Activation of blood platelets after percutaneous transluminal coronary angioplasty and coronary artery bypass graft surgery.

We have evaluated the activation of platelets in blood samples taken from patients with stable angina undergoing balloon angioplasty (percutaneous transluminal coronary angioplasty [PTCA]) (n=11) or coronary artery bypass grafting (CABG) under hypothermic (n=11) or normothermic conditions (n=11). We have found that surface expression of P-selectin on platelets in whole blood from PTCA patients upon thrombin treatment was significantly reduced, as compared with control platelets from healthy subjects. This effect was partially reversed when platelets washed from the same blood sample were used, but even then P-selectin expression was significantly lower in PTCA patients than it was in control subjects. There was a significant increase in basal expression of P-selectin in blood platelets taken from patients who underwent CABG under normothermic conditions (warm blood cardioplegia) as opposed to hypothermic patients (cold crystalloid cardioplegia). These platelets retain the ability to respond to agonists, although to a much lower extent than do those from healthy control donors. The surface exposure of P-selectin on resting and thrombin-treated platelets isolated from CABG surgery patients was not different from that of the control platelets. The adhesion to fibrinogen of resting and thrombin-treated platelets from patients who underwent balloon angioplasty as well as CABG surgery under normothermic and hypothermic conditions was significantly reduced when compared with the fibrinogen of the control platelets. These results suggest that the function of platelet fibrinogen receptor is impaired in patients with stable angina pectoris and that PTCA and CABG surgery activates platelets.

Comparison of platelet aggregability and P-selectin surface expression on platelets isolated by different methods.

Three methods commonly used for isolation of blood platelets from plasma were compared. Platelets were isolated by: 1) a washing method; 2) a method of metrizamide-gradient centrifugation; 3) a modified method of gel-filtration. The last method employed BSA-Sepharose gel instead of routinely used Sepharose gel saturated with BSA. BSA-Sepharose gel was prepared by covalent binding of thermally deactivated BSA to CNBr-activated Sepharose 2B. In contrast to platelets isolated by the other methods, an aggregability of the gel-filtered platelets and control platelets in plasma, both activated with ADP, were comparable. When expression of P-selectin on the surface of freshly isolated platelets was examined, the gel-filtered platelets exhibited the same extent of fluorescence signal as platelets in the citrated blood, whereas platelets isolated by the other methods exhibited twice the extent of the signal. The methods involving the centrifugation process cause a low but a significant platelet activation.

Activity of pp60c-src and association of pp60c-src, pp54/58lyn, pp60fyn, and pp72syk with the cytoskeleton in platelets activated by collagen.

Collagen stimulation of platelets induced an increase in the specific activity of pp60c-src immunoprecipitated from the Triton-soluble fraction. The earliest time after collagen stimulation that an increase in pp60c-src activity was observed was 30 s. However, the maximum activity of pp60c-src in the Triton-soluble fraction was observed 60 s after collagen stimulation. At this time an approximately twofold increase of pp60c-src activity towards phosphorylation of KVEKIGEGTYGVVKK specific peptide and enolase and a 4.5-fold increase towards phosphorylation of pp60c-src itself was measured. Furthermore, the majority of pp60c-src as well as pp54/58lyn, pp60fyn, and pp72syk were found in the Triton-soluble fraction in resting platelets. Collagen induced, to different extents and velocities, translocation of all of these proteins from the Triton-soluble fraction to the Triton-insoluble, cytoskeleton-rich, platelets fraction. These results provide direct evidence that collagen stimulation of platelets increases the tyrosine kinase activity of pp60c-src and suggest that the platelet cytoskeleton plays an important role in collagen-induced signal transduction by localizing signaling molecules.

Tyrosine phosphorylation events during different stages of collagen-platelet activation.

Three groups of phosphoproteins have been distinguished, basing on the velocity and extent of phosphorylation in platelets stimulated with collagen. pp60c-src constituted the first group; the increase in its phosphorylation was the highest and most rapid (maximal in 30 s after the addition of collagen). pp80/85 and non-identified protein of 65 kDa formed the second group; the increase in their phosphorylation was twice smaller than that of pp60c-src, and reached its maximum 60 s after the addition of collagen. pp120, pp72syk, and two non-identified phosphoproteins of 90 and 75 kDa constituted the third group; the increase in their phosphorylation was 4-10-fold lower than that of pp60c-src and reached its maximum after 180 s. We conclude that the phosphorylation of pp60c-src is important for the change of shape of platelets, the phosphorylation of pp80/85 and pp65 for the initiation of the formation of aggregates and the phosphorylation of the third group of phosphoproteins for the formation of massive aggregates. This conclusion was supported by using a monoclonal anti-GPIb antibody, which did not inhibit the shape change of platelets and did not inhibit pp60c-src phosphorylation. This antibody inhibited aggregate formation as well as tyrosine phosphorylation of proteins belonging to the second and the third group of phosphoproteins.

Association of pp60c-src with alpha IIb beta 3 in resting platelets.

To detect whether 125I-alpha IIb beta 3 is associated with tyrosine kinases in platelets, antibodies specific to pp60c-src, pp54/58lyn, and pp62Fyn were used to precipitate their homologous antigens. In contrast to Lyn and Fyn kinases, pp60c-src appears to be complexed with alpha IIb beta 3. Both proteins, pp60c-src and alpha IIb beta 3, coprecipitated when antibodies to pp60c-src were used in the immunoprecipitation experiments. This conclusion was further supported by immunoprecipitation of alpha IIb beta 3 from Triton X-100 extracts of nonlabelled platelets with P2 antibodies. There was no pp60c-src detectable in immunoprecipitates obtained with antibodies specific to alpha 2 beta 1 or GPIb. Since PGE1 was used to prevent platelet activation in buffers throughout all procedures and there was no phosphorylation of pp72syk we assume that the platelets were in the resting state. Therefore, we conclude that alpha IIb beta 3 and pp60c-src can form a complex in resting platelets suggesting that pp60c-src is directly involved in initiating the outside-in signaling cascades in blood platelets.

Differential effects of the tyrosine kinase inhibitors on collagen type 1-induced platelet aggregation and adhesion to this protein.

Herbimycin A, lavendustin A, and methyl 2,5-dihydroxycinnamate were used to study the role of protein tyrosine kinases in collagen-platelet interaction. All three compounds produced a concentration dependent inhibition of platelet aggregation induced by collagen type I, characterized by values of IC50 equaled to 0.9, 10.0, and 5.0 microM, respectively. This effect was accompanied by strong inhibition of phosphorylation of p125FAK, p90, p72syk, p60c-arc, and p56lyn. In the absence of the inhibitors, phosphorylation of these proteins is evoked by aggregation of platelets. In addition to the antiaggregatory effect, the tyrosine kinase inhibitors reduced adhesion of platelets to collagen although to much lower extent than aggregation. Platelets which adhered to collagen showed also the presence of phosphorylated p125FAK, p90, p72syk, p60c-arc, and p56lyn. Of these proteins, the extent of phosphorylation of p90 was particularly high. Adhesion of platelets was associated with inhibition of phosphorylation of p125FAK, p60c-arc, and p56lyn only when high concentration of lavendustin A and methyl 2,5-dihydroxycinnamate were used. Herbimycin A did not affect adhesion-evoked protein tyrosine phosphorylation. Phosphorylation of p90 and p72syk was not affected by inhibitors. This study indicates that collagen type I can induce different transmembrane signalling dependent upon whether platelet aggregates formation or adhesion of platelets to this protein occurs.

Role of tyrosine kinase enzymes in TNF-alpha and IL-1 induced expression of ICAM-1 and VCAM-1 on human umbilical vein endothelial cells.

The aim of this study was to analyse the potential roles of protein kinase enzymes in tumour necrosis factor-alpha (TNF-alpha) and interleukin-1 (IL-1) induced expression of the adhesion molecules intercellular adhesion molecule-1 (ICAM-1) and vascular cell adhesion molecule-1 (VCAM-1) on human umbilical vein endothelial cells (HUVEC). The authors observed a marked increase in ICAM-1 and VCAM-1 expression on HUVEC stimulated for 24 h by TNF-alpha (10 ng/ml) or IL-1 (20 ng/ml). Pre-treatment of HUVEC for 30 min with protein tyrosine kinase (PTK) inhibitors genistein and herbimycin A (10 micrograms/ml and 0.5 microgram/ml, respectively) before stimulation with IL-1 did not affect the expression of these molecules. Similar results were observed with respect to VCAM-1 expression on HUVEC stimulated by TNF-alpha. In contrast, pre-incubation of HUVEC with PTK inhibitors prior to the addition of TNF-alpha significantly enhanced subsequent expression of ICAM-1, although spontaneous expression of ICAM-1 on unstimulated HUVEC was unaffected. Western blot analysis demonstrated a significant increase in phosphorylated tyrosine protein levels in HUVEC stimulated by TNF-alpha, and significantly lower levels of these proteins in TNF-alpha stimulated HUVEC pre-treated with PTK inhibitors. These results demonstrate that IL-1 induced ICAM-1 and VCAM-1 expression does not result from activation of PTK-dependent pathways. In the case of TNF-alpha induced responses, the selective co-stimulatory effect of this cytokine in combination with PTK inhibitors on ICAM-1 expression suggests a complicated intracellular pathway of TNF-alpha induced ICAM-1 expression, possibly involving down-modulation of increases in ICAM-1 by PTK enzymes.

The immunosuppressory and adhesive miniregion of the human major histocompatibility protein, human leukocyte antigen DQ.

Our previous studies showed that the nonapeptide fragment of human leukocyte antigen DQ with the TPQRGDVYT sequence strongly suppresses the immune response [Z. Szewczuk, I. Z. Siemion, and Z. Wieczorek (1996) Molecular Immunology, 33, 903-9081]. The fragment contains the RGDVY sequence, which is very similar to thymopentin (pentapeptide RKDVY, an active fragment (32-36) of thymopoietin, an immune system activator produced in thymi), and at the same time contains the RGD sequence, known as an inhibitor of adhesion processes. In the present study we tested an influence of the nonapeptide and its shorter fragments on binding the activated platelets and K562 cells to fibrinogen and fibronectin, respectively. We also designed and synthesized a cyclic thymopentin-like peptide. C*RGDVYC* (where C* indicates Cys participating in disulfide bridge) to restrict its conformation. The cyclization product strongly suppresses the humoral and cellular immune response and selectively inhibits the adhesion of K562 cells to fibronectin. The results are discussed in the light of CD conformational studies.

Microenvironment changes in human blood platelet membranes associated with binding of tissue-type plasminogen activator.

Whole washed platelets were labelled with the free radicals [2-(14-carboxytetradecyl)-2-ethyl-4,4-dimethyl-3-oxazolidinylox y] (16-DOXYL-Ste) or [2-(3-carboxypropyl)-4,4-dimethyl-2-tridecyl-3- oxazolidinyloxy] (5-DOXYL-Ste) and incubated with recombinant tissue-type plasminogen activator (rt-PA). Changes in the membrane fluidity caused by rt-PA were detected by alterations in h+1/h0 calculated from the ESR spectra for 16-DOXYL-Ste and 5-DOXYL-Ste incorporated into the lipid bilayer (h+1 and h0 are the heights of the low-field and middle-field lines of the spectra, respectively). Interaction of rt-PA with both resting and stimulated platelets resulted in increased rigidity of the membrane lipid bilayer as indicated by the reduced value of h+1/h0. This phenomenon can be explained either by conformational changes of membrane receptors caused by the attachment of rt-PA and the subsequent rearrangement of the lipid matrix of platelet membranes, or by the direct association of rt-PA with membrane phospholipids and thus partial embedding of protein molecules into the lipid bilayer restraining lipid mobility.

Interaction of human platelets with laminin and identification of the 67 kDa laminin receptor on platelets.

A microtitre adhesion assay has been developed to define parameters affecting the adherence of washed platelets to laminin. Adherence was optimally supported by Mg2+ and was inhibited by Ca2+ and by anti-laminin Fab fragments, but significant adhesion (75-90% of control) was found both in heparinized plasma containing physiological levels of bivalent cations and in plasma anti-coagulated with EGTA. Adherence was unaffected by platelet activation with ADP but was decreased by 50% by treatment with alpha-thrombin (1 unit/ml, 5 min). Adherence was unaffected by monospecific polyclonal antibodies to glycoprotein (GP) Ib and GPIV, and was normal with platelets from two patients with Glanzmann's thrombasthaenia, indicating that GPIb, the GPIIb/IIIa complex and GPIV are not involved in platelet-laminin interaction. Affinity chromatography of Triton-solubilized membranes on laminin-Sepharose followed by elution with 0.2 M-glycine/HCl (pH 2.85) identified a major band with a molecular mass of 67 kDa in the reduced and of 53 kDa in the unreduced form. This protein gave a positive reaction on Western blotting with a monospecific polyclonal antibody raised against the high-affinity laminin receptor isolated from human breast carcinoma tissue. The adhesion of platelets to laminin was inhibited by two monoclonal IgM antibodies specific to the LR-1 domain of the 67 kDa receptor. The binding protein was surface-oriented, as shown by flow cytofluorimetry and by the fact that it could be iodinated in intact platelets, but it was not labelled by the periodate-borotritide procedure, suggesting that it did not contain terminal sialic acid. The laminin-derived peptides Tyr-Ile-Gly-Ser-Arg and Cys-Asp-Pro-Gly-Tyr-Ile-Gly-Ser-Arg-NH2, which constitute a complementary binding domain in laminin for the 67 kDa receptor, themselves supported platelet adhesion, bound to the receptor and inhibited the adhesion of platelets to laminin. In addition, Fab fragments of anti-Tyr-Ile-Gly-Ser-Arg antibody inhibited platelet adhesion to laminin. These results demonstrate that the high-affinity 67 kDa laminin receptor previously identified in a range of normal and transformed cells and its complementary Tyr-Ile-Gly-Ser-Arg binding domain play an important role in the interaction of platelets with laminin.

A platelet membrane glycoprotein (GP) deficiency in healthy blood donors: Naka- platelets lack detectable GPIV (CD36).

It has recently been shown that the Naka antigen, which is absent in 3% to 11% of Japanese blood donors, is expressed on platelet glycoprotein IV (GPIV; CD36) (Tomiyama et al, BLOOD, 75:684, 1990). In the present studies, flow cytometry was used to distinguish differences in the reactivity of Naka+ and Naka- platelets with both OKM5, a monoclonal antibody that recognizes an epitope on GPIV, and with polyclonal anti-GPIV antibody. OKM5 was also used to screen 871 platelet concentrates prepared from healthy US blood donors. Three of these showed markedly deficient binding of 125I-OKM5 or an incidence of 0.34%. Two of these donors were re-accessed and showed less than 1% binding of 125I-OKM5 as compared with 10,300 +/- 1,500 binding sites per platelet in controls (n = 4). Platelets from these two US donors were radiolabeled (125I, 3H) and compared with control platelets and with platelets from Japanese Naka+ and Naka- donors by crossed immunoelectrophoresis, protein blots, immunoprecipitation, and two-dimensional gel electrophoresis. GPIV could not be detected by any of these techniques in the Naka- platelets nor in the donors whose platelets showed deficient binding of OKM5. These results suggest that GPIV functions as an isoantigen rather than an alloantigen in immunizing Naka- platelet recipients. This is the first report of the absence of a major platelet membrane GP in healthy blood donors.

Identification of glycoprotein IV (CD36) as a primary receptor for platelet-collagen adhesion.

The role of glycoprotein IV (GPIV) in platelet activation processes has been examined by several different approaches: (i) Fab fragments of a monospecific polyclonal antibody to purified platelet GPIV (approximately 20 micrograms/ml) completely inhibited platelet shape change, aggregation, and secretion induced by collagen. Aggregation and secretion by ADP (but not shape change) and by epinephrine were also inhibited, but there was no effect on platelet activation induced by thrombin, arachidonate, or ionophore A23187. (ii) Purified GPIV was able to compete completely with membrane-bound GPIV to inhibit platelet activation induced by collagen, including shape change, but not in activation induced by any of the other platelet agonists. 50% inhibition of collagen-induced activation and secretion were obtained at GPIV concentrations of approximately 10 nM (1 micrograms/ml). (iii) Purified GPIV bound rapidly and reversibly to collagen Type I fibrils, and binding was not inhibited by adhesive proteins such as denatured collagen, fibronectin, fibrinogen, or von Willebrand factor. The direct binding of purified GPIV to collagen Type I fibrils fit best to a single site model with Kd 0.34 +/- 0.10 nM. (iv) Using a microtiter assay, platelet adhesion to collagen was shown to be inhibited by Fab fragments of monospecific polyclonal anti-GPIV antibodies, but adhesion to other adhesive proteins was unaffected. (v) When anti-GPIV was added at various times during adhesion the time dependence of inhibition was seen to be biphasic. Anti-GP antibody was able to reverse adhesion that occurred within the first 5-8 min and to inhibit adhesion occurring thereafter. These results demonstrate that GPIV mediates the early stages of platelet recognition by and attachment to collagen but that there may be a second GPIV-independent mechanism that mediates the subsequent anchorage of these adherent platelets.

Effect of DMSO and a freezing-thawing process on the 'ecto-ATPase' and AChE activities of human platelets.

The activity of two membrane-bound enzymes of human platelets subjected to DMSO and a freezing-thawing process were analysed. Following treatment of fresh platelets with DMSO (5-20%) quantitative differences of AChE and 'ecto-ATPase' activities were seen. After cryopreservation of platelets in 5% DMSO the enzymatic activities of AChE and Mg2+-ATPase were no different from those obtained for fresh platelets. The lack of any changes in the activities of the enzymes of frozen platelets subjected to a washing procedure to remove DMSO, indicates that the mechanism of the DMSO-induced effect is reversible and that freezing-thawing process had no additional detrimental effects.