Creatine kinase inhibits ADP-induced platelet aggregation.

Bleeding risk with antiplatelet therapy is an increasing clinical challenge. However, the inter-individual variation in this risk is poorly understood. We assessed whether the level of plasma creatine kinase, the enzyme that utilizes ADP and phosphocreatine to rapidly regenerate ATP, may modulate bleeding risk through a dose-dependent inhibition of ADP-induced platelet activation. Exogenous creatine kinase (500 to 4000 IU/L, phosphocreatine 5 mM) added to human plasma induced a dose-dependent reduction to complete inhibition of ADP-induced platelet aggregation. Accordingly, endogenous plasma creatine kinase, studied in 9 healthy men (mean age 27.9 y, SE 3.3; creatine kinase 115 to 859 IU/L, median 358), was associated with reduced ADP-induced platelet aggregation (Spearman's rank correlation coefficient, -0.6; p < 0.05). After exercise, at an endogenous creatine kinase level of 4664, ADP-induced platelet aggregation was undetectable, normalizing after rest, with a concomitant reduction of creatine kinase to normal values. Thus, creatine kinase reduces ADP-induced platelet activation. This may promote bleeding, in particular when patients use platelet P2Y12 ADP receptor inhibitors.

Complement activation on the surface of cell-derived microparticles during cardiac surgery with cardiopulmonary bypass - is retransfusion of pericardial blood harmful?

OBJECTIVES: To investigate whether cell-derived microparticles play a role in complement activation in pericardial blood of patients undergoing cardiac surgery with cardiopulmonary bypass (CPB) and whether microparticles in pericardial blood contribute to systemic complement activation upon retransfusion. METHODS: Pericardial blood of 13 patients was retransfused in 9 and discarded in 4 cases. Microparticles were isolated from systemic blood collected before anesthesia (T1) and at the end of CPB (T2), and from pericardial blood. The microparticles were analyzed by flow cytometry for bound complement components C1q, C4 and C3, and bound complement activator molecules C-reactive protein (CRP), serum amyloid P-component (SAP), immunoglobulin (Ig)M and IgG. Fluid-phase complement activation products (C4b/c, C3b/c) and activator molecules were determined by ELISA. RESULTS: Compared with systemic T1 blood, pericardial blood contained increased C4b/c and C3b/c, and increased levels of microparticles with bound complement components. In systemic T1 samples, microparticle-bound CRP, whereas in pericardial blood, microparticle-bound SAP and IgM were associated with complement activation. At the end of CPB, increased C3b/c (but not C4b/c) was present in systemic T2 blood compared with T1, while concentrations of microparticles binding complement components and of those binding complement activator molecules were similar. Concentrations of fluid-phase complement activation products and microparticles were similar in patients whether or not retransfused with pericardial blood. CONCLUSIONS: In pericardial blood of patients undergoing cardiac surgery with CPB, microparticles contribute to activation of the complement system via bound SAP and IgM. Retransfusion of pericardial blood, however, does not contribute to systemic complement activation.

Cell-derived microparticles and complement activation in preeclampsia versus normal pregnancy.

BACKGROUND: Inflammation plays a major role in the vascular dysfunction seen in preeclampsia, and several studies suggest involvement of the complement system. OBJECTIVES: To investigate whether complement activation on the surface of microparticles is increased in plasma of preeclamptic patients versus healthy pregnant controls. METHODS: Microparticles from plasma of preeclamptic (n=10), healthy pregnant (n=10) and healthy nonpregnant (n=10) women were analyzed by flow cytometry for bound complement components (C1q, C4, C3) and complement activator molecules (C-reactive protein [CRP], serum amyloid P component [SAP], immunoglobulin [Ig]M, IgG). Fluid phase complement activation products and activator molecules were also determined. RESULTS: Levels of microparticles with bound complement components showed no increase in complement activation on the microparticle surface in preeclamptic women, in line with levels of fluid phase complement activation products. In healthy nonpregnant and pregnant women, bound CRP was associated with classical pathway activation on the microparticle surface, and in healthy pregnant women IgM and IgG molecules also contributed. In preeclamptic women, microparticles with bound SAP and those with IgG seemed to contribute to C1q binding without a clear association to further classical pathway activation. Furthermore, significantly increased levels of microparticles with bound CRP were present in preeclamptic compared with healthy pregnant women (median 178x10(6)/L versus 47x10(6)/L, P<0.01), but without concomitant increases in complement activation. CONCLUSIONS: We found no evidence of increased complement activation on the microparticle surface in preeclamptic women. Microparticles with bound CRP were significantly increased, but in contrast to healthy pregnant and nonpregnant women, this was not associated with increased classical pathway activation on the surface of the microparticles.

Congenital deficiency of factor XIII caused by two missense mutations in a Dutch family.

We present the clinical, biochemical and genomic findings of a family with congenital factor XIII (FXIII) deficiency. Congenital FXIII deficiency is a very rare autosomal recessive bleeding disorder, characterized by umbilical cord bleeding at birth and spontaneous intracranial haemorrhage. Routine clotting tests are normal, which may delay the diagnosis, leading to an increased chance of severe sequelae. The propositus and her brother, known with haemorrhagic diathesis, were found to be compound heterozygous with a known missense mutation (1050 G --> T transversion in exon 7, Val316Phe substitution) and a novel mutation 889 G --> A in exon 6, which predicts a Gly262Glu substitution. As these mutations were known in the family, DNA obtained from cord blood of the youngest sister was analysed for mutations in exons 6 and 7 only. We postulate that the diagnosis was facilitated by determining the two different mutations in the genotype of this family. The analysis showed that she was heterozygous for the exon 7 mutation. Hence, she was not at risk of experiencing haemorrhagic diathesis. This diagnosis avoided the administration of FXIII concentrate to the newborn.

Two novel and one recurrent missense mutation in the factor XIII A gene in two Dutch patients with factor XIII deficiency.

Congenital factor XIII (FXIII) deficiency is a rare autosomal recessive disorder, usually attributed to a defect in the FXIII A subunit, whose genetic basis has been studied in a number of cases. We describe here the genetic variations found in two unrelated patients with FXIII deficiency. Both patients, under prophylactic substitution with FXIII concentrate, showed low plasma FXIII A subunit antigen levels with undetectable A subunit antigen in the platelets and normal plasma B antigen levels, which indicate that the defects are present in the A subunit of the molecule. Both probands were heterozygous for a previously reported G-->A transversion in exon 8 of the FXIII A subunit gene (Arg326Gln substitution). Proband 1 was also heterozygous for a novel G-->T transversion in exon 7, which predicts a Val316Phe substitution. Two of her sons were heterozygous for this mutation and showed low FXIII activity and FXIII A subunit antigen levels. Val316 is a well-conserved amino acid among the transglutaminase family, located within the core domain, close to the Cys314 member of the catalytic triad. Proband 2 had a unique 2-bp (TT) insertion in one of the alleles within or adjacent to the -7 to -20 T tail of intron A. This insertion was not found in 50 healthy individuals, which supports this being the second mutation in this patient.

C-reactive protein is a strong but nonspecific risk factor of fatal stroke in elderly persons.

An elevated level of C-reactive protein is a strong predictor of cardiovascular events in elderly persons. Whether C-reactive protein has direct adverse vascular effects or is a marker of aspecific systemic inflammation remains to be determined. The aim of this study was to investigate the relation between C-reactive protein and the occurrence of fatal strokes in elderly persons. In the Leiden 85-Plus Study, a population-based prospective follow-up study, we studied the levels of C-reactive protein in 80 participants who died from stroke within the first 5 years of follow-up. Levels of C-reactive protein were determined in serum samples at baseline. Levels of C-reactive protein were also determined in 82 control subjects who survived for the first 5 years of follow-up and in 83 participants who died from noncardiovascular causes. Mortality risks were estimated with logistic regression and adjusted for differences in age, sex, smoking, medication, total cholesterol, history of diabetes or hypertension, and previous cardiovascular events. Levels of C-reactive protein at baseline were 2-fold higher in subjects who died from stroke than in control subjects (median 5.7 versus 2.7 mg/L, P<0.005). The levels of C-reactive protein in subjects who died from stroke or from noncardiovascular causes were similar (median 5.7 versus 4.9 mg/L, P=0.7). The risk of death from stroke as well as from noncardiovascular causes increased linearly up to 10-fold in subjects with the highest levels of C-reactive protein at baseline (P<0.001). The levels of C-reactive protein were lower when more time had elapsed between blood sampling and time of death during follow-up (P=0.01). C-reactive protein is a strong but nonspecific risk factor of fatal stroke in old persons. The data do not support the idea that C-reactive protein has direct vascular effects that underlie fatal cerebrovascular disease.

Disappearance of glycoprotein Ib from the platelet surface in pericardial blood during cardiopulmonary bypass.

OBJECTIVES: Several investigators have reported decreased expression of glycoprotein Ib on the platelet surface during coronary artery bypass grafting, but others could not confirm this finding. Because platelet glycoprotein Ib functions as an adhesion receptor for von Willebrand factor and other adhesive proteins, this decreased expression may explain excessive postoperative blood loss. In this study the expressions of glycoprotein Ib and other platelet activation markers were studied in the systemic and pericardial blood of seven patients undergoing coronary artery bypass grafting. Pericardial blood was recently shown to have high activation levels of fibrinolytic and coagulation pathways; we hypothesized that this local blood activation might be paralleled by extensive platelet activation and associated disappearance of glycoprotein Ib. METHODS: Expression of platelet surface antigens was determined by whole-blood double-label flow cytometry. RESULTS: Glycoprotein Ib expression in systemic blood decreased 10% (p = 0.03) from preoperative levels at the start of cardiopulmonary bypass and 30% (p = 0.04) before release of the aortic crossclamp. Expression in pericardial blood at these times decreased by 50% and 51%, respectively (p = 0.003, p = 0.009). No changes were observed in the expression of the platelet activation antigens CD62P (P-selectin, indicating platelet alpha-granular release) and CD63 (indicating lysosomal release) or in binding of monoclonal antibody PAC-1 (detecting the fibrinogen-binding receptor conformation of the glycoprotein IIb-IIIa complex). CONCLUSION: Glycoprotein Ib disappeared from the platelet surface during bypass grafting, most notably in pericardial blood. No increased expression of CD62P, CD63, or PAC-1 was found, indicating the absence of general platelet activation.

Lorenzo's oil and platelet activation in adrenomyeloneuropathy and asymptomatic X-linked adrenoleukodystrophy.

X-linked adrenoleukodystrophy (X-ALD) is an inherited disorder of peroxisomal beta-oxidation, which results in accumulation of very long-chain fatty acids, causing damage to the nervous system, adrenal cortex and testis. The two most frequent phenotypes are childhood cerebral adrenoleukodystrophy (CCALD) and adrenomyeloneuropathy (AMN). Some affected males demonstrate no clinical signs (asymptomatic ALD), whereas female carriers can also be affected. Patients with X-ALD have been treated with Lorenzo's oil, a 4:1 combination of oleic acid and erucic acid, with thrombocytopenia as the main side effect and sometimes leading to a hemorrhagic diathesis. We studied platelet count, size and membrane surface exposure of platelet activation antigens in 17 adult X-ALD patients. Eight patients used the prescribed amount of erucic acid (as glyceroltrierucate) or more (very compliant), five used less(compliant), and four did not use the diet. All eight very compliant patients had highly enlarged platelets and seven manifested thrombocytopenia. An enhanced in vivo platelet activation status was established by increased platelet surface expression of P-selectin (CD62P, PADGEM, GMP-140) in five of the seven thrombocytopenic patients, and of increased fibrinogen receptor exposure (measured with the antibody PAC-1) in three of these five patients. The other nine compliant or untreated patients had normal platelet counts and, generally, normal P-selection and fibrinogen receptor expression. A diet-induced 7- to 27-fold enrichment of erucic acid was observed in the platelets of the four patients studied. We conclude that the thrombocytopenia in AMN patients using Lorenzo'soil is associated with circulating platelets that have an increased erucic acid content, size and activation status. We hypothesize that the erucic acid in some way induces the increased size and thus, directly or indirectly, increased platelet activation or instability in vivo. This then causes the thrombocytopenia, with circulating platelets representing a population that has not yet been sufficiently changed to be removed, but has clear signs of activation.

Can flow cytometric detection of platelet activation early in pregnancy predict the occurrence of preeclampsia? A prospective study.

OBJECTIVES: An increased platelet activation status is present in patients with preeclampsia. Our purpose was (1) to establish by means of flow cytometry whether platelets circulate in an activated state during the first and second trimesters of pregnancy and (2) to establish whether early platelet activation predicts the onset of preeclampsia. STUDY DESIGN: Consecutively, 244 pregnant women were included in a prospective study design. Platelets in whole blood samples from the pregnant women in the first trimester, the second trimester, and after delivery were labeled with the following antibodies associated with platelet activation: anti-CD62P (P-selectin, alpha-granule secretion), anti-CD63 (GP53, lysosomal secretion), anti-CD31 (GPIIa', platelet endothelial cell adhesion molecule-1). The surface antigen exposure was determined by double-label flow cytometry with anti-CD42b (GPIb, a platelet-specific monoclonal glycoprotein) to select platelets and platelet-derived materials. Preeclampsia was defined as a diastolic blood pressure > or = 90 mm Hg and proteinuria > or = 0.3 gm in a 24-hour urine sample (International Society for Study of Hypertension in Pregnancy criteria). RESULTS: Seventeen of 244 patients had preeclampsia (6.9%). Only first-trimester CD63 expression had an area under the curve > 0.5 by receiver-operator characteristic curve analysis and was selected as a possible predictor of preeclampsia. We found a sensitivity of 47% and a specificity of 76% with use of a percentage of activated platelets above 2% as a positive test. Likelihood ratios were 1.94 for positive likelihood and 0.69 for negative likelihood. Univariate logistic regression analysis results were odds ratio 2.8 (95% confidence interval 1.0 to 7.6). Multivariate logistic regression analysis results were odds ratio 2.9 (95% confidence interval 0.92 to 8.9). However, the odds ratio of first antenatal diastolic blood pressure was two to four times higher than the odds ratio of first-trimester CD63 expression. The combination of first-trimester CD63 and first antenatal diastolic blood pressure increases the positive likelihood ratio from 1.94 to 9.4, with a sensitivity of 41%, a specificity of 96%, and a negative likelihood ratio of 0.62. CONCLUSIONS: Increased first-trimester CD63 expression is an independent risk factor for development of preeclampsia. CD63 expression might be useful to identify a subgroup of patients with a high risk for development of preeclampsia, especially in combination with first-trimester antenatal diastolic blood pressure. This method of patient selection may enable more efficient intervention studies in patients at risk than do the selection methods used so far.

Extensive platelet activation in preeclampsia compared with normal pregnancy: enhanced expression of cell adhesion molecules.

OBJECTIVES: Platelets play an important role in the pathophysiologic mechanisms of preeclampsia. Our purpose was to investigate by means of flow cytometry to what extent platelets circulate in an activated state during normal pregnancy and whether this activation is more extensive in preeclampsia. STUDY DESIGN: Platelets in whole blood from 10 preeclamptic third-trimester pregnant women (highest diastolic blood pressure range 100 to 130 mm Hg, proteinuria range 0.59 to 11.5 gm/24 hr) and from 10 normotensive third-trimester pregnant controls were analyzed with the following activation markers: anti-P-selectin (alpha-granule secretion), anti-CD63 (lysosomal secretion), PAC-1 (monoclonal antibody against fibrinogen receptor conformation of the glycoprotein IIb/IIIa complex), anti-platelet endothelial cell adhesion molecule-1, and annexin-V (a placental protein that binds to negatively charged phospholipids, present on the outside of the platelet plasma membrane after activation). The differences in surface antigen exposure between the two groups were determined by double-label flow cytometry. Flow cytometric data were analyzed in two ways: first, the percentages of activated platelets above a certain threshold compared with a nonpregnant control sample were determined, indicative for activation of a subpopulation of cells, and, second, the mean fluorescence intensities were determined, indicative of the mean surface antigen expression of the total platelet population. RESULTS: Analysis of the percentage of activated platelets proved most informative. With this analysis an enhanced platelet activation status was present in 4 of 10 normotensive patients and a more extensive platelet activation status in all 10 preeclamptic patients, as indicated by P-selectin (p = 0.008) and CD63 (p = 0.03) expression. Increased platelet endothelial cell adhesion molecule-1 (p = 0.005) expression was also observed in preeclampsia. CONCLUSIONS: Flow cytometric analysis clearly indicated that platelets circulate in a more extensively activated state during preeclampsia than during normal pregnancy. The increased platelet endothelial cell adhesion molecule-1 expression in preeclamptic patients demonstrates that, besides alpha-granular and lysosomal release, other hitherto unknown mechanisms are involved. Platelet endothelial cell adhesion molecule-1 appears to be the best marker to distinguish preeclamptic patients from normotensive pregnant women. Only a subpopulation of the platelets appears to be activated.

A point mutation in an invariant splice acceptor site results in a decreased mRNA level in a patient with severe coagulation factor XIII subunit A deficiency.

Amplification and sequencing of exons I-XV of the gene encoding subunit A of coagulation factor XIII (FXIII) in a patient with severe subunit A deficiency revealed a single G-->A base substitution at the last position of intron E, mutating the invariant AG dinucleotide splice acceptor site to AA. Northern blot analysis of FXIII subunit A mRNA levels in peripheral mononuclear leukocytes showed that this mutation leads to an undetectable FXIII subunit A mRNA level, suggesting that the mutant transcript is either highly unstable or only spliced at low efficiency. Despite this low mRNA level we were able to amplify cDNA fragments containing the exonV-exonVI junction. Sequence analysis showed that the AA dinucleotide is not recognized by the splicing machinery. Instead, an AG dinucleotide located seven bases downstream of the mutated splice acceptor site is used as alternative acceptor. The resulting, alternatively spliced, FXIII subunit A transcript contains a deletion of the first seven bases of exon VI, while translation continues out of frame and leads to a premature stop codon 27 bases thereafter.

The influence of insulin, beta-estradiol, dexamethasone and thyroid hormone on the secretion of coagulant and anticoagulant proteins by HepG2 cells.

As a basis for regulatory studies on the influence of hormones on (anti)coagulant protein production by hepatocytes, we examined the amounts of the plasma proteins antithrombin III (AT III), protein C, protein S, factor II, factor X, fibrinogen, and prealbumin produced by the hepatoma cell line HepG2, into the culture medium, in the absence and presence of insulin, beta-estradiol, dexamethasone and thyroid hormone. Without hormones these cells produced large amounts of fibrinogen (2,452 +/- 501 ng/mg cell protein), AT III (447 +/- 16 ng/mg cell protein) and factor II (464 +/- 31 ng/mg cell protein) and only small amounts of protein C (50 +/- 7 ng/mg cell protein) and factor X (55 +/- 5 ng/mg cell protein). Thyroid hormone had a slight but significant effect on the enrichment in the culture medium of the anticoagulant protein AT III (1.34-fold) but not on protein C (0.96-fold) and protein S (0.91-fold). This hormone also significantly increased the amounts of the coagulant proteins factor II (1.28-fold), factor X (1.45-fold) and fibrinogen (2.17-fold). Insulin had an overall stimulating effect on the amounts of all the proteins that were investigated. Neither dexamethasone nor beta-estradiol administration did substantially change the amounts of these proteins. We conclude that the HepG2 cell is a useful tool to study the hormonal regulation of the production of (anti)coagulant proteins. We studied the overall process of protein production, i.e., the amounts of proteins produced into the culture medium. Detailed studies have to be performed to establish the specific hormonal effects on the underlying processes, e.g., transcription, translation, cellular processing and transport, and secretion.

Preeclampsia is not associated with altered platelet vasopressin binding and cytosolic Ca++ concentration.

OBJECTIVES: Preeclampsia is an important cause of fetal and maternal morbidity and mortality. Recently it was described that platelet cytosolic Ca++ levels could be used to screen for preeclampsia. The current study investigated platelet arginine vasopressin receptor characteristics, platelet cytosolic Ca++ levels, plasma- and platelet-bound arginine vasopressin in white pregnant women. STUDY DESIGN: In a cross-sectional study nine third-trimester nulliparous pregnant women with gestational hypertension (seven with proteinuria, two with excessive weight gain without proteinuria) were compared with nine healthy nulliparous pregnant women matched for gestation length and age and 10 healthy age-matched nonpregnant women. Determined were (1) platelet arginine vasopressin receptor number and affinity, (2) platelet cytosolic Ca++ levels, both basal and on arginine vasopressin or thrombin stimulation, and (3) plasma- and platelet-bound arginine vasopressin levels. RESULTS: None of the measured parameters differed significantly among the three groups studied. Mean arginine vasopressin receptor number and affinity ranged from 108 to 143 receptors per platelet and 0.35 to 0.40 nmol/L, respectively. A single population of binding sites was found (Hill number 0.96). Basal Ca++ levels ranged from 113.4 to 133.3 nmol/L, on arginine vasopressin stimulation from 199 to 250 nmol/L. Median arginine vasopressin levels in platelet-poor plasma were between 1.2 and 2.4 pg/ml, with circulating platelets being estimated to possess two to five molecules of arginine vasopressin per platelet. A significant correlation was found between platelet cytosolic Ca++ levels before and after arginine vasopressin stimulation (r = 0.69, p < 0.001) and a weak correlation between platelet receptor density and arginine vasopressin-stimulated platelet cytosolic Ca++ levels (r = 0.38, p < 0.05). CONCLUSIONS: The studied parameters, platelet cytosolic Ca++ levels, whether basal or after stimulation with arginine vasopressin and vasopressin platelet receptor density and affinity, do not discriminate already hypertensive or preeclamptic white women from nondiseased subjects. A valuable test to screen for preeclampsia awaits further prospective studies.

Synthesis of platelet-activating factor by human blood platelets and leucocytes. Evidence against selective utilization of cellular ether-linked phospholipids.

Synthesis of platelet activating factor (PAF) in blood platelet suspensions may be due to leucocyte contamination. We therefore investigated PAF synthesis in human blood platelet suspensions and granulocyte- (PMN)-enriched leucocyte suspensions upon stimulation by thrombin and Ca2+-ionophore A23187, both in the presence and absence of the presumed PAF catabolism inhibitor phenylmethylsulfonyl fluoride (PMSF). PAF synthesis was measured by aggregation of washed rabbit platelets and by [3H]acetate incorporation. In contrast to A23187, thrombin was unable to stimulate PAF synthesis by leucocytes. As thrombin did induce PAF synthesis by platelet suspensions, this was evidently not due to leucocyte contamination. A23187 also induced PAF synthesis by platelets, but this was dependent upon the platelet isolation method and possibly associated activation. The ratio of [3H]acetate incorporation into 1-alkyl- versus 1-acyl-2-acetylglycerophosphocholine upon stimulation of non-PMSF-treated leucocytes and platelets amounted to 12.8 and 1.2, respectively. These values are at least 10-fold higher than the ratio of 1-alkyl versus 1-acyl species in the cellular phosphatidylcholine precursor for PAF. By PMSF pretreatment, the distribution of incorporated [3H]acetate between 1-ether- and 1-ester-linked species became similar to that in the precursor phosphatidylcholines of the respective cell type, due to increased recovery of [3H]acetate in the acyl compounds. Both leucocyte and platelet homogenates rapidly degraded acylacetylglycerophosphocholine to (acetyl)glycerophosphocholine, and this deacylation was inhibited by PMSF pretreatment of the cells. We conclude that upon cell stimulation a phospholipase A2 converts both alkylacylglycerophosphocholine and diacylglycerophosphocholine to the 2-lysoanalogs in a ratio similar to the occurrence of the parent compounds. The acetyltransferase subsequently acetylates both compounds to acylacetylglycerophosphocholine and alkylacetylglycerophosphocholine (PAF), respectively. Deacylation of the 1-ester-linked species, either before or after acetylation, gives the impression of selective utilization of 1-ether-linked species for PAF production. It is only after inhibition of the deacylation by pretreatment of the cells with PMSF that a mainly nondiscriminative use of 1-ether- and 1-ester-linked species by both phospholipase A2 and acetyltransferase becomes evident.

Age-related deficiency of the synthesis of platelet activating factor by leukocytes from Zellweger patients.

Ca2+-ionophore A23187-induced synthesis of the alkoxyether lipid platelet activating factor (PAF) by leukocytes from Zellweger patients was undetectable in two patients studied at 3 and 4 weeks of age, reduced in a third patient studied at 2 months of age, and in the low normal range in four patients studied between 4 months and 5 years of age. We have previously reported that plasmalogen-type phosphatidylethanolamine (PE) levels of erythrocytes are reduced in Zellweger patients up to 20 weeks of age, but normal in older patients. These levels were reduced in the three patients with abnormal PAF synthesis, and normal in the other four patients. The results suggest a close relationship between the age of the patients at sampling, and both the A23187-induced capacity of leukocytes to synthesize PAF and the plasmalogen PE levels in their erythrocytes.

Platelet-activating factor: mediator of the third pathway of platelet aggregation? A study in three patients with deficient platelet-activating factor synthesis.

Thrombin, collagen, and Ca2+-ionophore A23187 aggregate platelets in the presence of inhibitors of the first (ADP-mediated) and second (cyclooxygenase-dependent) pathway of platelet activation. This aggregation, via a third pathway, was hypothesized to be mediated by the alkoxyether lipid platelet-activating factor (PAF). We recently demonstrated virtual absence of plasmalogen-type alkoxyether lipids and deficiency in key enzymes of their biosynthesis in Zellweger patients. We hypothesized that PAF synthesis might also be impaired. We report two Zellweger patients with an undetectable A23187-induced PAF synthesis of leukocytes (patients, less than 3 pmol PAF/10(8) granulocytes (PMN); four age-matched controls, 249-2,757 pmol PAF/10(8) PMN; five adult controls, 291-5,433 pmol PAF/10(8) PMN). In a third patient, residual PAF synthesis was detected. However in all patients the thrombin-induced third mechanism of platelet aggregation was present. We therefore conclude that PAF may not be the mediator of the third pathway.

Synergistic effects of platelet-activating factor and other platelet agonists in human platelet aggregation and release: the role of ADP and products of the cyclooxygenase pathway.

The synergistic effects of platelet-activating factor (PAF) with ADP, collagen, thrombin, A23187, adrenaline, sodium arachidonate and ristocetin in human platelet aggregation and beta-thromboglobulin (beta-TG) release were investigated in citrated platelet-rich plasma (PRP). Synergism in both aggregation and release was present with all agonists except ristocetin. Upon oral intake of aspirin (ASA) the PAF-induced irreversible aggregation as well as the synergistic irreversible aggregation became reversible. Both prior to and after ASA ingestion ADP removal by creatine phosphate/creatine phosphokinase (CP/CPK) resulted in a reduced, reversible platelet aggregation induced by PAF alone or in combination with the other agonists. The ADP-removal and ASA-ingestion also strongly inhibited the beta-TG release. The synergistic aggregation and release were also inhibited by ASA and indomethacin in vitro as well as by the competitive ADP-inhibitor ATP. It is concluded that not only the activation of human platelets by low doses of PAF itself, but also the synergism of PAF and other platelet agonists is highly dependent upon ADP and products of the cyclooxygenase pathway.