Calcium-dependent self-association of the C-type lectin domain of versican.

Versican is a large (1-2 x 10(6) Da) chondroitin-sulfate proteoglycan that can form large aggregates by means of interaction with hyaluronan and also binds to a series of other extracellular matrix proteins, chemokines and cell-surface molecules. Versican is a multifunctional molecule with roles in cell adhesion, matrix assembly, cell migration and proliferation. Characterization of the binding interactions mediated by the various domains of versican is a first step towards understanding the functions of versican and interacting molecules in the extracellular matrix. In this study we investigated a recombinant construct corresponding to the C-type lectin domain of versican and demonstrated a calcium-dependent self-association of this region by blot overlay and plasmon surface resonance assays. Electron microscopy provided further evidence of the relevance of the binding reaction by demonstrating a mixture of monomers, dimers and complex aggregates of recombinant versican C-type lectin domain. This binding reaction could contribute to the ability of versican to organize formation of the proteoglycan extracellular matrix by inducing binding of individual versican molecules or by modulating binding reactions to other matrix components.

Prion protein stimulates tissue-type plasminogen activator-mediated plasmin generation via a lysine-binding site on kringle 2.

Recombinant human prion-protein (PrP23-231) stimulates plasminogen activation by tissue-type plasminogen activator (t-PA). The stimulatory activity is conserved in the N-terminal fragment (PrP23-110). It has further been shown by others that PrP(c) binds to kringle-domains of plasminogen. We compared the stimulatory activity of recombinant PrP23-231 and PrP23-110 on plasminogen activation catalyzed by t-PA, urokinase (u-PA), streptokinase and Desmodus salivary plasminogen activator (DSPAalpha1). As these plasminogen activators are distinct, with respect to their kringle domains we studied their binding to immobilized PrP23-110. Plasminogen activation was measured in a chromogenic assay in vitro and binding studies were carried out using surface plasmon resonance technology. We found that recombinant full-length prion protein, PrP23-231, and PrP23-110 specifically stimulate t-PA mediated plasminogen activation. Two hundred nanomoles per liter of PrP23-110 stimulated 1.8 nmol L(-1) t-PA 48-fold, 180 nmol L(-1) DSPA(alpha1) 2.5-fold, 1.8 nmol L(-1) u-PA 1.1-fold, and 1.8 nmol L(-1) streptokinase 1.8-fold. Our data show no specific binding for streptokinase. In contrast all plasminogen activators carrying a kringle domain bound to PrP23-110. We further studied the effect of lysine on binding to PrP23-110 and on plasminogen activation by DSPA(alpha1) or t-PA. Lysine decreased both the binding of t-PA to PrP23-110 and the stimulation of plasmin generation by t-PA. Both binding and plasminogen activation of DSPA(alpha1) were not influenced by the presence of lysine. All plasminogen activators tested bearing kringle domains bind to PrP23-110. Binding to PrP23-110 is not sufficient for stimulation of plasmin generation. Thus the lysine-binding site of kringle 2 that is unique to t-PA appears to mediate the specific stimulation of plasminogen activation by the cellular prion protein.

Both lysine-clusters of the NH2-terminal prion-protein fragment PrP23-110 are essential for t-PA mediated plasminogen activation.

We have recently shown that the NH(2)-terminal fragment (PrP23-110) of the human cellular prion protein (PrP(c) ) stimulates t-PA mediated plasminogen activation. PrP23-110 contains an N-terminal lysine cluster (LC1; K(23),K(24), K(27)) and a C-terminal one (LC2; K(101),K(104),K(106),K(110)). To study their biological function we have substituted all lysine residues of each cluster by alanine and generated the recombinant PrP proteins PrP23-110sLC1 and PrP23-110sLC2. The ability of the mutant proteins to stimulate plasminogen activation was assayed. We found that both lysine clusters are essential for t-PA mediated plasminogen activation. We further studied the binding of soluble PrP23-110 to immobilized t-PA or plasminogen using surface plasmon resonance. The recorded binding curves could not be modeled by classical 1:1 binding kinetics suggesting oligomerisation of PrP23-110. Further plasmon resonance studies show that indeed PrP23-110 binds to itself and that glycosaminoglycans modify this interaction. Binding of t-PA or plasminogen to PrP23-110 was no longer influenced by glycosaminoglycans when PrP23-110 was immobilized on the chip surface. Thus a possible role of heparin as a cofactor in the stimulation of plasminogen activation by t-PA could be the generation of a PrP23-110 form with both lysine clusters accessible for binding of t-PA and plasminogen.

Stimulation of plasminogen activation by recombinant cellular prion protein is conserved in the NH2-terminal fragment PrP23-110.

The cellular prion protein (PrP(c)), tissue-type plasminogen activator (t-PA) and plasminogen are expressed in synaptic membranes in vivo. In the central nervous system the fibrinolytic system is associated with excitotoxin-mediated neurotoxicity and Alzheimer's disease. Recently binding of the disease associated isoform of the prion protein (PrP(Sc)) to plasminogen and stimulation of t-PA activity have been reported. In this study the interaction of PrP(c) and plasminogen was investigated using chromogenic assays in vitro. We found that plasmin is able to cleave recombinant PrP(c) at lysine residue 110 generating an NH(2)-terminal truncated molecule that has previously been described as a major product of PrP(c) metabolism. We further characterized the proteolytic fragments with respect to their ability to stimulate plasminogen activation in vitro. Our results show that the NH(2)-terminal part of PrP(c) spanning amino acids 23-110 (PrP23-110) together with low molecular weight heparin stimulates t-PA mediated plasminogen activation in vitro. The apparent rate constant was increased 57 fold in the presence of 800 nM PrP23-110. Furthermore, we compared the stimulation of t-PA activity by PrP(c) and beta-amyloid peptide (1-42). While the activity of the beta-amyloid was independent of low molecular weight heparin, PrP23-110 was approximately 4- and 37 fold more active than beta-amyloid in the absence or presence of low molecular weight heparin. In summary, plasmin cleaves PrP(c) in vitro and the liberated NH(2)-terminal fragment accelerates plasminogen activation. Cleavage of PrP c has previously been reported. Thus cleavage of PrP(c) enhancing plasminogen activation at the cell surface could constitute a regulatory mechanism of pericellular proteolysis.