Short-term glucocorticoid treatment of piglets causes changes in growth plate morphology and angiogenesis.

OBJECTIVE: Glucocorticoid treatment of children often leads to growth retardation, and the precise target(s) in the growth plate responsible for this effect are unknown. Angiogenesis is an important part of the endochondral ossification process, and VEGF expressed in the growth plate is essential for proper angiogenesis to occur. Since glucocorticoid treatment down-regulates VEGF expression in cultured chondrocytes, we hypothesized that in vivo glucocorticoid treatment could result in VEGF down-regulation in the growth plate and disturbed angiogenesis, thus contributing to the growth retardation. DESIGN: We treated 6-week-old prepubertal piglets (10 kg) for 5 days with prednisolone (50 mg/day). Tibial growth plate sections were studied for apoptosis and the expression of VEGF protein and mRNA and MMP-9 protein. Capillaries in the metaphysis were visualized by CD31 immunostaining. Growth plate morphology (width of various zones) was determined by interactive measurements on hematoxylin/eosin stained sections and apoptotic cells were detected by TUNEL assay. RESULTS: In the prednisolone-treated animals, the total width of the growth plate decreased to 81% of controls (P<0.02), which was explained by a decrease of the width of the proliferative zone to 73% (P<0.05). The treatment had no effect on the orderly organization of the chondrocyte columns. In the growth plates of control animals, apoptosis was shown in 5.8% of the hypertrophic chondrocytes and was limited to the terminal hypertrophic chondrocytes. In prednisolone-treated animals, 40.5% of the hypertrophic chondrocytes was apoptotic (P<0.02), with apoptotic chondrocytes also appearing higher in the hypertrophic zone. We observed fewer capillaries and loss of their parallel organization in the metaphysis in the prednisolone-treated animals. The capillaries were shorter and chaotic in appearance. In contrast to controls, in prednisolone-treated animals VEGF mRNA and protein could not be detected in the hypertrophic zone of the growth plate. Trabecular bone length in the primary spongiosa was also diminished by the treatment. No changes were observed in the expression pattern of MMP-9, a matrix metalloproteinase, which is also important for angiogenesis and bone formation. CONCLUSIONS: These results indicate that short-term glucocorticoid treatment of growing piglets severely disturbs the width of the growth plate, apoptosis of chondrocytes, VEGF expression by hypertrophic chondrocytes, the normal invasion of blood vessels from the metaphysis to the growth plate and bone formation at the chondro-osseous junction. These effects could alter the dynamics of endochondral ossification and thus contribute to glucocorticoid-induced growth retardation.

Short-term glucocorticoid treatment of prepubertal mice decreases growth and IGF-I expression in the growth plate.

The insulin-like growth factor (IGF) system is an important mediator of postnatal longitudinal growth, and the growth inhibiting effects of glucocorticoid (GC) treatment are suggested to be due to impaired action of the IGF system. However, the precise changes of the IGFs and the IGF-binding proteins (IGFBPs) in the growth plate, occurring upon short-term GC treatment have not been characterized. Prepubertal mice treated daily with dexamethasone (DXM) for 7 days, showed significant growth inhibition of total body length and weight and weight of the liver, thymus and spleen, whereas the weight of the kidneys was not affected. Analysis of the tibial growth plate showed that the total growth plate width significantly decreased to 84.5% of control values, caused by a significant decrease in the proliferative zone. The number of proliferating cell nuclear antigen (PCNA)-positive chondrocytes in the proliferative zone decreased significantly (to 40%) and TUNEL staining showed a significant 1.6-fold increase in apoptotic hypertrophic chondrocytes. In the growth plates, both IGF-I and IGF-II, as well as IGFBP-2 mRNAs were detected, mainly in the proliferative and prehypertrophic zones. DXM treatment significantly decreased the number of chondrocytes expressing IGF-I, whereas the number of chondrocytes expressing IGF-II and IGFBP-2 were not affected. The decrease in IGF-I expression in the growth plate indicates that GC treatment affects IGF-I at the local level of the growth plate, which could contribute to the GC-induced growth retardation.

IGF and IGF-binding protein expression in the growth plate of normal, dexamethasone-treated and human IGF-II transgenic mice.

Glucocorticoid (GC) treatment in childhood can lead to suppression of longitudinal growth as a side effect. The actions of GCs are thought to be mediated in part by impaired action of the insulin-like growth factors (IGF-I and IGF-II) and their binding proteins (IGFBP-1 to -6). We have studied the effects of GCs on IGF and IGFBP expression at the local level of the growth plate, using non-radioactive in situ hybridization. We treated 3-week-old normal mice for 4 weeks with dexamethasone (DXM). We also treated human IGF-II (hIGF-II) transgenic mice in order to investigate whether IGF-II could protect against the growth retarding effect of this GC. DXM treatment resulted in general growth retardation in both mice strains, however, only in normal mice was tibial length decreased. In both normal and hIGF-II trangenic mice, the total width of the growth plate was not affected, whereas the width of the proliferative zone decreased as a result of the DXM treatment. Additionally, only in normal mice, the width of the hypertrophic zone thickened. Only expression of IGF-I, IGF-II and IGFBP-2 could be detected in the growth plates of 7-week-old normal mice. IGFBP-1, -3, -4, -5 and -6 mRNAs were not detected. DXM treatment of normal mice induced a significant 2.4-fold increase in the number of cells expressing IGF-I mRNA, whereas IGF-II and IGFBP-2 mRNA levels were not affected. In hIGF-II transgenic mice, IGF-I mRNA levels were significantly increased, while endogenous IGF-II and IGFBP-2 mRNAs were unaffected, compared to normal animals. DXM treatment of the hIGF-II transgenic mice induced a further increase of IGF-I mRNA expression, to a similar extent as in DXM-treated normal mice. The increase of IGF-I due to DXM treatment in normal mice might be a reaction in order to minimize the GC-induced growth retardation. Another possibility could be that the increase of IGF-I would contribute to the GC-induced growth retardation by accelerating the differentiation of chondrocytes, resulting in accelerated ossification. In the growth plates of hIGF-II transgenic mice, the higher basal level of IGF-I, might be responsible for the observed partial protection against the adverse effects of GCs on bone.

Dexamethasone-induced growth inhibition of porcine growth plate chondrocytes is accompanied by changes in levels of IGF axis components.

High (pharmacological) doses of glucocorticoids inhibit the proliferation of growth plate chondrocytes, which leads to one of the side-effects of these steroids, namely suppression of longitudinal growth. Growth inhibition by glucocorticoids is thought to be mediated in part by impaired action of components of the IGF axis, which are important for chondrocyte regulation and hence for longitudinal growth. The aim of the present study was to determine whether glucocorticoid-induced growth retardation involves changes in IGF axis components. Chondrocytes were isolated from epiphyseal growth plates of neonatal piglets and treated with pharmacological doses of dexamethasone (DXM) for 24 h to study glucocorticoid-induced growth retardation. Under IGF-I-supplemented (10 nM) culture conditions, IGF-binding proteins (IGFBPs)-2, -4 and -5 were secreted by the growth plate chondrocytes and IGFBP-2 protein and mRNA levels were decreased by the DXM treatment, whereas IGFBP-4 and -5 were not affected. Proliferation of the chondrocytes, as measured by [(3)H]thymidine incorporation, was 3.5-fold higher in serum-supplemented medium in contrast to IGF-I-supplemented (10 nM) medium. In the presence of serum, DNA synthesis was significantly inhibited by 50-63% when treated with 100 nM DXM, which was prevented by the glucocorticoid-receptor antagonist Org34116. mRNA levels of IGF axis components were determined using Northern blot analysis. IGFBP-2 to -6 were expressed in the chondrocytes, IGFBP-1 was absent and both IGF-I and IGF-II, and the type I and type II IGF receptors were expressed. Treatment with DXM (100 nM) resulted in a 2-fold increase in mRNA levels of both IGFBP-5 and the type I IGF receptor, whereas IGFBP-2 mRNA levels decreased by 55%, in concert with the decrease in protein level observed under IGF-I-supplemented culture conditions. The changes in mRNA levels due to the DXM treatment were prevented by the glucocorticoid receptor antagonist. Our data show that exposure to pharmacological doses of DXM results in inhibition of proliferation and changes in components of the IGF axis, IGFBP-2 and -5 and the type I IGF receptor, suggesting a role for these components in glucocorticoid-induced growth retardation at the local level of the growth plate.

Differential regulation of IGF-binding proteins in rabbit costal chondrocytes by IGF-I and dexamethasone.

Cartilage is a primary target tissue for the IGFs. The mitogenic activity of these peptides is regulated by a family of high-affinity IGF-binding proteins (IGFBP-1 to -6). We characterized the IGFBPs produced by cultured chondrocytes derived from rib cartilage of prepubertal rabbits. Culture medium, which had been conditioned by these cells for 48 h showed bands of 22 kDa, 24 kDa and a 31/32 kDa doublet by Western ligand blotting with [(125)I]IGF-II. When the cells were grown in the presence of increasing amounts of IGF-I or IGF-II, the 31/32 kDa doublet increased in intensity (reaching a plateau of about 11-fold stimulation between 2 and 10 nM IGF-I). The 22 kDa and 24 kDa bands increased only slightly while a 26 kDa band became faintly visible. By Western immunoblotting the 31/32 kDa doublet was identified as IGFBP-5. An IGF-I analog with reduced affinity for IGFBPs, Long-R3 IGF-I, also induced IGFBP-5, while insulin was less effective (2.2-fold stimulation at 10 nM). IGF-I protected IGFBP-5 against proteolytic degradation by conditioned medium. IGF-I also enhanced the level of IGFBP-5 mRNA. LY294002, a specific inhibitor of the intracellular signaling molecule phosphatidylinositol 3-kinase, inhibited stimulation of IGFBP-5 by IGF-I. Dexamethasone suppressed IGFBP-5 (by 70% at 20 nM) but, at the same time, a 39/41 kDa doublet (presumably IGFBP-3) was induced. IGFBP-5 has been shown in several cell types to enhance the mitogenic activity of IGF-I. IGFBP-3 generally acts as a growth inhibitor. Therefore, the differential effects of dexamethasone on these regulatory proteins could account, at least in part, for the growth-arresting effect of this glucocorticoid.

Human insulin-like growth factor (IGF) binding protein-1 inhibits IGF-I-stimulated body growth but stimulates growth of the kidney in snell dwarf mice.

The actions of insulin-like growth factor-I (IGF-I) are modulated by IGF binding proteins (IGFBPs). The effects of IGFBP-1 in vivo are insufficiently known, with respect to inhibitory or stimulatory actions on IGF-induced growth of specific organs. Therefore, we studied the effects of IGFBP-1 on IGF-I-induced somatic and organ growth in pituitary-deficient Snell dwarf mice. Human GH, IGF-I, IGFBP-1, and a preequilibrated combination of equimolar amounts of IGF-I and IGFBP-1 were administered sc during 4 weeks. Treatment with IGF-I alone induced a significant increase in body length (108% of control) and weight (112%) as well as an increase in weight of the submandibular salivary glands (135%), kidneys (124%), femoral muscles (111%), testes (129%), and spleen (126%) compared with saline-treated controls. IGFBP-1 alone induced a significant increase in weight of the kidneys (152% of control). Coadministration of IGF-I with IGFBP-1 neutralized the stimulating effects of IGF-I on body length and weight as well as on the femoral muscles and testes. In contrast, the weights of the submandibular salivary glands (143%) were not significantly different from those of IGF-I-treated animals, whereas the weights of the kidneys (171%) and spleen (156%) were significantly increased compared with IGF-I-treated mice. The effect of IGFBP-1 plus IGF-I on kidney weight was not significantly greater than the effect of IGFBP-1 alone. Western ligand blotting showed induction of the IGFBP-3 doublet as well as IGFBPs with molecular masses of 24 kDa, most probably IGFBP-4, by human GH, IGF-I alone, and IGF-I in combination with IGFBP-1. Our data show that coadministration of IGFBP-1 inhibits IGF-I-induced body growth of GH-deficient mice but significantly stimulates the growth promoting effects of IGF-I on the kidneys and the spleen. These data warrant further investigation because differences in concentrations of IGFBP-1 occurring in vivo may influence IGF-I-induced anabolic processes.

Insulin-like growth factor-binding protein-3 protease activity in Snell normal and Pit-1 deficient dwarf mice.

Partial proteolysis of insulin-like growth factor-binding protein-3 (IGFBP-3) lowers its affinity for IGFs. Presumably, this leads to destabilization of the ternary IGF-IGFBP-3-acid-labile subunit complex in the circulation and an increased bioavailability of IGFs. We investigated the effect of GH on IGFBP-3 proteolysis by comparing serum from normal mice and GH-deficient dwarf mice. While normal mouse serum degraded 125I-IGFBP-3, this activity declined with age. In contrast, serum from dwarf mice displayed strong proteolytic activity at all ages tested (up to 10 weeks). In dwarf mice of 4 weeks and older, this activity could not be inhibited by EDTA and 1,10-phenanthroline, indicating the presence of a divalent cation-independent protease. Prolonged treatment with GH (4 weeks) did not decrease the overall potency of the serum to degrade IGFBP-3, but partially restored the ability of EDTA to inhibit IGFBP-3 protease activity. GH deficiency therefore appears to induce a new kind of IGFBP-3 protease. Similarly, serum from hypophysectomized rats displayed enhanced IGFBP-3 protease activity compared with control rat serum. These results suggest that a protease induced under conditions of severe GH deficiency may contribute to making IGFs optimally available to the tissues.

Insulin-like growth factor binding proteins-3 and -5 form sodium dodecyl sulfate-stable multimers.

Insulin-like growth factor binding proteins (IGFBPs) are important modulators of IGF actions. IGFBP-3 and IGFBP-5 can bind to the extracellular matrix of a number of cell types. We now describe a new posttranslational structural modification of IGFBP-3 and IGFBP-5, which could play a role in determining their localization. We incubated radioiodinated forms of all six IGFBPs in the presence of a redox buffer consisting of 10 mM reduced glutathione and 0.2 mM oxidized glutathione. Under these conditions IGFBP-3 and IGFBP-5, but not the other IGFBPs, formed high molecular weight disulfide-linked multimers. Heparin and a peptide encompassing the high-affinity heparin-binding site in the C-terminal portion of IGFBP-3 were capable of blocking the multimerization of IGFBP-3. IGFBP-3, but not IGFBP-1, was shown to be able to self-associate non-covalently, which could be a requisite first step in the formation of covalent multimers. The self-association of IGFBP-3 required the high-affinity heparin-binding site in the C-terminal portion of the molecule.

Solid-phase assay for insulin-like growth factor (IGF) binding to IGF-binding protein-3: application to the study of the effects of antibodies and heparin.

The availability and activity of insulin-like growth factors (IGF-I and IGF-II) are largely determined by a group of IGF-binding proteins (IGFBPs). We have developed a new assay to characterize the interaction between the IGFs and IGFBP-3. In this assay, recombinant IGFBP-3 (5 ng) was immobilized on plastic microtitre wells, after which radiolabelled IGF-I or -II was allowed to bind. The assay is highly specific, since neither IGF bound to control wells blocked with albumin. By constructing both saturation and competition binding curves, equivalence of binding between the radiolabelled and native IGF ligands could be demonstrated. From these curves, reliable specific activities of the tracers were calculated. Scatchard plots of both types of data produced identical results for dissociation constants and number of binding sites. The affinity of IGF-II was twice as high as the affinity of IGF-I (dissociation constants of 44 and 102 pM respectively). The assay was used to show that polyclonal anti-IGFBP-3 antibodies could block binding of IGF. Alkylating agents and NaCl were without effect, but chaotropic salts such as CaCl2 and NaSCN decreased IGF binding to IGFBP-3. IGFBP-1 and IGFBP-3, but not an N-terminal fragment of IGFBP-3, could effectively block binding of both IGF-I and IGF-II to the solid-phase IGFBP-3. Increasing concentrations of heparin had little or no effect on IGF-I binding, but strongly inhibited IGF-II binding. This was shown to be a consequence of a decrease in both the affinity and the number of binding sites. Possibly, the interaction of IGFBP-3 with heparin or heparin-like structures in vivo can lead to the selective release of IGF-II from this binding protein. Our results with heparin also suggest that the binding sites on IGFBP-3 for IGF-I and IGF-II are not completely identical. This assay can be applied to the study of various aspects of the interaction between the IGFs and IGFBP-3, such as the effects of interfering substances and structure-function relationships of both moieties of the complex.

Characterization of O-glycosylated precursors of insulin-like growth factor II by matrix-assisted laser desorption/ionization mass spectrometry.

High molecular weight precursors of insulin-like growth factor II (IGF-II) were isolated from Cohn fraction IV of human plasma by ultrafiltration, affinity chromatography and reversed-phase high-performance liquid chromatography. Molecular weight determination by matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS) of two high molecular weight IGF-II preparations revealed heterogeneous glycosylation. A combination of enzymatic degradation and MALDI-MS were applied for further structural characterization of the glycosylated precursors of IGF-II. The first step was molecular weight determination of intact high molecular weight IGF-IIs prior to and after treatment with neuraminidase and O-glycosidase. This, together with a comparison of molecular weight information available from the cDNA, revealed that both high molecular weight IGF-II species contain an identical C-terminal extension of 20 residues but different degrees of glycosylation. Second, comparative Endo Glu-C digestion of the preparations prior to and after enzymatic release of carbohydrates and subsequent remeasurement of the molecular weight by MALDI-MS confirmed the primary structure of precursor IGF-II1-87. The O-linked carbohydrates were found to be associated with the C-terminal extension and the heterogeneity was identified as varied sialylated forms of one and two HexNAc-Hex groups.

von Willebrand factor as a regulator of intrinsic factor X activation.

Factor VIII is an important cofactor in the intrinsic activation of factor X. To function effectively as a cofactor, factor VIII must be activated. In plasma, factor VIII circulates in a complex with von Willebrand factor, and although thrombin can activate complexed factor VIII, the activation by activated factor X is inhibited by von Willebrand factor. In this study, the effect of von Willebrand factor on the generation of factor Xa by the factor IXa-VIII complex was investigated. Purified human factors VIII, IXa, and X were incubated on human umbilical vein endothelial cells or phospholipid vesicles in the presence of calcium ions, and the generation of factor Xa was followed. In the presence of von Willebrand factor, a prolonged lag-phase and a dose-dependent inhibition of factor X activation was observed. These effects were not observed when von Willebrand factor was preincubated with a monoclonal antibody directed against von Willebrand factor that blocks factor VIII binding. When factor VIII was activated with thrombin before the incubation, neither the monoclonal antibody nor von Willebrand factor had an effect on the rate of factor X activation. Preincubation of endothelial cells with the monoclonal antibody resulted in a somewhat higher rate of factor X activation. When endothelial cells from a patient with von Willebrand's disease type I were used, preincubation of the monoclonal antibody had no effect on the rate of factor X activation. We conclude that von Willebrand factor on the surface of endothelial cells can modulate the intrinsic factor X activation. This effect is greatly enhanced, however, by the addition of exogenous von Willebrand factor.(ABSTRACT TRUNCATED AT 250 WORDS)

Insulin-like growth factors (IGFs) and IGF binding protein-3 display disulfide isomerase activity.

Insulin-like growth factors (IGF-I and -II) bind with high affinity to IGF-binding proteins (IGFBPs). IGFBP-3 contains vicinal cysteines in sequence which is similar to the active sites in thioredoxin and protein disulfide isomerase. We tested if, in analogy with these redox enzymes, IGFBP-3 could catalyze the isomerization of intramolecular disulfide bridges in protein substrates. IGFBP-3 (30 microM) was able to reactivate reduced ribonuclease at a rate of 38% of that of thioredoxin. Also recombinant IGF-I induced the regeneration of ribonuclease activity. Thiol redox reactions are known to play a role in regulating conformational changes in the insulin receptor and possibly also in the IGF-I receptor. Therefore, the intrinsic isomerase activities of IGF-I may be important in the activation of its receptor. The observed effects of IGFBP-3 may help to elucidate the mechanism by which this binding protein can modulate the actions of IGF-I.

Divergent fates of P- and E-selectins after their expression on the plasma membrane.

P-selectin and E-selectin are related adhesion receptors for monocytes and neutrophils that are expressed by stimulated endothelial cells. P-selectin is stored in Weibel-Palade bodies, and it reaches the plasma membrane after exocytosis of these granules. E-selectin is not stored, and its synthesis is induced by cytokines. We studied the fate of the two proteins after their surface expression by following the intracellular routing of internalized antibodies to the selectins. By immunofluorescent staining, P-selectin antibody was first seen in endosomes, then in the Golgi region, and finally in Weibel-Palade bodies. In contrast, the E-selectin antibody was detected only in endosomes and lysosomes. Subcellular fractionation of cells after 4 h chase confirmed the localization of P-selectin antibody in storage granules and of the E-selectin antibody in lysosomes. In AtT-20 cells, a mouse pituitary cell line, transfected with P- or E-selectin, only P-selectin was delivered to the endogenous adrenocorticotrophic hormone storage granules after endocytosis. Deletion of the cytoplasmic domain abolished internalization. In summary, after a brief surface exposure, internalized E-selectin is degraded in the lysosomes, whereas P-selectin returns to the storage granules from where it can be reused.

P-selectin, a granule membrane protein of platelets and endothelial cells, follows the regulated secretory pathway in AtT-20 cells.

P-selectin (PADGEM, GMP-140, CD62) is a transmembrane protein specific to alpha granules of platelets and Weibel-Palade bodies of endotheial cells. Upon stimulation of these cells, P-selectin is translocated to the plasma membrane where it functions as a receptor for monocytes and neutrophils. To investigate whether the mechanism of targeting of P-selectin to granules is specific for megakaryocytes and endothelial cells and/or dependent on von Willebrand factor, a soluble adhesive protein that is stored in the same granules, we have expressed the cDNA for P-selectin in AtT-20 cells. AtT-20 cells are a mouse pituitary cell line that can store proteins in a regulated fashion. By double-label immunofluorescence, P-selectin was visible as a punctate pattern at the tips of cell processes. This pattern closely resembled the localization of ACTH, the endogenous hormone produced and stored by the AtT-20 cells. Fractionation of the transfected cells resulted in the codistribution of P-selectin and ACTH in cellular compartments of the same density. Immunoelectron microscopy using a polyclonal anti-P-selectin antibody demonstrated immunogold localization in dense granules, morphologically indistinguishable from the ACTH granules. Binding experiments with radiolabeled monoclonal antibody to P-selectin indicated that there was also surface expression of P-selectin on the AtT-20 cells. After stimulation with the secretagogue 8-Bromo-cAMP the surface expression increased twofold, concomitant with the release of ACTH. In contrast, the surface expression of P-selectin transfected into CHO cells, which do not have a regulated pathway of secretion, did not change with 8-Br-cAMP treatment. In conclusion, we provide evidence for the regulated secretion of a transmembrane protein (P-selectin) in a heterologous cell line, which indicates that P-selectin contains an independent sorting signal directing it to storage granules.

Characterization of human factor VIII and interaction with von Willebrand factor. An electron microscopic study.

Blood coagulation factor VIII is a large glycoprotein that circulates in plasma at relative low concentration (0.1 microgram/ml). It consists of a heterogeneous mixture of a series heavy-chain peptides (90-200 kDa), each associated with a light chain of 80 kDa. To gain insight into the physical properties of the protein, we have characterized purified human factor VIII by electron microscopy and rotary shadowing. Electron microscopy of rotary shadowed factor VIII molecules showed predominantly a single globular domain structure, with a somewhat asymmetric shape, while two-domain structures were also encountered. The overall dimensions of the globular domains ranged from 4 x 6 nm to 8 x 12 nm. EDTA treatment of factor VIII reduced the overall dimensions (2.5 x 5 nm to 6 x 10 nm) while treatment with thrombin reduced the dimensions to a small extent. In complexes with von Willebrand factor, factor VIII appeared localized at the globular domains of von Willebrand factor multimers. In addition, incubation of factor VIII with Staphylococcus aureus V8 protease fragments SpII and SpIII revealed only binding to the globular domains of SpIII. In this study, the first morphological characterization of human factor VIII is presented, together with its direct localization on von Willebrand factor multimers.

The effect of von Willebrand factor on activation of factor VIII by factor Xa.

Factor VIII has to be activated before it can serve efficiently as a cofactor in the intrinsic pathway of blood coagulation. This activation occurs through specific proteolytic cleavages in the molecule by either thrombin or factor Xa. In this study, we show that von Willebrand factor inhibits the activation of factor VIII by factor Xa. Incubation of factor VIII (30 U/ml) with 0.1 microgram/ml factor Xa resulted in a 1.6-fold activation followed by a decay of coagulant activity. In the presence of 10 micrograms/ml von Willebrand factor, activation and inactivation of factor VIII was completely inhibited. In contrast, the activation of factor VIII by thrombin was not influenced by von Willebrand factor. At high concentrations of factor Xa (10 micrograms/ml), von-Willebrand-factor-bound factor VIII could be cleaved and activated. The generated proteolytic fragments were identical to the fragments produced in the absence of von Willebrand factor and all fragments were released from von Willebrand factor. The major products were light-chain-derived fragments of molecular mass 66/68 kDa and 60 kDa and heavy-chain-derived fragments of 40 and 42 kDa. Also minor products of 12, 20/21, 23, 27 and 30 kDa were observed, most of which were specific for cleavage of factor VIII by factor Xa.

Inactivation of human factor VIII by activated protein C. Cofactor activity of protein S and protective effect of von Willebrand factor.

Activated protein C (APC) acts as a potent anticoagulant enzyme by inactivating Factor V and Factor VIII. In this study, protein S was shown to increase the inactivation of purified Factor VIII by APC ninefold. The reaction rate was saturated with respect to the concentration of protein S when protein S was present in a 10-fold molar excess over APC. The heavy chain of Factor VIII was cleaved by APC and protein S did not alter the degradation pattern. Factor VIII circulates in a complex with the adhesive protein von Willebrand factor. When purified Factor VIII was recombined with von Willebrand factor, the inactivation of Factor VIII by APC proceeded at a 10-20-fold slower rate as compared with Factor VIII in the absence of von Willebrand factor. Protein S had no effect on the inactivation of the Factor VIII-von Willebrand factor complex by APC. After treatment of this complex with thrombin, however, the actions of APC and protein S towards Factor VIII were completely restored. In hemophilia A plasma, purified Factor VIII associated with endogenous von Willebrand factor, resulting in a complete protection against APC (4 nM). By mixing hemophilic plasma with plasma from a patient with severe von Willebrand's disease, we could vary the amount of von Willebrand factor. 1 U of von Willebrand factor was needed to provide protection of 1 U Factor VIII. Also in plasma from patients with the IIA-type variant of von Willebrand's disease, Factor VIII was protected. In von Willebrand's disease plasma, which was depleted of protein S, APC did not inactivate Factor VIII. These results indicate that protein S serves as a cofactor in the inactivation of Factor VIII and Factor VIIIa by APC and that von Willebrand factor can regulate the action of these two anticoagulant proteins.

The effect of thrombin on the complex between factor VIII and von Willebrand factor.

Purified human factor FVIII (FVIII; 6000-8000 U/mg) was radiolabeled and bound to immobilized von Willebrand factor (vWF). The complex was incubated with human thrombin. Thrombin induced a release of 65% of the radioactivity initially bound. Released FVIII fragments and fragments remaining bound during incubation with thrombin were analyzed using gel electrophoresis. This led to the following observations. Released fragments largely consisted of Mr-70000 and Mr-50000 fragments; Mr-90000 and Mr-80000 fragments were only found in the fractions remaining bound to vWF and decreased with time. In contrast to these digestion products of FVIII, the Mr-42000 heavy-chain fragment remained bound to vWF, comprising the larger part of the radioactivity after a 2-h incubation. No thrombin-induced cleavages were observed in vWF. Furthermore, vWF-coated wells preincubated with thrombin were still able to bind 125I-FVIII. These results implicate a new concept for the activation of vWF-bound FVIII. Activation is a multistep process in which several cleavages are necessary to produce and release a coagulant-active FVIII molecule (FVIIIa), which is probably an Mr-50000/70000 heterodimer. Inactivation of FVIIIa is likely to be the result of a nonproteolytic dissociation due to loss of the joining divalent cation(s).

Factor VIII binds to von Willebrand factor via its Mr-80,000 light chain.

The formation of a complex between factor VIII (FVIII) and von Willebrand factor (vWF) was studied using purified radiolabeled human FVIII and purified human vWF. A binding assay was developed in which vWF was coated on microtiter wells. FVIII was shown to bind specifically and reversibly to the immobilized vWF. At a coating of 70 pg vWF/well, binding was half-maximal at a FVIII concentration of 70 +/- 10 pM. In order to ascertain which part of FVIII interacted with vWF, eight monoclonal antibodies, directed against FVIII, were tested for their ability to inhibit FVIII-vWF interaction. One of the eight antibodies, CLB-CAg:58, inhibited binding completely. This antibody was demonstrated to react with the Mr-80,000 light chain of FVIII. Direct evidence for the involvement of this chain in vWF binding was obtained by studying the binding of isolated, radiolabeled FVIII heavy and light chains. In a typical experiment 23-30% of the radioactivity bound when the FVIII light chain was added and less than 1% when the FVIII heavy chain was added.

Human factor VIII: purification from commercial factor VIII concentrate, characterization, identification and radiolabeling.

Human factor VIII was purified from commercial factor VIII concentrate with a 12% yield. The specific coagulant activity of purified factor VIII was 8,000 units/mg. In the presence of SDS the purified factor VIII consisted of a variety of polypeptides on polyacrylamide gels, ranging between Mr 80,000 and Mr 208,000. In the absence of SDS the purified factor VIII showed an apparent molecular weight of 270,000 upon Sephadex G200 gel-filtration. The purified factor VIII could be activated by thrombin, which resulted in the disappearance of Mr 108,000-208,000 polypeptides in favor of an Mr 92,000 polypeptide. Treatment with factor Xa also activated factor VIII, whereas treatment with activated protein C resulted in the inactivation of coagulant activity. Coagulant-active 125I-factor VIII was prepared using a lactoperoxidase radioiodination procedure. This 125I-factor had the same characteristics as unlabeled factor VIII. All polypeptides could be precipitated with monoclonal antibodies directed against factor VIII. With 125I-factor VIII a pIapp of 5.7 was found in the presence of urea.