Nonsyndromic hearing impairment is associated with a mutation in DFNA5.

Nonsyndromic hearing impairment is one of the most heterogeneous hereditary conditions, with more than 40 loci mapped on the human genome, however, only a limited number of genes implicated in hearing loss have been identified. We previously reported linkage to chromosome 7p15 for autosomal dominant hearing impairment segregating in an extended Dutch family (DFNA5). Here, we report a further refinement of the DFNA5 candidate region and the isolation of a gene from this region that is expressed in the cochlea. In intron 7 of this gene, we identified an insertion/deletion mutation that does not affect intron-exon boundaries, but deletes five G-triplets at the 3' end of the intron. The mutation co-segregated with deafness in the family and causes skipping of exon 8, resulting in premature termination of the open reading frame. As no physiological function could be assigned, the gene was designated DFNA5.

Mutations in the human alpha-tectorin gene cause autosomal dominant non-syndromic hearing impairment.

The tectorial membrane is an extracellular matrix of the inner ear that contacts the stereocilia bundles of specialized sensory hair cells. Sound induces movement of these hair cells relative to the tectorial membrane, deflects the stereocilia, and leads to fluctuations in hair-cell membrane potential, transducing sound into electrical signals. Alpha-tectorin is one of the major non-collagenous components of the tectorial membrane. Recently, the gene encoding mouse alpha-tectorin (Tecta) was mapped to a region of mouse chromosome 9, which shows evolutionary conservation with human chromosome 11q (ref. 3), where linkage was found in two families, one Belgian (DFNA12; ref. 4) and the other, Austrian (DFNA8; unpublished data), with autosomal dominant non-syndromic hearing impairment. We determined the complete sequence and the intron-exon structure of the human TECTA gene. In both families, mutation analysis revealed missense mutations which replace conserved amino-acid residues within the zona pellucida domain of TECTA. These findings indicate that mutations in TECTA are responsible for hearing impairment in these families, and implicate a new type of protein in the pathogenesis of hearing impairment.

Involvement of T-cell receptor-beta alterations in the development of otosclerosis linked to OTSC2.

Otosclerosis is a common form of hearing loss, characterized by disordered bone remodeling in the otic capsule. Within the otosclerotic foci, several immunocompetent cells and immune-modulating factors can be found. Different etiological theories involving the immune system have been suggested. However, a genetic component is clearly present. In large otosclerosis families, seven autosomal-dominant loci have been found, but none of the disease-causing genes has been identified. This study focused on the exploration of the second otosclerosis locus on chromosome 7q34-36 (OTSC2), holding the T-cell receptor beta locus (TRB locus). A significantly lower T-cell receptor-beta (TCR-beta) mRNA expression and percentage of blood circulating TCR-alphabeta(+) T cells was detected in OTSC2 patients compared with controls and patients with the complex form of the disease. Further analysis illustrated more significant disturbances in specific T-cell subsets, including an increased CD28(null) cell population, suggesting a disturbed T-cell development and ageing in OTSC2 patients. These disturbances could be associated with otosclerotic bone remodeling, given the known effects of immunocompetent cells on bone physiology. These data implicate the TRB locus as the causative gene in the OTSC2 region and represent an important finding in the elucidation of the disease pathology.

Hearing impairment and neurological dysfunction associated with a mutation in the mitochondrial tRNASer(UCN) gene.

We studied a large Dutch family with maternally inherited, progressive, sensorineural hearing loss in 27 patients. Only in a single family member was the hearing loss accompanied by neurological symptoms including ataxia and dysarthria. DNA analysis of the mitochondrial genome revealed the insertion of a C at nucleotide position 7472 in the tRNASer(UCN) gene (7472insC mutation). We determined the percentage of mutant DNA (heteroplasmy) in blood from all family members, and found no correlation between hearing loss and leucocyte heteroplasmy. The 7472insC mutation was previously identified in a smaller family from Sicily with sensorineural hearing loss in 9 family members, six of them also presenting neurologically with ataxia and myoclonus. The presence of the 7472insC mutation in two different pedigrees strongly supports its pathogenicity. However, the interfamilial difference in penetrance of the neurologic abnormalities is most likely to be strongly influenced by secondary factors different from the 7472insC mutation, as heteroplasmy or age of the patients were similar in both families. This mutation should therefore be analysed in families with maternally inherited hearing loss, irrespective of whether the hearing loss is non-syndromic or accompanied by neurologic abnormalities.

Refined mapping of a gene for autosomal dominant progressive sensorineural hearing loss (DFNA5) to a 2-cM region, and exclusion of a candidate gene that is expressed in the cochlea.

A gene for an autosomal dominant form of progressive sensorineural hearing loss (DFNA5) was previously assigned by us to a 15-cM region on chromosome 7p15. In this study, the DFNA5 candidate region was refined to less than 2 cM, and completely cloned in a YAC contig. The HOXA1 gene located in 7p15 was considered to be a good candidate gene for DFNA5 as it harbours mutations leading to developmental defects of the inner ear in mice. However, the refinement of the candidate region of DFNA5 excludes the HOXA1 gene as a candidate for DFNA5. We cloned a novel candidate gene (CG1, candidate gene 1), which is expressed in human fetal cochlea, from the DFNA5 candidate region. The complete cDNA sequence of CG1, encoding a 423 amino acid protein of unknown function, was determined. Mutation analysis of the CG1 gene in DFNA5 patients, however, could not reveal a disease-causing mutation.

Otosclerosis: a genetically heterogeneous disease involving at least three different genes.

Otosclerosis is caused by abnormal bone homeostasis of the otic capsule, resulting in hearing impairment in 0.3%-0.4% of the white population. The etiology of the disease remains unclear and environmental as well as genetic factors have been implicated. We localized the first autosomal-dominant locus to chromosome 15 in 1998 (OTSC1) in an Indian family and, recently, we reported the localization of a second gene for otosclerosis to a 16 cM interval on chromosome 7q (OTSC2). In this study, we recruited and analyzed nine additional families (seven Belgian and two Dutch families with 53 affected and 20 unaffected subjects) to investigate the importance of these loci in autosomal-dominant otosclerosis. We completed linkage analysis with three microsatellite markers of chromosome 15 (D15S652, D15S1004, D15S657) and five microsatellite markers of chromosome 7 (D7S495, D7S2560, D7S684, D7S2513, D7S2426). In two families, results compatible with linkage to OTSC2 were found, but in the seven remaining families OTSC1 and OTSC2 were excluded. Heterogeneity testing provided significant evidence for genetic heterogeneity, with an estimated 25% of families linked to OTSC2. These results indicate that, besides OTSC1 and OTSC2, there must be at least one additional otosclerosis locus.

Immunoassay of murine t-PA, u-PA and PAI-1 using monoclonal antibodies raised in gene-inactivated mice.

Three enzyme-linked immunosorbent assays for the quantitation of murine tissue-type plasminogen activator (t-PA), urokinase-type plasminogen activator (u-PA) and plasminogen activator inhibitor 1 (PAI-1), were developed using monoclonal antibodies raised against the autologous proteins in gene-inactivated mice. Dose-response was linear for t-PA and PAI-1 between 5 and 0.1 ng/ml and for u-PA between 50 and 1 ng/ml, with intra-assay, inter-assay and inter-dilution coefficients of variation of 6 to 14%. Assay recoveries of proteins (5 to 100 ng/ml) added to plasma were 73 to 95% for t-PA and PAI-1. Linear correlations (r = 0.65, r = 0.91 and r = 0.92, for t-PA, u-PA and PAI-1 respectively) were found between antigen and activity in plasma, urine and tissue extracts. Levels of t-PA and PAI-1 antigen in murine plasma were 2.5 +/- 1.0 ng/ml (mean +/- SD, n=9) and 1.9 +/- 0.6 ng/ml (mean +/- SD, n = 8), respectively, in wild-type mice and undetectable in gene-inactivated mice. Bradykinin injection in mice provoked a 12-fold increase (p < 0.0002) of t-PA and endotoxin injection an 80-fold increase (p < 0.005) of PAI-1 levels. u-PA antigen levels in urine from wild-type mice ranged between 0.2 and 8.2 micrograms/ml (1.8 +/- 1.9 micrograms/ml, mean +/- SD, n = 17) and were undetectable in gene-inactivated mice. Thus, these assays may be useful for studies on the role of these proteins in tissue remodeling, atherosclerosis, embryogenesis, etc., in established mouse models. Gene-inactivated mice may constitute a general approach for the generation of monoclonal antibodies against the deficient translation products and for the development of specific immunoassays for murine proteins.

An enzyme-linked immunosorbent assay (ELISA) for the measurement of plasmin-alpha 2-antiplasmin complex in human plasma--application to the detection of in vivo activation of the fibrinolytic system.

An enzyme-linked immunosorbent assay (ELISA) is developed for the measurement of plasmin-alpha 2-antiplasmin complex in human plasma. Microtiter plates were coated with a mixture of two murine monoclonal antibodies directed against human alpha 2-antiplasmin and bound plasmin-alpha 2-antiplasmin complex was quantitated with a peroxidase-conjugated monoclonal antibody directed against human plasminogen. The lower limit of sensitivity of the assay was 0.01 nM of plasmin-alpha 2-antiplasmin complex in 100-fold diluted human plasma, allowing detection of 1 nM in undiluted plasma samples. After 100-fold dilution of the plasma samples, the assay was no longer influenced by the presence of the precursors plasminogen and alpha 2-antiplasmin. At a concentration of 2.0 nM of plasmin-alpha 2-antiplasmin complex in plasma, intra- and interassay variation coefficients were 4.2 and 5.5 percent respectively. In plasma samples of 25 control subjects the levels of plasmin-alpha 2-antiplasmin complex were below 1 nM. Extensive in vivo activation of the fibrinolytic system during thrombolytic therapy with streptokinase resulted in the generation of elevated levels of plasmin-alpha 2-antiplasmin complex up to 690 +/- 150 nM. No measurable levels of plasmin-alpha 2-antiplasmin complex were found in the plasma of 32 patients with acute deep vein thrombosis nor in the plasma of 11 patients with recurrent deep vein thrombosis. These findings indicate that plasmin-alpha 2-antiplasmin complex is generated during in vivo activation of the fibrinolytic system and that its assay may be useful to monitor thrombolytic therapy but not for the diagnosis of venous thrombosis.

An enzyme-linked immunosorbent assay for urokinase-type plasminogen activator (u-PA) and mutants and chimeras containing the serine protease domain of u-PA.

An enzyme-linked immunosorbent assay (ELISA) for quantitation of natural and recombinant plasminogen activators containing the serine protease domain (B-chain) of urokinase-type plasminogen activator (u-PA) was developed, based on two murine monoclonal antibodies, MA-4D1E8 and MA-2L3, raised against u-PA and reacting with non-overlapping epitopes in the B-chain. MA-4D1E8 was coated on microtiter plates and bound antigen was quantitated with MA-2L3 conjugated with horseradish peroxidase. The intra-assay, inter-assay and inter-dilution coefficients of variation of the assay were 6%, 15% and 9%, respectively. Using recombinant single-chain u-PA (rscu-PA) as a standard, the u-PA-related antigen level in normal human plasma was 1.4 +/- 0.6 ng/ml (mean +/- SD, n = 27). The ELISA recognized the following compounds with comparable sensitivity: intact scu-PA (amino acids, AA, 1 to 411), scu-PA-32k (AA 144 to 411), a truncated (thrombin-derived) scu-PA comprising AA 157 to 411, and chimeric t-PA/u-PA molecules including t-PA(AA1-263)/scu-PA(AA144-411), t-PA(AA1-274)/scu-PA(AA138-411) and t-PA(AA87-274)/scu-PA(AA138-411). Conversion of single-chain to two-chain forms of u-PA or inhibition of active two-chain forms with plasminogen activator inhibitor-1 or with the active site serine inhibitor phenyl-methyl-sulfonyl fluoride, did not alter the reactivity in the assay. In contrast, inactivation with alpha 2-antiplasmin or with the active site histidine inhibitor Glu-Gly-Arg-CH2Cl resulted in a 3- to 5-fold reduction of the reactivity.(ABSTRACT TRUNCATED AT 250 WORDS)

Role of alpha 2-antiplasmin in fibrin-specific clot lysis with single-chain urokinase-type plasminogen activator in human plasma.

The role of plasma alpha 2-antiplasmin (alpha 2-AP) in the fibrin-specificity of clot lysis by recombinant single-chain urokinase-type plasminogen activator (rscu-PA) and in the conversion of rscu-PA to its two-chain derivative (rtcu-PA, urokinase) was investigated in an in vitro human plasma clot lysis system. Fifty % lysis in 2 h of a 0.1 ml 125I-fibrin labeled human plasma clot immersed in 0.5 ml normal human plasma was obtained with 1.4 +/- 0.15 micrograms/ml rscu-PA (mean +/- SD, n = 8). This was associated with degradation of 23 +/- 7% of fibrinogen and generation of 0.20 +/- 0.09 micrograms/ml rtcu-PA. In alpha 2-AP-depleted plasma 50% clot lysis in 2 h required 2-fold less rscu-PA which was associated with 3-fold more extensive fibrinogen degradation and 2-fold more rtcu-PA generation. Fifty % lysis in 2 h, of a 0.1 ml alpha 2-AP-depleted plasma clot, submersed in 0.5 ml normal plasma, was obtained with 0.80 +/- 0.05 micrograms/ml rscu-PA (n = 3, p less than 0.001 vs normal clot) and was associated with 17 +/- 6% fibrinogen breakdown (p = 0.22 vs normal clot) and 0.08 +/- 0.02 micrograms/ml rtcu-PA generation (p less than 0.05 vs normal clot). In alpha 2-AP-depleted plasma the equipotent rscu-PA concentration was 4-fold lower than in normal plasma and was associated with 3-fold more fibrinogen degradation and a similar extent of rtcu-PA generation.(ABSTRACT TRUNCATED AT 250 WORDS)

Hereditary otovestibular dysfunction and Ménière's disease in a large Belgian family is caused by a missense mutation in the COCH gene.

OBJECTIVE: To report the clinical, auditory, and vestibular characteristics of a nonsyndromic otovestibular dysfunction in a large Belgian family caused by a missense mutation of the DFNA9 gene: COCH. STUDY DESIGN: Retrospective study of the clinical, audiologic, and vestibular data of 60 genetically affected cases. SETTING: Tertiary referral center. PATIENTS: All members of a Belgian kindred who carry the genetic (P51S) defect linked to the inherited hearing and vestibular impairment. INTERVENTIONS: Diagnostic otologic, audiometric, and vestibular analysis and imaging. MAIN OUTCOME MEASURES: Pure tone audiometry, supraliminary audiometry. and vestibular investigation. RESULTS: The autosomal dominant inherited impairment was characterized by peripheral degeneration of the inner ear, leading to total deafness and bilateral vestibular areflexia. CONCLUSIONS: The genetically affected persons of a Belgian family shared a progressive sensorineural hearing loss starting between the third and sixth decade. Vestibular symptoms started at about the same age as the hearing loss. The vestibular symptoms consisted of instability in darkness, a tendency to fall sideways, light-headiness, a drunken feeling, and attacks of vertigo. Most of the patients reported tinnitus, and half of them reported pressure in the ears. Clinically, 9 of the 60 patients met the criteria for definite Ménière's disease, and another 13 and 17 patients met the criteria for probable or possible Ménière's disease, respectively. All 9 were older than the age of 35, but only 1 was older than 55 years, so more than 30% of the patients were between 35 and 55 years old. A specific pattern could be recognized in the evolution of the otovestibular impairment. Under the age of 35 years, almost all the affected family members had normal hearing, whereas above the age of 55 years, the hearing loss was at least moderate, and vestibular hypofunction occurred. In between, there was a transition period of two to three decades, when deterioration of the cochleovestibular function occurred, with a temporary audiometric and vestibular asymmetry.

Transient depression of inner ear function after stapedotomy: skeeter versus CO(2) laser technique.

Performing stapes surgery for otosclerosis is known to be potentially irreversibly harmful to the inner ear function in about 1% of the cases. An early postoperative transient depression of the bone conduction thresholds is frequently detected after stapes surgery. The purpose of this study was to compare the evolution of bone conduction thresholds after primary stapedotomy with two different techniques: skeeter versus CO(2) laser stapedotomy. Audiological data of 336 otosclerosis operations performed by 2 surgeons between 1997 and 2003 were subjected to analysis. The calibrated hole in the footplate was performed randomly either with the skeeter drill or with the CO(2) laser. Preoperative bone conduction thresholds were compared with the postoperative levels (day 2-3, week 2, week 6 and month 6) in all patients. Evolution of the bone conduction was compared for the two studied subgroups (laser versus skeeter).

Osteogenesis imperfecta and hearing loss.

A review of the literature of osteogenesis imperfecta (OI) and hearing loss in the last century is given. This includes the first association in 1912 by Adair-Dighton of OI with hearing loss, the triad of VAN DER HOEVE and DE KLEYN and different clinical and audiometric studies and classifications of the disease. A case report of a 14 year old patient is also included. The genetics of OI and the complexity in correlating genetics with clinical classifications is elucidated. Finally, the findings reported after middle ear surgery, the results of surgery and comparisons with otosclerosis are reviewed.

Screening panels of monoclonal antibodies using phage-displayed antigen.

A procedure is described to screen panels of hybridomas or purified monoclonal antibodies using antigen displayed on the surface of filamentous bacteriophage. In this system, samples containing murine monoclonal antibodies are incubated with phage-displayed antigen in microtiter plates coated with rabbit anti-mouse IgG, and bound antibody-phage complex is detected with horseradish peroxidase-sheep anti-phage M13 conjugate. The assay has been validated with a panel of 16 monoclonal antibodies directed against human plasminogen, using phage-displayed miniplasmin-(ogen) (amino acids Ala444 through Asn791 comprising kringle 5 and the proteinase domain of plasminogen) or microplasminogen (amino acids Ala543 through Asn791 comprising the proteinase domain). Six monoclonal antibodies were identified directed against miniplasminogen and miniplasmin; this was confirmed using a microtiter plate coated with antigens. One of these monoclonal antibodies (MA-42B12) did not react with microplasminogen, suggesting that its epitope is comprised within the kringle 5 domain. This test is rapid and sensitive (detecting 10-20 ng/ml of monoclonal antibody), and screening can be performed using phage-displayed zymogens or active enzymes or selected domains thereof. The procedure eliminates the need for large amounts of purified antigen for screening. Furthermore, immunization can be performed with partially purified antigen because only antibodies raised against the antigen of interest will be identified with the use of phage-displayed antigen. Therefore, this test may offer distinct advantages over the classical one-site enzyme-linked immunosorbent assay using antigen-coated microtiter plates.

Generation of monoclonal antibodies against autologous proteins in gene-inactivated mice.

Induction of an immune response is strongly dependent on the phylogenetic distance between antigen and recipient. In general, antibodies will not be raised against self-antigens nor against highly conserved domains. In the present study we describe the production and characterization of murine monoclonal "auto-antibodies" against murine tissue-type plasminogen activator (t-PA) raised in "knock-out" mice, homozygously deficient of the functional gene. 203 stable hybridomas were obtained producing murine monoclonal antibodies against murine t-PA. Analysis of the species reactivity revealed that 182 cross-reacted with one or more (t-)PAs originating from other species including rat t-PA, human t-PA, and vampire bat-PA. 121 reacted with epitopes conserved among murine, rat, and human t-PA. In addition, 31 of the monoclonal antibodies were directed against domains present in all four species. Epitope mapping indicated a high frequency of specificity toward diverse epitopes that are highly conserved across species. Comparative analysis of their influence on the enzymatic activity of t-PA and their species cross-reactivity clearly demonstrated that the domains required for the biological activity of plasminogen activators are more conserved (p < 0.02) than non-functional domains. The availability of such unique antibodies against a wide variety of conserved epitopes may facilitate studies on the structural homologies between (t-)PAs isolated from various species. The present approach should also apply to various other classes of proteins, allowing the generation of monoclonal antibodies, against conserved epitopes, which could not be raised in wild-type animals because of their "self-antigen" nature.

Depletion of circulating alpha(2)-antiplasmin by intravenous plasmin or immunoneutralization reduces focal cerebral ischemic injury in the absence of arterial recanalization.

In the absence of arterial recanalization, thrombolytic agents induce a dose-related extension of focal cerebral ischemic injury (FII) in experimental animals. However, FII is smaller in mice lacking alpha(2)-antiplasmin (alpha(2)-AP), the physiologic inhibitor of plasmin, suggesting its depletion might reduce FII in the absence of reperfusion. Therefore, the effect of human plasmin (Pli), human miniplasmin (mPli), and an Fab fragment neutralizing murine alpha(2)-AP (Fab-4H9) on FII after middle cerebral artery (MCA) ligation was studied in mice and in hamsters. In BALB/c mice, the median FII after 24 hours was 28 microL (range, 20-34) (n = 10) with saline and 23 microL (range, 17-26) (n = 9) with a single bolus of 0.07 mg Pli, given after MCA ligation (P =.010), which reduced alpha(2)-AP to 44% and fibrinogen from 0.75 to 0.44 g/L. FII was 20 microL (range, 13-26) (n = 6, P =.025) with 0.2 mg mPli and was 24 microL (range, 20-27) (n = 6, P =.020) with 1.7 mg Fab-4H9. Neuronal atrophy and reduction of laminin immunoreactivity were comparably observed in the infarct area after saline and Pli. In hamsters, a single bolus injection of 1 mg Pli, after MCA ligation, depleted alpha(2)-AP and fibrinogen and reduced FII at 24 hours from 20 microL (range, 9.9-38) (n = 6) to 7.0 microL (range, 0.44-31) (n = 7, P =.032). Thus, reduction of circulating alpha(2)-AP, with a single bolus of plasmin or of a neutralizing antibody fragment, significantly reduced FII after MCA ligation in mouse and hamster models, suggesting that, provided these observations can be extrapolated to human beings, transient depletion of circulating alpha(2)-AP might reduce ischemic stroke in the absence of reperfusion.

A monoclonal antibody specific for two-chain urokinase-type plasminogen activator. Application to the study of the mechanism of clot lysis with single-chain urokinase-type plasminogen activator in plasma.

A murine monoclonal antibody (MA-12E6A8) was raised against human urokinase-type plasminogen activator (u-PA), which, in an enzyme-linked immunosorbent assay (ELISA), reacted 15,000-fold better with recombinant two-chain u-PA (rtcu-PA) than with recombinant single-chain u-PA (rscu-PA). The antibody had no effect on the activity of rtcu-PA or on its inhibition by a chloromethylketone, but reduced the inhibition of rtcu-PA by recombinant plasminogen activator inhibitor-1 (rPAI-1) at least 10-fold. The dissociation constant of the rtcu-PA/MA-12E6A8 complex was 7 nmol/L. An ELISA was developed using MA-12E6A8 as capture antibody and a horseradish peroxidase conjugated u-PA specific antibody for tagging. It recognized free and active site blocked rtcu-PA but not rtcu-PA in complex with rPAI-1 or with alpha 2-antiplasmin. This ELISA was used to monitor the generation of rtcu-PA during fibrin clot lysis with rscu-PA in human plasma. Addition of 5 micrograms/mL rscu-PA to 3 mL plasma containing a 0.2 mL 125I-fibrin labeled plasma clot caused 50% clot lysis in 62 +/- 13 minutes (mean +/- SD, n = 6), at which time 99 +/- 28 ng/mL rtcu-PA was detected but no fibrinogen breakdown had occurred. Fifty percent fibrinogen breakdown did occur only when rtcu-PA had reached a level of 1,000 +/- 270 ng/mL (at 150 +/- 21 minutes). rscu-PA, 2 micrograms/mL, induced 50% clot lysis in 160 +/- 41 minutes (n = 6); no fibrinogen degradation occurred within 4 hours and rtcu-PA levels did not exceed 80 ng/mL. In the absence of a fibrin clot, 5 micrograms/mL rscu-PA added to human plasma did not result in significant generation of rtcu-PA (less than 50 ng/mL after 4 hours) and no fibrinogen degradation was observed. These results indicate that clot lysis with rscu-PA in a plasma milieu does not require extensive systemic conversion of rscu-PA to rtcu-PA, and that fibrinogen degradation occurs secondarily to systemic conversion of rscu-PA to rtcu-PA.

Measurement of plasminogen activator inhibitor 1 in biologic fluids with a murine monoclonal antibody-based enzyme-linked immunosorbent assay.

An enzyme-linked immunosorbent assay for plasminogen activator inhibitor-1 (PAI-1) in biologic fluids was developed on the basis of two murine monoclonal antibodies raised against PAI-1 purified from HT-1080 fibrosarcoma cells. The lower limit of sensitivity of the assay in plasma is 2 ng/mL. The assay is 12 times less sensitive toward the PAI-1/human tissue-type plasminogen activator (t-PA) complex as compared with free PAI-1. The intraassay, interassay, and interdilution coefficients of variation are 5.2%, 8.0%, and 7.1%, respectively. The level of PAI-1 in platelet-poor plasma of healthy subjects is 18 +/- 10 ng/mL (mean +/- SD, n = 45). In platelet-rich plasma after freezing and thawing, 92% of PAI-1 antigen is released from platelets, whereas only 8% is found in the corresponding platelet-poor plasma. In platelet-poor plasma from healthy subjects, a linear correlation (r = 0.80) was found between PAI activity and PAI-1 antigen. In plasma approximately two thirds of the PAI-1 antigen was functionally active, whereas only 5% of the PAI-1 antigen released from platelets was active. During pregnancy a progressive increase of PAI-1 antigen levels up to three- to sixfold the control value was observed. In plasma of patients with recurrent deep vein thrombosis, PAI-1 levels were 44 +/- 20 ng/mL (mean +/- SD, n = 7), during a clinically silent phase. Four of these patients had a level above 38 ng/mL (mean +/- 2 SD of normal). The present assay, based on stable and reproducible reagents, allows the specific determination of PAI-1 antigen in biologic fluids. It may facilitate interlaboratory comparisons and be useful for further investigations of the role of PAI-1 in clinical conditions associated with impaired fibrinolysis and/or a tendency to thrombosis and investigations of the role of PAI-1 in platelets.