Cold storage of pooled, buffy-coat-derived, leucoreduced platelets in plasma.

BACKGROUND: This study was designed to determine which in vitro assays would be most useful for studying the effects of cold storage on platelet concentrates and to establish an in vivo model for platelet recovery and survival. STUDY DESIGN AND METHODS: Paired, plasma-suspended, leucoreduced, buffy-coat-derived platelet concentrates were stored either at 22 or 4 degrees C. Prior to storage and after 18 h, 5 days and 7 days, samples were taken and various assays were performed. On day 6, in vivo studies were carried out using a model system. Galactosylation of the platelets, prior to cold storage, was also tested. RESULTS: Hypotonic shock response, collagen-induced aggregation, RANTES and P-selectin binding site measurements demonstrated differences between platelets stored at 22 and 4 degrees C. The glycocalicin assay was able to demonstrate microvesicle formation at 4 degrees C. The in vivo model showed that there was at least a 50% decrease in recovery and survival when the platelets were stored in the cold. Galactosylation did not improve these results. CONCLUSIONS: Several assays, both in vitro and in vivo, were able to detect differences in platelet-storage characteristics and in vivo recovery and survival in a model system. Galactosylation did not correct these cold-induced changes.

Platelet storage in Fresenius/NPBI polyolefin and BTHC-PVC bags: a direct comparison.

SUMMARY: New platelet storage systems, such as changes in the plastic of the storage bags, require validation. In this study, pooled buffy coat platelets stored in Fresenius/NPBI polyolefin bags were compared with those stored in Fresenius/NPBI butyryl-trihexyl citrate (BTHC) plasticized polyvinyl chloride (PVC). The CompoSelect thrombocyte polishing filter system (1000 mL polyolefin bag) and the CompoStop F730 system (1300 mL BTHC-PVC bag) were used to prepare paired, plasma-suspended, buffy coat platelet concentrates. Samples were taken up to day 7 for in vitro analysis. In a separate experiment, 12 units were prepared using the CompoStop F730 system and samples taken after leucofiltration for FXIIa assay. By day 7, platelet concentrates stored in BTHC-PVC demonstrated significantly higher pH levels (7.32 +/- 0.05 vs. 7.26 +/- 0.05) and a greater degree of cell lysis as shown by increased lactate dehydrogenase levels (497 +/- 107 vs. 392 +/- 81 U L(-1)). The supernatants contained higher concentrations of soluble P-selectin and the chemokine 'regulated on activation, normal T-cell expressed and presumably secreted', which are released from the alpha-granules during activation. The ATP concentrations were significantly lower in BTHC-PVC. Platelet counts, mean platelet volume and hypotonic shock response were similar for both bags. FXIIa antigen concentrations were 0.6 +/- 0.2 ng mL(-1) indicating that activation of the contact factor pathway had not occurred. Although the CompoStop F730 leucoreduction filter did not activate the contact system, platelets stored in 100% plasma in BTHC-PVC bags demonstrated different in vitro characteristics from those stored in polyolefin. Further work is required to demonstrate whether these differences will affect in vivo recovery and survival.

The anti-heparin properties of human low-density lipoprotein.

High-density (HDL), low-density (LDL) and very low-density lipoproteins (VLDL) have been purified from normal human plasma by a combination of ultracentrifugation in high-density salt and agarose gel filtration. The ability of these lipoproteins to inhibit different molecular weight heparin fractions has been compared, using incubation mixtures comprised of antithrombin III and factor Xa. Residual factor Xa activity was measured using the chromogenic peptide substrate Bz-Ile-Glu-Gly-Arg-pNA. LDL inhibited the high molecular weight (but not low molecular weight) heparin accelerated neutralisation of factor Xa by antithrombin III. VLDL showed a similar, though much reduced anti-heparin activity, while the addition of HDL to the factor Xa incubation mixture produced no measurable anti-heparin activity. These observations suggest that certain plasma lipoproteins may selectively modulate the inhibitory action of heparin against factor Xa.

The heparin-accelerated neutralisation of bovine alpha and beta thrombins by antithrombin III.

The heparin-accelerated neutralisation of bovine alpha and beta thrombins has been examined using a peptide substrate H-D-phenylalanyl-pipecolyl-arginine-paranitroanilide-HCl to measure thrombin amidase activity. Alpha and beta thrombins were both neutralised by antithrombin III and this neutralisation was further accelerated by the presence of small amounts of heparin. Low and high molecular weight heparin and heparins fractionated by their affinity for antithrombin III were all able to accelerate the neutralisation of alpha and beta thrombin. This work is therefore unable to confirm reports that alpha and beta thrombins have different heparin sensitivities.

Evidence for a plasma inhibitor of the heparin accelerated inhibition of factor Xa by antithrombin III.

The ability of heparin fractions of different molecular weight to potentiate the action of antithrombin III against the coagulation factors thrombin and Xa has been examined in purified reaction mixtures and in plasma. Residual thrombin and Xa have been determined by their peptidase activities against the synthetic peptide substrates H-D-Phe-Pip-Arg-pNA and Bz-Ile-Gly-Arg-pNA. High molecular weight heparin fractions were found to have higher anticoagulant activities than low molecular weight heparin when studied with both thrombin and Xa incubation mixtures in purified mixtures and in plasma. The inhibition of thrombin by heparin fractions and antithrombin III was unaffected by other plasma components. However, normal human plasma contained a component that inhibited the heparin and antithrombin III inhibition of Xa particularly when the high molecular weight heparin fraction was used. Experiments using a purified preparation of platelet factor 4 suggested that the platelet-derived heparin-neutralizing protein was not responsible for the inhibition.

Impacts and concerns for vCJD in blood transfusion: current status.

The impact of vCJD upon blood transfusion practice hinges on its lymphoreticular involvement. B lymphocytes play a key supporting role for the capture and replication of infectivity by follicular dendritic cells of the lymphoid tissue in animal models of transmissible spongiform encephalopathies (TSE) and tonsils, spleen and appendix in man can harbour vCJD infectivity, a situation not seen with the other human TSEs. Leucodepletion of blood donations in the UK was implemented to reduce possible vCJD transmission and preliminary data suggests that white cell associated infectivity will be effectively removed although plasma infectivity will not. Blood screening assays are under development but none yet are ready for application. The conformation dependant immunoassay, based on differences in secondary and tertiary structure between normal and TSE-associated abnormal prion protein, has a sensitivity now approaching the best bioassay. Even so further development is needed to detect the fg/ml levels likely in the event that vCJD blood does contain abnormal prion, which is as yet unproven. Surrogate assays, such as for erythroid associated factor, may provide additional means of identifying donors harbouring vCJD. Validation of clearance of TSEs from pooled plasma products consistently demonstrates effective removal of the agents in downscaled systems and studies comparing vCJD, BSE and scrapie agents yield similar results. Many approaches to therapy are under investigation, in cell culture and animal models, targeted to normal or abnormal prion metabolism, including chemical and immunological interventions. Efficacy of quinacrine/chlorpromazine and pentosan polysulphate in a clinical setting, and agents yet to be used, will be more accurately known following recent agreement of clinical drug evaluation protocols.

Neoantigens and antibodies to factor VIII.

It is well known that 10-20% of severe haemophiliacs are likely to develop inhibitors to factor VIII, usually soon after the commencement of therapy. Two recent inhibitor outbreaks have occurred in patients treated for a number of years on switching to a product subjected to additional virus inactivation. Hence the incidence of inhibitor formation may be affected by the type of product used for treatment and the potential for processing to result in 'neoantigens'. Examination of the parts of factor VIII interacting with inhibiting antibodies, and the effect of various therapies on these, can teach us something about the mechanisms involved in antibody formation. However, the development of pre-clinical assays to assess products and processes for neoantigen formation should allow the prevention of inhibitor outbreaks. This review summarizes current in vitro and in vivo approaches to this problem, concluding that most available assays are inadequate for this purpose, with competitive immunoassay and phospholipid binding providing the most hopeful route forward.

An enzyme-linked immunosorbent assay (ELISA) used to study the cellular secretion of endothelial plasminogen activator inhibitor (PAI-1).

A two-site sandwich ELISA was developed to measure PAI-1 antigen and utilised a polyclonal antiserum produced against PAI-1 purified from human endothelial cell secretory products. The assay was calibrated against a preparation of pure PAI-1 whose protein concentration had been determined by amino acid analysis and the detection limit was 30 pg PAI-1 ml-1 sample. PAI-1 was detected in primate sera but not in a wide range of non-primate sera and no cross-reactivity with alpha 2-antiplasmin or antithrombin III was observed. The ELISA was used to study cellular secretion of PAI-1 which was confirmed as a major secretory protein in human umbilical vein endothelial cells (HUVEC). PAI-1 antigen accumulated in the medium in a linear fashion with time and accounted for approximately equal to 10% of total secreted protein. Specific activity of intracellular PAI-1 was typically 20-fold greater than that of PAI-1 in 24 h conditioned medium and a t1/2 for inactivation of secreted PAI-1 of 0.53 h was calculated. Purified endotoxin stimulated the secretion of PAI-1 antigen and raised the intracellular levels in HUVEC cultures showing that the anti-fibrinolytic actions of endotoxin are effected by increasing the rate of synthesis and secretion of PAI-1.

A novel rat model of thrombogenicity: its use in evaluation of prothrombin complex concentrates and high purity factor IX concentrates.

A non-stasis rodent model of thrombogenicity has been used for dose-ranging studies with a conventional prothrombin complex concentrate (PCC) and to evaluate high purity factor IX concentrates from different manufacturers. Fibrin monomer (soluble fibrin) and fibrinopeptide A (FPA) were monitored before and after infusion of test solution. FPA was found to be the more sensitive and reproducible indicator of thrombogenicity and exhibited a dose-related elevation after infusion of the PCC at doses of between 100-300 IU/kg. In contrast the amounts of FPA generated after 300 IU/kg of the high purity factor IX products were similar to control infusions of albumin.

Characterisation of epitopes on human tissue plasminogen activator recognised by a group of monoclonal antibodies.

Seven mouse monoclonal antibodies have been produced against human melanoma tissue plasminogen activator (t-PA). They were specifically bound to 125I t-PA but not 125I urokinase (u-PA) and inhibited t-PA, but not u-PA, activity in plasminogen-rich 125I fibrin wells. Three of the antibodies directly inhibited the amidolytic activity of t-PA and the two most effective also bound near the active site histidine residue as determined by competition experiments using active site blocking agents. Several antibodies interfered with the fibrin binding properties of t-PA. One antibody neither interacted with the active site nor inhibited fibrin binding but still effectively quenched t-PA activity in fibrin wells suggesting that it masks another region of the molecule necessary for effective biological activity.

Quantitation by radioimmunoassay of thrombospondin synthesised and secreted by human endothelial cells.

Synthesis and secretion of thrombospondin by cultured human endothelial cells was quantified using a sensitive radioimmunoassay. Two endothelial lines were examined; those from umbilical veins and those from saphenous veins. The foetal line secreted approximately 17 micrograms/10(6) cells of the antigen at 24 hr, whereas the saphenous line secreted five times less in the same period. Intracellular values were similar in both lines. Thrombin and calcium ionophore A23187 were added to umbilical vein endothelial cultures. In contrast to platelets where both agents cause a rapid release reaction, no such thrombospondin release was observed in endothelial cells.

Comparison of high purity factor IX concentrates and a prothrombin complex concentrate in a canine model of thrombogenicity.

A non-stasis canine model of thrombogenicity has been used to evaluate batches of high purity factor IX concentrates from 4 manufacturers and a conventional prothrombin complex concentrate (PCC). Platelets, activated partial thromboplastin time (APTT), fibrinogen, fibrin(ogen) degradation products and fibrinopeptide A (FPA) were monitored before and after infusion of concentrate. Changes in FPA were found to be the most sensitive and reproducible indicator of thrombogenicity after infusion of batches of the PCC at doses of between 60 and 180 IU/kg, with a dose related delayed increase in FPA occurring. Total FPA generated after 100-120 IU/kg of 3 batches of PCC over the 3 h time course was 9-12 times that generated after albumin infusion. In contrast the amounts of FPA generated after 200 IU/kg of the 4 high purity factor IX products were in all cases similar to albumin infusion. It was noted that some batches of high purity concentrates had short NAPTTs indicating that current in vitro tests for potential thrombogenicity may be misleading in predicting the effects of these concentrates in vivo.

Fibrinolytic and haemostatic responses to desamino-D-arginine vasopressin (DDAVP) administered by intravenous and subcutaneous routes in healthy subjects.

DDAVP was administered at 0.4 microgram kg-1 intravenous (i.v.) and subcutaneous (s.c.) routes to 6 healthy subjects in a double blind crossover study. Both study treatments were well tolerated. Flushing occurred after both treatments but was more prominent after i.v. than after s.c. DDAVP. Mild transient local discomfort at the s.c. injection site occurred in 5 of 6 subjects. The mean peak factor VIII (FVIII) response was 369% and 247% of baseline after i.v. and s.c. DDAVP respectively and the maximum increase in FVIII occurred earlier with the i.v. route. Changes in FVIII antigen (FVIII:Ag) and von Willebrand factor antigen (vWF:Ag) were also monitored. Tissue-type plasminogen activator (t-PA) activity measured by a chromogenic assay employing soluble fibrin had a median peak value of 2.9 IU ml-1 at 20 min after i.v. and of 1.9 IU ml-1 at 60 min after s.c. DDAVP. t-PA antigen was also measured so that the specific activity of circulating t-PA could be determined. Preinjection median values of 14,650 and 13,700 IU mg-1 increased to peak median values of 236,200 IU mg-1 at 20 min after i.v. and 202,400 IU mg-1 at 60 min after s.c. DDAVP. Plasminogen activator inhibitor (PAI) activity fell following DDAVP and became undetectable in some subjects during the sampling period. The ratio of maximum fibrinolytic response was similar to the ratio of maximum haemostatic responses obtained by two routes of injection. Our results indicate that s.c. DDAVP might successfully replace i.v. DDAVP in several applications such as confirmation of haemostatic or fibrinolytic responsiveness in patient groups; for obtaining FVIII enriched plasma; as well as its obvious potential usefulness in home treatment of haemophilia A and von Willebrand's disease.

Metabolism of sodium pentosan polysulphate in man measured by a new competitive binding assay for sulphated polysaccharides--comparison with effects upon anticoagulant activity, lipolysis and platelet alpha-granule proteins.

Three human volunteers were injected with a range of doses of pentosan polysulphate, SP54, i.v. or s.c. A competitive binding assay (CBA) for sulphated polysaccharides was used to detect circulating SP54 after doses as low as 1 mg i.v. and a linear relationship was observed between the peak plasma concentration of SP54 measured by CBA and the administered dose. A comparison was made between the clearance of SP54 measured by CBA and its anticoagulant and lipolytic activities. SP54 was detectable by CBA after doses which caused no alteration in activated partial thromboplastin time (APTT) or anti-factor Xa activity but after which a small increase of lipase activity was measurable. After SP54 at 10 mg i.v. or 100 mg s.c. anti-factor Xa activity was 4-6 times greater than would be expected from the in vitro activity of the concentrations of SP54 measured by CBA. Like heparin and other heparin analogues, SP54 caused an increase in plasma concentrations of platelet factor 4 (PF4) without a concomitant rise in beta-thromboglobulin (beta-TG). It is concluded that the newly developed CBA will provide a more sensitive means than conventional bioassays for the determination of plasma concentrations of SP54.

Metabolism of sodium pentosan polysulphate in man--catabolism of iodinated derivatives.

An iodinated derivative of the heparin analogue SP54 has been prepared and used in conjunction with unlabelled SP54 to study the catabolism and organ distribution of this potential antithrombotic agent in healthy human volunteers. As observed previously with 125I-heparin, we found that the 125I-SP54 was rapidly cleared from the circulation, returning later in a desulphated form. Organ distribution studies with 123I-SP54 suggested that the liver and spleen were major sites of desulphation. Gel filtration and Polybrene binding showed the presence of sulphated macromolecular SP54 and desulphated macromolecular and depolymerised SP54 in post-injection urines. No depolymerised material was present in plasma suggesting depolymerisation occurs in the kidney.

High purity factor IX and prothrombin complex concentrate (PCC): pharmacokinetics and evidence that factor IXa is the thrombogenic trigger in PCC.

Recent studies using assays for surrogate markers of thrombogenicity in man have demonstrated that activation of the coagulation system occurs following infusion of clinical doses of prothrombin complex concentrates (PCC) but not after the same doses of high-purity factor IX concentrates (HP-FIX) in patients with haemophilia B. Here we have investigated the mechanism of such thrombogenesis by applying assays that detect early-through to late-events in coagulation system activation in a pharmacokinetic cross-over study of 50 IU/kg PCC and a new HP-FIX product in haemophilia B patients. Satisfactory recoveries and half-lives were observed for both concentrates. HP-FIX caused no increases in thrombin-antithrombin III complex (TAT), prothrombin activation peptide fragment F1+2 (F1+2), factor X activation peptide (FXAP) or factor VIIa (FVIIa). In contrast the same dose of factor IX in the form of PCC was followed by significant increases over pre-infusion levels of TAT, F1+2 and FXAP, but not FVIIa. Elevations of FIXAP occurred after both HP-FIX and PCC but did not reach normal levels and were attributed to normalisation of the FIX concentration in those patients whose levels of FIXAP were initially low. We conclude that the thrombogenic trigger associated with PCC infusion occurs at the level of factor X activation. In the absence of any increase in FVIIa, we would attribute this to the likely presence of FIXa in the PCC.

A comparative study using immunological and biological assay of the haemostatic responses to DDAVP infusion venous occlusion and exercise in normal men.

In a group of six normal male volunteers, infusion of DDAVP, venous occlusion and exercise were shown to increase plasma levels of factor VIII and plasminogen activator, activity and antigen, to different extents and at differing rates. Any mechanisms suggested to explain release of these proteins by various stimuli should account for such differences. All three stimuli could also increase plasma levels of prostacyclin metabolites, although this was only significant for high doses of DDAVP. Other potential endothelial markers, such as fibronectin and thrombospondin, showed no specific increase after any of the stimuli.

Tissue plasminogen activator in human plasma measured by radioimmunoassay.

A radioimmunoassay (RIA) has been set up using purified human melanoma tissue plasminogen activator and a specific antiserum raised against it. Concentration of antigen in citrated plasma of resting human subjects was found to be 7.1 +/- 1.6 ng/ml (mean +/- S.D.) while the lower limit of sensitivity of the RIA was 2 ng/ml. After venous occlusion or DDAVP infusion the levels of antigen were invariably elevated and a significant increase in the proportion of antigen having fibrin binding properties was observed. Decay studies in vitro and gel filtration of post-stimulus plasma indicated that the RIA detects active and inactive forms of antigen and the possible nature of the inactive antigen is discussed.

Neutrophil activation in chronic liver disease.

OBJECTIVE: To assess the relationship between neutrophil activation and indices of disease severity in patients with chronic liver disease. METHODS: Plasma neutrophil elastase was measured by radioimmunoassay as a marker of neutrophil activation, and disease severity assessed by standard clinical, biochemical, haematological and histological techniques. PATIENTS: Eighty-eight patients with chronic liver disease were studied, Thirty-nine had alcohol-induced liver disease (ALD), 18 autoimmune chronic hepatitis, 13 cryptogenic cirrhosis, seven primary biliary cirrhosis, six primary sclerosing cholangitis, three haemochromatosis and two secondary biliary cirrhosis. Seventy-three patients were cirrhotic and 15 were non-cirrhotic, confirmed by biopsy. RESULTS: Levels of neutrophil elastase were raised in Childs C cirrhotic patients with ALD compared with Childs A or B patients with ALD (P < 0.01), Childs A or B patients with non-ALD (P < 0.01), and Childs C patients with non-ALD (P = 0.02). In patients with ALD, neutrophil elastase correlated with prothrombin time (r = 0.679, P = 0.001), bilirubin (r = 0.587, P < 0.001), Child-Pugh score (r = 0.546, P < 0.001) and inversely with serum albumin (r = -0.511, P < 0.001). In patients with non-ALD, there were no correlations with the measurements of with transaminase levels. CONCLUSION: Neutrophil activation, as measured by plasma neutrophil elastase, is a marker of disease severity in patients with alcohol-induced chronic liver damage, but not in those with other causes of liver disease.

Study of endothelial t-PA and vWf in normal subjects and in von Willebrand's disease.

Some well-described similarities exist between tissue plasminogen activator (t-PA) and von Willebrand factor (vWf) which may suggest a link in either the synthesis or release of both proteins from endothelial cells. To investigate this relationship further immunocytochemical localization of t-PA and vWf was performed in normal tissues and in skin obtained from patients with type I and type III von Willebrand's disease (vWd). Components of the fibrinolytic system were measured at baseline and after venous occlusion in healthy controls and patients with vWd. Patients with severe vWd received intravenous vWf concentrate, followed by desmopressin (DDAVP), to study the plasma response of vWf and t-PA. By immunocytochemical staining, t-PA was demonstrated in endothelial cells of normal skin, kidney and liver and also in the skin of patients with type I and type III vWd. vWf was localized in endothelial cells of all tissues except the specimens from an individual with severe vWd. Basal plasma levels of fibrinolytic components were normal in patients with vWd. Venous occlusion resulted in a rise of fibrinolytic activity in controls and patients with type I, but not type III, vWd. No rise in plasma t-PA was observed following DDAVP in severe vWd, even though near-normalization of plasma vWf levels had been obtained by prior infusion of vWf concentrate. It is concluded that the synthesis of t-PA and vWf is probably regulated by independent processes. Constitutive and regulated release of both proteins occur through different mechanisms and the basal secretion of t-PA is intact in severe vWd.(ABSTRACT TRUNCATED AT 250 WORDS)