Recombinant human granulocyte colony-stimulating factor (rhG-CSF) promotes in vitro platelet aggregation

Abstract

rhG-CSF is increasingly used for stimulation of granulopoiesis and stem cell mobilization in healthy donors for allogeneic stem cell transplantation. However, a possible association between thrombosis and rhG-CSF administration has been reported. For that reason, in this study, we investigated the effect of rhG-CSF on platelet aggregation in whole blood of 10 healthy volunteers. Three concentrations of rhG-CSF solution (1, 10 and 100 ng/ml) were prepared. Each concentration of rhG-CSF solution and a control diluent without rhG-CSF were incubated with whole blood. Incubation with rhG-CSF solutions would result in 0.1, 1.0 and 10 ng/ml rhG-CSF concentrations in the blood. After incubation, aggregation responses were evaluated with ADP (5 and 10 μM) and collagen (2 and 5 μg/ml) in whole blood. When compared to control, preincubation with all dilutions of rhG-CSF augmented aggregation of platelets induced by ADP and collagen in a statistically significant manner (p < 0.05 for all comparisons). There was also a relationship between rhG-CSF concentration (1, 10 and 100 ng/ml) and augmentation of platelet aggregation response (p < 0.0001 for 5–10 μM ADP; p < 0.0001 for 2–5 μg/ml collagen). In conclusion, this study with an in vitro model showed that rhG-CSF administration may lead to platelet hyperaggregability.     

Recombinant human granulocyte colony-stimulating factor (rhG-CSF) promotes in vitro platelet aggregation

rhG-CSF is increasingly used for stimulation of granulopoiesis and stem cell mobilization in healthy donors for allogeneic stem cell transplantation. However, a possible association between thrombosis and rhG-CSF administration has been reported. For that reason, in this study, we investigated the effect of rhG-CSF on platelet aggregation in whole blood of 10 healthy volunteers. Three concentrations of rhG-CSF solution (1, 10 and 100 ng/ml) were prepared. Each concentration of rhG-CSF solution and a control diluent without rhG-CSF were incubated with whole blood. Incubation with rhG-CSF solutions would result in 0.1, 1.0 and 10 ng/ml rhG-CSF concentrations in the blood. After incubation, aggregation responses were evaluated with ADP (5 and 10 microM) and collagen (2 and 5 microg/ml) in whole blood. When compared to control, preincubation with all dilutions of rhG-CSF augmented aggregation of platelets induced by ADP and collagen in a statistically significant manner (p < 0.05 for all comparisons). There was also a relationship between rhG-CSF concentration (1, 10 and 100 ng/ml) and augmentation of platelet aggregation response (p < 0.0001 for 5-10 microM ADP; p < 0.0001 for 2-5 microg/ml collagen). In conclusion, this study with an in vitro model showed that rhG-CSF administration may lead to platelet hyperaggregability.     

Total inhibition of high shear stress induced platelet aggregation by homodimeric von Willebrand factor A1-loop fragments

Under high shear stress, the binding of von Willebrand factor (VWF) A1-loop domain to platelet glycoprotein (GP) Ib alpha occurs as the earliest event in thrombus formation. Therefore recombinant VWF A1-loop fragments could be of therapeutic use in blocking this interaction as competing ligands. We have prepared three homodimeric VWF A1-loop fragments [315 kD Fr III (a homodimer of amino acid [aa] residues 1-1365 of the subunit), 220 kD Fr (a homodimer of aa residues 1-708 of the subunit), and 116 kD Fr (a homodimer of aa residues 449-728 of the subunit) and two monomeric fragments [39/34 kD Fr (a monomer of aa residues 480/481-718 of the subunit] and His-rVWF465-728 (a monomer of aa residues 465-728 of the subunit)], and assessed their potency as inhibitors of botrocetin-induced VWF binding to GPIb alpha and high shear stress induced platelet aggregation mediated by intact VWE All these fragments completely inhibited botrocetin-induced VWF binding to GPIb alpha at a final concentration of 40-200 microM. The homodimeric A1-loop fragments also totally inhibited high shear stress induced platelet aggregation at a final concentration of 0.45-2.0 microM in the following order: 315 kD Fr > or = 220 kD Fr > 116 kD Fr. In contrast, the monomeric A1-loop fragments were only partial inhibitors of high shear stress induced platelet aggregation even at a final concentration of 20 microM, and their IC50s were 13-39 times higher than those of the homodimers. These results indicate that the homodimeric structure of the A1-loop fragment is important for optimal molecular interaction with GPIb alpha under high shear stress.     

Platelet-induced granulocyte aggregation in vitro.

Recent studies have shown that the response of granulocytes to stimulation is increased by the presence of platelets. Using an assay based on counting aggregates with a Coulter Counter, we found that heparinized or citrated platelet-rich plasma directly caused aggregation of isolated granulocytes. Platelets themselves were frequently present within the granulocyte aggregates. The degree of aggregation depended on the platelet:granulocyte ratio, and increased over a 2--5 min period of mixing, with a slower increase thereafter. Aggregation was calcium-dependent, increased by stimulation of the platelets with ADP or collagen but not serotonin, and only slightly reduced by a thrombin inhibitor (hirudin, at concentrations up to 0.9 U/ml). A similar adhesive interaction in vivo might be relevant to thrombotic and ischaemic pathology.     

Mechanism of dexamethasone inhibition of plasminogen activator in rat hepatoma cells.

Glucocorticoids rapidly and completely inhibit intracellular plasminogen activator activity in rat hepatoma cells, as assayed by the solubilization of 125I-labeled fibrin. Experiments in which extracts of dexamethasone-treated cells are mixed with extracts of control cells demonstrate an inhibitor of protease activity. Plasminogen activator is found primarily in the particulate fraction of cell lysates, whereas the inhibitor is localized in the soluble fraction. Variant cells have been isolated previously that are fully resistant to the dexamethoasone inhibition of plasminogen activator activity. These variants have no demonstrable inhibitor activity, whereas plasminogen activator in these cells is fully sensitive to inhibitor from wild-type cells. Thus, the basis for hormone resistance appears to be the failure of dexamethasone to induce an inhibitor.     

Acute arterial thrombosis due to platelet aggregation in a patient receiving granulocyte colony-stimulating factor

We describe a 44-year-old man with non-Hodgkin's lymphoma receiving granulocyte colony-stimulating factor (G-CSF) who developed an acute arterial thrombosis. The removed thrombus contained large amounts of platelet aggregation. A rapid increase of platelets and increased adenosine diphosphate (ADP)- and collagen-induced platelet aggregation were observed at the time of the thrombotic event. A challenge test of G-CSF showed an increase in the platelet count and an augmentation of ADP- and collagen-induced platelet aggregation. In the use of G-CSF, patients who produce a rapid increase in platelet levels could be at greater risk for thrombotic events and need to be followed-up carefully.     

Ethanol inhibition of granulocyte adherence

Ethanol inhibits granulocyte adherence (GA) in whole blood. To determine the mechanism, GA was measured by the nylon fiber assay. In vitro addition of 300 mg/dl ethanol to whole blood caused a 28.1% decrease of GA compared with the control value (p less than 0.009) but did not affect GA of pure PMNs suspended in plasma or in HBSS. Similarly, GA measured in whole blood from intoxicated rabbits fell 48.8% from preintoxication values, but adherence of isolated postintoxication PMNs suspended in postintoxication plasma was not inhibited. Therefore, inhibition of GA by ethanol requires the presence of other cellular components of whole blood. To determine which cell(s) mediate the inhibition, PMNs were suspended with isolated blood components: platelet-rich and platelet-poor plasma, with and without the addition of mononuclear leukocytes and RBCs. Although the reconstitution of whole blood from its components allowed alcohol-induced inhibition of GA (decrease of 32.9%), none of the incomplete blood component mixtures demonstrated this effect. It was concluded that inhibition of GA by ethanol is mediated by an interaction between the components of whole blood.     

Acute arterial thrombosis due to platelet aggregation in a patient receiving granulocyte colony-stimulating factor

We describe a 44-year-old man with non-Hodgkin's lymphoma receiving granulocyte colony-stimulating factor (G-CSF) who developed an acute arterial thrombosis. The removed thrombus contained large amounts of platelet aggregation. A rapid increase of platelets and increased adenosine diphosphate (ADP)- and collagen-induced platelet aggregation were observed at the time of the thrombotic event. A challenge test of G-CSF showed an increase in the platelet count and an augmentation of ADP- and collagen-induced platelet aggregation. In the use of G-CSF, patients who produce a rapid increase in platelet levels could be at greater risk for thrombotic events and need to be followed-up carefully.     

淫羊藿甙逆转地塞米松抑制成骨细胞分化及其机制

目的研究淫羊藿甙对地塞米松诱导的成骨细胞分化抑制的作用及其分子机制。方法采用0.25%胰蛋白酶和0.1%Ⅱ型胶原酶两步消化法,由出生24h以内的新生SD大鼠头盖骨分离成骨细胞,用含10%胎牛血清的DMEM培养；细胞80%～90%汇合时改用含0.2%牛血清白蛋白的DMEM孵育12h后分组试验,抑制试验在丝裂原激活的蛋白激酶激酶(MEK)抑制剂PD89059预培养1h后加入相关干预药物,pNPP法检测碱性磷酸酶(ALP)活性,RT-PCR检测相关基因表达,Westem印迹检测蛋白表达及磷酸化水平。结果(1)与对照组比较,1μmol/L的地塞米松显著抑制成骨细胞ALP活性(0.72±0.05vs 0.78±0.03,P＜0.05)；20μg/L淫羊藿甙对成骨细胞ALP活性具有促进作用(0.91±0.07vs 0.78±0.03,P＜0.05),并逆转了地塞米松对成骨细胞ALP活性的抑制作用(0.83±0.04vs 0.72±0.05,P＜0.05)。(2)与对照组比较,1μmol/L地塞米松刺激MKP-1、RANKL mRNA表达,抑制GSK-3B、OPG mRNA表达,20μg/L淫羊藿甙的作用与之相反,两者共同作用时,淫羊藿甙能逆转地塞米松的作用。(3)1μmol/L地塞米松抑制细胞外信号调节激酶(ERK)磷酸化,1h达峰值；20μg/L淫羊藿甙刺激ERK磷酸化,峰值在10min。(4)MEK抑制剂PD98059显著降低20μg/L淫羊藿甙刺激ERK磷酸化的同时,降低了其刺激的ALP活性。结论淫羊藿甙可以逆转地塞米松抑制的成骨细胞分化,这种作用可能与丝裂原激活的蛋白激酶途径有关。     

Platelet aggregation induced by type IIb platelet von Willebrand factor

Platelet lysates from five patients with a form of type IIb von Willebrand's disease (vWd), associated with spontaneous platelet aggregation and thrombocytopenia, induced platelet aggregation of normal and other vWd's platelet-rich plasma (PRP). Platelet lysate from normals, type I or type IIa vWd did not cause platelet aggregation of normal PRP. When polyclonal monospecific antibodies directed against plasma von Willebrand factor (vWf) were incubated with the type IIb platelet lysate, they inhibited the platelet aggregation. Monoclonal antibodies directed against the glycoprotein (GP) Ib binding domain of plasma vWf incubated with the type IIb platelet lysate did not inhibit the platelet aggregation. Normal platelets suspended in afibrinogenaemic plasma did not aggregate when type IIb vWd platelet lysate was added. Normal platelets incubated with monoclonal antibodies directed against the fibrinogen and vWf binding site(s) on the GPIIb/IIIa were not aggregated by the type IIb platelet lysate. Bernard-Soulier PRP aggregated when type IIb vWd platelet lysate was added, while Glanzmann's thrombasthenic platelets did not. Peptides containing the RGDS sequence or the sequence of the carboxy terminal 15 amino acids of the gamma chain of fibrinogen inhibited the type IIb vWd platelet lysate-induced platelet aggregation. These data suggest that type IIb platelet vWf can cause platelet aggregation of PRP without the addition of any agonist. This interaction is different from that observed with the plasma vWf from these patients.     

Proteolytic cleavage of granulocyte colony-stimulating factor and its receptor by neutrophil elastase induces growth inhibition and decreased cell surface expression of the granulocyte colony-stimulating factor receptor

Neutrophil elastase (NE) is a serine protease stored in the primary granules of neutrophils that proteolytically cleaves multiple cytokines and cell surface proteins on release from activated neutrophils. Recent reports of mutations in the gene encoding this enzyme in some patients with neutropenic syndromes prompted us to investigate whether granulocyte colony-stimulating factor (G-CSF) and its receptor (G-CSFR) are also substrates for NE. To further address this, we examined the effect of NE on G-CSF and the G-CSFR both in solution and on intact cells. Incubation of recombinant G-CSF or a G-CSFR form corresponding to its extracellular domain with purified NE resulted in rapid proteolytic cleavage of both proteins. Addition of NE to tissue culture medium or pretreatment of G-CSF with NE before its addition to media suppressed the growth of G-CSF-responsive cells. NE also cleaved the G-CSFR on the surface of intact cells resulting in a time-dependent reduction in cell surface expression of the G-CSFR. Notably, decreased G-CSFR surface expression resulting from treatment of cells with NE was also associated with a reduction in cell viability and proliferation in response to G-CSF. These results are the first to demonstrate that G-CSF and G-CSFR are proteolytically cleaved by NE and that NE-induced degradation of these proteins correlates with a reduction in the biologic activity of the cytokine and a decrease in the signaling function of the receptor because of decreased G-CSFR surface expression. These findings provide additional insights into mechanisms by which G-CSF/G-CSFR interactions may be modulated.     

Transplantation potential of peripheral blood stem cells induced by granulocyte colony-stimulating factor.

The major effect of granulocyte colony-stimulating factor (G-CSF) is to induce neutrophilia in previously untreated animals or after chemotherapy or marrow transplantation in humans, primates and rodents. In addition, it has been reported that migration of committed progenitor cells to the blood occurs during G-CSF therapy. In this article, by using sex mismatched transplants and a molecular probe for Y-chromosome specific DNA sequences, we show that among the peripheral blood cells during G-CSF therapy are substantial numbers of primitive stem cells capable of (1) reconstituting the hematopoietic system in the long term, and (2) making a contribution to the lymphoid populations of the thymus, in radiation ablated recipients. These data suggest that blood from patients treated with G-CSF may provide a convenient source of the most primitive stem cells for autologous or allogeneic bone marrow transplantation.     

Chlorpromazine inhibition of granulocyte superoxide production

Superoxide production by granulocytes is a result of the activation of an NAD(P)H-dependent oxidase present in the plasma membrane. Chlorpromazine (5-50 muM) prolongs the time necessary to activation of the superoxide generating system and inhibits the extent of activation. When chlorpromazine is added after activation, there is an inhibition of further superoxide production. These effects are seen with digitonin, phorbol myristate acetate, and opsonized zymosan stimulated guinea pig and human granulocytes. Other phenothiazines (1-20 muM) and tetracaine (0.1-1.0 muM) produce similar effects. Lidocaine (1-10 mM) inhibits superoxide production but has no effect on the rate of activation. The effect of chlorpromazine on the rate of activation is reversible, but its effect on extent of activation is unaffected by extensive washing. Incubation of granulocytes with chlorpromazine results in decreased activation of the plasma membrane superoxide generating NADPH oxidase. Chlorpromazine also competes with NADPH for the membrane oxidase. These data and previously published results provide the basis of a model for the activation of the superoxide generating system.     

A role for sulfhydryl groups in granulocyte aggregation

Polymorphonuclear leukocytes (PMN) stimulated by high concentrations of the complement component C5A form cellular aggregates, and both the rate and degree of aggregation are influenced by changes in the PMN plasma membrane and cytoskeleton. Since sulfhydryls are important constituents of the plasma membrane and cytoskeleton, we investigated the effect of agents that oxidize and bind sulfhydryls on C5A-induced aggregation. PMN incubated with diamide, a nonspecific sulfhydryl- oxidizing agent, had a marked increase in their aggregation response to C5A. Tertiary butyl hydroperoxide (BHP), which reacts specifically with the soluble sulfhydryl glutathione (GSH), had no effect on aggregation. The enhancement of PMN aggregation by diamide, but not BHP, suggested that oxidation of non-GSH sulfhydryls contributes to the aggregation response. To test the requirement for sulfhydryls in PMN aggregation, PMN were treated with the sulfhydryl-binding agent N-ethylmaleimide (NEM). NEM markedly impaired aggregation without affecting resting or methylene blue-stimulated [14C]-L-glucose oxidation of the granulocytes. P-chloromercuriphenyl sulfonic acid (PCMPSA), an external sulfhydryl-binding agent, had no effect on aggregation. These studies suggest that cellular sulfhydryls are required for optimal PMN aggregation and that oxidation of these sulfhydryls may be one of the biochemical changes that contributes to aggregation.     

Bacterial species-dependent inhibition of human granulocyte elastase

Lung infection with Streptococcus pneumoniae causes the recruitment of many granulocytes (PMN) to the alveoli, but little damage to lung structure ensues, as opposed to infections with other bacterial species that induce a similar PMN response and cause lung damage. Elastase, a proteolytic enzyme of PMN, has been implicated as an agent of lung injury. We studied the interaction of different bacterial species with human PMN in vitro to determine if PMN elastase activity is affected by the species of bacteria ingested. Pseudomonas aeruginosa, Staphylococcus aureus, Klebsiella pneumoniae, and S. pneumoniae were grown, opsonized, and incubated with human PMN. After 1 h of incubation, intracellular and extracellular elastase activity was measured. S. pneumoniae reduced PMN elastase activity by 48%, whereas the other 3 species tested had only minimal effects on elastase activity. Loss of elastase activity occurred with S. pneumoniae: PMN ratios as low as 2:1. Adherence or ingestion of bacteria by the PMN was necessary for the decrease in elastase activity to occur; interventions that decreased phagocytosis, such as not opsonizing the bacteria, pretreatment of the PMN with cytochalasin B, and separation of bacteria from PMN by 0.22-mu filters, increased elastase activity. ELISA and Western blot analysis of elastase levels in these experiments suggested that normal amounts of elastase were present. Thus, the loss of elastase activity we observed may be due to an elastase inhibitor present in S. pneumoniae.