appears that a membrane surface component that has lectin activity mediates platelet aggregation.

A novel method for purification of plasma fibronectin.

Plasma fibronectin was purified from a gelatin-affinity chromatography column by elution with glucose. This procedure was effective only if the gelatin was particulate when it was attached to the Sepharose 4B. Glucose could not elute fibronectin from the gelatin if the gelatin was melted before it was attached to the Sepharose 4B. This new purification technique has the advantage of using very mild conditions for the isolation of plasma fibronectin.

AlphaIIbbeta3-mediated outside-in signaling induced by the agonist peptide LSARLAF utilizes ADP and thromboxane A2 receptors to cause alpha-granule secretion by platelets.

The peptide LSARLAF (LSA) causes alphaIIbbeta3-dependent platelet activation that results in alpha-granule secretion and aggregation. LSARLAF-induced, alphaIIbbeta3-mediated outside-in signaling causing alpha-granule secretion and platelet aggregation was studied using washed mouse platelets. ADP receptor antagonists, enzyme inhibitors, normal platelets and platelets from mice that lack either Galphaq or thromboxane (Tx) A2 receptors were used for this investigation. The results demonstrate that LSA-induced alphaIIbbeta3-mediated signaling producing aggregation of washed platelets is mediated through the release of ADP and thromboxane, which cause alpha-granule release by mediating their effects though Galphaq and/or Gi depending on the level of LSA used to activate the platelets. Specifically, alphaIIbbeta3 elicited aggregation of washed platelets in response to a low level of LSA requires signaling through the ADP receptor P2Y1 and Galphaq, and the ADP receptor P2Y12 and Gi as well as TxA2 receptors. However, this aggregation is independent of Galphaq and TxA2 signaling in response to high LSA concentrations, but is dependent on ADP signaling through its receptor P2Y12, and therefore presumably Gi, regardless of the level of LSA used to activate the platelets. PKC function is required for ADP secretion and the subsequent signaling through P2Y12 regardless of the level of LSA used to activate the platelets. The end point of the LSA-induced alphaIIbbeta3-mediated signaling characterized in this study is alpha-granule secretion, which provides the fibrinogen required for aggregation of washed platelets.

The roles of ADP and TXA in botrocetin/VWF-induced aggregation of washed platelets.

BACKGROUND: Binding of von Willebrand factor (VWF) to the platelet membrane glycoprotein (GP) Ib-IX-V complex initiates a cascade of events leading to alphaIIbbeta3 activation and platelet aggregation. The roles of ADP and thromboxane A2 (TXA2) in agglutination-induced GPIbalpha-mediated platelet activation have not been fully described. METHODS: Botrocetin and human VWF were used to stimulate washed mouse platelets. Platelets deficient in TXA2 receptors, Galphaq, or alphaIIbbeta3, and inhibitors and chelating agents were used to investigate the roles of TXA2, ADP, alphaIIbbeta3 and Ca2+ in botrocetin/VWF-induced signaling. RESULTS: Our data demonstrate that botrocetin/VWF/GPIbalpha-mediated agglutination results in calcium-independent protein kinase C (PKC) and phospholipase A2 (PLA2) activities required for GPIbalpha-elicited TXA2 production that in turn causes dense granule secretion. Aggregation of washed platelets requires TXA2-induced alphaIIbbeta3 activation and ADP signaling. TXA2 or ADP can activate alphaIIbbeta3, but both are required for alpha-granule secretion and aggregation. Botrocetin/VWF-induced dense granule secretion is Galphaq-dependent. alpha-Granule secretion requires initial ADP signaling through P2Y1 and subsequent signaling through P2Y12. Signaling initiated by agglutination is propagated and amplified in an alphaIIbbeta3-dependent manner. CONCLUSIONS: In contrast to adhesion or shear stress-induced GPIb-elicited signaling, agglutination-elicited GPIb signaling that activates alphaIIbbeta3 requires TXA2. Agglutination-elicited TXA2 production is independent of Ca2+ influx and mobilization of internal Ca2+ stores. Therefore, our results demonstrate that agglutination-elicited GPIb signaling causes alphaIIbbeta3 activation by a mechanism that is distinct from those used by adhesion, or shear stress-induced GPIb signaling.

Platelet-aggregation is stimulated by lactose-inhibitable snake venom lectins.

Five lactose-specific lectins from snake venoms were tested for the ability to stimulate the aggregation of human platelets. Three of the lectins, bushmaster (Lachesis muta), cottonmouth (Ancistrodon piscivorous leukostoma) and rattlesnake (Crotalus atrox) lectins, consistently stimulated secretion and aggregation. Thrombolectin (Bothrops atrox) occasionally caused aggregation. Copperhead (Agkistrodon contortrix contortrix) lectin did not by itself cause platelet aggregation. Lactose, a specific inhibitor of hemagglutination mediated by these lectins was a potent inhibitor of lectin-induced aggregation. Antiserum specific for bushmaster lectin inhibited aggregation by bushmaster lectin. In contrast, the same antiserum and anti-cottonmouth lectin serum enhanced aggregation by low levels of the other lectins. A variety of substances were assayed in the aggregometer for the ability to inhibit aggregation in response to these lectins. Both secretion and aggregation were inhibited by PGI2 and PGE1. Furthermore, lectin-induced aggregation was completely blocked by trifluoperazine and partially blocked by indomethacin. Monoclonal antibodies specific for GP IIb/IIIa (AP2, A2A9, LJP5, LJCP8) but not monoclonals directed against other platelet membrane proteins (AP1 and AP3) inhibited lectin-induced aggregation. The peptide Arg-Gly-Asp-Ser but not Arg-Ala-Asp-Ser was a potent inhibitor of aggregation.

Effect of anti-thrombospondin antibodies on the hemagglutination activities of the endogenous platelet lectin and thrombospondin.

The proposal that thrombospondin is the endogenous platelet lectin was evaluated using antisera and monoclonal antibodies to thrombospondin. The platelet-bound hemagglutinin activity of human platelets stimulated with A23187 was inhibited by rabbit anti-thrombospondin sera and by a monoclonal anti-thrombospondin IgG. A second monoclonal IgG did not inhibit platelet-bound agglutinin activity. Preparations of purified platelet thrombospondin differed in their hemagglutination activities. The hemagglutination activity of an active preparation of thrombospondin was inhibited by the monoclonal antibody that inhibited platelet-bound lectin activity. The hemagglutination activity of an almost inactive preparation of thrombospondin was enhanced by the anti-thrombospondin monoclonal antibody that did not block platelet-bound lectin activity. The results demonstrate that expression of the platelet-bound form of the endogenous lectin is thrombospondin-dependent and suggest that thrombospondin must become part of a larger complex, either by binding to the platelet surface or by becoming aggregated in solution, before hemagglutination activity can be expressed.

Evidence for new endothelial cell binding sites on fibrinogen.

In vitro assays were used to characterize adhesion of human aortic, microvascular and umbilical vein endothelial cells to various forms of immobilized fibrinogen. All three types of endothelial cells adhered to fibrinogen in a manner that was independent of the Aalpha-chain 572-574 RGD cell binding site. In fact, all three adhered to a fragment of the molecule which is composed of only one D domain (D1) of fibrinogen. A time course study revealed that extensive adhesion of endothelial cells on the ligand coated surface occurred between one and two hours incubation. The anti-fibrinogen gammaA-chain monoclonal antibody 4A5 as well as 4A5 Fabs, blocked adhesion of endothelial cells to fibrinogen, not vitronectin. The inhibitory effects of 4A5 seemed to be indirect because the endothelial cells adhered to the recombinant fibrinogen gamma407 (which lacks the gamma-chain AGDV sequence of the carboxyl terminal 4A5 binding site) as well as they did to normal recombinant fibrinogen. A recombinant fibrinogen lacking the gamma-chain AGDV sequence, containing RGE in place of RGD at the gamma-chain 572-574 and 95-97 positions, also supported endothelial cell adhesion. The anti-alphavbeta3 antibody, LM609, blocked adhesion of endothelial cells to fibrinogen. The peptide GRGDSP inhibited endothelial cell adhesion on fibrinogen and vitronectin. These results demonstrate that alphavbeta3 mediated adhesion (attachment and spreading) of HUVECs to fibrinogen may use a site in the D domain of fibrinogen and is not dependent on the Aalpha-chain RGD (95-97 and 572-574) sequences, as has been shown in shorter term (where cells were rounded) experiments, or the alphaA-chain 408-411 cell binding sites. Thus, the data reveal the existence of another unidentified site(s) on fibrinogen which can support the irreversible adhesion (attachment and spreading) of endothelial cells.

Distinct domains of alphaIIbbeta3 support different aspects of outside-in signal transduction and platelet activation induced by LSARLAF, an alphaIIbbeta3 interacting peptide.

The peptide LSARLAF causes alphaIIbeta3-dependent platelet activation exemplified by secretion, aggregation, spreading and adhesion on fibrinogen, and tyrosine phosphorylation. alphaIIIbeta3-dependent outside-in signal transduction induced by LSARLAF was investigated in variant thrombasthenic platelets which lack most of the cytoplasmic domain of the integrin beta3 subunit (alphaIIbbeta3 delta724). These studies revealed that only certain aspects of this alphaIIbbeta3-dependent outside-in signaling were affected by the beta3 truncation. Specifically, alphaIIbbeta3 delta724 supported LSARLAF-induced platelet aggregation, agglutination and secretion, but failed to trigger cytoskeletal reorganization and platelet spreading on fibrinogen, despite the fact that PMA-induced non alphaIIbbeta3 mediated signaling caused spreading of these platelets on fibrinogen. Thus, distinct domains of alphaIIbbeta3 are required to support different aspects of LSARLAF-induced platelet activation. Furthermore, these studies suggest that not all alphaIIbbeta3-dependent platelet responses require an intact beta3 cytoplasmic tail.

Isolation and characterization of lactose-binding lectins from the venoms of the snakes Lachesis muta and Dendroaspis jamesonii.

Two lectins have been isolated: one from the venom of Lachesis muta (bushmaster lectin) and one from Dendroaspis jamesonii venom (Jameson's mamba lectin). The lectin from bushmaster venom (BML) is similar to the lactose-binding lectins previously isolated from snake venoms (Gartner et al. (1980) FEBS Lett. 117, 13-16; Gartner & Ogilvie (1984) Biochem. J. 224, 301-307) in that it is calcium-dependent, lactose inhibitable, and is a dimer of molecular weight 28,000. In contrast, the lactose-blockable lectin from Jameson's mamba venom (JML) has an apparent molecular weight of 26,000 and agglutinates erythrocytes in the presence of EDTA. The absorption spectra of BML were affected by the binding of calcium, or calcium and lactose to the lectin. However, JML spectra were not affected by these conditions. While the hemagglutination activity of each of the previously described lactose-binding snake venom lectins is inhibited by reducing agent, the activities of BML and JML are not affected by reducing agent. Antiserum against bushmaster lectin cross-reacts with thrombolectin, cottonmouth lectin (CML), rattlesnake lectin (RSL), and copperhead lectin (CuHL) but not lectin from Jameson's mamba venom. This evidence plus a comparison of atomic absorption spectra, isoelectric points and amino acid analyses of the lectins demonstrate that JML and BML are different from thrombolectin, CML, RSL, and CuHL.

Peptides and monoclonal antibodies which bind to platelet glycoproteins IIb and/or IIIa inhibit clot retraction.

The rate of clot retraction in platelet-rich plasma was decreased by synthetic peptide analogues of fibrinogen and monoclonal antibodies each of which bind to the platelet plasma membrane glycoproteins IIb and/or IIIa. These and related data demonstrate that intact complexes of the glycoproteins IIb and IIIa are required for platelet-mediated clot retraction.

Secreted platelet thrombospondin binds monovalently to platelets and erythrocytes in the absence of free Ca2+.

Washed human platelets suspended in Ca2+-free buffer bind thrombospondin secreted in response to stimulation by the calcium ionophore A23187. Under these conditions, the secreted thrombospondin binds to the surface of the platelets monovalently, that is, the thrombospondin does not agglutinate the platelets. In addition, the secreted thrombospondin can bind monovalently to the surface of the erythrocytes used to assay the endogenous lectin of human platelets. These findings resolve the contradictions resulting from the apparent requirement for free Ca2+ in the binding of secreted thrombospondin to the plasma membranes of platelets, the behavior of purified thrombospondin in hemagglutination assays and the characteristics of the endogenous lectin (thrombospondin) expressed by platelets stimulated with A23187 or gamma-thrombin.

TFP inhibits the expression of the endogenous platelet lectin and secretion of alpha-granules.

Activated platelets that have not undergone secretion do not express the endogenous platelet lectin whereas activated platelets that have undergone secretion do express this lectin. Expression of the lectin, secretion of beta-TG and PF-4, and secondary aggregation are inhibited by TFP. These results demonstrate that expression of the endogenous platelet lectin is apparently secretion dependent. Our results also reveal that inhibition of platelet aggregation by TFP apparently is not caused by inhibition of secretion and that there are TFP sensitive and TFP insensitive mechanisms of secretion.

The amino acid sequence Gly-Ala-Pro-Leu appears to be a fibrinogen binding site in the platelet integrin, glycoprotein IIb.

The amino acid sequence Gly-Ala-Pro-Leu-Arg-Val is predicted by the anticomplementarity hypothesis to be a fibrinogen binding site on human platelet fibrinogen receptors. The peptide Ala-Pro-Leu-Arg-Val binds fibrinogen and inhibits platelet aggregation and clot retraction. The peptide Gly-Ala-Pro-Leu is the shortest sequence within the predicted sequence which potently inhibits the adhesion of platelets to fibrinogen and platelet aggregation. The sequence Gly-Ala-Pro-Leu is present as residues 309-312 in glycoprotein IIb, the alpha-subunit of the glycoprotein IIb/IIIa complex, the fibrinogen receptor. The sequence Gly-Ala-Pro-Leu is present in 4 of 8 integrin alpha-subunits and Gly-Ala-Pro is present in 8 of 8 integrin alpha-subunits.

Inhibition of platelet aggregation and endogenous lectin activity by oligoamines.

Amino sugars and basic amino acids inhibit platelet aggregation and the activity of the endogenous platelet lectin, yet, relatively high concentrations (approximately 30 mM) are required for the inhibition. If cooperative interactions are involved in these platelet surface activities, oligomers of primary amines should be more potent inhibitors than their individual component amines. Accordingly, series of oligomers of basic amino acids, of the polyamines (putrescine, spermidine and spermine) and of aliphatic diamines differing in chain length were tested for potency of inhibition of platelet aggregation and endogenous platelet lectin activity. Indeed, oligoamines were much more potent inhibitors of platelet aggregation than their corresponding monomers or shorter oligomers, more than accountable by an additive effect. For example, 60, 3 and 0.3 mM were needed for 50% inhibition of platelet aggregation by lysine, (lys)3 and (lys)5, respectively. A similar pattern was observed for the effect of the oligoamines on the activity of the endogenous platelet lectin. The inhibition of platelet aggregation by spermine is competitive, since the effect of a given dose of spermine decreased with increasing platelet concentration. Neither inhibition of platelet-inducer interaction nor Ca2+ insufficiency explain the inhibitory effects of the oligoamines. The results are consistent with the hypothesis that cooperative surface interactions underlie platelet aggregation and platelet lectin activity. The cooperative effects may reflect the formation of patches or clusters of positively charged groups on the surface of activated platelets.

Peptide LSARLAF activates alpha(IIb)beta3 on resting platelets and causes resting platelet aggregate formation without platelet shape change.

Adhesion of resting platelets to fibrinogen was enhanced by a peptide which was designed to bind near the presumptive fibrinogen gamma-chain binding site of the alpha subunit of the integrin alpha(IIb)beta3. This peptide, but not a scrambled control peptide, induced adhesion of resting platelets to fibronectin, vitronectin, von Willebrand factor, and monovalent (lacks one functional gamma-chain) fibrinogen. Resting platelets not treated with the agonist peptide did not adhere to these ligands. Agonist peptide induced adhesion of resting platelets to Fg was not secretion dependent and was inhibited by the monoclonal antibody 7E3. The agonist peptide caused aggregation of resting platelets on resting platelets adherent to immobilized Fg without causing platelet shape change. Therefore, the agonist peptide may activate alpha(IIb)beta3 by directly inducing a conformation change in the receptor on resting platelets.

External Na+ is not required for Ca2+ mobilization in platelets stimulated with rattlesnake lectin or alpha-thrombin.

The effects of extracellular sodium on platelet aggregation and calcium mobilization in platelets stimulated with either rattlesnake (Crotalus atrox) lectin (RSL) or alpha-thrombin were compared. The absence of extracellular sodium had no effect on platelet aggregation or calcium mobilization in response to all levels of RSL tested. In contrast platelet aggregation was sodium-dependent in response to less than or equal to .2 units/ml alpha-thrombin. Surprisingly, calcium mobilization occurred in platelets treated with a threshold level of alpha-thrombin in the absence of external sodium. Thus sodium-dependent platelet aggregation in response to a low dose of thrombin apparently is not the result of sodium-dependent calcium mobilization.

Characterization of adhesion of "resting" and stimulated platelets to fibrinogen and its fragments.

Adhesion of resting and stimulated platelets to immobilized fibrinogen (Fg) was characterized using various forms of Fg, receptor peptide mimics, and antibodies to glycoprotein (GP) IIb/IIIa and Fg. Resting platelets adhered to Fg, but to less than half the extent of the same platelets stimulated with epinephrine/ADP. The adhesion of resting and stimulated platelets to Fg was inhibited by a receptor peptide mimic (G13, a peptide corresponding to residues 300-312 of GPIIb), anti-GPIIb/IIIa antibodies, and a monoclonal antibody (4A5) against the carboxyl terminus of the gamma chain of Fg. The results presented here demonstrate that the alpha chain RGD platelet recognition sites are not required to mediate the adhesion of either stimulated or resting platelets to immobilized Fg. Although stimulated platelets can adhere extensively to monomeric Fg containing one functional gamma chain, resting platelets require bivalent Fg containing two functional gamma chains to mediate irreversible adhesion to Fg.

Characterization of adhesion of non-exogenously stimulated and resting platelets in normal plasma to fibrinogen and its fragments.

Platelet adhesion to various forms of fibrinogen was studied using platelets in plasma and washed platelets. The study was designed to determine if platelets prepared with minimal handling in plasma at physiological pH containing normal levels of Ca2+ have different requirements for adhesion to immobilized fibrinogen than do washed platelets tested in the absence of plasma. Exposure of platelets to citrate and low pH did not seem to affect the requirements of washed platelets for adhesion to fibrinogen. Nonetheless, behavioural differences between these two types of platelets were seen. Surprisingly, in the absence of exogenous activation normal platelets in plasma behaved qualitatively as stimulated washed platelets. That is, both types of platelets adhered to all forms of fibrinogen which possessed at least one gamma-chain carboxyl terminal platelet binding site. Platelets in plasma treated with prostaglandin E1 (resting platelets) adhered only to forms of fibrinogen which contained two gamma-chain platelet binding sites. These observations also demonstrate that the fibrinogen alpha-chain arginine-glycine-aspartic acid-phenylalanine and arginine-glycine-aspartic acid-serine sequences are not necessary or sufficient to mediate the adhesion of resting or stimulated platelets in plasma to fibrinogen. The presence of endogenous adenosine diphosphate appears to account, at least in part, for the ability of normal platelets in plasma to adhere to forms of fibrinogen which have only one gamma-chain platelet binding site.