Plasminogen Activator Inhibitor-1 Variants Research Articles

Deep mutational scanning and massively parallel kinetics of plasminogen activator inhibitor-1 functional stability to probe its latency transition

Plasminogen activator inhibitor-1 (PAI-1), a member of the serine protease inhibitor superfamily of proteins, is unique among serine protease inhibitors for exhibiting a spontaneous conformational change to a latent or inactive state. The functional half-life for this transition at physiologic temperature and pH is ∼1 to 2h. To better understand the molecular mechanisms underlying this transition, we now report on the analysis of a comprehensive PAI-1 variant library expressed on filamentous phage and selected for functional stability after 48h at 37 °C. Of the 7201 possible single amino acid substitutions in PAI-1, we identified 439 that increased the functional stability of PAI-1 beyond that of the WT protein. We also found 1549 single amino acid substitutions that retained inhibitory activity toward the canonical target protease of PAI-1 (urokinase-like plasminogen activator), whereas exhibiting functional stability less than or equal to that of WT PAI-1. Missense mutations that increase PAI-1 functional stability are concentrated in highly flexible regions within the PAI-1 structure. Finally, we developed a method for simultaneously measuring the functional half-lives of hundreds of PAI-1 variants in a multiplexed, massively parallel manner, quantifying the functional half-lives for 697 single missense variants of PAI-1 by this approach. Overall, these findings provide novel insight into the mechanisms underlying the latency transition of PAI-1 and provide a database for interpreting human PAI-1 genetic variants.

Resolving distinct molecular origins for copper effects on PAI-1

Components of the fibrinolytic system are subjected to stringent control to maintain proper hemostasis. Central to this regulation is the serpin plasminogen activator inhibitor-1 (PAI-1), which is responsible for specific and rapid inhibition of fibrinolytic proteases. Active PAI-1 is inherently unstable and readily converts to a latent, inactive form. The binding of vitronectin and other ligands influences stability of active PAI-1. Our laboratory recently observed reciprocal effects on the stability of active PAI-1 in the presence of transition metals, such as copper, depending on the whether vitronectin was also present (Thompson et al. Protein Sci 20:353–365, 2011). To better understand the molecular basis for these copper effects on PAI-1, we have developed a gel-based copper sensitivity assay that can be used to assess the copper concentrations that accelerate the conversion of active PAI-1 to a latent form. The copper sensitivity of wild-type PAI-1 was compared with variants lacking N-terminal histidine residues hypothesized to be involved in copper binding. In these PAI-1 variants, we observed significant differences in copper sensitivity, and these data were corroborated by latency conversion kinetics and thermodynamics of copper binding by isothermal titration calorimetry. These studies identified a copper-binding site involving histidines at positions 2 and 3 that confers a remarkable stabilization of PAI-1 beyond what is observed with vitronectin alone. A second site, independent from the two histidines, binds metal and increases the rate of the latency conversion.

Vitronectin-binding PAI-1 protects against the development of cardiac fibrosis through interaction with fibroblasts

Plasminogen activator inhibitor-1 (PAI-1) promotes or abates fibrotic processes occurring in different organs. Binding of PAI-1 to vitronectin, an extracellular matrix component, may inhibit vitronectin–integrin complex-mediated cellular responses in pathophysiological conditions. To investigate the importance of plasmin suppression vs vitronectin-binding pathways of PAI-1 in cardiac fibrosis, we studied uninephrectomized mice fed a high salt diet and infused with angiotensin II (Ang II) together with different PAI-1 variants, including PAI-1AK (AK) that inhibits plasminogen activators but does not bind vitronectin, PAI-1RR (RR) that binds vitronectin but does not have protease inhibitory effects or control PAI-1 (CPAI), the control mutant that has similar molecular backbone and half-life as AK and RR while retaining all functions of native PAI-1. Compared with RR and CPAI, non-vitronectin-binding AK significantly increased expression of cardiac fibroblast marker, periostin (Ang+AK 8.40±3.55 vs Ang+RR 2.23±0.44 and Ang+CPAI 2.33±0.12% positive area, both P<0.05) and cardiac fibrosis (Ang+AK 1.79±0.26% vs Ang+RR 0.91±0.18% and Ang+CPAI 0.81±0.12% fibrotic area, both P<0.05), as well as Col1 mRNA (Ang+AK 12.81±1.84 vs Ang+RR 4.04±1.06 and Ang+CPAI 5.23±1.21 fold increase, both P<0.05). To elucidate mechanisms underlying the protective effects of vitronectin-binding PAI-1 against fibrosis, fibroblasts from normal adult human ventricles were stimulated with Ang and different PAI-1 variants. Protease inhibitory AK and CPAI increased supernatant fibronectin, while decreasing plasminogen activator/plasmin activities and matrix metalloproteinase. RR and CPAI variants significantly reduced fibroblast expression of integrin β3, vitronectin level in the supernatant and fibroblast adhesion to vitronectin compared with the non-vitronectin-binding AK. Further, RR and CPAI preserved apoptotic, decreased anti-apoptotic and proliferative activities in fibroblasts. Thus, PAI-1 promotes or protects against development of cardiac fibrosis differentially through the protease inhibitory pathway or through its binding to vitronectin.

Hydrogen/Deuterium Exchange Mass Spectrometry Reveals Specific Changes in the Local Flexibility of Plasminogen Activator Inhibitor 1 upon Binding to the Somatomedin B Domain of Vitronectin

The native fold of plasminogen activator inhibitor 1 (PAI-1) represents an active metastable conformation that spontaneously converts to an inactive latent form. Binding of the somatomedin B domain (SMB) of the endogenous cofactor vitronectin to PAI-1 delays the transition to the latent state and increases the thermal stability of the protein dramatically. We have used hydrogen/deuterium exchange mass spectrometry to assess the inherent structural flexibility of PAI-1 and to monitor the changes induced by SMB binding. Our data show that the PAI-1 core consisting of β-sheet B is rather protected against exchange with the solvent, while the remainder of the molecule is more dynamic. SMB binding causes a pronounced and widespread stabilization of PAI-1 that is not confined to the binding interface with SMB. We further explored the local structural flexibility in a mutationally stabilized PAI-1 variant (14-1B) as well as the effect of stabilizing antibody Mab-1 on wild-type PAI-1. The three modes of stabilizing PAI-1 (SMB, Mab-1, and the mutations in 14-1B) all cause a delayed latency transition, and this effect was accompanied by unique signatures on the flexibility of PAI-1. Reduced flexibility in the region around helices B, C, and I was seen in all three cases, which suggests an involvement of this region in mediating structural flexibility necessary for the latency transition. These data therefore add considerable depth to our current understanding of the local structural flexibility in PAI-1 and provide novel indications of regions that may affect the functional stability of PAI-1.

Interactions of Plasminogen Activator Inhibitor-1 with Vitronectin Involve an Extensive Binding Surface and Induce Mutual Conformational Rearrangements

In order to explore early events during the association of plasminogen activator inhibitor-1 (PAI-1) with its cofactor vitronectin, we have applied a robust strategy that combines protein engineering, fluorescence spectroscopy, and rapid reaction kinetics. Fluorescence stopped-flow experiments designed to monitor the rapid association of PAI-1 with vitronectin indicate a fast, concentration-dependent, biphasic binding of PAI-1 to native vitronectin but only a monophasic association with the somatomedin B (SMB) domain, suggesting that multiple phases of the binding interaction occur only when full-length vitronectin is present. Nonetheless, in all cases, the initial fast interaction is followed by slower fluorescence changes attributed to a conformational change in PAI-1. Complementary experiments using an engineered, fluorescently silent PAI-1 with non-natural amino acids showed that concomitant structural changes occur as well in native vitronectin. Furthermore, we have measured the effect of vitronectin on the rate of insertion of the reactive center loop into beta-sheet A of PAI-1 during reaction with target proteases. With a variety of PAI-1 variants, we observe that both full-length vitronectin and the SMB domain have protease-specific effects on the rate of loop insertion but that the two exhibit clearly different effects. These results support a model for PAI-1 binding to vitronectin in which the interaction surface extends beyond the region of PAI-1 occupied by the SMB domain. In support of this model are recent results that define a PAI-1-binding site on vitronectin that lies outside the somatomedin B domain (Schar, C. R., Blouse, G. E., Minor, K. H., and Peterson, C. B. (2008) J. Biol. Chem. 283, 10297-10309) and the complementary site on PAI-1 (Schar, C. R., Jensen, J. K., Christensen, A., Blouse, G. E., Andreasen, P. A., and Peterson, C. B. (2008) J. Biol. Chem. 283, 28487-28496).

Transgenic overexpression of a stable Plasminogen Activator Inhibitor-1 variant

Introduction Plasminogen Activator Inhibitor-1 (PAI-1) is a member of the Serine Protease Inhibitor (SERPIN) gene family and a key regulator of fibrinolysis. PAI-1 is unique among SERPINs in its spontaneous transition to a latent, inactive state, with a half-life of approximately 2 hours under physiologic conditions. The biologic importance of the PAI-1 transition to latency is unknown. This study aimed to engineer transgenic overexpression of a stable murine PAI-1 variant to examine the physiologic effects in vivo from delayed transition of PAI-1 to latency. Materials and Methods Ten independent transgenic lines were generated with expression of a stable PAI-1 variant driven by the hybrid CMV/chicken β-actin promoter. Results Plasma PAI-1 levels in the transgenic founders ranged from 3.1 ± 0.1 ng/mL to 1268.8 ± 717.0 ng/mL. Quantitative PCR analysis in 3 transgenic lines demonstrated elevated PAI-1 mRNA in multiple tissues, with the highest increases observed in liver, brain, heart, and kidney. The fold-increase in PAI-1 mRNA over wild-type ranged from 2-fold to > 2000-fold. Immunohistochemistry showed increased PAI-1 in liver, kidney, heart, spleen, and lung. Histologic examination of transgenic mice showed no evidence of thrombosis. The two founders with the highest plasma PAI-1 levels failed to produce any transgenic offspring that survived to weaning, although genotyping of expired pups revealed successful transmission of the transgene. Conclusion These results suggest that high expression of a stable variant of PAI-1 may be lethal in mice, while more moderate expression is generally well tolerated and produces no apparent thrombosis.

Association of PAI-1 4G/5G and -844G/A Gene Polymorphism and Changes in PAI-1/tPA Levels in Stroke: A Case-Control Study

Mutations in the plasminogen activator inhibitor-1 (PAI-1) gene, along with altered PAI-1 and tissue-type plasminogen activator (tPA) levels, have been implicated in stroke pathogenesis. We investigated the association of PAI-1 and tPA levels with stroke as a function of PAI-1 4G/5G and -844G/A genotypes, as well as the link between these PAI-1 gene variants and stroke risk, in a case-control study of 135 ischemic stroke patient, diagnosed according to clinical and radiologic findings and confirmed by computed tomography scan. Controls (n = 118) were age- and sex-matched and had no personal/family history of stroke. PAI-1 4G/5G and -844G/A genotyping were done by polymerase chain reaction-restriction fragment length polymorphism, and PAI-1 and tPA levels were measured by enzyme immunoassay. Significant elevation in PAI-1 and marked reduction in tPA levels were seen in stroke patients and were correlated with 4G/5G, but not with -844G/A, PAI-1 variants. Whereas the frequencies of 4G or -844A alleles were comparable between patients and controls, 4G/4G carriers had reduced risk of stroke compared with other genotypes (odds ratio [OR] = 0.54; 95% confidence interval [CI] = 0.31-0.95). The 4G/-844A haplotype also was more closely associated with reduced stroke risk (OR = 0.43; 95% CI = 0.20-0.97) than 5G/-844A or 4G/-844G haplotypes. Regression analysis demonstrated that 4G homozygosity (OR = 0.176), hypertension (OR = 6.288), and body mass index (OR = 1.325) were independent predictors of stroke. The protective effect of 4G allele against stroke suggests involvement of PAI-1 4G/5G polymorphism in stroke through a mechanism not related to fibrinolysis, possibly involving altered plaque stabilization, and/or through antagonism of tPA effects.

Evidence for a Pre-latent Form of the Serpin Plasminogen Activator Inhibitor-1 with a Detached β-Strand 1C

Latency transition of plasminogen activator inhibitor-1 (PAI-1) occurs spontaneously in the absence of proteases and results in stabilization of the molecule through insertion of its reactive center loop (RCL) as a strand in beta-sheet A and detachment of beta-strand 1C (s1C) at the C-terminal hinge of the RCL. This is one of the largest structural rearrangements known for a folded protein domain without a concomitant change in covalent structure. Yet, the sequence of conformational changes during latency transition remains largely unknown. We have now mapped the epitope for the monoclonal antibody H4B3 to the cleft revealed upon s1C detachment and shown that H4B3 inactivates recombinant PAI-1 in a time-dependent manner. With fluorescence spectroscopy, we show that insertion of the RCL is accelerated in the presence of H4B3, demonstrating that the loss of activity is the result of latency transition. Considering that the epitope for H4B3 appears to be occluded by s1C in active PAI-1, this finding suggests the existence of a pre-latent conformation on the path from active to latent PAI-1 characterized by at least partial detachment of s1C. Functional characterization of mutated PAI-1 variants suggests that a salt-bridge between Arg273 and Asp224 may stabilize the pre-latent conformation. The binding of H4B3 and of a peptide targeting the cleft revealed upon s1C detachment was hindered by the glycans attached to Asn267. Conclusively, we have provided evidence for the existence of an equilibrium between active PAI-1 and a pre-latent form, characterized by reversible detachment of s1C and formation of a glycan-shielded cleft in the molecule.

Additivity in Effects of Vitronectin and Monoclonal Antibodies against α-Helix F of Plasminogen Activator Inhibitor-1 on Its Reactions with Target Proteinases

The serpin plasminogen activator inhibitor-1 (PAI-1) is a potential therapeutic target in cardiovascular and cancerous diseases. PAI-1 circulates in blood as a complex with vitronectin. A PAI-1 variant (N-((2-(iodoacetoxy)ethyl)-N-methyl)amino-7-nitrobenz-2-oxa-3-diazole (NBD) P9 PAI-1) with a fluorescent tag at the reactive center loop (RCL) was used to study the effects of vitronectin and monoclonal antibodies (mAbs) directed against alpha-helix F (Mab-2 and MA-55F4C12) on the reactions of PAI-1 with tissue-type and urokinase-type plasminogen activators. Both mAbs delay the RCL insertion and induce an increase in the stoichiometry of inhibition (SI) to 1.4-9.5. Binding of vitronectin to NBD P9 PAI-1 does not affect SI but results in a 2.0-6.5-fold decrease in the limiting rate constant (klim) of RCL insertion for urokinase-type plasminogen activator at pH 6.2-8.0 and for tissue-type plasminogen activator at pH 6.2. Binding of vitronectin to the complexes of NBD P9 PAI-1 with mAbs results in a decrease in klim and in a 1.5-22-fold increase in SI. Thus, vitronectin and mAbs demonstrated additivity in the effects on the reaction with target proteinases. The same step in the reaction mechanism remains limiting for the rate of RCL insertion in the absence and presence of Vn and mAbs. We hypothesize that vitronectin, bound to alpha-helix F on the side opposite to the epitopes of the mAbs, potentiates the mAb-induced delay in RCL insertion and the associated substrate behavior by selectively decreasing the rate constant for the inhibitory branch of PAI-1 reaction (ki). These results demonstrate that mAbs represent a valid approach for inactivation of vitronectin-bound PAI-1 in vivo.

Plasminogen activator inhibitor-1 impairs alveolar epithelial repair by binding to vitronectin.

The pathogenesis of pulmonary fibrosis is thought to involve alveolar epithelial injury that, when successfully repaired, can limit subsequent scarring. The plasminogen system participates in this process with the balance between urokinase-type plasminogen activator (uPA) and plasminogen activator inhibitor-1 (PAI-1) being a critical determinant of the extent of collagen accumulation that follows lung injury. Because the plasminogen system is known to influence the rate of migration of epithelial cells, including keratinocytes and bronchial epithelial cells, we hypothesized that the balance of uPA and PAI-1 would affect the efficiency of alveolar epithelial cell (AEC) wound repair. Using an in vitro model of AEC wounding, we show that the efficiency of repair is adversely affected by a deficiency in uPA or by the exogenous administration of PAI-1. By using PAI-1 variants and AEC from mice transgenically deficient in vitronectin (Vn), we demonstrate that the PAI-1 effect requires its Vn-binding activity. Furthermore, we have found that cell motility is enhanced by the availability of Vn in the matrix and that the AEC-Vn interaction is mediated, in part, by the alpha(v)beta(1) integrin. The significant effect of uPA and PAI-1 on epithelial repair suggests a mechanism by which the plasminogen system may modulate pulmonary fibrosis.

Conservation of Critical Functional Domains in Murine Plasminogen Activator Inhibitor-1

Plasminogen activator inhibitor-1 is the main physiological regulator of tissue-type plasminogen activator in normal plasma. In addition to its critical function in fibrinolysis, plasminogen activator inhibitor-1 has been implicated in roles in other physiological and pathophysiological processes. To investigate structure-function aspects of mouse plasminogen activator inhibitor-1, the recombinant protein was expressed in Escherichia coli and purified. Five variant recombinant murine proteins (R76E, Q123K, R346A, R101A, and Q123K/R101A) were also generated using site-directed mutagenesis. The variant (R346A) was found to be defective in its inhibitory activity against tissue plasminogen activator relative to its wild-type counterpart. Enzyme-linked immunosorbent assay and surface plasmon resonance experiments demonstrated reduced vitronectin-binding affinity of the (Q123K) variant (K(D) = 1800 nm) relative to the wild-type protein (K(D) = 5.4 nm). Kinetic analyses indicated that the (Q123K) variant had a slower association (k(on) = 2.92 x 10(4) m(-1) s(-1)) to, and a faster dissociation from, vitronectin (k(off) = 5.3 x 10(-2) s(-1)), (wild-type k(on) = 1.03 x 10(6) m(-1) s(-1) and k(off) = 5.27 x 10(-3) s(-1)). The Q123K/R101A variant demonstrated an even lower vitronectin-binding ability. Low density lipoprotein receptor-related protein binding was decreased for the (R76E) variant. It was also demonstrated that the plasminogen activator inhibitor-1/vitronectin complex decreased the interaction of plasminogen activator inhibitor-1 with low density lipoprotein receptor-related protein. These results indicate that the complex interactions traditionally associated with different plasminogen activator inhibitor-1 functions apply to the murine system, thus showing a commonality of subtle functions among different species and evolutionary conservation of this protein. Further, this study provides additional evidence that the human hemostasis system can be studied effectively in the mouse, which is a great asset for investigations with gene-altered mice.

Construction of a plasminogen activator inhibitor-1 variant without measurable affinity to vitronectin but otherwise normal

Vitronectin (VN) and plasminogen activator inhibitor-1 (PAI-1) have important functional interactions: VN stabilises the protease inhibitory activity of PAI-1 and PAI-1 inhibits binding of adhesion receptors to VN. Having previously mapped the PAI-1 binding area for VN, we have now constructed a PAI-1 variant, R103A-M112A-Q125A, without measurable affinity to VN, but with full protease inhibitory activity and endocytosis receptor binding. As a tool for evaluating the physiological and pathophysiological functions of the PAI-1–VN interaction, our new variant is far superior to the previously widely used PAI-1 variant Q125K, which we have found possesses an only about 10-fold reduced affinity to VN.

Mutational analysis of plasminogen activator inhibitor-1.

The serpin plasminogen activator inhibitor-1 (PAI-1) is a fast and specific inhibitor of the plasminogen activating serine proteases tissue-type and urokinase-type plasminogen activator and, as such, an important regulator in turnover of extracellular matrix and in fibrinolysis. PAI-1 spontaneously loses its antiproteolytic activity by inserting its reactive centre loop (RCL) as strand 4 in beta-sheet A, thereby converting to the so-called latent state. We have investigated the importance of the amino acid sequence of alpha-helix F (hF) and the connecting loop to s3A (hF/s3A-loop) for the rate of latency transition. We grafted regions of the hF/s3A-loop from antithrombin III and alpha1-protease inhibitor onto PAI-1, creating eight variants, and found that one of these reversions towards the serpin consensus decreased the rate of latency transition. We prepared 28 PAI-1 variants with individual residues in hF and beta-sheet A replaced by an alanine. We found that mutating serpin consensus residues always had functional consequences whereas mutating nonconserved residues only had so in one case. Two variants had low but stable inhibitory activity and a pronounced tendency towards substrate behaviour, suggesting that insertion of the RCL is held back during latency transition as well as during complex formation with target proteases. The data presented identify new determinants of PAI-1 latency transition and provide general insight into the characteristic loop-sheet interactions in serpins.

Resolution of Michaelis complex, acylation, and conformational change steps in the reactions of the serpin, plasminogen activator inhibitor-1, with tissue plasminogen activator and trypsin.

Michaelis complex, acylation, and conformational change steps were resolved in the reactions of the serpin, plasminogen activator inhibitor-1 (PAI-1), with tissue plasminogen activator (tPA) and trypsin by comparing the reactions of active and Ser 195-inactivated enzymes with site-specific fluorescent-labeled PAI-1 derivatives that report these events. Anhydrotrypsin or S195A tPA-induced fluorescence changes in P1'-Cys and P9-Cys PAI-1 variants labeled with the fluorophore, NBD, indicative of a substrate-like interaction of the serpin reactive loop with the proteinase active-site, with the P1' label but not the P9 label perturbing the interactions by 10-60-fold. Rapid kinetic analyses of the labeled PAI-1-inactive enzyme interactions were consistent with a single-step reversible binding process involving no conformational change. Blocking of PAI-1 reactive loop-beta-sheet A interactions through mutation of the P14 Thr --> Arg or annealing a reactive center loop peptide into sheet A did not weaken the binding of the inactive enzymes, suggesting that loop-sheet interactions were unlikely to be induced by the binding. Only active trypsin and tPA induced the characteristic fluorescence changes in the labeled PAI-1 variants previously shown to report acylation and reactive loop-sheet A interactions during the PAI-1-proteinase reaction. Rapid kinetic analyses showed saturation of the reaction rate constant and, in the case of the P1'-labeled PAI-1 reaction, biphasic changes in fluorescence indicative of an intermediate resembling the noncovalent complex on the path to the covalent complex. Indistinguishable K(M) and k(lim) values of approximately 20 microM and 80-90 s(-1) for reaction of the two labeled PAI-1s with trypsin suggested that a diffusion-limited association of PAI-1 and trypsin and rate-limiting acylation step, insensitive to the effects of labeling, controlled covalent complex formation. By contrast, differing values of K(M) of 1.7 and 0.1 microM and of k(lim) of 17 and 2.6 s(-1) for tPA reactions with P1' and P9-labeled PAI-1s, respectively, suggested that tPA-PAI-1 exosite interactions, sensitive to the effects of labeling, promoted a rapid association of PAI-1 and tPA and reversible formation of an acyl-enzyme complex but impeded a rate-limiting burial of the reactive loop leading to trapping of the acyl-enzyme complex. Together, the results suggest a kinetic pathway for formation of the covalent complex between PAI-1 and proteinases involving the initial formation of a Michaelis-type noncovalent complex without significant conformational change, followed by reversible acylation and irreversible reactive loop conformational change steps that trap the proteinase in a covalent complex.

Elucidation of the Binding Regions of PAI-1 Neutralizing Antibodies Using Chimeric Variants of Human and Rat PAI-1

Increased levels of plasminogen activator inhibitor-1 (PAI-1), the main physiological inhibitor of tissue-type plasminogen activator (t-PA) in plasma, are a known risk factor for thromboembolic and cardiovascular diseases. The elucidation of the binding site of inhibitory monoclonal antibodies may contribute to the rational design of PAI-1 modulating therapeutics. In this study, homolog-scanning mutagenesis was used to identify the binding region of a variety of human PAI-1 inhibitory antibodies, lacking cross-reactivity with rat PAI-1. Therefore. eight chimeric human/rat PAI-1 variants, containing rat PAI-1 substitutions at the N-terminal or C-terminal end with splicing sites at positions 26, 81, 187, 277 or 327, were generated and purified. Biochemical characterization revealed that all chimeras were folded properly. Subsequently, surface plasmon resonance was used to determine the affinity of various monoclonal antibodies for these chimera. Comparative analysis of the affinity and ELISA data allowed the identification of the major binding region of the inhibitory antibodies MA-8H9D4, MA-33B8F7, MA-44E4, MA-42A2F6 and MA-56A7C10. Thus, three segments in human PAI-1 containing each at least one site involved in the neutralization of PAI-1 activity could be identified, i.e. (1) the segment from residue 81 to residue 187 (comprising alpha-helices hD, hE and hF, beta-strands s4C, s3A, s2A and s1A and the loops connecting these elements). (2) the segment between residues 277 and 327 (hI, thIs5A, s5A and s6A) and (3) the region C-terminal from amino acid 327, including the reactive site loop. The current data. together with previous data, indicate that PAI-1 contains at least four different regions that could be considered as putative targets to modulate its activity.

Different structural requirements for plasminogen activator inhibitor 1 (PAI-1) during latency transition and proteinase inhibition as evidenced by phage-displayed hypermutated PAI-1 libraries

Plasminogen activator inhibitor type 1 (PAI-1) is a member of the serine protease inhibitor (serpin) superfamily. Its highly mobile reactive-center loop (RCL) is thought to account for both the rapid inhibition of tissue-type plasminogen activator (t-PA), and the rapid and spontaneous transition of the unstable, active form of PAI-1 into a stable, inactive (latent) conformation ( t 1/2at 37 °C, 2.2 hours). We determined the amino acid residues responsible for the inherent instability of PAI-1, to assess whether these properties are independent and, consequently, whether the structural basis for inhibition and latency transition is different. For that purpose, a hypermutated PAI-1 library that is displayed on phage was pre-incubated for increasing periods (20 to 72 hours) at 37 °C, prior to a stringent selection for rapid t-PA binding. Accordingly, four rounds of phage-display selection resulted in the isolation of a stable PAI-1 variant (st-44: t 1/2 450 hours) with 11 amino acid mutations. Backcrossing by DNA shuffling of this stable mutant with wt PAI-1 was performed to eliminate non-contributing mutations. It was shown that the combination of mutations at positions 50, 56, 61, 70, 94, 150, 222, 223, 264 and 331 increases the half-life of PAI-1 245-fold. Furthermore, within the limits of detection the stable mutants isolated are functionally indistinguishable from wild-type PAI-1 with respect to the rate of inhibition of t-PA, cleavage by t-PA, and binding to vitronectin. These stabilizing mutations constitute largely reversions to the stable “serpin consensus sequence” and are located in areas implicated in PAI-1 stability (e.g. the vitronectin-binding domain and the proximal hinge). Collectively, our data provide evidence that the structural requirements for PAI-1 loop insertion during latency transition and target proteinase inhibition can be separated.

Inactivation of Plasminogen Activator Inhibitor-1 by Specific Proteolysis with Stromelysin-1 (MMP-3)

Matrix metalloproteinase-3 (MMP-3 or stromelysin-1) specifically hydrolyzes the Ser(337)-Ser(338) (P10-P9) and Val(341)-Ile(342) (P6-P5) peptide bonds in human plasminogen activator inhibitor-1 (PAI-1). Cleavage is completely abolished in the presence of the metal chelators EDTA or 1,10-phenanthroline. A stabilized active PAI-1 variant was also cleaved by MMP-3. At an enzyme/substrate ratio of 1/10 at 37 degrees C, PAI-1 protein cleavage occurred with half-lives of 27 or 14 min for active or stable PAI-1 and was associated with rapid loss of inhibitory activity toward tissue-type plasminogen activator with half-lives of 15 or 13 min, respectively. A substrate-like variant of PAI-1, lacking inhibitory activity but with exposed reactive site loop, was cleaved with a half-life of 23 min, whereas latent PAI-1 in which a major part of the reactive site loop is inserted into the molecule, was resistant to cleavage. Biospecific interaction analysis indicated comparable binding of active, stable, and substrate PAI-1 to both proMMP-3 and MMP-3 (K(A) of 12-22 x 10(6) m(-1)), whereas binding of latent PAI-1 occurred with lower affinity (1.7-2.3 x 10(6) m(-1)). Stable PAI-1 bound to vitronectin was cleaved and inactivated by MMP-3 in a manner comparable with that of free PAI-1; however, the cleaved protein did not bind to vitronectin. Cleavage and inactivation of PAI-1 by MMP-3 may thus constitute a mechanism decreasing the antiproteolytic activity of PAI-1 and impairing the potential inhibitory effect of vitronectin-bound PAI-1 on cell adhesion and/or migration.

PAI-1 stability: the role of histidine residues

The role of the 13 histidine residues in plasminogen activator inhibitor 1 (PAI-1) for the stability of the molecule was studied by replacing these residues by threonine, using site-directed mutagenesis. The generated mutants were expressed in Escherichia coli, purified and characterized. All variants had a normal activity and formed stable complexes with tissue-type plasminogen activator. Most of these PAI-1 variants displayed a similar pH-dependency in stability as wild-type PAI-1, with increased half-lives at lower pH. However, the variant His364Thr had a half-life of about 50 min at 37°C and had almost completely lost its pH-dependency. Therefore, our data suggest that His 364, in the COOH-terminal end of the molecule might be responsible for the pH-dependent stability of PAI-1.

Vitronectin and Substitution of a β-Strand 5A Lysine Residue Potentiate Activity-neutralization of PA Inhibitor-1 by Monoclonal Antibodies against α-Helix F

Some monoclonal antibodies against plasminogen activator inhibitor-1 (PAI-1) are able to inhibit its reaction with its target proteinases. We have characterized the effect on PAI-1 of two monoclonal antibodies, Mab-2 and Mab-6, with overlapping epitopes in a sequence encompassing beta-strand 1A, alpha-helix F, and the loop connecting alpha-helix F and beta-strand 3A (the hF/s3A loop). Mab-2 reduced the inhibitory activity of wild type PAI-1 and almost totally abolished the inhibitory activity of a PAI-1 variant harboring an Ala substitution of Lys 325 (335 in the alpha1-proteinase inhibitor template residue numbering) in beta-strand 5A. In both cases, the neutralizing effect of the antibody was strongly potentiated by vitronectin. Mab-6 had no effect on wild type PAI-1, but reduced the inhibitory activity of the K325A variant. The effect of Mab-6 was not potentiated by vitronectin. With both Mab-2 and Mab-6, the neutralization of PAI-1 activity was associated with PAI-1 substrate behaviour. Mab-2, but not Mab-6, prevented vitronectin from rescuing PAI-1 from cold-induced substrate behaviour. We propose that the antibodies act by weakening the anchoring of alpha-helix F to the adjacent structures, resulting in an increased flexibility of beta-strand 5A and the hF/s3A loop and a changed conformational response to the binding of vitronectin in the alpha-helix E region. The potentiating effect of vitronectin on neutralization of PAI-1 by antibodies is a novel concept in the development of compounds for neutralizing PAI-1 in vivo.

Induction of conformational changes within crystals of plasminogen activator inhibitor-1 (PAI-1)

Plasminogen activator inhibitor-1 (PAI-1) is a member of the serpin (serine protease inhibitor) superfamily, that can adopt three different conformations: active, latent and substrate. It was recently shown that the non-ionic detergent Triton X-100 (TX-100) can accelerate the conformational transitions in wild-type PAI-1 in solution. Recently, we have crystallized a stable PAI-1 variant (PAI-1-stab) in its active conformation (Acta Cryst D 55, 574–576, 1999). In this study, we examined the effect of TX-100 on the conformational transitions in PAI-1-stab, in solution as well as in crystals. Within the crystal (at t = 0: 75 ± 3% active, 10 ± 4% non-reactive and 15 ± 4% substrate; mean ± SD, n = 3) a time-dependent increase of the substrate form was observed, with a concomitant decrease of the active form. A steady state situation with active (52 ± 2%) and substrate (34 ± 5%) forms was reached within 4 h. Under those conditions, the amount of non-reactive PAI-1 remained essentially unchanged (14 ± 5%). The conformational changes induced by TX-100 in PAI-1-stab in solution (at t = 0: 59 ± 2% active, 0% non-reactive and 41 ± 2% substrate, n = 3) were characterized mainly by a decrease of the active form in favour of the non-reactive form, reaching a steady state between 15 and 24 h resulting in active (20 ± 3%, n = 4), substrate (29 ± 3%) and non-reactive (48 ± 5%) forms. Thus, this study demonstrates that, even though characterized with a restricted mobility, conformational changes can be induced within protein crystals. Importantly, in both cases studied a steady state situation with various functional forms was observed.