A lifelong bleeding disorder associated with a deficiency of plasminogen activator inhibitor type 1

A 36-year-old patient was investigated for a lifelong history of epistaxis and delayed bleeding after minor surgeries. Deficiencies or abnormalities of the coagulation system, of platelet function, or of factor XIII and alpha-2-antiplasmin were excluded. Consistently, however, over a period of 7 years, a high basal euglobulin fibrinolytic activity was observed that was characterized by a high tissue-type plasminogen activator (t-PA) activity, normal t-PA antigen, and undetectable plasminogen activator inhibitor type-1 (PAI-1) antigen and activity. The high specific activity of t-PA (640,000 IU/mg) and the minimal amounts of t-PA/PAI-1 complexes detected by fibrin zymography suggest that in this patient all t-PA was active. This is in striking contrast to normal plasma, where the majority of t-PA is complexed to PAI-1. Thus, in this patient, a severe deficiency of PAI-1 is associated with a delayed type bleeding tendency. Our observation underscores the importance of plasma PAI-1 for the stabilization of the hemostatic plug.     

Conversion of the active to latent plasminogen activator inhibitor from human endothelial cells

The plasminogen activator inhibitor from human endothelial cells (PAI- 1) exists in two forms in the culture medium: an active form that binds to and inactivates plasminogen activators and a latent form that in its native state has no anti-activator activity. Inhibitor activity associated with the latent form can be generated by treatment with protein denaturants and makes up more than 98% of the total inhibitor activity in conditioned medium. Plasminogen activator inhibitor activity is also found in cell cytosol. This inhibitor activity is stable to SDS-treatment but is not enhanced by it. We investigated the relationship between this active cell-associated inhibitor and the latent PAI-1 found in the conditioned medium. Both intracellular and extracellular inhibitors were immunoprecipitated by a monoclonal antibody produced against the latent inhibitor from HT1080 fibrosarcoma cells and electrophoresis on SDS gels of various acrylamide concentrations demonstrated that both forms had the same Mr. Incubation of cytosol inhibitor at 37 degrees C resulted in a decline in inhibitor activity with a half-life of approximately 4 hours, a rate of decline similar to that of the active PAI-1 in conditioned medium, with less than 10% of the original activity present after eight hours. This decline is accelerated at higher temperatures and is not affected by the presence of a variety of protease inhibitors. Approximately 90% of the activity can be regenerated after SDS treatment suggesting that the cell associated inhibitor, during incubation at 37 degrees C, converts to a form similar to that found in conditioned medium. Despite these similarities, the apparent Stoke's radii of the active intracellular inhibitor and the latent inhibitor in conditioned medium were significantly different with values of 2.77 nm and 2.40 nm for active and latent PAI-1, respectively. Incubation of the active form at 37 degrees C resulted in the shift of the Stoke's radius to that similar to the latent PAI-1 (2.45 nm). Thus, the active and latent PAI-1, while being immunologically similar and of the same apparent Mr, can be differentiated by their behavior on gel permeation columns. This suggests that the intracellular inhibitor is a precursor to the latent form.     

Platelets synthesize large amounts of active plasminogen activator inhibitor 1

Previous studies have suggested that plasminogen activator inhibitor 1 (PAI-1) released from platelets convey resistance of platelet-rich blood clots to thrombolysis. However, the majority of PAI-1 in platelets is inactive and therefore its role in clot stabilization is unclear. Because platelets retain mRNA and capacity for synthesis of some proteins, we investigated if platelets can de novo synthesize PAI-1 with an active configuration. PAI-1 mRNA was quantified with real-time polymerase chain reaction and considerable amounts of PAI-1 mRNA were detected in all platelet samples. Over 24 hours, the amount of PAI-1 protein as determined by an enzyme-linked immunosorbent assay increased by 25% (P = .001). Metabolic radiolabeling with (35)S-methionine followed by immunoprecipitation confirmed an ongoing PAI-1 synthesis, which could be further stimulated by thrombin and inhibited by puromycin. The activity of the newly formed PAI-1 was investigated by incubating platelets in the presence of tissue-type plasminogen activator (tPA). This functional assay showed that the majority of the new protein was in an active configuration and could complex-bind tPA. Thus, there is a continuous production of large amounts of active PAI-1 in platelets, which could be a mechanism by which platelets contribute to stabilization of blood clots.     

Independent regulation of plasminogen activator inhibitor 2 and plasminogen activator inhibitor 1 in human synovial fibroblasts.

OBJECTIVE. To study the plasminogen activator inhibitor(s) (PAI) produced in vitro by human synovial fibroblast-like cells. METHODS. Human synovial cell explant cultures were established using cells from nonrheumatoid donors. PAI-2 and PAI-1 antigens were measured by enzyme-linked immunosorbent assay, and messenger RNA (mRNA) levels were determined by Northern blot. RESULTS. The synovial fibroblasts produced both PAI-2 and PAI-1. Interleukin-1 (IL-1) increased PAI-2 but decreased PAI-1 formation, both at the protein and the mRNA levels. Using cyclooxygenase inhibitors, evidence was obtained that an endogenous cyclooxygenase product(s) in the IL-1-treated cultures inhibited formation of both PAIs; exogenous prostaglandin E2 (10(-7) M) reversed the effect of cyclooxygenase inhibition. The glucocorticoid dexamethasone (10(-6) to 10(-7) M) inhibited IL-1-stimulated PAI-2 formation but reversed the suppressive effect of IL-1 on PAI-1 production. CONCLUSION. PAI-2 formation and PAI-1 formation can be regulated independently in human synoviocytes, illustrating the complexity of the modulation of the net PA activity expressed by these cells.     

Tissue-type plasminogen activator and its inhibitor plasminogen activator inhibitor type 1 are coordinately expressed during ovulation in the rhesus monkey

Ovulation is a gonadotropin-controlled process that is essential for the propagation of all mammalian species. In the present study, we used a pregnant mare serum gonodotropin/human chorionic gonadotropin (hCG)-induced, synchronized ovulation model in rhesus monkeys and systematically investigated the roles of the plasminogen activator (PA) system in the ovulatory process of the primate. At different follicular developmental stages throughout the periovulatory period, samples of ovaries, granulosa cells, and theca-interstitial cells as well as follicular fluid were collected, and levels of PA and PA inhibitor type-1 (PAI-1) were evaluated by fibrin overlay, reverse fibrin overlay, Northern blot analysis, and in situ hybridization, respectively. We showed that in response to an injection of ovulation-triggering hCG, which mimics the preovulatory surge of LH in the circulation, granulosa cell-derived tissue-type PA (tPA) was substantially elevated in preovulatory follicles and reached its maximum level just before ovulation. Although theca-interstitial cell-derived PAI-1 was also stimulated by pregnant mare serum gonodotropin and hCG treatments, however, the maximum level of PAI-1 appeared 12 h earlier than that of tPA. When ovulation approached, accompanying the highest tPA level in the preovulatory follicles, the follicular PAI-1 level declined dramatically to its minimum value. Moreover, our data on the expression of follicular PA and PAI-1 over the periovulatory period were reinforced by results obtained at the mRNA level. Our data suggest that the coordinated expression of tPA and PAI-1 may be of importance for the follicular rupture process during ovulation in the primate.     

Amino acid residues that affect interaction of tissue-type plasminogen activator with plasminogen activator inhibitor 1.

Fibrinolysis is regulated in part by the interaction between tissue-type plasminogen activator (t-PA) and plasminogen activator inhibitor 1 (PAI-1, a serine protease inhibitor of the serpin family). It is known from our earlier work that deletion of a loop of amino acids (residues 296-302) from the serine protease domain of t-PA suppresses the interaction between the two proteins without altering the reactivity of t-PA towards its substrate, plasminogen. To define more precisely the role of individual residues within this loop, we have used site-directed mutagenesis to replace Lys-296, Arg-298, and Arg-299 with negatively charged glutamic residues. Replacement of all three positively charged amino acids generates a variant of t-PA that associates inefficiently with PAI-1 and is highly resistant to inhibition by the serpin. Two t-PAs with point mutations (Arg-298----Glu and Arg-299----Glu) are partially resistant to inhibition by PAI-1 and associate with the serpin at intermediate rates. Other point mutations (Lys-296----Glu, His-297----Glu, and Pro-301----Gly) do not detectably affect the interaction of t-PA with PAI-1. None of these substitutions has a significant effect on the rate of catalysis by t-PA or on the affinity of the enzyme for its substrate, plasminogen. On the basis of these results, we propose a model in which positively charged residues located in a surface loop near the active site of t-PA form ionic bonds with complementary negatively charged residues C-terminal to the reactive center of PAI-1.     

Involvement of plasminogen activator and plasminogen activator inhibitor type 1 in spermatogenesis, sperm capacitation, and fertilization

Increasing evidence suggests that local fibrinolysis generated by plasminogen activator (PA) and modulated by plasminogen activator inhibitor type 1 (PAI-1) is essential for mammalian spermatogenesis, sperm capacitation, and fertilization. Tissue-type PA (t-PA), urokinase-type PA (u-PA), and PAI-1 have been reported in the testes of various animals. Sertoli cells within the seminiferous epithelium are believed to play a central role in the control and maintenance of spermatogenesis by producing regulatory factors, including PA/PAI-1. Fertilization is a unique and exquisitely choreographed cellular interaction between male and female gametes, in which some basic biochemical mechanisms remain unresolved. A key issue is the molecular basis of sperm-egg recognition, binding, and penetration. Sperm capacitation and the acrosomal reaction appear to rely on local fibrinolysis generated by the PA/PAI-1 system. Ejaculated spermatozoa from various species carry u-PA activity. The u-PA receptor (uPAR) and the inhibitor PAI-1 have also been reported to bind on the sperm membrane surface. Thus, it is possible that uPAR and PAI-1 function in a counterbalanced and coordinated way on the surface of spermatozoa to regulate the u-PA binding capacity. This review summarizes evidence for the involvement of PA/PAI-1 system in spermatogenesis, sperm capacitation, and fertilization.     

Extracellular matrix of cultured bovine aortic endothelial cells contains functionally active type 1 plasminogen activator inhibitor

The extracellular matrix (ECM) of cultured bovine aortic endothelial cells (BAEs) was analyzed by immunoblotting and reverse fibrin autography and shown to contain type 1 plasminogen activator inhibitor (PAI-1). Most PAI-1 in the ECM formed complexes with exogenously added tissue-type plasminogen activator (tPA), demonstrating that this PAI-1 was functionally active. The resulting tPA/PAI-1 complexes were recovered in the reaction solution, indicating that the PAI-1 in such complexes no longer bound to ECM. The PAI-1 could not be removed by incubating ECM in high salt (2 mol/L NaCl), sugars (1 mol/L galactose, 1 mol/L mannose), glycosaminoglycans (10 mmol/L heparin, 10 mmol/L dermatan sulfate), or epsilon-aminocaproic acid (0.1 mol/L). However, PAI-1 could be extracted from ECM by treatment with either arginine (0.5 mol/L) or potassium thiocyanate (2 mol/L), or by incubation under acidic conditions (pH 2.5). ECM depleted of PAI-1 by acid extraction was able to bind both the active and latent forms of PAI-1. In this instance, most of the bound PAI-1 did not form complexes with tPA, indicating that the latent form was not activated as a consequence of binding to ECM. Although the PAI-1 activity in conditioned medium decayed with a half-life (t 1/2) of less than 3 hours, the t 1/2 of ECM- associated PAI-1 was greater than 24 hours. These data suggest that PAI- 1 is produced by cultured BAEs in an active form and is then either released into the medium where it is rapidly inactivated or into the subendothelium where it binds to ECM. The specific binding of PAI-1 to ECM protects it from this inactivation.     

Human visceral adipose tissue and the plasminogen activator inhibitor type 1

OBJECTIVE: The objective of this study was to systematically evaluate the molecular basis of the association between visceral fat mass and plasma plasminogen activator inhibitor-1 (PAI-1) levels in man. DESIGN: A comprehensive approach comprising observational, in vitro, and human intervention studies. MEASUREMENTS AND RESULTS: We confirmed an exclusive relationship between visceral fat and plasma PAI-1 levels (r=0.79, P<0.001) and corroborated preferential PAI-1 release from adipose tissue explants. Yet, messenger RNA analysis and in vivo measurement of PAI-1 release from visceral fat (AV-differences over the omentum) not only excluded visceral adipose tissue as a relevant source of circulating PAI-1, but also excluded visceral fat as a significant source of proinflammatory mediators such as tumor necrosis factor-alpha, IL-1 or transforming growth factor-beta that could induce PAI-1 expression in tissues other than visceral fat. Short-term interventions with acipimox and growth hormone (GH) as well as statistical evaluation excluded free fatty acids and GH as metabolic links. Further analysis of the metabolic data in a stepwise regression model indicated that plasma PAI-1 levels and visceral fat rather are co-correlates that both relate to impaired lipid handling. CONCLUSION: Our PAI-1 studies show that visceral fat mass and plasma PAI-1 levels are co-correlated rather than causatively related, with lipid load as common denominator.     

Antigen levels of plasminogen activator inhibitor type 1 in citrated plasma: determination of the contribution of platelets

To estimate the extent of the release of plasminogen activator inhibitor type 1 (PAI-1) from platelets following their accidental activation in vitro during blood collection, we studied the antigen levels of PAI-1 and beta-thromboglobulin (beta TG) in plasma and sera from 38 healthy volunteers. For each volunteer, we calculated the ratio of platelet-derived PAI-1/platelet-derived beta TG from the difference in their respective concentrations in serum and plasma. The mean +/- SD (0.012 +/- 0.003) and median (0.012) ratios were identical before and after 10 or 20 min of venous occlusion. The platelet contribution of PAI-1 in plasma was estimated by multiplying the plasma concentration of beta TG (as a marker of platelet activation) with the above-mentioned individual ratio. Under our conditions of blood collection (1/10 vol of 0.1 M citrate, pH 4.5, as anticoagulant, and handling of blood at 4 degrees C and within 30 min), the platelet contribution to the PAI-1 concentration in plasma was on average 12% and always negligible in plasma with high levels of PAI-1.     

Laboratory evidence of hyperfibrinolysis in association with low plasminogen activator inhibitor type 1 activity.

Low activity of plasminogen activator inhibitor type 1 (PAI-1) has been associated with bleeding complications in surgery. We earlier reported a higher prevalence of low PAI-1 activity among patients with bleeding tendency as compared with normal control individuals. The present study evaluated whether low PAI-1 activity actually is associated with markers of increased fibrinolytic activity in plasma from patients with a history of bleeding. PAI-1 activity, plasmin-antiplasmin complex (PAP) and D-dimer were analyzed in plasma samples from 424 consecutive patients referred to the Coagulation Unit for investigation of bleeding symptoms. The median PAI-1 activity was 4.0 U/ml [interquartile range (IQR), 1-10 U/ml], the median PAP level was 1.59 mg/l (IQR, 1.40-1.91 mg/l) and the median D-dimer level was 71 microg/l (IQR, 46-111 microg/l). The median PAP concentration for patients with PAI-1 less than 1.0 U/ml was 1.73 mg/l (IQR, 1.53-2.30 mg/l), and that for PAI-1 of at least 1.0 U/ml was 1.54 mg/l (IQR, 1.36-1.83 mg/l) (P < 0.0001). There was also a significant difference between the PAP levels in patients with normal PAI-1 (1-15 U/ml) versus elevated PAI-1 (> 15 U/ml) (P = 0.024). The level of D-dimer did not correlate with PAI-1 activity. In conclusion, the activation of plasminogen measured as PAP was higher in patients with bleeding symptoms in combination with PAI-1 activity less than 1.0 U/ml than in those with PAI-1 activity of at least 1.0 U/ml. The coagulation activity under normal conditions, as measured by D-dimer, did not differ between the two patient subsets. The results support our previous definition of low PAI-1 as activity below 1.0 U/ml.     

Molecular advances in plasminogen activator inhibitor 1 interaction with thrombin and tissue-type plasminogen activator

Plasminogen activator inhibitor 1 (PAI-1) is a glycoprotein that controls the activity of the key enzymes of the fibrinolytic system, the serine proteases tissue-type plasminogen activator (t-PA) and urokinase-type plasminogen activator (u-PA). Inhibition is accomplished by rapid formation of inactive, equimolar PAI-1/PA complexes. The physiological importance of PAI-1 for the fibrinolytic system has been underscored by the observation that in humans, a homozygous defect results in hemorrhagic episodes. In addition to its function in surveillance of the integrity of clots, PAI-1 efficiently inhibits the serine protease thrombin in vitro, provided that either the high molecular weight glycosaminoglycan heparin or the glycoprotein vitronectin is present. These cofactors accelerate the rate of thrombin inhibition by PAI-1 by more than two orders of magnitude. Inhibition of thrombin by PAI-1 proceeds according to a "suicide substrate mechanism," typified by a branched reaction pathway, leading either to stable PAI-1/thrombin complexes or to degradation of the inhibitor and recycling of enzyme. The cofactors heparin and vitronectin, although increasing inhibition through different mechanisms, essentially promote PAI-1 degradation by thrombin. In view of the multitude of functions attributed to thrombin, the authors propose that the relevance of thrombin inhibition by PAI-1 is to restrict its mitogenic activity, rather than to affect its coagulation function in plasma. (Trends Cardiovasc Med 1997;7:47-51). ? 1997, Elsevier Science Inc.     

Induction of plasminogen activator inhibitor by products released from platelets

Activation of platelets and augmentation of plasma plasminogen activator inhibitor (PAI) type I activity accompany acute myocardial infarction. To determine whether the two may be related, platelet compounds including epidermal growth factor and transforming growth factor beta as well as platelet lysates were studied in rabbits in vivo. After intravenous infusion of epidermal growth factor (1 and 5 micrograms/kg), plasma PAI activity increased sevenfold and 20-fold, peaking at 2 hours. After infusions of transforming growth factor beta (0.2 and 0.5 microgram/kg), plasma PAI activity increased sevenfold and 12-fold but peaked more slowly (at 5 hours). After infusion of platelet lysates (lysates from 2.8 and 5.6 X 10(8) platelets/kg), the increase was 19-fold and 35-fold, with a peak at 4 hours. Platelet lysates induced a pronounced increase of plasma PAI type 1 messenger RNA (Northern blots) in aorta, liver, and myocardium. Anti-transforming growth factor beta neutralizing antibody markedly attenuated the plasma PAI increase. Concentrations in plasma of fibrinogen and alpha 2-antiplasmin were virtually unaffected under all conditions. Thus, platelet-associated growth factors and platelet lysates, shown previously to increase plasma PAI type 1 messenger RNA expression and protein production in cultured hepatocytes and vascular endothelial cells in vitro, augment plasma PAI in vivo as well. Accordingly, activation of platelets and release of platelet-associated growth factors appear to contribute to the increased plasma PAI seen after myocardial infarction.     

Determination of plasma tissue type plasminogen activator and plasminogen activator inhibitor activity in patients with ischemic stroke

Plasma tissue-type plasminogen activator and plasminogen activator inhibitor were determined during the acute, recovery and sequelae stages of patients with ischemic stroke by chromophoric substrate assay. The result showed that t-PA activity was elevated during the acute phase, remained elevated during the recovery stage and declined during the sequelae stage. Lowering of PAI activity was found during acute phase, which reversed during recovery phase and remained significantly elevated during sequelae stage. As a result, the ratio of PAI/t-PA fluctuated during different stages of the disease. Significant elevation of PAI and PAI/t-PA ratio during sequelae stage may be one of the risk factors of further thrombosis and contribute partly to the high relapsing rate of the disease. In addition, a positive correlation was found between PAI and serum cholesterol content.     

Circulating tissue-type plasminogen activator and plasminogen activator inhibitor type 1 in proliferative diabetic retinopathy: a pilot study

Several haemostatic abnormalities are associated with proliferative diabetic retinopathy. While abnormalities in plasma fibrinolytic activity have been described in diabetic retinopathy, platelets (a rich source of plasminogen activator inhibitor type 1, PAI-1) have received little attention. As a result, little is known about the fibrinolytic potential of circulating whole blood in diabetic retinopathy. The concentrations of tissue-type plasminogen activator (t-PA) and of its fast-acting inhibitor. PAI-1 were measured in plasma from eight patients with type 1 diabetes complicated by proliferative retinopathy, and from eight patients with type 1 diabetes and background or no retinopathy, matched for age, sex and duration of diabetes. The concentration of PAI-1 in platelets was also measured. The ratio of t-PTA to PAI-1 in plasma was significantly higher in patients with proliverative retinopathy than in those without (0.66 vs. 0.37, p < 0.02). The average quantitiy of PAI-1 per patelet was significantly lower in the group with proliferative retinopathy (0.33 vs. 0.50 ng/10⁶ platelets, p < 0.02). These data suggest that among patients with type 1 diabetes, total circulating fibrinolytic potential is higher in those with proliferative retinopathy. Received: 14 October 1998 / Accepted in revised form: 30 September 1999     

Association of protein C and type 1 plasminogen activator inhibitor with primary graft dysfunction

BACKGROUND: Acute lung injury is characterized by hypercoagulability and impaired fibrinolysis. We hypothesized that lower protein C and higher type 1 plasminogen activator inhibitor (PAI-1) levels in plasma would be associated with primary graft dysfunction (PGD) after lung transplantation. Design: Prospective, multicenter cohort study. METHODS: We measured plasma levels of protein C and PAI-1 before lung transplantation and 6, 24, 48, and 72 h after allograft reperfusion in 128 lung transplant recipients at six centers. The primary outcome was grade 3 PGD (Pa(O(2))/Fi(O(2)) < 200 with alveolar infiltrates) 72 h after transplantation. Biomarker profiles were evaluated using logistic regression and generalized estimating equations. RESULTS: Patients who developed PGD had lower protein C levels 24 h posttransplantation than did patients without PGD (mean +/- SD [relative to control]: 64 +/- 27 vs. 92 +/- 41%, respectively; p = 0.002). Patients with PGD also had PAI-1 levels that were almost double those of patients without PGD at 24 h (213 +/- 144 vs. 117 +/- 89 ng/ml, respectively; p < 0.001). Throughout the 72-h postoperative period, protein C levels were significantly lower (p = 0.007) and PAI-1 levels were higher (p = 0.026) in subjects with PGD than in others. These differences persisted despite adjustment for potential confounders in multivariate analyses. Higher recipient pulmonary artery pressures, measured immediately pretransplantation, were associated with higher PAI-1 levels and increased risk of PGD. CONCLUSION: Lower postoperative protein C and higher PAI-1 plasma levels are associated with PGD after lung transplantation. Impaired fibrinolysis and enhanced coagulation may be important in PGD pathogenesis.     

Independent regulation of plasminogen activator inhibitor 2 and plasminogen activator inhibitor 1 in human synovial fibroblasts

Objective. To study the plasminogen activator inhibitor(s) (PAI) produced in vitro by human synovial fibroblast-like cells. Methods. Human synovial cell explant cultures were established using cells from nonrheumatoid donors. PAI-2 and PAI-1 antigens were measured by enzyme-linked immunosorbent assay, and messenger RNA (mRNA) levels were determined by Northern blot. Results. The synovial fibroblasts produced both PAI-2 and PAI-1. Interleukin-1 (IL-1) increased PAI-2 but decreased PAI-1 formation, both at the protein and the mRNA levels. Using cyclooxygenase inhibitors, evidence was obtained that an endogenous cyclooxygenase product(S) in the IL-1—treated cultures inhibited formation of both PAIs; exogenous prostaglandin E₂ (10⁻⁷M) reversed the effect of cyclooxygenase inhibition. The glucocorticoid dexamethasone (10⁻⁶ to 10⁻⁷M) inhibited IL-1—stimulated PAI-2 formation but reversed the suppressive effect of IL-1 on PAI-1 production. Conclusion. PAI-2 formation and PAI-1 formation can be regulated independently in human synoviocytes, illustrating the complexity of the modulation of the net PA activity expressed by these cells.     

Increased type 1 plasminogen activator inhibitor gene expression in atherosclerotic human arteries.

Decreased fibrinolytic capacity has been suggested to accelerate the process of arterial atherogenesis by facilitating thrombosis and fibrin deposition within developing atherosclerotic lesions. Type 1 plasminogen activator inhibitor (PAI-1) is the primary inhibitor of tissue-type plasminogen activator and has been found to be increased in a number of clinical conditions generally defined as prothrombotic. To investigate the potential role of this inhibitor in atherosclerosis, we examined the expression of PAI-1 mRNA in segments of 11 severely diseased and 5 relatively normal human arteries obtained from 16 different patients undergoing reconstructive surgery for aortic occlusive or aneurysmal disease. Densitometric scanning of RNA (Northern) blot autoradiograms revealed significantly increased levels of PAI-1 mRNA in severely atherosclerotic vessels (mean densitometric value, 1.7 +/- 0.28 SEM) compared with normal or mildly affected arteries (mean densitometric value, 0.63 +/- 0.09 SEM; P less than 0.05). In most instances, the level of PAI-1 mRNA was correlated with the degree of atherosclerosis. Analysis of adjacent tissue sections from the same patients by in situ hybridization demonstrated an abundance of PAI-1 mRNA-positive cells within the thickened intima of atherosclerotic arteries, mainly around the base of the plaque. PAI-1 mRNA could also be detected in cells scattered within the necrotic material and in endothelial cells of adventitial vessels. In contrast to these results, PAI-1 mRNA was visualized primarily within luminal endothelial cells of normal-appearing aortic tissue. Our data provide initial evidence for the increased expression of PAI-1 mRNA in severely atherosclerotic human arteries and suggest a role for PAI-1 in the progression of human atherosclerotic disease.