Novel assay demonstrates that coronary artery disease patients have heightened procoagulant platelet response

Essentials Procoagulant platelets can be detected using GSAO in human whole blood. Stable coronary artery disease is associated with a heightened procoagulant platelet response. Agonist-induced procoagulant platelet response is not inhibited by aspirin alone. Collagen plus thrombin induced procoagulant platelet response is partially resistant to clopidogrel.

SUMMARY

Background Procoagulant platelets are a subset of highly activated platelets with a critical role in thrombin generation. Evaluation of their clinical utility in thrombotic disorders, such as coronary artery disease (CAD), has been thwarted by the lack of a sensitive and specific whole blood assay. Objectives We developed a novel assay, utilizing the cell death marker, GSAO [(4-(N-(S-glutathionylacetyl)amino)phenylarsonous acid], and the platelet activation marker, P-selectin, to identify procoagulant platelets in whole blood by flow cytometry. Patients/Methods Using this assay, we characterized the procoagulant platelet population in healthy controls and a cohort of patients undergoing elective coronary angiography. Results In patients with CAD, compared with patients without CAD, there was a heightened procoagulant platelet response to thrombin (25.2% vs. 12.2%), adenosine diphosphate (ADP) (7.8% vs. 2.7%) and thrombin plus collagen (27.2% vs. 18.3%). The heightened procoagulant platelet potential in CAD patients was not associated with other markers of platelet function, including aggregation, dense granule release and activation of α2b β3 integrin. Although dual antiplatelet therapy (DAPT) was associated with partial suppression of procoagulant platelets, this inhibitory effect on a patient level could not be predicted by aggregation response to ADP and was not fully suppressed by clopidogrel. Conclusions We report for the first time that procoagulant platelets can be efficiently detected in a few microliters of whole blood using the cell death marker, GSAO, and the platelet activation marker, P-selectin. A heightened procoagulant platelet response may provide insight into the thrombotic risk of CAD and help identify a novel target for antiplatelet therapies in CAD.

Encryption and decryption of tissue factor

Tissue factor (TF) is a transmembrane cofactor that binds and promotes the catalytic activity of factor (F) VIIa. The TF/VIIa complex activates FX by limited proteolysis to initiate blood coagulation and helps provide the thrombin burst that is important for a stable thrombus. TF is present both in the extravascular compartment, where it functions as a hemostatic envelope, and the intravascular compartment, where it contributes to thrombus formation, particularly when endothelial disruption is minimal. The regulation of its cofactor function appears to differ in the two compartments. Intravascular TF derives predominately from leucocytes, with either monocytes or neutrophils implicated in different models of thrombosis. This TF exists mostly in a non-coagulant or cryptic form and acute events lead to local decryption of TF and FX activation. A variety of experimental observations imply that decryption of leucocyte surface TF involves both a dithiol/disulfide switch and exposure of phosphatidylserine. The dithiol/disulfide switch appears to involve the Cys186-Cys209 disulfide bond in the membrane-proximal domain of TF, although this has not been demonstrated in vivo. Activation of a purinergic receptor or complement has recently been observed to decrypt TF on myeloid cells and a dithiol/disulfide switch and the oxidoreductase, protein disulfide isomerase, have been implicated in both systems. The molecular mechanism of action of protein disulfide isomerase in TF encryption/decryption, though, remains to be determined.

Optical imaging of cell death in traumatic brain injury using a heat shock protein-90 alkylator

Traumatic brain injury is a major public health concern and is characterised by both apoptotic and necrotic cell death in the lesion. Anatomical imaging is usually used to assess traumatic brain injuries and there is a need for imaging modalities that provide complementary cellular information. We sought to non-invasively image cell death in a mouse model of traumatic brain injury using a near-infrared fluorescent conjugate of a synthetic heat shock protein-90 alkylator, 4-(N-(S-glutathionylacetyl) amino) phenylarsonous acid (GSAO). GSAO labels both apoptotic and necrotic cells coincident with loss of plasma membrane integrity. The optical GSAO specifically labelled apoptotic and necrotic cells in culture and did not accumulate in healthy organs or tissues in the living mouse body. The conjugate is a very effective imager of cell death in brain lesions. The optical GSAO was detected by fluorescence intensity and GSAO bound to dying/dead cells was detected from prolongation of the fluorescence lifetime. An optimal signal-to-background ratio was achieved as early as 3 h after injection of the probe and the signal intensity positively correlated with both lesion size and probe concentration. This optical GSAO offers a convenient and robust means to non-invasively image apoptotic and necrotic cell death in brain and other lesions.

Redox control of β2-glycoprotein I-von Willebrand factor interaction by thioredoxin-1

BACKGROUND

 β(2) -Glycoprotein I (β(2) GPI) is an abundant plasma protein that is closely linked to blood clotting, as it interacts with various protein and cellular components of the coagulation system. However, the role of β(2) GPI in thrombus formation is unknown. We have recently shown that β(2) GPI is susceptible to reduction by the thiol oxidoreductases thioredoxin-1 and protein disulfide isomerase, and that reduction of β(2) GPI can take place on the platelet surface.

METHODS

 β(2) GPI, reduced by thioredoxin-1, was labeled with the selective sulfhydryl probe N(a)-(3-maleimidylpropionyl)biocytin and subjected to mass spectrometry to identify the specific cysteines involved in the thiol exchange reaction. Binding assays were used to examine the affinity of reduced β(2) GPI for von Willebrand factor (VWF) and the effect of reduced β2GPI on glycoprotein (GP)Ibα binding to VWF. Platelet adhesion to ristocetin-activated VWF was studied in the presence of reduced β(2) GPI.

RESULTS

We demonstrate that the Cys288-Cys326 disulfide in domain V of β(2) GPI is the predominant disulfide reduced by thioredoxin-1. Reduced β(2) GPI in vitro displays increased binding to VWF that is dependent on disulfide bond formation. β(2) GPI reduced by thioredoxin-1, in comparison with non-reduced β(2) GPI, leads to increased binding of GPIbα to VWF and increased platelet adhesion to activated VWF.

CONCLUSIONS

Given the importance of thiol oxidoreductases in thrombus formation, we provide preliminary evidence that the thiol-dependent interaction of β(2) GPI with VWF may contribute to the redox regulation of platelet adhesion.

Contribution of allosteric disulfide bonds to regulation of hemostasis

Protein disulfide bonds are covalent links between pairs of Cys residues in the polypeptide chain. Acquisition of disulfide bonds is an important way that proteins have evolved and are continuing to evolve. These bonds serve either a structural or functional role. There are two types of functional disulfide: the catalytic bonds that reside in the active sites of oxidoreductases and the allosteric bonds. Allosteric disulfides are defined as bonds that have evolved to control the manner in which proteins function by breaking or forming in a precise way. The known allosteric bonds have a particular configuration known as the -RHStaple. Several hemostasis proteins contain -RHStaple disulfides and there is increasing evidence that some of these bonds may be involved in the functioning of the protein in which they reside. The best studied of these to date is the -RHStaple disulfide in tissue factor and its role in de-encryption of the cofactor.

Allosteric disulfide bonds in thrombosis and thrombolysis

Allosteric disulfide bonds control protein function by mediating conformational change when they undergo reduction or oxidation. The known allosteric disulfide bonds are characterized by a particular bond geometry, the -RHStaple. A number of thrombosis and thrombolysis proteins contain one or more disulfide bonds of this type. Tissue factor (TF) was the first hemostasis protein shown to be controlled by an allosteric disulfide bond, the Cys186-Cys209 bond in the membrane-proximal fibronectin type III domain. TF exists in three forms on the cell surface: a cryptic form that is inert, a coagulant form that rapidly binds factor VIIa to initiate coagulation, and a signaling form that binds FVIIa and cleaves protease-activated receptor 2, which functions in inflammation, tumor progression and angiogenesis. Reduction and oxidation of the Cys186-Cys209 disulfide bond is central to the transition between the three forms of TF. The redox state of the bond appears to be controlled by protein disulfide isomerase and NO. Plasmin(ogen), vitronectin, glycoprotein 1balpha, integrin beta(3) and thrombomodulin also contain -RHStaple disulfides, and there is circumstantial evidence that the function of these proteins may involve cleavage/formation of these disulfide bonds.

Identification and characterisation of a platelet GPIb/V/IX-like complex on human breast cancers: implications for the metastatic process

The glycoprotein (GP) Ib /V/IX receptor complex is an important adhesion molecule, originally thought to be unique to the megakaryocytic lineage. Recent evidence now indicates that GPIb /V/IX may be more widely expressed. In this study we report the presence of all subunits of the complex on four breast cancer cell lines, and 51 / 80 primary breast tumours. The surface expression of GPIb /V/IX was confirmed by flow cytometry, and by immunoprecipitation of biotin surface-labelled tumour cells. Western blotting of cell lysates under reducing conditions revealed that tumour cell-GPIb alpha had a relative molecular weight of 95 kDa as compared to 135 kDa on platelets. Despite the discrepant protein size, molecular analyses on the tumour cell-GPIb alpha subunit using RT-PCR and DNA sequencing revealed 100% sequence homology to platelet GPIb alpha. Tumour cell-GPIb /V/IX was capable of binding human von Willebrand factor (vWf), and this binding caused aggregation of tumour cells in suspension. Tumour cells bound to immobilised vWf in the presence of EDTA and demonstrated prominent filapodial extensions indicative of cytoskeletal reorganisation. Furthermore, in a modified Boyden chamber assay, prior exposure to vWf or a GPIb alpha monoclonal antibody, AK2, enhanced cell migration. The presence of a functional GPIb /V/IX-like complex in tumour cells suggests that this complex may participate in the process of haematogenous breast cancer metastasis.

Mechanism of transfer of NO from extracellular S-nitrosothiols into the cytosol by cell-surface protein disulfide isomerase

N-dansylhomocysteine (DnsHCys) is quenched on S-nitrosation. The product of this reaction, N-dansyl-S-nitrosohomocysteine, is a sensitive, direct fluorogenic substrate for the denitrosation activity of protein disulfide isomerase (PDI) with an apparent K(M) of 2 microM. S-nitroso-BSA (BSA-NO) competitively inhibited this reaction with an apparent K(I) of 1 microM. The oxidized form of DnsHCys, N,N-didansylhomocystine, rapidly accumulated in cells and was reduced to DnsHCys. The fluorescence of DnsHCys-preloaded human umbilical endothelial cells and hamster lung fibroblasts were monitored as a function of extracellular BSA-NO concentration via dynamic fluorescence microscopy. The observed quenching of the DnsHCys fluorescence was an indirect measure of cell surface PDI (csPDI) catalyzed denitrosation of extracellular S-nitrosothiols as decrease or increase in the csPDI levels in HT1080 fibrosarcoma cells correlated with the rate of quenching and the PDI inhibitors, 5,5'-dithio-bis-3-nitrobenzoate and 4-(N-(S-glutathionylacetyl) amino)phenylarsenoxide inhibited quenching. The apparent K(M) values for denitrosation of BSA-NO by csPDI ranged from 12 microM to 30 microM. Depletion of membrane N(2)O(3) with the lipophylic antioxidant, vitamin E, inhibited csPDI-mediated quenching rates of DnsHCys fluorescence by approximately 70%. The K(M) for BSA-NO increased by approximately 3-fold and V(max) decreased by approximately 4-fold. These findings suggest that csPDI catalyzed NO released from extracellular S-nitrosothiols accumulates in the membrane where it reacts with O2 to produce N(2)O(3). Intracellular thiols may then be nitrosated by N2O3 at the membrane-cytosol interface.

Lupus antibody bivalency is required to enhance prothrombin binding to phospholipid

Lupus anticoagulants (LA) are a family of autoantibodies that are associated with in vitro anticoagulant activity but a strong predisposition to in vivo thrombosis. They are directed against plasma phospholipid-binding proteins including prothrombin. We have proposed that LA propagates coagulation in flowing blood by facilitating prothrombin interaction with the damaged blood vessel wall. A murine monoclonal anti-prothrombin Ab and three of three LA IgGs enhanced prothrombin binding to 75:25 phosphatidyl choline:phosphatidyl serine vesicles measured by either ultracentrifugation or right-angle light scattering. The assembly of prothrombin and LA IgG on phospholipid vesicles was estimated by surface plasmon resonance. The on rates for prothrombin and LA IgG were approximately the same as the on rate for prothrombin alone. In contrast, the off rates for prothrombin and LA IgG were 2- to 3-fold slower than the off rate for prothrombin. LA IgG bivalency was required for enhanced prothrombin binding to phospholipid vesicles, as Fab of the LA IgGs did not influence prothrombin binding at concentrations up to 40 microM. Modeling of the interactions of prothrombin, LA IgG and phospholipid vesicles indicated that augmentation of prothrombin binding to phospholipid vesicles by LA IgG could be accounted for by the bivalency of the LA IgG and the elevated microenvironmental concentration of prothrombin on the surface of phospholipid vesicles.

Phosphoglycerate kinase acts in tumour angiogenesis as a disulphide reductase

Disulphide bonds in secreted proteins are considered to be inert because of the oxidizing nature of the extracellular milieu. An exception to this rule is a reductase secreted by tumour cells that reduces disulphide bonds in the serine proteinase plasmin. Reduction of plasmin initiates proteolytic cleavage in the kringle 5 domain and release of the tumour blood vessel inhibitor angiostatin. New blood vessel formation or angiogenesis is critical for tumour expansion and metastasis. Here we show that the plasmin reductase isolated from conditioned medium of fibrosarcoma cells is the glycolytic enzyme phosphoglycerate kinase. Recombinant phosphoglycerate kinase had the same specific activity as the fibrosarcoma-derived protein. Plasma of mice bearing fibrosarcoma tumours contained several-fold more phosphoglycerate kinase, as compared with mice without tumours. Administration of phosphoglycerate kinase to tumour-bearing mice caused an increase in plasma levels of angiostatin, and a decrease in tumour vascularity and rate of tumour growth. Our findings indicate that phosphoglycerate kinase not only functions in glycolysis but is secreted by tumour cells and participates in the angiogenic process as a disulphide reductase.

Presence of closely spaced protein thiols on the surface of mammalian cells

It has been proposed that certain cell-surface proteins undergo redox reactions, that is, transfer of hydrogens and electrons between closely spaced cysteine thiols that can lead to reduction, formation, or interchange of disulfide bonds. This concept was tested using a membrane-impermeable trivalent arsenical to identify closely spaced thiols in cell-surface proteins. We attached the trivalent arsenical, phenylarsenoxide, to the thiol of reduced glutathione to produce 4-(N-(S-glutathionylacetyl)amino)phenylarsenoxide (GSAO). GSAO bound tightly to synthetic, peptide, and protein dithiols like thioredoxin, but not to monothiols. To identify cell-surface proteins that contain closely spaced thiols, we attached a biotin moiety through a spacer arm to the primary amino group of the gamma-glutamyl residue of GSAO (GSAO-B). Incorporation of GSAO-B into proteins was assessed by measuring the biotin using streptavidin-peroxidase. Up to 12 distinct proteins were labeled with GSAO-B on the surface of endothelial and fibrosarcoma cells. The pattern of labeled proteins differed between the different cell types. Protein disulfide isomerase was one of the proteins on the endothelial and fibrosarcoma cell surface that incorporated GSAO-B. These findings demonstrate that the cell-surface environment can support the existence of closely spaced protein thiols and suggest that at least some of these thiols are redox active.

Reduction of von Willebrand factor by endothelial cells

The haemostatic activity of plasma von Willebrand factor (vWF) is a function of multimer size. Only the large vWF multimers are effective in promoting platelet adhesion to a site of vascular injury. We observed that the conditioned medium of cultured human umbilical vein, human microvascular and bovine aortic endothelial cells contained an activity which reduced the average multimer size of plasma or purified vWF. The average multimer size of vWF produced endogenously by human umbilical vein endothelial cells was similarly reduced following secretion. The reducing activity was ablated by pre-treatment with heat or the thiol blocking agents. iodoacetamide, N-ethylmaleimide or E-64, but not by a range of specific serine-, cysteine-, aspartic-, or metalloproteinase inhibitors. Reduction in vWF multimer size was associated with formation of new thiols in vWF and there was no evidence for additional proteolytic processing of vWF. The reducing activity was associated with a protein with an anionic pi that binds heparin and contains reactive thiol(s). These results suggested that the interchain disulfide bonds that link the vWF homodimers near the N-termini were being reduced by a vWF reductase secreted by endothelial cells. In support of this hypothesis, incubation of vWF with the protein reductants, protein disulfide isomerase and thioredoxin, resulted in formation of new thiols in vWF and reduction in the average multimer size of vWF. These findings may have consequences for control of vWF haemostatic activity.

Physical proximity and functional association of glycoprotein 1balpha and protein-disulfide isomerase on the platelet plasma membrane

Platelet function is influenced by the platelet thiol-disulfide balance. Platelet activation resulted in 440% increase in surface protein thiol groups. Two proteins that presented free thiol(s) on the activated platelet surface were protein-disulfide isomerase (PDI) and glycoprotein 1balpha (GP1balpha). PDI contains two active site dithiols/disulfides. The active sites of 26% of the PDI on resting platelets was in the dithiol form, compared with 81% in the dithiol form on activated platelets. Similarly, GP1balpha presented one or more free thiols on the activated platelet surface but not on resting platelets. Anti-PDI antibodies increased the dissociation constant for binding of vWF to platelets by approximately 50% and PDI and GP1balpha were sufficiently close on the platelet surface to allow fluorescence resonance energy transfer between chromophores attached to PDI and GP1balpha. Incubation of resting platelets with anti-PDI antibodies followed by activation with thrombin enhanced labeling and binding of monoclonal antibodies to the N-terminal region of GP1balpha on the activated platelet surface. These observations indicated that platelet activation triggered reduction of the active site disulfides of PDI and a conformational change in GP1balpha that resulted in exposure of a free thiol(s).

Lupus anticoagulants form immune complexes with prothrombin and phospholipid that can augment thrombin production in flow

Lupus anticoagulants (LA) are a family of autoantibodies that are associated with in vitro anticoagulant activity but a strong predisposition to in vivo thrombosis. They are directed against plasma phospholipid binding proteins, including prothrombin. We found that a murine monoclonal antiprothrombin antibody and 7 of 7 LA IgGs tested enhanced binding of prothrombin to 25:75 phosphatidyl serine:phosphatidyl choline vesicles in a concentration-dependent manner. We hypothesized that enhanced binding of prothrombin to phospholipid in the presence of LA IgG might result in increased thrombin production when reactions are performed in flow. Thrombin production by purified prothrombinase components was measured in a phospholipid-coated flow reactor. The flow reactor was incubated with prothrombin, calcium ions, and the IgGs and then perfused with prothrombin, calcium ions, the IgGs, factor Va, and factor Xa. A murine monoclonal antiprothrombin antibody and 4 of 6 LA IgGs from patients with a history of thrombosis increased thrombin production up to 100% over control in the first 15 minutes. In summary, LA IgGs concentrate prothrombin on a phospholipid surface that can augment thrombin production by prothrombinase in flow. These observations suggest that LA might propagate coagulation in flowing blood by facilitating prothrombin interaction with the damaged blood vessel wall.

An Arg/Ser substitution in the second epidermal growth factor-like module of factor IX introduces an O-linked carbohydrate and markedly impairs activation by factor XIa and factor VIIa/Tissue factor and catalytic efficiency of factor IXa

Factor IXR94S is a naturally occurring hemophilia B defect, which results from an Arg 94 to Ser mutation in the second epidermal growth factor (EGF)-like module of factor IX. Recombinant factor IXR94S was activated by factor XIa/calcium with an approximately 50-fold reduced rate and by factor VIIa/tissue factor/phospholipid/calcium with an approximately 20-fold reduced rate compared with wild-type factor IX. The apparent molecular mass of the light chain of factor IXaR94S was approximately 6 kD higher than that of plasma or wild-type factor IX, which was not corrected by N-glycosidase F digestion. This result indicated the presence of additional O-linked carbohydrate in the mutant light chain, probably at new Ser 94. The initial rate of activation of factor X by factor IXaR94S in the presence of polylysine was 7% +/- 1% of the initial rate of activation of factor X by plasma factor IXa, and the kc/Km for activation of factor X by factor IXaR94S/factor VIIIa/phospholipid/calcium was 4% +/- 1% of the kc/Km for activation of factor X by plasma factor IXa/factor VIIIa/phospholipid/calcium. The reduced efficiency of activation of factor X by factor IXaR94S in the tenase enzyme complex was due to a 58-fold +/- 12-fold decrease in kcat with little effect on Km. In conclusion, the R94S mutation had introduced an O-linked carbohydrate, which markedly impaired both activation by factor XIa and turnover of factor X in the tenase enzyme complex.

Microheterogeneity of beta-2 glycoprotein I: implications for binding to anionic phospholipids

Considerable interest is currently focused on the interactions of beta-2 glycoprotein I (beta2GPI) and anti-phospholipid antibodies with anionic phospholipids in an attempt to understand the association between these antibodies and clinical diseases such as thrombosis. The interactions of beta2GPI and anionic phospholipids have only been characterized partially, and the physiological role of this glycoprotein remains uncertain. In this study we have explored in detail the physical and phospholipid-binding characteristics of a number of beta2GPI preparations. We have found (i) that perchloric acid-purification methods are damaging to beta2GPI during purification, (ii) that the dissociation constants of the various preparations for phosphatidylserine vary between 0. 1-2 microM and are considerably weaker than previously reported, (iii) that considerable differences in affinity of the various beta2GPI preparations for anionic phospholipids are obtained when comparing anionic phospholipids immobilized to a solid-phase versus phospholipid assembled in unilamellar vesicles, (iv) that the integrity of the fifth domain of beta2GPI is important for binding immobilized anionic phospholipid but not especially important in binding vesicular anionic phospholipid, and (v) that beta2GPI preparations with differing isoelectric species content bind anionic phospholipids differently, suggesting that varying glycosylation and/or protein polymorphisms impact upon phospholipid binding. These results highlight the importance of assessing the determinants of the interaction of beta2GPI with anionic phospholipids assembled in unilamellar vesicles.

Angiostatin formation involves disulfide bond reduction and proteolysis in kringle 5 of plasmin

Plasmin is processed in the conditioned medium of HT1080 fibrosarcoma cells producing fragments with the domain structures of the angiogenesis inhibitor, angiostatin, and microplasmin. Angiostatin consists of kringle domains 1-4 and part of kringle 5, while microplasmin consists of the remainder of kringle 5 and the serine proteinase domain. Our findings indicate that formation of angiostatin/microplasmin involves reduction of plasmin by a plasmin reductase followed by proteolysis of the reduced enzyme. We present evidence that the Cys461-Cys540 and Cys511-Cys535 disulfide bonds in kringle 5 of plasmin were reduced by plasmin reductase. Plasmin reductase activity was secreted by HT1080 and Chinese hamster ovary cells and the human mammary carcinoma cell lines MCF-7, MDA231, and BT20 but not by the monocyte/macrophage cell line THP-1. Neither primary foreskin fibroblasts, blood monocyte/macrophages, nor macrovascular or microvascular endothelial cells secreted detectable plasmin reductase. In contrast, cultured bovine and rat vascular smooth muscle cells secreted small but reproducible levels of plasmin reductase. Reduction of the kringle 5 disulfide bonds triggered cleavage at either Arg529-Lys530 or two other positions C-terminal of Cys461 in kringle 5 by a serine proteinase. Plasmin autoproteolysis could account for the cleavage, although another proteinase was mostly responsible in HT1080 conditioned medium. Three serine proteinases with apparent Mr of 70, 50, and 39 were purified from HT1080 conditioned medium, one or more of which could contribute to proteolysis of reduced plasmin.

Redox control of exofacial protein thiols/disulfides by protein disulfide isomerase

Protein disulfide isomerase (PDI) facilitates proper folding and disulfide bonding of nascent proteins in the endoplasmic reticulum and is secreted by cells and associates with the cell surface. We examined the consequence of over- or underexpression of PDI in HT1080 fibrosarcoma cells for the redox state of cell-surface protein thiols/disulfides. Overexpression of PDI resulted in 3.6-4. 2-fold enhanced secretion of PDI and 1.5-1.7-fold increase in surface-bound PDI. Antisense-mediated underexpression of PDI caused 38-53% decreased secretion and 10-33% decrease in surface-bound PDI. Using 5,5'-dithio-bis(2-nitrobenzoic acid) to measure surface protein thiols, a 41-50% increase in surface thiols was observed in PDI-overexpressing cells, whereas a 29-33% decrease was observed in underexpressing cells. Surface thiol content was strongly correlated with cellular (r = 0.998) and secreted (r = 0.969) PDI levels. The pattern of exofacial protein thiols was examined by labeling with the membrane-impermeable thiol reactive compound, 3-(N-maleimidylpropionyl)biocytin. Fourteen identifiable proteins on HT1080 cells were labeled with 3-(N-maleimidylpropionyl)biocytin. The intensity of labeling of 11 proteins was increased with overexpression of PDI, whereas the intensity of labeling of 3 of the 11 proteins was clearly decreased with underexpression of PDI. These findings indicated that secreted PDI was controlling the redox state of existing exofacial protein thiols or reactive disulfide bonds.

Evidence for multiple enzyme site involvement in the modulation of thrombin activity by products of prothrombin proteolysis

Kinetic evidence is presented for the interaction of prothrombin with several distinctive topological regions of the thrombin molecule. Modulations of thrombin catalytic activity on the protein substrates prothrombin and prethrombin 1 are demonstrated that involve the fragment 1 and fragment 2 portions. The inhibitory effects are demonstrably non-competitive. In addition to exhibiting non-competitive inhibition, fragment 2 is capable of enhancing proteolysis by thrombin; and therefore to react with a second region of the enzyme. On the basis of the crystallographic studies of the complex between fragment 2 and thrombin (Arni et al., Biochemistry 32 (1992) 4727), this activating site is proposed to be associated with exosite II. The allosteric switch between procoagulant and anticoagulant activities identified from studies by Di Cera (Dang et al., Proc. Natl. Acad. Sci USA 92 (1995) 5977) could be 'thrown' by a macromolecular effector that is generated during thrombin formation--a plausible mechanism for switching that deserves further investigation.

Exposure of the cryptic Arg-Gly-Asp sequence in thrombospondin-1 by protein disulfide isomerase

Thrombospondin-1 is a matrix protein that inhibits proliferation, motility and sprouting of endothelial cells in vitro and angiogenesis in vivo. One mechanism by which thrombospondin-1 may influence endothelial cell biology is through interaction with the endothelial cell alphav beta3 integrin receptor. This interaction is mediated via a cryptic Arg-Gly-Asp sequence in the C-terminal Ca2+-binding region of thrombospondin-1. Exposure of the Arg-Gly-Asp sequence is controlled by disulfide interchange events in the Ca2+-binding loops and C-globular domain. Limited reduction of thrombospondin-1 by dithiothreitol exposes the Arg-Gly-Asp sequence which can bind to the alphav beta3 integrin receptor and support endothelial cell spreading (X. Sun, K. Skorstengaard, D.F. Mosher, J. Cell Biol. 118 (1992) 693-701). Our aim was to identify possible physiological reductants that can mediate Arg-Gly-Asp exposure. We now report that protein disulfide isomerase, which is known to catalyze disulfide interchange in thrombospondin-1 and change its enzyme inhibitory properties and its binding to monoclonal antibodies, was secreted by bovine aortic endothelial cells and deposited on the cell surface. There was an average of approximately 2.2 fg of protein disulfide isomerase on the surface of a bovine aortic endothelial cell. Treatment of thrombospondin-1 with purified protein disulfide isomerase enhanced adhesion of endothelial cells to thrombospondin-1 in an Arg-Gly-Asp-dependent manner through the alphav beta3 integrin receptor and supported cell spreading. Both Ca2+-depleted and Ca2+-replete thrombospondin-1 were substrates for protein disulfide isomerase. These results suggest that endothelial cell derived protein disulfide isomerase may regulate Arg-Gly-Asp-dependent binding of thrombospondin-1.

Significant correlation between thrombospondin 1 and serine proteinase expression in rheumatoid synovium

OBJECTIVE

Thrombospondin 1 (TSP1) is a potent active site inhibitor of leukocyte elastase and cathepsin G. This effect is markedly dependent on the disulfide-bond conformation of TSP1, with one isoform, TSP1(0.1), being the most potent. The aims of this study were to examine the expression of different disulfide-bonded isoforms of TSP1 in inflammatory environments in which elastase and cathepsin G are present in variable amounts, and to determine the relationship between these proteinases and their potential inhibitor.

METHODS

Immunohistochemical staining and histomorphometric analysis were used to examine adjacent sections of synovial tissue from patients with rheumatoid arthritis (RA), osteoarthritis (OA), and meniscal trauma (MT), for expression of TSP1 and the TSP1(0.1) isoform, elastase, cathepsin G, and chymase.

RESULTS

TSP1 localized to vessels and cells within the synovium. TSP1 expression was highly up-regulated in RA (mean density 98 cells and vessels/mm2, compared with 13/mm2 in OA and 17/mm2 in MT). The TSP1(0.1) isoform was found virtually exclusively in RA, with 44% of vascular TSP1 staining being due to the TSP1(0.1) isoform in RA, as compared with 7% in OA (P = 0.0047). Elastase- and cathepsin G-positive cells were abundant in RA, with mean densities of 106 cells/mm2 and 103 cells/mm2, respectively, compared with 2 cells/mm2 and 11 cells/mm2 in OA. There was a wide range of both TSP1 and proteinase expression within the RA group, but samples containing large numbers of elastase- and cathepsin G-positive cells also showed high expression of TSP1, especially TSP1(0.1). A strong correlation was found between elastase or cathepsin G densities and TSP1(0.1) expression in blood vessels (r = 0.86 and r = 0.76 respectively, P < 0.01).

CONCLUSION

TSP1(0.1), with the most potent inhibitory activity in vitro, is specifically up-regulated in RA, and this up-regulation is in proportion to the numbers of surrounding leukocytes containing elastase and cathepsin G. One role of TSP1 may be to act as a matrix-based regulator of leukocyte-derived serine proteinases in vivo.

Interaction of platelet-derived growth factor with thrombospondin 1

Key factors that mediate vascular smooth muscle cell proliferation and migration are platelet-derived growth factor (PDGF) and thrombospondin 1 (TSP1). We now report that PDGFBB bound tightly and specifically to TSP1, that this interaction was markedly dependent on the disulphide bond arrangement in TSP1, and that binding of PDGFBB to TSP1 did not preclude PDGFBB from binding to its receptor on rat aortic vascular smooth-muscle cells. At physiologic ionic strength and pH, PDGFBB bound to Ca2+-depleted TSP1 with a dissociation constant of 11 +/- 2 nM and to Ca2+-replete TSP1 with a dissociation constant of 32 +/- 5 nM. Binding was specific, as both soluble TSP1 and unlabelled PDGFBB competed for binding of iodinated PDGFBB to immobilized TSP1, whereas other platelet alpha-granule proteins did not compete. The tertiary structure of TSP1 is regulated by intramolecular disulphide interchange; we found that catalysis of disulphide interchange in TSP1 by protein disulphide isomerase ablated the binding of PDGFBB. The interaction of PDGFBB with TSP1 was weakened by increasing salt concentration and essentially ablated at 0.65 ionic strength; it was inhibited by heparin with a half-maximal effect at 20 i.u./ml, implying that the binding was mediated largely by ionic interactions. An anti TSP1 monoclonal antibody decreased the binding of iodinated PDGFBB to PDGF receptor on rat aortic vascular smooth-muscle cells by 37 +/- 2%, whereas platelet TSP1 non-competitively inhibited binding of iodinated PDGFBB. Uncomplexed PDGFBB bound to PDGF receptor with an affinity 5 +/- 2 times that of PDGFBB-TSP1 complexes. These results suggest that TSP1 might assist in the targeting of PDGF to its receptor on vascular smooth-muscle cells.

Generation of angiostatin by reduction and proteolysis of plasmin. Catalysis by a plasmin reductase secreted by cultured cells

Extracellular manipulation of protein disulfide bonds has been implied in diverse biological processes, including penetration of viruses and endotoxin into cells and activation of certain cytokine receptors. We now demonstrate reduction of one or more disulfide bonds in the serine proteinase, plasmin, by a reductase secreted by Chinese hamster ovary or HT1080 cells. Reduction of plasmin disulfide bond(s) triggered proteolysis of the enzyme, generating fragments with the domain structure of the angiogenesis inhibitor, angiostatin. Two of the known reductases secreted by cultured cells are protein disulfide isomerase and thioredoxin, and incubation of plasmin with these purified reductases resulted in angiostatin fragments comparable with those generated from plasmin in cell culture. Thioredoxin-derived angiostatin inhibited proliferation of human dermal microvascular endothelial cells with half-maximal effect at approximately 0.2 microg/ml. Angiostatin made by cells and by purified reductases contained free sulfhydryl group(s), and S-carbamidomethylation of these thiol group(s) ablated biological activity. Neither protein disulfide isomerase nor thioredoxin were the reductases used by cultured cells, because immunodepletion of conditioned medium of these proteins did not affect angiostatin generating activity. The plasmin reductase secreted by HT1080 cells required a small cofactor for activity, and physiologically relevant concentrations of reduced glutathione fulfilled this role. These results have consequences for plasmin activity and angiogenesis, particularly in the context of tumor growth and metastasis. Moreover, this is the first demonstration of extracellular reduction of a protein disulfide bond, which has general implications for cell biology.

Binding of fibrin monomer and heparin to thrombin in a ternary complex alters the environment of the thrombin catalytic site, reduces affinity for hirudin, and inhibits cleavage of fibrinogen

Interaction of the blood clotting proteinase, thrombin, with fibrin monomer and heparin to form a thrombin.fibrin monomer.heparin ternary complex is accompanied by a change in thrombin catalytic specificity. Equilibrium binding interactions in the assembly of the ternary complex were characterized quantitatively using thrombin labeled at the active site with a fluorescent probe and related to changes in thrombin specificity toward exosite I-dependent binding of hirudin and cleavage of fibrinogen. Changes in the active site environment accompanying binding of heparin or fibrin to thrombin in binary complexes were reported by fluorescence enhancements which contributed additively to the perturbation accompanying formation of the ternary complex. Quantitative analysis of the interactions supports a preferentially ordered path of ternary complex assembly, in which initial binding of heparin to thrombin facilitates binding of fibrin monomer with an approximately 40-fold increased affinity. Binding of fibrin monomer in the ternary complex decreased the affinity of native thrombin for hirudin by >100-fold and inhibited cleavage of fibrinogen, but this inhibition was overcome when fibrin(ogen)-fibrin interactions occurred. These results support a ternary complex model in which heparin binding through exosite II of thrombin facilitates fibrin monomer binding via exosite I, with accompanying changes in thrombin catalytic specificity resulting from perturbations in the active site and reduced accessibility of exosite I to hirudin and fibrinogen.

Identification of monoclonal antibodies that recognize different disulfide bonded forms of thrombospondin 1

Thrombospondin 1 (TSP1) is a multidomain glycoprotein from platelets and cells which functions in cell-cell and cell-matrix interactions. The structure of TSP1 is regulated by sulfhydryl-disulfide interchange in the carboxy-terminal Ca2(+)-binding loops and globular domain which markedly influence its interaction with cell surface integrins and its inhibition of neutrophil enzymes. We have identified murine monoclonal antibodies that recognized different disulfide-bonded forms of TSP1, made by preparing TSP1 in buffers containing either 0.1 mM or 2 mM Ca2+. Antibody HB8432 recognizes TSP1 prepared in buffers containing either 0.1 or 2 mM Ca2+, while antibodies D4.6 and A65M recognized only TSP1 prepared in buffers containing 0.1 mM Ca2+. The antibodies recognize these different TSP1 preparations either adsorbed to plastic or extracellular matrix. Immunohistochemistry of human rheumatoid synovial tissue using HB8432 resulted in staining of numerous blood vessel walls and matrix cells, while D4.6 and A65M stained a subset of the HB8432 positive blood vessels and only occasionally stained matrix cells. These results suggested that different disulfide-bonded forms of TSP1 were being expressed in different areas of inflamed tissue.

Extracellular matrix is a source of mitogenically active platelet-derived growth factor

Platelet-derived growth factor (PDGF) is a chemotactic and mitogenic agent for fibroblasts and smooth muscle cells and plays a key role in the development of atherosclerotic lesions. PDGF is produced by a number of normal and transformed cell types and occurs as homo- or heterodimers of A and B polypeptide chains. Using Chinese hamster ovary (CHO) cells transfected with various forms of PDGF, we have previously shown that PDGF A(s) (short splice version) is secreted, PDGF A(l) (long splice version) predominantly extracellular matrix-associated, and PDGF B divided between medium, cells, and matrix. In the present study we have demonstrated the mitogenic activity of matrix-localized PDGF in artificial and more physiologically relevant models by culturing Balb/c-3T3 cells (3T3), human foreskin fibroblasts (HFF), and rabbit aortic smooth muscle cells (SMC) on extracellular matrix (ECM) laid down by PDGF-expressing CHO cells and human umbilical vein endothelial cells (HUVEC). These cells responded to the local growth stimulus of PDGF-containing CHO ECM and HUVEC ECM. We showed that 3T3 cells required proteolytic activity to utilize matrix-localized PDGF, as aprotinin and epsilon-ACA inhibited growth and 3T3 cells were shown to possess plasminogen activator activity. HFF and SMC did not appear to require proteolytic activity (including metalloproteinase and serine protease activity) as a prerequisite for mitogenesis but were able to access immobilized PDGF by contact with the matrix. An understanding of the mechanisms whereby the utilization of stored PDGF is controlled in situations of excessive cellular proliferation will aid in the development of therapy for these conditions.

Catalysis of disulfide isomerization in thrombospondin 1 by protein disulfide isomerase

Thrombospondin 1 is a multidomain glycoprotein from platelets and most cells that participates in diverse biological processes. The structure and some functional properties of thrombospondin 1 are regulated by disulfide interchange in the Ca(2+)-binding repeats and C-globular domain. The recent identification of the enzyme, protein disulfide isomerase, on the platelet surface suggested that protein disulfide isomerase may catalyze disulfide isomerization in platelet thrombospondin 1. Protein disulfide isomerase was found to form disulfide-linked complexes with thrombospondin 1, which is consistent with protein disulfide isomerase-mediated rearrangement of disulfide bonds in thrombospondin 1. To quantitate disulfide interchange in thrombospondin 1, perturbation of the enzyme inhibitory properties of platelet thrombospondin 1 were measured, specifically changes in the apparent dissociation constant for inhibition of neutrophil cathepsin G by thrombospondin 1. The inhibition constant increased > or = 10-14-fold following incubation of either Ca(2+)-replete or Ca(2+)-depleted thrombospondin 1 with protein disulfide isomerase and reduced glutathione. The rate of protein disulfide isomerase-catalyzed disulfide interchange in thrombospondin 1 increased linearly with protein disulfide isomerase concentration and the K(m) for reduced glutathione was 0.4 +/- 0.2 mM. Disulfide isomerization in both platelet and fibroblast thrombospondin 1 was probed by measuring perturbation in epitopes for two anti-thrombospondin 1 monoclonal antibodies. Antibody D4.6 binds to the C-terminal Ca(2+)-binding domains which are involved in disulfide interchange, whereas antibody HB8432 binds toward the N-terminus of the thrombospondin 1 subunit. In accordance with the location of these epitopes, incubation of platelet thrombospondin 1 or fibroblast thrombospondin 1 with protein disulfide isomerase and reduced glutathione resulted in 2-fold enhancement of binding of D4.6, whereas binding of HB8432 did not significantly change. In summary, protein disulfide isomerase catalyzes disulfide interchange in thrombospondin 1 which alters binding of neutrophil cathepsin G and antibody D4.6 to thrombospondin 1.

Beta 2-glycoprotein I in thrombosis: evidence for a role as a natural anticoagulant

Although the physiological role of beta2-glycoprotein (B2GPI) is unknown, in vitro evidence indicates that B2GPI may be a natural anticoagulant. In this study we have examined whether fluctuations of plasma B2GPI occur in in vivo coagulation. Serial measurements of B2GPI and other anticoagulant proteins were performed in 51 patients with thrombotic (group 1: six patients with disseminated intravascular coagulation (DIC), group 2: venous (n = 4) or arterial (n = 170 thrombosis) and non-thrombotic disease (group 3: 24 patients undergoing elective surgery). Reductions in plasma B2GPI levels were seen in most patients which were roughly proportional to the severity of their illness. Particularly striking reductions of B2GPI, protein C (PC) and antithrombin III (AT-III) (mean +/- 95% CI: 42.7 +/- 8.6%, 42.1 +/- 14.8%, 39.1 +/- 28.4% respectively) were seen in group 1. The reductions in plasma B2GPI were significantly greater in group 1 than in the other groups. Dilutional factors explain most of the reductions in B2GPI, PC and AT-III in groups 2 and 3, but contribute little to group 1. In conclusion, although B2GPI behaves as a 'negative acute phase reactant', the magnitude of reduction of plasma B2GPI levels, accompanied by reductions in other anticoagulant proteins in patients with DIC, suggests specific consumption of B2GPI in in vivo coagulation. This study provides further evidence that B2GPI is an anticoagulant of physiological importance.

Urokinase binding and catabolism by Hep G2 cells is plasminogen activator inhibitor-1 dependent, analogous to interactions of tissue-type plasminogen activator with these cells

The adherent human hepatoma cell line Hep G2 exhibits receptor mediated endocytosis and catabolism of tissue-type plasminogen activator.plasminogen activator inhibitor type-1 (t-PA.PAI-1) complexes formed when exogenous t-PA combines with endogenous PAI-1 in the extracellular matrix. To determine whether the other major PA, urokinase (u-PA), which also complexes with PAI-1, is metabolised via the same mechanism, 125I-labelled high (hmw) and low (lmw) molecular weight forms of u-PA were incubated with Hep G2 cells at 4 degrees C for 2 hr in the absence and presence of a 100-fold excess of unlabelled ligand in order to detect specific binding. Both hmw and lmw 125I-u-PA formed complexes with PAI-1 and these bound specifically and with high affinity (apparent Kd 3.9 and 4.1 nM, with Bmax 78 x 10(3) and 83 x 10(3) binding sites/cell respectively). Binding by each form of radiolabelled u-PA was inhibited in a dose-dependent fashion by unlabelled t-PA, hmw-u-PA, lmw-u-PA, and by monoclonal anti-PAI-1 antibody. At 37 degrees C, bound hmw and lmw 125I-u-PA.PAI-1 complexes were internalised and degraded rapidly. These findings indicate that the specificity of the previously described receptor which mediates PAI-1 dependent catabolism of t-PA by Hep G2 cells extends to complexes of u-PA with this inhibitor.

Inhibition of heparin activity in plasma by soluble fibrin: evidence for ternary thrombin-fibrin-heparin complex formation

The ability of heparin to dramatically enhance the inactivation of thrombin (IIa) by antithrombin III (ATIII) in buffer is negated through formation of a IIa-fibrin-heparin ternary complex (Hogg and Jackson, Proc Natl Acad Sci USA 86:3619, 1989; Hogg and Jackson, J Biol Chem 265:241, 1990). IIa, in this ternary complex, is protected from inactivation by ATIII. Our aim was to determine whether fibrin also compromises heparin efficacy in plasma. We found that soluble fibrin ablated the heparin-mediated prolongation of the thrombin time with half-maximal effect at 60 nmol/L fibrin. The heparin-mediated prolongation of the activated partial thromboplastin time (APTT) was also reduced by fibrin with half-maximal effects at 140 nmol/L fibrin using 0.12 U/mL heparin and 500 nmol/L fibrin using 0.25 U/mL heparin. The mechanism of inhibition of heparin activity by fibrin in plasma was determined by measuring IIa-ATIII complexes by enzyme-linked immunosorbent assay (ELISA). Fibrin was found to inhibit the heparin-catalyzed inactivation of IIa by ATIII with half-maximal effect at 97 +/- 19 nmol/L fibrin. Fibrin had no effect on the heparin-catalyzed inactivation of factor Xa by ATIII in plasma, using either standard heparin, a heparinoid preparation (Orgaran; Organon, Lane Cove, Sydney, Australia), or low-molecular weight heparin. These findings imply that fibrin is a potent modulator of heparin activity in vivo by inhibiting heparin-catalyzed IIa-ATIII complex formation through formation of ternary IIa-fibrin-heparin complexes.

Identification of possible inhibitory reactive centers in thrombospondin 1 that may bind cathepsin G and neutrophil elastase

Thrombospondin 1 is a multidomain trimeric glycoprotein from platelets and a variety of normal and transformed cells of both mesenchymal and epithelial origin, which functions in cell adhesion and cell-cell interactions. We have recently shown that human thrombospondin 1 binds and inhibits the neutrophil enzymes, neutrophil elastase [Hogg, P.J., Owensby, D.A., Mosher, D.F., Misenheimer, T.M., & Chesterman, C.N. (1993a) J. Biol. Chem. 268, 7139-7146] and cathepsin G [Hogg, P.J., Owensby, D.A., & Chesterman, C.N. (1993b) J. Biol. Chem. 268, 21811-21818]. One mole of thrombospondin 1 trimer binds 3 mol of neutrophil elastase or up to 6 mol of cathepsin G, with site-binding dissociation constants around the nanomolar range, and the enzymes have been shown to interact with thrombospondin 1 in the vicinity of the calcium-binding type 3 repeats. None of the protein modules in this region, or within the whole thrombospondin 1 molecule, have previously been implicated in the inhibition of proteinases. We noted that there are two stretches of eight amino acids each in the human thrombospondin 1 type 3 repeats, residues 735-742 and 794-801, that have striking similarity to a reactive-site consensus sequence derived from selected members of the Kazal and Streptomyces subtilisin inhibitor families. Synthetic peptides corresponding to the putative P5 through P4' residues of both proposed reactive centers interacted efficiently with the active site of cathepsin G and were competitive inhibitors of the fibronectin-degrading and platelet-activating activities of this enzyme, while only the peptide corresponding to residues 793-801 efficiently interacted with the active site of neutrophil elastase and competitively inhibited its fibronectin-degrading activity.(ABSTRACT TRUNCATED AT 250 WORDS)

The gamma-carboxyglutamic acid and epidermal growth factor-like modules of factor IXa beta. Effects on the serine protease module and factor X activation

Blood coagulation factors IX and X are two serine proteases with a similar modular structure. The non-catalytic part of each protein consists of a gamma-carboxyglutamic acid (Gla)-containing module and two modules homologous to the epidermal growth factor (EGF) precursor. We have now found that the NH2-terminal EGF-like module of both factors IX and X inhibits factor Xa formation in a Gla-independent manner, both in the presence and absence of phospholipid and the cofactor, factor VIIIa. In contrast, the COOH-terminal EGF-like module has no such effect. Our data indicate that the NH2-terminal EGF-like module of factor IXa beta interacts either with the corresponding module or with the serine protease module in the substrate, factor X, without affecting the hydrolysis of low molecular weight substrates. Using antibodies as structural probes, we found that Ca2+ binding to the Gla module of factor IXa beta induces a conformational transition in the serine protease module. No evidence was found for a direct interaction between the Gla module and factor VIIIa. We therefore propose that the Gla module in factor IXa beta is indirectly involved in the cofactor interaction, in that Ca2+ binding to sites in this module induces a conformation in the serine protease module that is commensurate with factor VIIIa interaction. In addition, the immunochemical approach revealed a Gla-independent Ca2+ binding site in the serine protease module (apparent Kd of approximately 120 microM) that also might influence its conformation. Antibodies against the EGF-like modules of factor IX were used to probe Ca2+ binding to these modules in intact and in Gla-domainless factor IXa beta. The data indicate a Ca2+ binding site with an apparent Kd of approximately 50 microM in the NH2-terminal EGF-like module of both factor IX species.

Effects of Ca2+ binding on the protease module of factor Xa and its interaction with factor Va. Evidence for two Gla-independent Ca(2+)-binding sites in factor Xa

The assembly of macromolecular complexes containing factors Xa and Va on suitable phospholipid surfaces is crucial for rapid activation of prothrombin. We have used quantitative affinity chromatography to characterize the interaction between factor Va and intact factor Xa on the one hand and between factor Va and factor Xa lacking the gamma-carboxyglutamic acid (Gla)-containing module on the other. The dissociation constants were found to be 1.0 +/- 0.1 and 9.5 +/- 1.8 microM, respectively. There was good agreement between these dissociation constants and the concentrations of active site-inhibited factor Xa and Gla-domainless factor Xa that caused half-maximal inhibition of prothrombin activation. To investigate whether the noncatalytic modules of factor Xa interacted directly with factor Va, intact modules were isolated from proteolytic digests of factor X and used as inhibitors of prothrombin activation. The inhibitory effect observed with the isolated Gla module in the absence of phospholipid was due to inhibition of the amidolytic activity of factor Xa rather than to an interaction with factor Va. The epidermal growth factor-like modules did not inhibit prothrombin activation. Using antibodies specific for calcium-dependent epitopes in the serine protease module of factor Xa we demonstrated that Ca2+ binding to the Gla module alters the conformation of the catalytic module. Half-maximal binding was observed at approximately 0.8 mM Ca2+. Evidence was also obtained for the presence of two Gla-independent Ca(2+)-binding sites in factor Xa. One of these sites, located in the NH2-terminal epidermal growth factor-like module, was half-saturated at approximately 60 microM Ca2+ in intact factor Xa and at approximately 1.2 mM Ca2+ in Gla-domainless factor Xa. This site appeared not to influence the conformation of the protease module. The second site, which was half-saturated at approximately 0.16 mM Ca2+, appeared to reside in the serine protease module and to alter its conformation as judged by binding of antibodies specific for calcium-dependent epitopes.

Thrombospondin 1 is a tight-binding competitive inhibitor of neutrophil cathepsin G. Determination of the kinetic mechanism of inhibition and localization of cathepsin G binding to the thrombospondin 1 type 3 repeats

Thrombospondin 1 was recently shown to bind to and inhibit the activity of neutrophil elastase (Hogg, P. J., Owensby, D. A., Mosher, D. F., Misenheimer, T. M., and Chesterman, C. N. (1993) J. Biol. Chem. 268, 7139-7146). This finding led us to question whether thrombospondin 1 also binds and inhibits the other major serine proteinase of neutrophils, cathepsin G. In a competitive binding assay, cathepsin G bound to thrombospondin 1 reversibly and saturably with a dissociation constant in the low nanomolar range. The kinetic mechanism of inhibition of cathepsin G activity by thrombospondin 1 was determined using the synthetic cathepsin G substrate, Suc-Ala-Ala-Pro-Phe-p-nitroanilide, and is consistent with hyperbolic tight-binding inhibition in which thrombospondin 1 binds cathepsin G and the Michaelis cathepsin G-substrate complex and weakens, but does not abolish, the efficiency of hydrolysis of Suc-Ala-Ala-Pro-Phe-p-nitroanilide. In the presence of 2 mM calcium ions, 2.9 +/- 0.4 mol of cathepsin G interacted with 1 mol of thrombospondin 1 trimer with a site-binding constant of 7.0 +/- 3.5 nM, which reduced the efficiency of hydrolysis of Suc-Ala-Ala-Pro-Phe-p-nitroanilide 8.5 +/- 1.4-fold. A lower limit for the on rate constant of 5 x 10(6) M-1 S-1 was established. The affinity of binding and stoichiometry for the interaction between cathepsin G and thrombospondin 1 was enhanced in the absence of calcium ions. In the presence of EDTA, 5.3 +/- 0.5 mol of cathepsin G interacted with 1 mol of thrombospondin 1 with a site-binding constant of 2.1 +/- 1.6 nM, implying the existence of two binding sites for cathepsin G on each subunit of thrombospondin 1, one or both of which is variably exposed and sensitive to calcium ions. Thrombospondin 1 protected fibronectin from cleavage by cathepsin G and blocked cathepsin G-mediated platelet aggregation. In summary, the binding of cathepsin G to thrombospondin 1 is tight, reversible, and close enough to the active site of cathepsin G to perturb the interactions of a small synthetic substrate and exclude a macromolecular protein substrate and platelets. Using defined proteolytic fragments and different conformers of thrombospondin 1, the binding sites for cathepsin G have been localized to the thrombospondin 1 type 3 repeats.

Anticardiolipin antibodies block the inhibition by beta 2-glycoprotein I of the factor Xa generating activity of platelets

Antiphospholipid antibodies, defined either by lupus anticoagulant (LA) activity or positive anticardiolipin immunoabsorbent assay (ACA) are associated with a predisposition to thromboses, recurrent fetal loss or thrombocytopenia. The mechanisms for these predispositions remain undefined. We have enriched immunoglobulin fractions from two patient plasmas to obtain antibodies with LA activity but no ACA, or conversely, with ACA positivity but no LA, in order to investigate in vitro characteristics which might explain a thrombotic propensity. beta 2-glycoprotein I (beta 2-GPI), the plasma cofactor required for ACA binding to negatively charged phospholipid, has previously been shown to inhibit prothrombinase generation in the presence of activated platelets (8). We now report that beta 2-GPI, at physiological concentrations, inhibits the generation of factor Xa in the presence of activated gel-filtered platelets. Further, ACA interferes with this inhibition, resulting in protracted, unopposed factor Xa generation. This interference with beta 2-GPI, a natural anticoagulant component of plasma, is potentially prothrombotic. LA immunoglobulins behave differently and inhibit factor Xa generation in a manner similar to beta 2-GPI. These findings provide the basis for a previously unsuspected mechanism for thrombosis in patients with aPL.

Thrombospondin is a tight-binding competitive inhibitor of neutrophil elastase

Thrombospondin, a glycoprotein of three identical disulfide-bonded subunits, is a constituent of platelet alpha-granules and a variety of normal and transformed cells and binds to cell surfaces and becomes incorporated into extracellular matrix. It has been implicated in processes such as wound healing and tumor growth and metastasis. In addition, thrombospondin was shown recently to be an inhibitor of the fibrinolytic enzyme, plasmin. In the cause of studying the effects of thrombospondin on other serine proteinases, we found that thrombospondin binds neutrophil elastase in an active-site-dependent manner and competitively inhibits the activity of the enzyme. In a competitive binding assay, neutrophil elastase bound to thrombospondin with a dissociation constant of 17 +/- 7 nM, expressed per mole of thrombospondin trimer, or 52 +/- 20 nM, expressed per mole of thrombospondin subunit. In kinetic studies of the inhibition of the amidolytic activity of neutrophil elastase by thrombospondin, 2.7 +/- 0.3 mol of elastase interacted with 1 mol of thrombospondin trimer with a site-binding constant of 57 +/- 13 nM. Lower limits for the on rate constant of 5 x 10(6) M-1 s-1 and off rate constant of 0.27 s-1 were established. Affinity of binding of neutrophil elastase to thrombospondin was sensitive to ionic strength and calcium ions. Thrombospondin was cleaved by neutrophil elastase, but the site(s) of the limited cleavage are independent of the competitive inhibition of elastase activity by thrombospondin. Neutrophil elastase inactivated with phenylmethylsulfonyl fluoride did not compete with active elastase for binding to thrombospondin, implying that a functional active site is important for the interaction of elastase with thrombospondin. Thrombospondin protected fibronectin from cleavage by neutrophil elastase. In summary, the binding of neutrophil elastase to thrombospondin is tight, reversible, and close enough to the active site of elastase to exclude small synthetic tripeptidyl p-nitroanilide substrates and macromolecular protein substrates. Two potential reactive centers that may be involved in binding elastase have been identified in the calcium-binding type 3 domains of thrombospondin. Neutrophil elastase is the enzyme primarily responsible for degrading and solubilizing connective tissue during inflammatory processes. These findings suggest a previously unsuspected mechanism for regulation of elastase activity at inflammatory sites.

Modulation of fibrinolysis by thrombospondin

Thrombospondin is a large, trimeric glycoprotein secreted by activated platelets and growing cells. Thrombospondin copolymerizes with fibrin during blood coagulation and deposits in extracellular matrix. We found that thrombospondin is a slow (rate constant approximately 6.3 x 10(3) M-1 sec-1), tight-binding (Kd < 10(-9) M) inhibitor of plasmin as determined by loss of amidolytic activity, loss of ability to degrade fibrinogen, and decreased lysis zones in fibrin plate assays (Biochemistry 31: 265-269, 1992). Thrombospondin also slowly inhibits urokinase plasminogen activator. The lysis zone when urokinase is put on fibrin plates made from whole plasma is less if thrombospondin is present. The stoichiometry of inhibition is approximately one mole plasmin:one mole thrombospondin trimer, a somewhat surprising result considering the trimeric nature of thrombospondin. These results indicate that thrombospondin is an important regulator of fibrinolysis and degradation of extracellular matrix, particularly when these processes are initiated by urokinase and even when other inhibitors of fibrinolysis are present.

Effects of gamma-carboxyglutamic acid and epidermal growth factor-like modules of factor IX on factor X activation. Studies using proteolytic fragments of bovine factor IX

Factor IX is a vitamin K-dependent zymogen of a serine protease. The NH2-terminal half of the molecule consists of a Ca(2+)-binding gamma-carboxyglutamic acid (Gla)-containing module and two modules homologous to the epidermal growth factor (EGF) precursor. To elucidate the role of these non-catalytic modules of factor IXa beta in factor X activation, we have isolated and characterized fragments of bovine factor IX, containing one or both of the EGF-like modules as well as these modules linked to the Gla module. The fragments were used as inhibitors of factor IXa beta-mediated factor X activation in a plasma clotting system and in systems with purified components of the Xase complex. Fragments consisting of either the two EGF-like modules of factor IX linked together or the NH2-terminal EGF-like module alone were found to inhibit factor Xa generation both in the presence and absence of the cofactor, factor VIIIa. Moreover, a fragment consisting of the corresponding modules of factor X had a similar effect. We therefore propose that factor IXa beta and factor X interact directly through their EGF-like modules on or in the vicinity of a phospholipid surface. We have also found that the isolated Gla module of factor IX inhibits the formation of factor Xa both in the presence and absence of phospholipid but not in the absence of factor VIIIa. Our results are compatible with a model of the Xase complex, in which both the serine protease part and the Gla module of factor IXa beta interact with factor VIIIa.