Interaction of amorphous calcium phosphate with fibrin in vitro causes decreased fibrinolysis and altered protease profiles: implications for atherosclerotic disease.

Abstract

Previously, we demonstrated that amorphous calcium phosphate (ACP), chemical precursor to apatite, strongly interacted with fibrin and facilitated binding of matrix metalloproteinase (MMP)-9, a type IV collagenase. Plasmin-dependent fibrinolysis resulted in coordinate MMP-9 activation. Here we report on the effect(s) of ACP on fibrin degradation and binding of endogenous plasma proteases. Electrophoresis (8.5% SDS-PAGE) revealed that fibrin formed in the presence of ACP demonstrated characteristic gamma-gamma dimers (90-kDa) and beta-monomers (55-kDa), but resisted spontaneous fibrinolysis (72 h, 37 degrees C) or degradation by plasminogen activators (uPA, tPA). Casein zymography revealed an ACP-dependent decrease in fibrin binding of a low molecular weight (Mw) protease triplet (47-, 43-, 42-kDa) and increased fibrin binding of two high Mw proteases (94- and 84-kDa). The low Mw triplet also possessed gelatinolytic activity, but was not an MMP since 1,10-phenanthroline was ineffective as an inhibitor. Fibrin-binding proteases were inhibited to some degree by the serine protease inhibitor aprotinin. Competition/dissociation experiments with epsilon-aminocaproic acid revealed that the low Mw triplet lacked kringle regions whereas the 94- and 84-kDa proteases were tentatively identified and glu-/lys-plasmin(ogen)s. The triplet may, however, represent one or more kringle deficient mini-plasminogen(s), since electrophoretic mobility and substrate specificity was similar to elastase-generated mini-plasminogen. To explore these findings in a clinically relevant setting, a series of plasma samples was collected from a patient with unstable angina prior to, during, and post coronary artery bypass graft (CABG) surgery. Fibrin formed from plasma collected during and immediately post CABG was associated with increased fibrinolytic capacity and enhanced binding of a) MMP-9, b) the low Mw protease triplet (described above), and c) PA (as putative 110-kDa tPA:PAI-1 complex). The relevance of these findings to pathologic calcification of atherosclerotic plaques is discussed.