Insights into the dynamic structure and inhibitory mechanism of plasminogen activator inhibitor type 1

Plasminogen activator inhibitor type 1 (PAI-1) has been found to play a role in several human diseases and disease processes including metabolic syndrome, atherosclerosis, tumor angiogenesis, and renal and pulmonary fibrosis. It is a protease inhibitor of the serpin type and targets tissue-type and urokinase type plasminogen activators (t-PA and u-PA, respectively). Serpins operate using a "suicide-substrate" mechanism, in which the inhibitor is stoichiometrically consumed during protease inactivation. Specificity for target proteases is accomplished via a flexible reactive center loop (RCL) that acts as bait for the enzyme active site. Exertion of catalytic activity on the scissile bond of the RCL leads to a conformational change in the serpin in which the RCL becomes inserted into a central beta-sheet, beta-sheet A, translocating the acyl-linked protease a distance of 70 A and concomitantly distorting the active site. Unable to catalyze the deacylation of the inhibitor, the protease is kinetically trapped in a covalent acyl-enzyme intermediate with the serpin. PAI-1 is unique among serpins due to its ability to rapidly undergo such a conformational change in the absence of proteolytic activity on its RCL, thereby rendering the serpin inactive in a so-called latent form. This conformational lability has made the structural study of PAI-1 challenging. Thus a stable PAI-1 variant, 14-1B, was developed, producing an X-ray crystal structure of an active PAI-1 molecule for the first time. 14-1B contains four amino acid substitutions: one in the distal hinge in close proximity to the RCL, and three in the distant alpha-helix F -- beta- strand 3A (hF/s3A) loop sub-domain. In addition to enhancing the functional stability of PAI-1, these mutations also perturb the stoichiometry and kinetics of t-PA inhibition.;The aim of this study was to delineate the biochemical differences between wild-type PAI-1 (wtPAI-1) and 14-1B and to infer structural and functional details about the naturally occurring inhibitor using the structural model of the stable variant as a starting point. Using conformationally-sensitive ligands, we found that the RCL of PAI-1 exhibited more partial insertion into the beta-sheet A and that beta-sheet A was more accessible to the mobile RCL than was evident from the crystal structure of 14-1B. These properties suggested that PAI-1 be grouped in a novel sub-class of serpins exhibiting partially-inserted RCLs and allowed further study of its structure-function relationships in a new light. The accessibility of beta-sheet A to the RCL in 14-1B is limited by mutations in the hF/s3A loop sub-domain, diminishing partial insertion of the RCL and presumably reducing the rate of full loop insertion during protease inhibition. We provide evidence strongly suggesting that these hindrances to RCL flexibility and mobility affect the mechanism by which PAI-1 inactivates t-PA at several steps, perturbing the initial non-covalent binding, slowing the formation of the acyl-enzyme complex, and compromising the stabilization of the acyl-enzyme intermediate. The work provides a more comprehensive structural model for the active conformation of PAI-1 and illuminates a new role for the hF/s3A loop sub-domain in the inhibitory mechanism of serpins.