| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1 Research & Development Department, Mitsubishi Kagaku Medical, Inc., 8-5-1 Cyuou, Ami-machi, Inashiki-gun, Ibaraki 300-0332, Japan
2 Management Department, Dia-Iatron Co., Ltd., Tokyo 101-0031, Japan
aauthor for correspondence: fax 81-298-87-6807, e-mail MED0030{at}cc.m-kagaku.co.jp
Plasminogen activator inhibitor-1 (PAI-1) is a key regulator of the fibrinolytic system (1). The continuously high PAI-1 concentrations make fibrins resistant to dissolution by tissue-type plasminogen activator (tPA), which leads to multiorgan dysfunction and a bad prognosis in patients (2). Several methods have been developed for measuring the functional activity of PAI-1 in plasma samples based on the principle of adding a specified amount of tPA in excess of the PAI-1 and measuring residual tPA activity after a short incubation period (3). These methods, however, are neither completely specific nor accurate, and many other protease inhibitors in plasma inhibit tPA (4). Although several assays have also been described for PAI-1 antigen that are based on two-step enzyme immunoassays using monoclonal and/or polyclonal antibodies (5)(6), they have a narrow measurement range (0–40 µg/L) and are time-consuming (7). Because PAI-1 is a labile molecule, the utmost care is needed during blood collection and sample handling to ensure accurate measurement of PAI-1 and tPA. The half-life for the transformation from the active form of PAI-1 into the inactive latent form is 
4 h in vitro and even shorter in vivo (8). It is therefore essential to correctly evaluate the amount of PAI-1 released from endothelial cells and adipose tissue (9) to make such methods clinically useful. We have developed a latex photometric immunoassay (LPIA) for total PAI-1 within a dynamic measurement range that can detect all forms of PAI-1 without the influence of conformational changes.
Samples were obtained from 47 patients with myocardial infarction (29 men and 18 women) and 276 hospital employees, who volunteered for this study, as controls [100 men (age range, 23–63 years) and 168 women (age range, 21–57 years)]. After informed consent, blood samples were collected in a 1/10 volume of a solution containing 38 g/L sodium citrate and centrifuged at 2000g for 15 min to obtain platelet-poor plasma. Blood samples from the healthy volunteers were taken between 0900 and 1100. All plasma samples were stored at -80 °C until the assay.
For the PAI-1 assay, suspended polystyrene latex particles were coated with polyclonal antibody F(ab')2 fragment against PAI-1 purified from the conditioned medium of dexamethasone-stimulated human fibroblast (WI38VA13/2RA) cells. Briefly, a 10 g/L latex particle suspension and 1/10 volume of 4 g/L anti PAI-1 polyclonal antibody F(ab')2 fragment were mixed with 0.1 mol/L sodium acetate (pH 4.0) by stirring for 30 min at room temperature. The latex particle size was 0.45 µm, and the amount of adsorbed anti-PAI-1 F(ab')2 fragment was 0.3 mg/mL of 10 g/L latex suspension. After centrifugation, we added 3 g/L bovine serum albumin (BSA) in 0.1 mol/L Tris (pH 8.0) to the mixture and stirred for 30 min at room temperature. After centrifugation, the precipitate was suspended with distilled water to make a 10 g/L latex suspension. The polyclonal antibody F(ab')2 fragment was obtained from rabbit antisera by ammonium sulfate precipitation and pepsin digestion. The antibody F(ab')2 fragments were purified by immunoadsorption to PAI-1 antigen immobilized on Sepharose 4B and elution with 0.1 mol/L glycine-HCl (pH 2.5) so that antibodies obtained were uniform and highly specific.
We used the LPIA-200 (Mitsubishi Chemical Co.), a fully automated quantitative latex photometric immunoassay instrument, to measure PAI-1. The LPIA-200 automatically pipetted 5 µL of plasma sample and 250 µL of reaction buffer into the reaction cuvette, and 40 µL of reagent containing 20 g/L latex was then stirred in. The change in absorbance at 950 nm produced by latex agglutination was monitored over 10 min, and the concentration of PAI-1 in the sample was calculated from the calibration curve. To prepare calibrators, we diluted the PAI-1 antigen, the concentration of which was determined by the Bradford method with BSA as a calibrator, to 15, 30, 90, 180, and 270 µg/L in a solution containing 0.1 mol/L Tris-HCl (pH 8.0), 0.15 mol/L NaCl, and 1 g/L BSA. The method was linear across the calibration curve (r = 0.999) and had a detection limit of 8 µg/L, which was defined as the PAI-1 concentration at which the mean agglutination rate minus 3 SD did not overlap the mean agglutination rate for the 0 µg/L calibrator plus 3 SD. The intraassay imprecision was determined by measuring three independent plasma samples at 10 different times. Their intraassay CV was 0.9–1.8% for samples containing 20–155 µg/L PAI-1. The interassay imprecision was determined by measuring the same samples on 10 different days. The interassay CV was 3.4–8.2%. In patients with overt disseminated intravascular coagulation and sepsis, total PAI-1 antigen is 10–20 times higher than in healthy volunteers (10)(11). To distinguish PAI-1 concentrations in healthy individuals from concentrations associated with disease, the assay needs to cover a dynamic range. The range of our assay (0–270 µg/L) is more than seven times wider than those of other assays, and it does not require dilution of samples.
To measure concentrations in healthy volunteers, we modified the method described above. Using the same reagent and testing system, we changed only sample volume (10 µL) and the dilutions of the calibrators (5, 10, 30, 60, and 90 µg/L). This modified method was linear across the calibration curve (r = 1.000) and had a detection limit of 2 µg/L. The reference interval (defined as the parametrically determined central 95% interval) for total plasma PAI-1 from 276 healthy volunteers was 3.18–49.86 µg/L. The interval for men was 3.26–67.40 µg/L, and that for women was 3.20–40.40 µg/L. Total PAI-1 was significantly higher in men (P <0.01) as was reported previously (12). None of the individuals enrolled had clinical evidence of disease.
We investigated the correlation between our assay, the conventional method for PAI-1 activity (Spectrolyse/fibrin; Biopool), and a commercialized total PAI-1 antigen assay based on polyclonal antibody (TintElize PAI-1; Biopool) in 47 myocardial infarction samples. Both comparisons gave linear correlation by Deming regression analysis. Comparison with the activity assay gave an equation with a mean (SE) slope of 1.628 (0.187) and a y-intercept of 3.37 (3.64) IU/mL (r = 0.79; Sy|x = 0.175 IU/mL), and comparison with the total PAI-1 assay gave an equation with a slope of 1.289 (0.075) and a y-intercept of -2.12 (2.13) µg/L (r = 0.93; Sy|x = 0.463 µg/L).
We confirmed the specificity of this assay by comparing the reactivity with various forms of PAI-1 with several measurement methods (Fig. 1
). Purified PAI-1 antigen was activated based on the method of Hekman and Loskutoff (13), and single-chain tPA was added to active PAI-1. The amount of tPA/PAI-1 complex increased with the addition of tPA, and the concentration of the tPA/PAI-1 complex was saturated when the molar ratio for tPA:PAI-1 was 
1:1. In contrast, PAI-1 activity decreased with increasing amounts of tPA. Although it is believed that the TintElize PAI-1 can measure all forms of PAI-1 with equal efficiency, our results showed that it reflected the amount of tPA/PAI-1 complex only when the tPA molar concentration was lower than the PAI-1 molar concentration. It appeared that this assay might be insensitive to active PAI-1. Total PAI-1 measured by the LPIA, however, was constant despite the addition of tPA. In addition, the amounts of PAI-1 detected by our assay did not change, although active PAI-1 changed to latent PAI-1 as a result of freezing and thawing.
|
Although it is thought that measuring active PAI-1 concentrations is the best way to evaluate the inhibitory ability of tPA, active PAI-1 quickly loses its activity with time as a result of conformational changes. We therefore believe that the activity assay does not reflect the real PAI-1 activity of plasma, especially in cases of retesting or multisample testing, which are influenced by the loss of activity with time. Furthermore, the plasma must be separated immediately after blood sampling, and active PAI-1 concentrations must be measured immediately after plasma separation, which is a cumbersome protocol for clinical laboratories. In our assay, total PAI-1 was stable at 10–25 °C for 5 days after separation of plasma, making it suitable for clinical laboratory use.
PAI-1 is also present in platelets and is released by stimulation. Our assay recognizes this platelet-derived PAI-1, which causes the total PAI-1 concentration to increase. We must therefore either use platelet-poor plasma or avoid activating the platelets in plasma samples.
In conclusion, we report the development of a fully automated LPIA for total PAI-1 that is simple, reproducible, and overcomes the drawbacks of other assays in terms of dynamic range, time, and specificity. Because this assay can detect all three forms of PAI-1 present in plasma with equal efficiency, it can estimate the exact amounts of PAI-1 that are released from endothelial cells by injury or stimulation and from adipose tissue as a result of hypertriglyceridemia or an insulin-dependent mechanism. This assay could therefore be a useful tool for investigating the pathophysiologic role of PAI-1.
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  H. Kimura, X. Li, K. Torii, T. Okada, N. Takahashi, H. Fujii, S. Ishihara, and H. Yoshida A natural PPAR-{gamma} agonist, 15-deoxy-delta 12,14-prostaglandin J2, may act as an enhancer of PAI-1 in human proximal renal tubular cells under hypoxic and inflammatory conditions Nephrol. Dial. Transplant., August 1, 2008; 23(8): 2496 - 2503. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Miyanishi, Y. Kamo, H. Ihara, T. Naka, M. Hirakawa, and Y. Sugioka Risk factors for dysbaric osteonecrosis Rheumatology, July 1, 2006; 45(7): 855 - 858. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Ono, J. Mimuro, S. Madoiwa, K. Soejima, Y. Kashiwakura, A. Ishiwata, K. Takano, T. Ohmori, and Y. Sakata Severe secondary deficiency of von Willebrand factor-cleaving protease (ADAMTS13) in patients with sepsis-induced disseminated intravascular coagulation: its correlation with development of renal failure Blood, January 15, 2006; 107(2): 528 - 534. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, formation of tPA/PAI-1 complex by tPAI · C test; 
, PAI activity by Spectrolyse/fibrin method; 
, total PAI-1 antigen by TintElize assay. Purified PAI-1 was activated by 4 mol/L guanidine for 30 min at 37 °C and added to tPA/PAI-1-depleted plasma to obtain active PAI-1. We added 0.2–15.2 µmol/L single-chain tPA was added to plasmas with active PAI-1 and incubated the samples for 30 min at 37 °C.