Heparinization of biological vascular graft reduces fibrin deposition.

An alternative graft is needed for coronary bypass operations in patients lacking suitable autologous vessels. We therefore studied Denaflex, a biologic graft, in a dog ex-vivo shunt model to determine whether heparin treatment makes this graft less thrombogenic. Comparison was also made to Bioflow, a nonheparinized biologic graft. Fibrinogen deposition during high flow (593 +/- 202 ml/min) decreased from 672 +/- 467 ng/mm2 in nonheparinized Denaflex grafts to 448 +/- 298 ng/mm2 (p < 0.05) in heparinized Denaflex grafts. At low flow (117 +/- 13 ml/min), heparinization of Denaflex grafts similarly decreased fibrinogen deposition from 1102 +/- 601 ng/mm2 to 703 +/- 405 ng/mm2 (p < 0.05). At both flow rates fibrinogen deposition in Bioflow grafts was less than in nonheparinized Denaflex, but was similar to heparinized Denaflex grafts. Platelet deposition was not influenced by heparinization of Denaflex grafts and was similar among Denaflex and Bioflow preparations. Whether Denaflex performs acceptably in vivo as a xenograft requires extensive study.

Antibody-detectable changes in fibrinogen adsorption affecting platelet activation on polymer surfaces.

Reactivity of platelets with an artificial surface exposed to whole blood is correlated with the concentration of adsorbed fibrinogen detectable by antifibrinogen antibodies. To examine the effect on platelets of the organization (distribution, orientation, conformation) of fibrinogen adsorbed on a hydrophobic surface, we studied the binding of polyclonal and monoclonal antifibrinogen antibodies to polyalkyl methacrylate polymers previously exposed to purified fibrinogen solution or diluted plasma and compared the results with platelet retention in methacrylate bead columns. There was an increase in platelet retention following diluted plasma pretreatment, which was eliminated by a polyclonal antibody against fibrinogen or against a gamma-(395-411) peptide from fibrinogen and was reduced by monoclonal antibodies (4A5, 4-2) against other COOH-terminal gamma-chain epitopes. Monoclonal antibody 10E5 against the fibrinogen receptor GpIIb/IIIa totally inhibited platelet retention in the bead columns. Our data suggest that different methacrylate polymers induce different changes in adsorbed fibrinogen, which may interfere with its interaction with platelets, and that platelet retention in a methacrylate bead column involves interaction of the COOH-terminal end of the gamma-chain of adsorbed fibrinogen with platelet GpIIb/IIIa receptors.

Rapid platelet separation and collection for postoperative autologous transfusion.

The authors describe a technique for rapid, on-line, precardiopulmonary bypass separation and collection of 10-20% of a patient's platelets for storage and postoperative reinfusion. In a three stage process, blood is diverted through a hollow fiber hemofilter and platelets are concentrated near the fiber walls; the core is flushed of whole blood and the concentrated platelets are collected. Process mechanics were investigated in vitro with bovine blood; efficacy was studied with in vivo ovine experiments; and platelet function and viability were examined with in vitro human blood tests. Using commercially available hemofilters with 13,000 polyacrylonitrile fibers, the authors collected 1 x 10(11) to 2 x 10(11) platelets in 250 ml in 3.5 min. With fresh human blood, collected platelets exhibited normal morphology and aggregation response to adenosine diphosphate, collagen, and thrombin for 4 hours after collection.

1H NMR sequential assignments and secondary structure analysis of human fibrinogen gamma-chain C-terminal residues 385-411.

The human fibrinogen gamma-chain, C-terminal fragment, residues 385-411, i.e., KIIPFNRLTIGEGQQHHLGGAKQAGDV, contains two biologically important functional domains: (1) fibrinogen gamma-chain polymerization center and (2) platelet receptor recognition domain. This peptide was isolated from cyanogen bromide degraded human fibrinogen and was investigated by 1H NMR (500 MHz) spectroscopy. Sequence-specific assignments of NMR resonances were obtained for backbone and side-chain protons via analysis of 2D NMR COSY, double quantum filtered COSY, HOHAHA, and NOESY spectra. The N-terminal segment from residues 385-403 seems to adopt a relatively fixed solution conformation. Strong sequential alpha CH-NH NOESY connectivities and a continuous run of NH-NH NOESY connectivities and several long-lived backbone NH protons strongly suggest the presence of multiple-turn or helix-like structure for residues 390 to about 402. The conformation of residues 403-411 seems to be much less constrained as evidenced by the presence of weaker and sequential alpha CH-NH NOEs, the absence of sequential NH-NH NOEs, and the lack of longer lived amides. Chemical shifts of resonances from backbone and side-chain protons of the C-terminal dodecapeptide, residues 400-411, differ significantly from those of the parent chain, suggesting that some preferred C-terminal conformation does exist.

Does the conformation of adsorbed fibrinogen dictate platelet interactions with artificial surfaces?

Platelet activation by polymer surfaces is thought to require preliminary adsorption of fibrinogen and perhaps changes in fibrinogen conformation. We measured fibrinogen adsorption by a series of polymers by two methods, using either 125I-labeled fibrinogen or 125I-labeled antifibrinogen antibodies, and correlated the results with platelet reactivity (retention and secretion) in columns of beads coated with the polymers. For polyalkyl methacrylates with 1 to 4 carbon side chains, platelet reactivity varied directly with increasing length of the alkyl side chain and with the quantity of bound fibrinogen recognizable by antifibrinogen antibody but not with the total quantity of fibrinogen adsorbed. The same pattern of results was seen with five antibody preparations, including affinity-purified Fab fragments against the D or E domain of fibrinogen. Tests of platelet retention and fibrinogen binding to four polyalkyl acrylates and to three unrelated polymers (polystyrene, polymethyl methacrylate, and a polyether polyurethane) indicated that platelet retention correlated positively with both total fibrinogen binding and with the amount of antibody-recognizable fibrinogen bound. Drugs that block platelet aggregation, but not adhesion, did not alter the hierarchy of platelet retention to the polyalkyl methacrylates. These data suggest that, contrary to previous views, platelet adhesion to artificial surfaces increases with increasing surface coverage of adsorbed fibrinogen if the bound fibrinogen maintains a conformation such that its functional domains remain recognizable by antibody probes.

Platelet aggregation by fibrinogen polymers crosslinked across the E domain.

There is evidence that platelet interactions with artificial surfaces are mediated by plasma proteins, especially fibrinogen, adsorbed on the surfaces. Multiple site interactions between fibrinogen molecules adsorbed in high concentration and receptors in the unactivated platelet may be sufficient for platelet adhesion and subsequent activation. To examine this hypothesis, we prepared soluble polymers of fibrinogen. Polymers produced by interaction of fibrinogen with Fab'2 fragments of antibodies against fibrinogen's E (central) domain (Fg-Fab'2(E] induced, in gel-filtered platelets, aggregation and serotonin release, which were blocked by monoclonal antibodies against the GPIIb/IIIa complex, by Fab fragments against the D domain, and by metabolic inhibitors; aggregation was attenuated but not abolished by enzymatic removal of ADP (with CP/CPK) or by blockage of ADP binding sites (with FSBA), and when secretion was inhibited by aspirin. Fg-Fab'2(E) also induced a dose-dependent elevation in cytoplasmic Ca2+ (measured by Aequorin luminescence) which was attenuated by CP/CPK and by FSBA, and was eliminated by metabolic inhibitors and by anti-IIb/IIIa antibody. Fibrinogen complexes crosslinked with dimethylsuberimidate or Factor XIII neither aggregated gel-filtered platelets nor inhibited platelet aggregation by ADP and fibrinogen, probably because of inaccessibility of lysine residues in the D (terminal) domain of fibrinogen, which are thought to be required for platelet binding. Thus, soluble complexes of fibrinogen having multiple available platelet receptor recognition sites activate gel-filtered platelets and may provide a useful model for platelet-surface interactions mediated by adsorbed fibrinogen.

Catalytic activity and platelet reactivity of heparin covalently bonded to surfaces.

Heparin was covalently bound to solid substrate surfaces by means of four different chemistries. It was coupled to polymethylacrylate (PMA) beads with glutaraldehyde, carbodiimide, or radical polymerization initiated by Ce4+, or to agarose beads with cyanogen bromide. Each of these chemistries produced measurable amounts of surface-bound heparin, which was minimally elutable in contact with plasma. Antithrombin (AT) binding by heparinized PMA materials (compared with PMA control beads) ranged from no AT binding for the material heparinized with carbodiimide (PMA-Alb-Hep(EDC] to 3.6 micrograms/ml packed beads for the material heparinized by radical polymerization (PMA-MA-Hep). Heparin-like catalytic activity of these materials (assayed by measuring the generation of thrombin-antithrombin complex in plasma) correlated well with the amount of heparin bound, but not as well with AT binding capacity. Heparinized agarose, which exhibited a large AT binding capacity (2.2 mg AT per milliliter of packed gel), had virtually no catalytic activity because of its inability to release thrombin-antithrombin complex from the surface. Platelet interaction with heparinized materials that exhibit high AT binding capacity was reduced by pretreatment with normal plasma but not by pretreatment with AT-depleted plasma. Platelet interaction with heparinized materials with low AT binding capacities was not reduced by pretreatment with normal plasma. We conclude that AT binding by heparin reduces the platelet reactivity of heparinized surfaces.

Platelet-compatible hydrophilic segmented polyurethanes from polyethylene glycols and cyclohexane diisocyanate.

A new type of segmented polyurethane (SPU) was synthesized from alpha, omega polyethylene oxide diols (PEG) of MW varying from 600 to 4500, by end capping with 1,4 trans cyclohexane diisocyanate (CHDI) and chain extending with ethylene diamine (ED) in toluene with dibutyl tin dilaurate as catalyst. These SPU are cast as films and coatings from hexafluroisopropanol (HFIP). Depending on PEG MW, these SPU swell two- to tenfold in water. Examined by an in vitro platelet retention test, these SPU are more bland (platelet retention -rho around 0.05) than most other polymers, whereas an alternating copolymer of CHDI and ED shows -rho around 0.80 (very active); x-ray photoelectron spectroscopy shows that the surfaces of these SPU to a depth of about 40 Ao are nearly pure PEO, unlike SPU synthesized from aromatic diisocyanates TDI and MDI.

Effect of heparin and heparin fractions on platelet aggregation.

Porcine intestinal mucosal heparin induced aggregation of platelets in citrated platelet-rich plasma and enhanced platelet aggregation and serotonin secretion induced by other agents. This action of heparin was blocked by substances that elevate platelet cyclic AMP and by EDTA but not by inhibitors of platelet cyclooxygenase. The effect was not inhibited by apyrase or by N-amylthio-5'-AMP and therefore did not require the action of ADP, nor was there activation of platelet phospholipase. Platelet aggregation by heparin required a plasma cofactor different from the cofactor required for ristocetin. Fractionation of heparin yielded preparations that varied in molecular weight and, within a given molecular weight fraction, in affinity for antithrombin III. Fractions of high molecular weight (average 20,000) were more reactive with platelets than were fractions of low molecular weight (7,000). Anticoagulant activity did not parallel the platelet reactivity of heparin fractions. Among high molecular weight fractions, preparations of high or low antithrombin affinity were equally active in induction of platelet aggregation. In low molecular weight fractions, there was an inverse relationship between platelet reactivity and anticoagulant activity in normal platelet-rich plasma, but, in platelet-rich plasma depleted of antithrombin, low molecular weight fractions of high and low antithrombin affinity reacted equally with platelets. These results suggest that formation of an antithrombin-heparin complex protected platelets from aggregation by heparin. Selection of heparin fractions of low molecular weight and high antithrombin affinity may improve anticoagulant therapy and development of thromboresistant heparin-coated artificial materials.

In vivo assessment in sheep of thromboresistant materials by determination of platelet survival.

The thromboresistance of 13 potentially blood-compatible polymers was assessed in sheep by determining survival of 51Cr-labeled platelets. Polymer tubing (120-150 cm x 2.0-2.3 mm i.d.) coiled around the neck was incorporated into the circulation through silicone rubber connectors as a carotid artery-external jugular vein shunt. The mean platelet half-life in control animals ("shunt control") was 78.2 +/- 2.8 (SEM) hours. Eleven of the 13 polymers tested significantly shortened platelet half-life. Polyvinyl chloride (T1/2 = 45.4 +/- 3.0 hours), polyperfluoro ethylene (T1/2 = 47.0 +/-1.6 hours), and a polymethylacrylate (PMA)/acrilonitrile copolymer (T1/2 = 50.7 +/- 7.0 hours) produced the greatest shortening. Only silica-free polydimethyl siloxane (T1/2 = 74.7 +/- 4.9 hours) and PMA (T1/2 = 81.5 +/- 3.4 hours) were indistinguishable from shunt controls. Pretreatment of PMA tubing with autologous plasma in a paired trial significantly increased platelet half-life (P less than 0.05 vs. untreated PMA). This system offers an economical, reproducible, sensitive, and biologically relevant method for assessment of the reactivity of artificial surfaces with platelets.

In vitro assessment of interaction of blood with model surfaces.

An in vitro system for the assessment of blood compatibility of artificial surfaces is described. Anticoagulated human whole blood was pumped through columns of beads and then analyzed for evidence of platelet retention and the release of platelet constitutents. Platelets were retained on the columns by adhesion to the beads and by aggregation. Experiments were conducted with PVAc, PC, PMA and PS beads to provide data on the reproducibility of the system and the effects of anticoagulants and RBC concentration on platelet adhesion and release. The system is versatile, inexpensive, reproducibile, and also suitable for studies of the effects of pharmacologic agents and other aspects of platelet-artificial surface interaction.

Effects of ions on ADP-induced aggregation of bovine or human platelets.

Effects of divalent cations on ADP-induced aggregation response were examined. Bovine platelets were transferred by Sepharose 2B gel filtration from citrate-PRP into citrate free buffer (buffer-GFP). Response increases, reaches a maximum and decreases with increasing calcium and/or magnesium concentration. For either calcium or magnesium alone, increasing response is proportional to a rate coefficient and, through an apparent ion-platelet association constant, to the fraction of platelet critical sites bound to cation. With both ions present, bound magnesium appears to inhibit bound calcium in excess of that accounted for by competition and a lower rate coefficient for bound magnesium. With citrate present in buffer-GFP, apparent association constants increase, excess magnesium inhibition is present, but systems are path dependent. Initial conditions appear to establish a response which is thereafter immutable to environmental magnesium alteration. Citrate-PRP resembles buffer-GFP: response is sensitive to the selective removal of calcium and excess magnesium inhibition is present. With heparin-PRP, response is immutable to the selective removal of approximately 90% of initial calcium. The dependency of response inhibition observed at high divalent cation concentrations indicates that aggregation is not due to interplatelet cross linking by ions. Ion effects are similar for bovine and human platelets.

Effects of matrix contact during gel filtration of human platelets in plasma.

When gel filtration is used to transfer platelets from plasma into an established environment, alterations in platelet characteristics may result from the change in environment or from the effects of platelet contact with the gel matrix. To approach the problem of evaluating the relative contributions from these sources, a Sepharose 2B matrix was employed and platelets transferred from citrate anticoagulated PRP into autologous PPP to yield plasma-GFP. Platelet recoveries averaged 93%. PRP: plasma-GFP pairs were found to be indistinguishable with respect to: morphology; ADP, thrombin or collagen-induced aggregation response; uptake of 5-hydroxytryptamine (5-HT) or adenosine; and thrombin or collagen-induced release of accumulated 5-HT or adenosine. Pairs are distinguishable by prostaglandin E2 synthesis assayed immediately after filtration.

Physiology and ultrastructure of the blood platelet following exposure to hydrogen peroxide.

Hydrogen peroxide in micromolar concentrations can induce shape change in human blood platelets, and can modify the aggregation and release reaction of these cells as induced by ADP or thrombin. In larger (millimolar) concentrations, H2O2 produces fusion of platelets with distortions in platelet morphology unlike those normally caused by aggregating agents. The production of H2O2 in vivo by granulocytes or other cells could influence the processes of haemostasis or thrombosis.

Cyclic AMP in human blood platelets: relation to platelet prostaglandin synthesis induced by centrifugation or surface contact.

Exposure of blood platelets to physical stimuli such as centrifugation, filtration through Sepharose gel, or stirring with particulate material leads to production of prostaglandins E2 and F2alpha and to a reduction in cAMP. These effects resemble those produced by ADP, epinephrine, collagen, and thrombin in association with the platelet release reaction but are less effectively inhibited by aspirin.