High Affinity Binding of Heparin by Necrotic Tumour Cells Neutralises Anticoagulant Activity

2001 ◽  
Vol 86 (08) ◽  
pp. 616-622 ◽  
Author(s):  
Sumihito Morita ◽  
Milena Gebska ◽  
Ajay Kakkar ◽  
Michael Scully

SummaryWe have observed a striking neutralisation of the anticoagulant activity of unfractionated heparin in the presence of a pancreatic carcinoma cell line (MIA PaCa-2) due to binding of around 9 g of heparin per 107 cells (apparent Kd, 30 nM). The loss of anticoagulant activity was less marked in the presence of low molecular weight forms of heparin. Binding to the cell blocked acceleration of the thrombin:anti-thrombin interaction by heparin. Neutralisation of heparin activity was also shown to occur in the presence of a number of other tumour cell lines. FACS analysis demonstrated that live cells did not bind heparin and high affinity binding only occurred to dead MIA PaCa-2 cells. Heparin binding proteins accumulating in cell medium were identified as histone and ribosomal proteins that will become exposed during necrosis. The release of these proteins from cells within the necrotic core of a tumour or from cells killed during chemotherapy may abrogate the heparan sulphate/antithrombin system and possibly contribute to the idiopathic thromboembolism often associated with cancer (Trousseau’s syndrome). The findings also suggest a reason for the reported advantage of LMWH over UFH in treating venous thromboembolism in cancer patients and in improving patient survival.

Blood ◽  
1983 ◽  
Vol 61 (2) ◽  
pp. 368-372 ◽  
Author(s):  
PT Bauer ◽  
R Machovich ◽  
P Aranyi ◽  
KG Buki ◽  
E Csonka ◽  
...  

Abstract The interaction of human alpha-thrombin with mini-pig aortic endothelial cells was studied using 125I-labeled enzyme. Equilibrium between bound and free thrombin was attained within 1 min, and the Klotz-Hunston equations indicated two populations of binding sites. Approximately 30,000 sites/cell belonged to the high-affinity class with a Kd of about 3 x 10(-8) M. Modification of two lysine residues of thrombin with pyridoxal 5′-phosphate (PLP2-thrombin) destroyed the high- affinity binding and about three-fourths of the low-affinity bindings. When the lysine residue of thrombin involved in heparin binding was protected with heparin against chemical modification (PLP-thrombin), the modified enzyme remained similar to the native one with respect to cellular binding, with some loss of low-affinity binding only. Heparin, in a tenfold molar excess to enzyme, inhibited the binding of the native as well as the PLP-thrombin, whereas it did not influence the interaction between PLP2-thrombin and the cell. Since heparin might interfere with both the enzyme and the cell, the binding of heparin to endothelial cells was also examined. The results revealed that 3H- heparin also bound to cells. This binding was characterized by a Kd of 3 x 10(-7) M, approximately 10(6) sites/cell. Furthermore, thrombin bound to endothelial cells was released by antithrombin III. On the basis of these and other data in the literature, a model is proposed for the mechanism of the binding of thrombin to endothelial cells.


Blood ◽  
1983 ◽  
Vol 61 (2) ◽  
pp. 368-372
Author(s):  
PT Bauer ◽  
R Machovich ◽  
P Aranyi ◽  
KG Buki ◽  
E Csonka ◽  
...  

The interaction of human alpha-thrombin with mini-pig aortic endothelial cells was studied using 125I-labeled enzyme. Equilibrium between bound and free thrombin was attained within 1 min, and the Klotz-Hunston equations indicated two populations of binding sites. Approximately 30,000 sites/cell belonged to the high-affinity class with a Kd of about 3 x 10(-8) M. Modification of two lysine residues of thrombin with pyridoxal 5′-phosphate (PLP2-thrombin) destroyed the high- affinity binding and about three-fourths of the low-affinity bindings. When the lysine residue of thrombin involved in heparin binding was protected with heparin against chemical modification (PLP-thrombin), the modified enzyme remained similar to the native one with respect to cellular binding, with some loss of low-affinity binding only. Heparin, in a tenfold molar excess to enzyme, inhibited the binding of the native as well as the PLP-thrombin, whereas it did not influence the interaction between PLP2-thrombin and the cell. Since heparin might interfere with both the enzyme and the cell, the binding of heparin to endothelial cells was also examined. The results revealed that 3H- heparin also bound to cells. This binding was characterized by a Kd of 3 x 10(-7) M, approximately 10(6) sites/cell. Furthermore, thrombin bound to endothelial cells was released by antithrombin III. On the basis of these and other data in the literature, a model is proposed for the mechanism of the binding of thrombin to endothelial cells.


1990 ◽  
Vol 63 (02) ◽  
pp. 193-203 ◽  
Author(s):  
John R Shainoff ◽  
Deborah J Stearns ◽  
Patricia M DiBello ◽  
Youko Hishikawa-Itoh

SummaryThe studies reported here probe the existence of a receptor-mediated mode of fibrin-binding by macrophages that is associated with the chemical change underlying the fibrinogen-fibrin conversion (the release of fibrinopeptides from the amino-terminal domain) without depending on fibrin-aggregation. The question is pursued by 1) characterization of binding in relation to fibrinopeptide content of both the intact protein and the CNBr-fragment comprising the amino-terminal domain known as the NDSK of the protein, 2) tests of competition for binding sites, and 3) photo-affinity labeling of macrophage surface proteins. The binding of intact monomers of types lacking either fibrinopeptide A alone (α-fibrin) or both fibrinopeptides A and B (αβ-fibrin) by peritoneal macrophages is characterized as proceeding through both a fibrin-specific low density/high affinity (BMAX ≃ 200–800 molecules/cell, KD ≃ 10−12 M) interaction that is not duplicated with fibrinogen, and a non-specific high density/low affinity (BMAX ≥ 105 molecules/cell, KD ≥ 10−6 M) interaction equivalent to the weak binding of fibrinogen. Similar binding characteristics are displayed by monocyte/macrophage cell lines (J774A.1 and U937) as well as peritoneal macrophages towards the NDSK preparations of these proteins, except for a slightly weaker (KD ≃ 10−10 M) high-affinity binding. The high affinity binding of intact monomer is inhibitable by fibrin-NDSK, but not fibrinogen-NDSK. This binding appears principally dependent on release of fibrinopeptide-A, because a species of fibrin (β-fibrin) lacking fibrinopeptide-B alone undergoes only weak binding similar to that of fibrinogen. Synthetic Gly-Pro-Arg and Gly-His-Arg-Pro corresponding to the N-termini of to the α- and the β-chains of fibrin both inhibit the high affinity binding of the fibrin-NDSKs, and the cell-adhesion peptide Arg-Gly-Asp does not. Photoaffinity-labeling experiments indicate that polypeptides with elec-trophoretically estimated masses of 124 and 187 kDa are the principal membrane components associated with specifically bound fibrin-NDSK. The binding could not be up-regulated with either phorbol myristyl acetate, interferon gamma or ADP, but was abolished by EDTA and by lipopolysaccharide. Because of the low BMAX, it is suggested that the high-affinity mode of binding characterized here would be too limited to function by itself in scavenging much fibrin, but may act cooperatively with other, less limited modes of fibrin binding.


1994 ◽  
Vol 72 (03) ◽  
pp. 465-474 ◽  
Author(s):  
Neelesh Bangalore ◽  
William N Drohan ◽  
Carolyn L Orthner

SummaryActivated protein C (APC) is an antithrombotic serine proteinase having anticoagulant, profibrinolytic and anti-inflammatory activities. Despite its potential clinical utility, relatively little is known about its clearance mechanisms. In the present study we have characterized the interaction of APC and its active site blocked forms with human umbilical vein endothelial cells (HUVEC). At 4° C 125I-APC bound to HUVEC in a specific, time dependent, saturable and reversible manner. Scatchard analysis of the binding isotherm demonstrated a Kd value of 6.8 nM and total number of binding sites per cell of 359,000. Similar binding isotherms were obtained using radiolabeled protein C (PC) zymogen as well as D-phe-pro-arg-chloromethylketone (PPACK) inhibited APC indicating that a functional active site was not required. Competition studies showed that the binding of APC, PPACK-APC and PC were mutually exclusive suggesting that they bound to the same site(s). Proteolytic removal of the N-terminal γ-carboxyglutamic acid (gla) domain of PC abolished its ability to compete indicating that the gla-domain was essential for cell binding. Surprisingly, APC binding to these cells appeared to be independent of protein S, a cofactor of APC generally thought to be required for its high affinity binding to cell surfaces. The identity of the cell binding site(s), for the most part, appeared to be distinct from other known APC ligands which are associated with cell membranes or extracellular matrix including phospholipid, thrombomodulin, factor V, plasminogen activator inhibitor type 1 (PAI-1) and heparin. Pretreatment of HUVEC with antifactor VIII antibody caused partial inhibition of 125I-APC binding indicating that factor VIII or a homolog accounted for ∼30% of APC binding. Studies of the properties of surface bound 125I-APC or 125I-PC and their fate at 4°C compared to 37 °C were consistent with association of ∼25% of the initially bound radioligand with an endocytic receptor. However, most of the radioligand appeared not to be bound to an endocytic receptor and dissociated rapidly at 37° C in an intact and functional state. These data indicate the presence of specific, high affinity binding sites for APC and PC on the surface of HUVEC. While a minor proportion of binding sites may be involved in endocytosis, the identity and function of the major proportion is presently unknown. It is speculated that this putative receptor may be a further mechanisms of localizing the PC antithrombotic system to the vascular endothelium.


Sign in / Sign up

Export Citation Format

Share Document