Donate Help Contact The AHA Sign In Home
American Heart Association
Circulation
Search: search_blue_button Advanced Search
« Previous Article | Table of Contents | Next Article »
Circulation. 1995;92:2984-2994

This Article
Right arrow Abstract Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Chandler, W. L.
Right arrow Articles by Stratton, J. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Chandler, W. L.
Right arrow Articles by Stratton, J. R.

(Circulation. 1995;92:2984-2994.)
© 1995 American Heart Association, Inc.

Articles

The Circulatory Regulation of TPA and UPA Secretion, Clearance, and Inhibition During Exercise and During the Infusion of Isoproterenol and Phenylephrine

Wayne L. Chandler, MD; Wayne C. Levy, MD; John R. Stratton, MD

From the Department of Laboratory Medicine (W.L.C.) and the Division of Cardiology, Department of Medicine (W.C.L., J.R.S.), University of Washington and Seattle VA Medical Center.

Correspondence to Wayne L. Chandler, MD, Department of Laboratory Medicine, SB-10, University of Washington, Seattle, WA 98195.


*    Abstract
up arrowTop
 *Abstract
down arrowIntroduction
 down arrowMethods
down arrowResults
down arrowDiscussion
down arrowReferences
 
Background Exercise to exhaustion and infusions of isoproterenol and phenylephrine were used to study interactions between plasminogen activator regulation and the control of regional blood flow in 10 healthy males.

Methods and Results Experimental measurements of cardiac output, heart rate, tissue plasminogen activator (TPA), urokinase plasminogen activator (UPA), plasminogen activator inhibitor (PAI-1), C1-inhibitor, and TPA/C1-inhibitor complex during the infusions and exercise were used to develop a comprehensive fluid-phase model of the circulatory regulation of fibrinolysis. {alpha}- and ß-adrenergic agonists increased TPA and UPA in plasma by different mechanisms: Phenylephrine decreased hepatic blood flow and thus clearance while isoproterenol stimulated increased secretion of TPA and UPA. Exercise to exhaustion increased TPA and UPA through a combination of increased secretion and decreased clearance. The time course of UPA and TPA release were similar, but the magnitude of their secretion responses differed. In vivo, C1-inhibitor bound to TPA at a rate of 553 mol-1 · s-1. C1-inhibitor contributed equally with PAI-1 to TPA inhibition when active PAI-1 levels were low (20 to 50 pmol/L) but was less important when active PAI-1 levels were high.

Conclusions We conclude that secretion, inhibition, clearance, and regional blood flow effects must all be taken into account when evaluating changes in plasminogen activator levels.


Key Words: plasminogen activators • exercise


*    Introduction
up arrowTop
 up arrowAbstract
 *Introduction
down arrowMethods
down arrowResults
down arrowDiscussion
down arrowReferences
 
The fibrinolytic system is important in regulating the formation and removal of thrombi and in maintaining vascular patency. Reduced fibrinolytic activity is associated with an increased risk of unstable angina, myocardial infarction, stroke, and mortality in heart disease patients, presumably due to a reduced ability to prevent thrombosis.1 2 3 4 5 6 7 8 It has been suggested that while long-term modification of the fibrinolytic system may be beneficial in reducing the risk of thrombotic cardiovascular disease, a better understanding of the systemic regulation of fibrinolysis is needed to predict which changes are likely to be most beneficial.9

Fibrinolysis is closely linked to the overall regulation of the cardiovascular system. Factors that increase fibrinolytic activity including adrenergic agonists, vasopressin, bradykinin and histamine also have more general cardiovascular effects, producing changes in cardiac output, regional blood flow, vascular tone, and permeability.10 11 Older studies using nonspecific global fibrinolytic assays found that both {alpha}- and ß-adrenergic agonists increased fibrinolytic activity in plasma, but the mechanism for each agent and the plasminogen activators involved were unknown.12 13 14 15 Fibrinolysis is initiated by TPA and UPA.16 17 18 UPA circulates as scu-PA. Only limited data are available on the source, secretion, and clearance of scu-PA,19 20 21 yet chronically increasing the level of scu-PA in plasma has been advocated as a possible treatment for thrombotic heart disease.22 Active TPA in plasma is primarily inhibited by PAI-1 and C1-inhibitor, yet the in vivo kinetics and relative role of the two inhibitors are unclear.23 24 25 While it is well known that exercise increases both TPA and scu-PA levels in plasma, the mechanisms behind these increases are still controversial. Both decreased plasminogen activator clearance and increased secretion have been proposed.19 26 27 28

The level of plasminogen activators in plasma can be affected by changes in secretion, clearance, inhibition, and regional blood flow.19 27 29 30 31 All of these factors operate simultaneously to regulate fibrinolytic activity in blood. The purpose of this study was to develop a comprehensive model of the fluid-phase circulatory regulation of the major fibrinolytic activators and inhibitors. This model incorporates experimental measurements of cardiac output, heart rate, and the plasma levels of active TPA, active PAI-1, total TPA, activatible scu-PA, total C1-inhibitor, TPA/PAI-1 complex, and TPA/C1-inhibitor complex during {alpha}- and ß-adrenergic agonist infusions and graded exercise to exhaustion; experimental conditions known to both increase fibrinolytic activity and alter cardiovascular dynamics. The experimental data were used to develop a computer simulation of the circulatory regulation of fibrinolysis resulting in an integrated model of how secretion, inhibition, clearance, and regional blood flow interact to regulate TPA and scu-PA levels in plasma. This is the first study to (1) model and evaluate the systemic regulation of scu-PA secretion and clearance, (2) determine the specificity and mechanisms of adrenergic agonist–induced increases in fibrinolytic activity, (3) quantitate the role of C1-inhibitor in relation to PAI-1 as an inhibitor of active TPA, and (4) incorporate all of these findings into a model of fibrinolytic regulation during exercise.


*    Methods
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 *Methods
down arrowResults
down arrowDiscussion
down arrowReferences
 
Human Subjects
The study group consisted of 10 healthy men aged 22 to 76 years (mean, 38±21). Informed consent was obtained from all participants, and the study was approved by the University of Washington Human Subjects Review Committee.

Materials
Human glu-plasminogen, single-chain UPA, two-chain high-molecular-weight UPA (80 000 IU/mg), one-chain melanoma-derived TPA (512 000 IU/mg), monoclonal anti-human urokinase (No. 394), and polyclonal anti-human TPA (No. 387) were obtained from American Diagnostica Inc. Human fibrinogen, cyanogen bromide (CNBr), and Triton X-100 were obtained from Sigma Chemical Co. CNBr-cleaved human fibrinogen fragments were prepared as previously described.32 Chromogenic substrate D-valyl-phenyl-alanyl-lysyl-p-nitro-analide (S-2390) was obtained from Pharmacia Hepar Inc. Enzyme-linked immunosorbent assay (ELISA) kits for the measurement of TPA/PAI-1 complex and total TPA antigen were obtained from American Bioproducts. Peroxidase-conjugated sheep anti-human C1-inhibitor F(ab')2 fragments and C1-inhibitor nephelometric kits were obtained from The Binding Site. All other materials not described below were reagent or analytical grade.

Exercise and Infusion Protocol
Samples were taken in the supine position after (1) 35 minutes of rest, (2) infusion of 7 and 35 ng/kg per minute of isoproterenol for 18 minutes each, (3) 60 minutes of rest, (4) infusion of 0.5, 1.0, and 1.5 mg/kg per minute of phenylephrine for 18 minutes each, (5) 60 minutes of rest, and (6) after each 3-minute supine bicycle exercise stage beginning at 33-W output and increasing by 33-W steps until exhaustion.

Hemodynamic Measurements
Cardiac outputs were determined with the use of radionuclide angiography.33 Heart rates and cardiac outputs were measured during the final 2 minutes of each infusion or bicycle exercise stage.

Blood Sampling and Sample Preparation
Blood was obtained from a forearm vein. Blood samples were anticoagulated by the addition of 4.5 mL whole blood to 0.5 mL of 130 mmol/L sodium citrate. To stabilize TPA activity, 0.5 mL of citrate anticoagulated whole blood was mixed with 0.25 mL of 0.5 mol/L sodium acetate, pH 4.2, within 1 minute after the sample was drawn.32 All samples were centrifuged for 10 minutes at 2500g at room temperature. Citrate and acidified plasma then were removed and frozen at -80°C until analyzed.

TPA and Activatible scu-PA Activity
TPA and activatable scu-PA activity were measured in acidified plasma using an amidolytic method as previously described.32 34 scu-PA was converted to active two-chain UPA by plasmin formed in the assay. TPA and UPA activity were separated using neutralizing concentrations (25 µg/mL) of anti-UPA and anti-TPA added to the plasminogen-substrate reagent. TPA and UPA activity results were converted into molar concentrations of active TPA and activatible scu-PA using specific molar activities of 4.48x1013 IU/mol (TPA units) and 6.4x1012 IU/mol (UPA units), respectively.35 36

TPA Antigen Assays
Total TPA antigen was determined in citrated plasma using an ELISA, as previously described.37 TPA/C1-inhibitor complex antigen was measured in citrated plasma by a modification of the ELISA method of Huisman et al25 using microtiter plates coated with monoclonal anti-human TPA F(ab')2 fragments and sheep polyclonal anti-human C1-inhibitor F(ab')2 fragments conjugated to peroxidase.

Active PAI-1
Active PAI-1 was measured at the end of each rest phase in citrated plasma with the use of an immunofunctional method.38 Active PAI-1 levels were determined by adding active one-chain TPA (final concentration, 50 IU/mL) to an aliquot of plasma followed by incubation for 20 minutes at 37°C. This converted active PAI-1 in the sample into TPA/PAI-1 complex. The concentration of TPA/PAI-1 complex was measured in the plasma before and after the addition of excess TPA.39 The concentration of active PAI-1 in the citrate plasma was equal to the difference in TPA/PAI-1 complex before versus after TPA addition.

As active TPA and active PAI-1 continued to react after the blood was drawn into citrate anticoagulant, the measured in vitro active PAI-1 level underestimated the in vivo active PAI-1 level by an amount approximately equal to the original TPA activity. To estimate the original active PAI-1 level in vivo, we added the measured active PAI-1 level in citrated plasma to the active TPA level measured in acidified plasma as previously described.35

Other Assays
C1-inhibitor antigen was measured in serum using an immunonephelometric method. Spun hematocrits were determined at each sample point. All protein measurements were corrected for hemoconcentration based on changes in hematocrit.40

Model Circulatory System
The effects of secretion, clearance, and inhibition reactions on the concentration of fibrinolytic factors were modeled throughout a simplified human circulatory system (Fig 1Down). The model circulation consisted of a cardiopulmonary segment that split into forearm, splanchnic, and systemic segments. The systemic segment consisted of all the noncardiopulmonary, nonforearm, nonsplanchnic circulatory beds, that is, cerebral, renal, lower extremity, and so forth.



View larger version (19K):
[in this window]
[in a new window]
  		Figure 1. Schematic of human circulatory model. Open boxes indicate large arteries or veins. Shaded boxes indicate capillary beds or hepatic sinusoids in which TPA, UPA, and PAI-1 secretion occurs. The area of each box is proportional to the plasma volume at rest in that segment. TPF indicates total plasma flow; FPF, forearm plasma flow; SPF, splanchnic plasma flow; and SysPF, systemic plasma flow. The systemic circulation in the model consists of all the noncardiopulmonary, nonforearm, nonsplanchnic beds, that is, cerebral, renal, lower extremity, and so forth.

The fibrinolytic system of the model contained active TPA, active PAI-1, active C1-inhibitor, TPA/PAI-1 complex, TPA/C1-inhibitor complex, and scu-PA. The concentrations of these proteins were regulated by three processes: (1) secretion of scu-PA, TPA, and PAI-1, (2) inhibition of TPA by PAI-1 and C1-inhibitor forming TPA/PAI-1 or t-PA/C1-I complex, and (3) hepatic clearance of all factors. Under physiological conditions, scu-PA is not inhibited by PAI-1.41 Total TPA was assumed to be equal to active TPA plus TPA/PAI-1 complex plus TPA/C1-I complex. C1-inhibitor levels did not change significantly during the experiment and were held constant at the resting level for each subject.

Blood Volumes
Resting total blood volume was estimated for each subject based on their height and weight.42 Blood volumes for different segments of the vascular system were calculated as a percentage of the total blood volume using the values in Table 1Down.43 44 Because fibrinolytic factors are in the plasma fraction and not in blood cells, the model simulated secretion, inhibition, and clearance of fibrinolytic factors in the plasma only. Resting plasma volume was estimated from total blood volume and resting hematocrit.42


View this table:
[in this window]
[in a new window]
  		Table 1. Resting Plasma Volume Assumptions in the Circulatory Model

Blood Flow Rates
Cardiac outputs and heart rates in the simulation were set to the measured level for each subject at each stage of the experiment. At rest, the splanchnic circulation in the simulation received 25% of total resting blood flow, both forearms together received 3.6%, and the systemic circulation (all remaining areas) received the remaining 71.4%.44 45 During isoproterenol infusions, forearm and splanchnic blood flows in the model increased in proportion to cardiac output.11 46 47 48 During phenylephrine infusions forearm blood flow decreased in proportion to cardiac output11 while splanchnic blood flow decreased proportional to the phenylephrine infusion rate:


(1)

where SBF is splanchnic blood flow and PE is the phenylephrine infusion rate in µg/kg per minute.47 49 During exercise, splanchnic blood flow in the model remained unchanged at the resting value up to a heart rate of 100 beats per minute, after which splanchnic flow decreased linearly with heart rate:


(2)

where HR is heart rate.27 29 45 Forearm blood flow in the model did not change for the first 6 minutes of exercise, doubled from 6 to 12 minutes, and tripled from 12 to 18 minutes, remaining constant at triple baseline levels from 18 minutes on.50

Secretion
TPA is primarily secreted from vascular endothelial cells.51 The majority of the endothelium are in the small vessels and capillaries.43 The rate of TPA secretion varies with different vascular beds. Keber et al44 have shown that the forearm vascular bed secretes TPA approximately 60% faster than the average rate for the entire body. TPA secretion in the simulation was simplified by assuming (1) TPA was secreted only by vascular endothelial cells, (2) essentially all of the vascular endothelium was located in four capillary beds: the pulmonary, splanchnic, forearm, and systemic capillary beds, and (3) TPA secretion was the same in all capillary beds except the forearm bed, which secreted TPA 60% faster than the average rate.

The source of scu-PA in blood is unknown. While endothelial cells secrete scu-PA in culture,52 53 studies in humans show no evidence of scu-PA secretion during venous occlusion.54 55 scu-PA secretion in the simulation was simplified by assuming (1) scu-PA was secreted in the capillary beds, but from an unknown source, and (2) all capillary beds secrete scu-PA at the same rate.

Brommer et al56 have shown by direct catheterization of the hepatic artery and hepatic vein that the liver is a net producer of active PAI-1. This is supported by other studies, which show that in healthy subjects, PAI-1 is secreted primarily by the liver and not endothelium and that human hepatocytes express the gene for PAI-1 in vivo.35 40 44 54 55 57 58 59 60 61 62 63 During the acute-phase response, PAI-1 may be secreted by both liver and endothelium.60 64 65 66 Since all subjects in this study were healthy, we started with an initial assumption that all PAI-1 was secreted by the liver. As a final study, we compared the effect of liver versus endothelial PAI-1 secretion on the predicted levels of TPA and PAI-1 in the model.

To minimize the effect of circadian variations, the infusion and exercise protocols were started in the late morning and continued into the early afternoon. A baseline sample was obtained just before starting each segment of the study, and the length of each segment was limited to 1 hour or less: {alpha}-agonist infusion average, 36 minutes; ß-agonist infusion average, 54 minutes; exercise average, 18 minutes.31 67 68 For the purposes of the simulation, it was assumed that PAI-1 secretion remained constant at the resting level obtained just before each experimental segment.

Inhibition
A second-order rate constant of 3.7x107 mol-1 · s-1 was used for the reaction between one-chain TPA and PAI-1.69 70 Initially, a second-order rate constant of 3.3 mol-1 · s-1 was used for the reaction between one-chain TPA and C1-inhibitor.71

Clearance
de Boer et al29 have shown that the rate of TPA clearance is directly proportional to liver blood flow. Plasma clearance is equal to


(3)

where E is the hepatic extraction fraction and Q is the hepatic plasma flow rate. Brommer et al56 measured the level of active and total TPA in the hepatic vein and hepatic artery, directly determining the hepatic extraction fractions. They found that active TPA was cleared at a faster rate than total TPA. As the majority of TPA in resting samples is in the form of TPA/PAI-1 or TPA/C1-inhibitor complexes, Brommer et al suggested that complexed TPA must have a hepatic extraction fraction similar to that for total TPA (39%).23 24 35 39 Huisman and coworkers25 reported similar half-lives for total TPA antigen and TPA/C1-inhibitor complex during TPA infusion.

At rest, the hepatic extraction fraction for active TPA is more difficult to determine because of the low level of circulating active TPA. Brommer et al used DDAVP to increase the plasma levels of active TPA; under these conditions they measured an average hepatic extraction fraction of 71% for active TPA. scu-PA is also rapidly cleared, with a half-life of 4 to 5 minutes, equivalent to a hepatic extraction fraction of approximately 55% in our model.19 20 21 Because liver is a net producer of active PAI-1, the hepatic extraction fraction must be low. Most active PAI-1 is cleared through reactions with active TPA.31 Unless otherwise noted, hepatic extraction fractions of 71% for active TPA, 39% for TPA/PAI-1 and TPA/C1-inhibitor complex, 55% for scu-PA, and 0% for active PAI-1 were used in the simulation.20 21 23 24 29 31 72 While these extraction fractions were the most likely based on current studies, additional simulations were run with active TPA clearance equal to and less than complexed TPA clearance to determine which clearance model fit the measured data best.

Reaction Kinetics
In large arteries and veins, the ratio of endothelial surface area to blood volume is low. It was assumed that no secretion of scu-PA, TPA, or PAI-1 occurred in large vessels. Because scu-PA does not react with PAI-1, the concentration of scu-PA in the model did not change in large vessels.41 The only changes in concentration were due to the reaction between TPA and either PAI-1 or C1-I:


(4)

where k1 and k2 are the second-order rate constants for the reactions.

In the capillary beds of the lung, splanchnic, forearm, and systemic circulation, changes in concentration occur as the result of both TPA and scu-PA secretion and the inhibition reaction between TPA and either PAI-1 or C1-I. Using TPA as an example again:


(5)

where STPA is the TPA secretion rate in pmol/L per second. In this article, scu-PA, TPA, and PAI-1 secretion rates are given as the rate averaged over the entire plasma volume. In the simulation, local secretion rates were used.

In the liver sinusoids, changes in concentration occur as the result of the secretion of active PAI-1, the reaction between TPA and either PAI-1 or C1-I, and the clearance of the appropriate fraction of each factor as the plasma flows into the hepatic veins.

Computer Simulation
Concentrations of fibrinolytic proteins in the forearm veins of the simulation were compared with measured levels of these factors in samples from the forearm of the subject. The simulation predicted splanchnic, forearm, and systemic blood flows based on prior studies of adrenergic agonist effects and exercise as described above. The computer model iteratively adjusted secretion rates for scu-PA, TPA, and PAI-1 to minimize the weighted squared error between simulated and measured levels of plasma scu-PA, active TPA, active PAI-1, and total TPA at each sample point in the experiment. The error for each factor was weighted using the between-run precision for that assay. Because scu-PA levels were a function of secretion and clearance only, they could be fit exactly. For active TPA, active PAI-1, and total TPA, the summed error for these three measurements was minimized. When a best-fit was achieved, the output from the model was an estimate of the secretion rates for scu-PA, TPA, and PAI-1 at each sample point in the experiment. The total sum of weighted errors for all active TPA, active PAI-1, and total TPA values for all steps in the simulation was also determined for each subject as an overall measure of how well the simulation fit the measured values.

TPA/C1-inhibitor levels were fit separately. The simulation was first run using the in vitro TPA versus C1-inhibitor second-order rate constant from the study by Ranby et al (3.3 mol-1 · s-1).71 Next, the TPA/C1-inhibitor rate constant was adjusted to produce the best fit between measured and simulated TPA/C1-inhibitor levels.

The simulation was run on a Macintosh Quadra 700 with SYMANTEC THINK PASCAL 4.0. A disk containing the program, a copy of the source code, and instructions for running the program are available from the authors at no charge. While the program is designed to run on the system described above, it can be adapted by a knowledgeable programmer to work on any system running standard PASCAL.


*    Results
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
*Results
down arrowDiscussion
down arrowReferences
 
Resting Values
At baseline, average total TPA levels were eightfold to ninefold higher than average scu-PA levels (Table 2Down). The simulation predicted that resting TPA secretion was on average 3 to 4 times higher than scu-PA secretion. But, because of the inhibitory action of PAI-1 and C1-inhibitor on TPA but not scu-PA, active TPA levels in plasma were similar to scu-PA levels. On average, 15% of the TPA circulated in the active form at rest.


View this table:
[in this window]
[in a new window]
  		Table 2. Average Measured and Predicted Variables During Infusions and Exercise

During the 60-minute rest periods after each adrenergic infusion, most cardiovascular and fibrinolytic parameters returned to the baseline status measured before the first infusion. Cardiac outputs, scu-PA levels, active PAI-1 levels, TPA/C1-I levels, and hematocrits were not significantly different than baseline levels at the end of each rest period. The kinetic model predicted that hepatic blood flow, forearm blood flow, TPA secretion, PAI-1 secretion, and scu-PA secretion also returned to baseline levels between experiments. Compared with baseline, heart rates were on average slightly higher during the rest periods after each infusion than at baseline. Active TPA levels were slightly higher, and total TPA levels slightly lower before exercise than at baseline.

Fibrinolytic Response to ß-Adrenergic Agonists
Infusion of 7 and 35 ng/kg per minute of isoproterenol resulted in significant increases in cardiac output, heart rate, active TPA, total TPA, and scu-PA (Table 2Up and Fig 2Down). Compared with baseline, the maximum infusion rate of isoproterenol increased the average cardiac output by 93% and the average heart rate by 73%. The percentage of TPA in the active form increased from an average of 15% active at baseline to 40% active at the maximum isoproterenol infusion rate. In addition to increasing plasma TPA levels, isoproterenol also increased the average plasma scu-PA level by 44%. The kinetic model predicted an increase in hepatic blood flow proportional to the increased cardiac output. Increased hepatic blood flow in turn cleared TPA and scu-PA faster. As TPA and scu-PA clearances were predicted to be increased by isoproterenol, the kinetic model indicated that the rise in TPA and scu-PA was due to increased secretion of these proteins. The maximum dose of isoproterenol was predicted to increase average TPA secretion threefold to fourfold while increasing average scu-PA secretion twofold to threefold.



View larger version (29K):
[in this window]
[in a new window]
  		Figure 2. Graphs show measured concentrations of active TPA and scu-PA (middle row), predicted hepatic blood flow (upper row) and predicted TPA and scu-PA secretion rates (lower row) during isoproterenol infusion (left column), phenylephrine infusion (middle column), and graded exercise to exhaustion (right column). Using measured cardiac outputs, heart rates and active TPA, active PAI-1, total TPA, and scu-PA concentrations, the simulation was used to predict hepatic and forearm blood flow and TPA, PAI-1, and scu-PA secretion rates for each subject. During the exercise experiment, individuals reached the point of exhaustion at different times. Because most of the fibrinolytic changes occurred in the last 6 minutes before exhaustion regardless of the total time of exercise, the data were normalized so that an exercise level of 200 W was assumed to be the highest exercise level for each subject. Error bars indicate mean±SEM (n=10). Note that the scale on the y-axis varies with each figure and is not the same across rows.

Fibrinolytic Response to {alpha}-Adrenergic Agonists
Infusion of 0.5 to 1.5 µg/kg per minute of phenylephrine resulted in significant decreases in cardiac output and heart rate while increasing plasma TPA and scu-PA levels. Compared with baseline, the maximum phenylephrine dose decreased average cardiac output 29% and heart rate 28%. Compared with baseline, the maximum infusion rate of phenylephrine increased average active TPA by 8 pmol/L but produced no significant change in average total TPA levels. On average, 19% of TPA was active before starting phenylephrine versus 32% at the maximum infusion rate. Phenylephrine infusion also increased scu-PA levels by 4 pmol/L. The kinetic model predicted a 32% reduction in hepatic blood flow and thus a similar decrease in TPA and scu-PA clearance. The increase in both TPA and scu-PA during phenylephrine infusion could be accounted for by reduced clearance of these factors alone. There was no predicted change in average TPA or scu-PA secretion.

Fibrinolytic Response to Graded Exercise
On average, the subjects exercised for 18 minutes, reaching a maximum exercise level of 200 W (range, 100 to 300 W). Graded exercise resulted in increased heart rate, cardiac output, and plasma TPA and scu-PA. Compared with baseline, graded exercise produced a rapid increase in cardiac output and heart rate that gradually leveled off at the point of exhaustion. In contrast to the rapid changes in cardiac output and heart rate, graded exercise had little effect on fibrinolytic parameters until the last 6 minutes before exhaustion when an exponential increase in TPA and scu-PA occurred. Total TPA levels increased twofold, active TPA levels increased fivefold to sixfold, while scu-PA levels increased threefold. The percentage of TPA in the active form increased from an average 24% before starting exercise to 58% at the point of exhaustion. The kinetic model predicted that graded exercise produced a progressive decline in hepatic blood flow and thus, TPA and scu-PA clearance. Average TPA secretion increased fourfold to fivefold, and scu-PA secretion increased sixfold to sevenfold at the point of exhaustion.

Effect of TPA Inhibitors
At baseline, active PAI-1 levels ranged from a low of 16 pmol/L to a high of 221 pmol/L. Active TPA levels were strongly influenced by the concentration of active PAI-1 (Fig 3Down). There was an inverse relation between the percent of TPA in the active form and the concentration of active PAI-1. In the subject with the highest active PAI-1 (221 pmol/L), less than 1% of his TPA was active at baseline, rising to a peak of 19% active at maximum exercise. Another subject with low active PAI-1 (19 pmol/L) had 36% active TPA at baseline, increasing to nearly 100% active TPA at maximum exercise.



View larger version (12K):
[in this window]
[in a new window]
  		Figure 3. Graph shows concentration of active PAI-1 at rest versus the percentage of TPA in the active form after the maximum isoproterenol infusion (I), the maximum phenylephrine infusion (P), and at the maximum exercise point (E).

Compared with PAI-1, C1-inhibitor produced a lower intensity but more constant inhibition of TPA. C1-inhibitor levels showed less variation among subjects (mean, 3.1±0.4 µmol/L) and essentially no change other than hemoconcentration during the experiments. At baseline, the average TPA/C1-inhibitor level was 13 pmol/L. On average, 16% of total TPA was inhibited by C1-inhibitor (Table 2Up). The only significant increase in TPA/C1-inhibitor levels occurred at the end of exercise when active TPA levels were highest.

Table 3Down shows the results of the two protocols used to simulate the different reaction rates between TPA and C1-inhibitor. In the first protocol a second-order rate constant of 3.3 mol-1 · s-1 was used for the reaction between TPA and C1-inhibitor. This rate constant was based on the in vitro reaction rate study of Ranby et al.71 Using this rate constant, the kinetic model predicted an average baseline TPA/C1-inhibitor level of 0.08 pmol/L, approximately 100-fold less than the measured level. This indicated that the in vivo rate constant was faster than the measured in vitro rate. In the second protocol, the TPA/C1-inhibitor rate constant was adjusted for each subject to produce the best fit between measured and simulated levels of TPA/C1-inhibitor. The kinetic model predicted an average rate constant of 553±100 (SEM) mol-1 · s-1 (range, 205 to 1139 mol-1 · s-1).


View this table:
[in this window]
[in a new window]
  		Table 3. Measured Versus Simulated Levels of TPA/C1-Inhibitor Complex

The use of higher rate constants for the TPA/C1-inhibitor reaction also significantly improved the fit for active TPA, active PAI-1, and total TPA. The average total error for these three factors for the entire simulation was 316±86 (SEM), using a TPA/C1-inhibitor rate constant of 3.3 mol-1 · s-1, which fell to an average summed error of 188±60 when the rate constant was optimized for each subject (paired t test, P=.006). Thus, higher TPA/C1-inhibitor rate constants improved the overall fit of the kinetic model in addition to improving the fit between measured and simulated TPA/C1-inhibitor levels.

Blood Volume, Clearance Fraction, and Regional Secretion Effects
While cardiac output in the model was measured for each subject at each data point, total and regional blood volumes and regional blood flows were estimated (Table 1Up).42 Fig 4Down shows the effect on predicted TPA secretion rates of increasing and decreasing total blood volumes in the model by 5% and 10%. Increasing blood volume while holding cardiac output constant at the measured value resulted in a proportional decrease in the predicted TPA secretion; decreasing blood volume had the opposite effect. Changes in blood volume had no effect on the relative change in TPA secretion due to isoproterenol infusion (or other experimental conditions). Thus, while changing the absolute secretion rate predictions, variations in blood volume did not effect the major predictions of the model.



View larger version (16K):
[in this window]
[in a new window]
  		Figure 4. Graph shows effect of estimated blood volume changes on the predicted rate of TPA secretion at rest and after infusion of 35 ng/kg per minute of isoproterenol (Iso). Blood volume for each of the 10 subjects was adjusted up or down while holding all other factors constant; the simulation then was repeated. Best-fit predicted TPA secretion rate (mean±SEM) is shown. Blood volume is expressed as a percentage of the volume estimated based on the height and weight of the subject (100%).42

Hepatic extraction fractions have been measured for total TPA and TPA activity but not for TPA/PAI-1 complex and TPA/C1-inhibitor complex. To determine the effect of different TPA extraction fractions on the model, three different simulations were run: The first, used in the simulations described above, used extraction fractions of 71% for active TPA and 39% for complexed TPA based on the measured hepatic extraction of TPA activity and total TPA antigen by Brommer et al.56 Second, both extraction fractions were set equal to 55%, and third the initial values were reversed: 39% for active TPA and 71% for complexed TPA. The best overall fit between measured and predicted levels of TPA and PAI-1 occurred when complexed TPA was cleared slower than active TPA, resulting in an average total simulation error of 188±60 (SEM). Equal clearance rates for active and complexed TPA produced a higher error of 277±92, with the worst fit occurring when complexed TPA was cleared faster than active TPA (error, 354±124). In summary, the model predicted that higher active PAI-1 levels result in more TPA complexed to PAI-1 in turn resulting in slower clearance of total TPA.

Both liver and endothelium have been proposed as sites for PAI-1 synthesis. Table 4Down shows the effect of 100% liver versus 100% endothelial PAI-1 secretion on predicted regional TPA and PAI-1 levels in the model vascular system. The site of PAI-1 secretion had little effect on predicted total TPA levels in the model. With liver secretion of PAI-1, the model predicted higher active PAI-1 levels in the hepatic vein versus the hepatic artery and a slightly negative arterial-venous gradient in the forearm due to the reaction between active TPA and active PAI-1. With endothelial PAI-1 secretion, a different pattern was seen, little or no change in active PAI-1 across the liver, with a slight increase in active PAI-1 in the forearm veins versus arteries.


View this table:
[in this window]
[in a new window]
  		Table 4. Effect of Liver Versus Endothelial Secretion of PAI-1 on TPA and PAI-1 Levels


*    Discussion
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
up arrowResults
*Discussion
down arrowReferences
 
Use of both experimental data and a comprehensive computer model of the circulatory regulation of fibrinolysis allowed us to evaluate how changes in regional blood flow and plasminogen activator secretion, clearance, and inhibition interact to regulate fibrinolytic activity in plasma.

Adrenergic Effects on the Fibrinolytic System
Prior studies reported that both {alpha}- and ß-adrenergic agonists increased fibrinolytic activity in plasma, but the mechanisms for each agonist and the specific plasminogen activators involved were unknown.12 13 14 15 Our results indicate that ß-agonists increase fibrinolytic activity by stimulating secretion of both TPA and scu-PA. {alpha}-Agonists, on the other hand, increase TPA and scu-PA in plasma by decreasing hepatic blood flow and thus clearance of these proteins.11 29 47 The ability to model both cardiovascular as well as fibrinolytic changes allowed us to separate these two mechanisms.

At baseline, TPA secretion was threefold to fourfold higher than scu-PA secretion (70 versus 20 fmol/L per second). Increased secretion of both TPA and scu-PA during isoproterenol infusion were dose dependent, rapid in onset, and sustained for at least 15 to 20 minutes. The time course of TPA and scu-PA secretion in response to isoproterenol infusion and exercise were similar, but the magnitude of their secretion responses differed: TPA secretion increased fourfold to fivefold with both, while scu-PA secretion increased sevenfold with exercise but only threefold with isoproterenol infusion. Prior studies indicated that TPA secretion during exercise was linearly related to the plasma epinephrine concentration.27 28 scu-PA secretion may involve different mechanisms or may show different sensitivities to epinephrine versus isoproterenol.

Regional Differences in Secretion
Recent studies suggest that major differences in regional TPA and PAI-1 secretion may exist. Keber et al59 reported reduced TPA secretion from the leg versus the forearm. In contrast, Gough et al73 found higher levels of plasminogen activator activity in the inferior vena cava versus forearm veins, suggesting higher TPA secretion in the abdomen versus the forearm. Our model predicted TPA gradients across the forearm and average TPA secretion rates similar to those reported at rest and after stimulation of TPA release.44 59 62 63 The only regional variation in secretion we included was higher than average TPA secretion in the forearm, based on the work of Keber et al, as that was the site of blood sampling in our study. Because all the subjects in this study were healthy, we initially assumed PAI-1 was secreted by the liver.35 40 44 54 55 57 58 59 60 61 62 63 Endothelial secretion of PAI-1 was also possible.60 64 65 66 The liver PAI-1 secretion model agreed well with the finding by Brommer et al of higher PAI-1 levels in the hepatic vein versus the hepatic artery and with other reports showing no arterial-venous gradient in total PAI-1 across the forearm and no increase in total PAI-1 during venous occlusion.44 56 59 62 63 The liver secretion model was also supported by Gough et al,73 who reported higher TPA antigen and total plasminogen activator activity, lower PAI activity, but no change in PAI-1 antigen in the inferior vena cava versus the abdominal aorta, suggesting TPA secretion from lower body capillary beds, leading to consumption of PAI activity but no secretion of PAI-1. Additional studies are needed on the regional release of TPA and PAI-1 both in healthy subjects and during the acute phase response.

Inhibition by PAI-1 and C1-Inhibitor
The second process that regulates plasminogen activator levels is inhibition by PAI-1, C1-inhibitor, and to a lesser extent, other serine protease inhibitors. Booth et al23 and Bennett et al24 have reported finding increases in plasma free TPA, TPA/PAI-1 complex, TPA/C1-inhibitor complex, and free UPA after venous occlusion, exercise, electrical shock, and obstetric complications but no UPA inhibitor complexes. Because scu-PA is not inhibited by PAI-1 (or C1-inhibitor), changes in the plasma levels of scu-PA are a function of secretion and clearance only. For TPA, the importance of PAI-1 versus C1-inhibitor was dependent on the concentration of active PAI-1. When active PAI-1 levels were low, in the range of 20 to 50 pmol/L, on average 22% of TPA circulated in an active form, with the remainder inhibited equally by PAI-1 and C1-inhibitor. When active PAI-1 levels were higher (100 to 200 pmol/L), PAI-1 became the principal inhibitor of TPA. As little as 5% of the TPA circulated in an active form, and only 5% of the TPA was inhibited by C1-inhibitor; the remaining 90% of TPA was inhibited by PAI-1. The percent TPA in the active form was inversely proportional to the active PAI-1 concentration in blood (Fig 3Up). This applied both at rest and at the point of maximum TPA secretion. In subjects with low active PAI-1 levels, the majority of their TPA was in the active form at the maximum isoproterenol infusion and maximum exercise points. In one subject with an active PAI-1 level of 221 pmol/L, only 19% of TPA remained active at the maximum TPA secretion point. It is important to note that there is a wide variation in the level of active PAI-1 among different subjects. Healthy individuals may have as little as 20 to 50 pmol/L active PAI-1 in plasma versus 500 to 1500 pmol/L active PAI-1 in some patients with persistently elevated levels and over 1500 pmol/L transiently during an acute phase response.74 The level of PAI-1 in plasma is important in determining the final active TPA level in plasma in response to various secretion stimuli.

The rate of TPA inhibition by C1-inhibitor was predicted to be slow, 3 to 5 mol-1 · s-1, based on in vitro kinetic measurements.25 71 Kinetic modeling using the measured levels of TPA, C1-inhibitor, and TPA/C1-inhibitor complex predicted a second-order rate constant in vivo of 553 mol-1 · s-1, almost 200 times faster than measured in vitro. This suggests that the mechanism of TPA inhibition by C1-inhibitor in vivo may be more complex than simple fluid phase binding, possibly involving catalysis by cell surfaces or receptors. Compared with PAI-1, C1-inhibitor was present in much higher concentration in plasma (about 3 µmol/L) and showed less interindividual variation but reacted much slower with TPA (4x107 versus 5x102 mol-1 · s-1) and was of greatest importance when PAI-1 activity was low.

Hepatic Clearance of TPA and scu-PA
The third process regulating plasminogen activator levels in plasma was hepatic clearance, which was as important as secretion or inhibition in regulating TPA and scu-PA levels. The rise in TPA and scu-PA during {alpha}-agonist infusion was entirely accounted for by decreased clearance due to reduced hepatic blood flow. Clearance was also changed during ß-agonist infusion but in the opposite direction. Isoproterenol infusion increased cardiac output, heart rate, hepatic blood flow, and thus, clearance.11 47 The model predicted that TPA and scu-PA secretion rose threefold to fourfold during isoproterenol infusion, overcoming increased clearance. If clearance had not been included in the model, predicted TPA and scu-PA secretion rates would have been lower.

Three TPA clearance models were studied, with active TPA cleared faster, equal to, and slower than complexed TPA. The best fit between measured and simulated levels of TPA and PAI-1 occurred when active TPA was cleared faster than complexed TPA. Brommer et al56 reported faster clearance for TPA activity compared with total TPA antigen, while Huisman et al25 reported similar clearance rates for TPA/C1-inhibitor complex and total TPA antigen during TPA infusion.

Increased TPA and scu-PA With Exercise
It is well known that exercise increases both TPA and scu-PA levels in plasma, but the mechanisms behind these increases are still controversial.19 26 Plasminogen activators may be increased in plasma during exercise in three ways: (1) hemoconcentration, which increases the concentration of most proteins in blood (corrected for in our data), (2) reduced hepatic blood flow, which decreases clearance of TPA and presumably scu-PA,29 and (3) release of epinephrine and other factors during strenuous exercise, which stimulate increased secretion.27 28

Depending on the exercise protocol used, either reduced clearance or increased secretion may be the most important cause of elevated plasminogen activators in plasma. Sustained submaximal exercise reduces hepatic blood flow and TPA clearance but has only minor effects on plasma epinephrine levels.25 26 29 75 TPA levels tend to rise gradually during submaximal exercise in parallel with reductions in clearance. When exercise ceases, clearance rapidly returns to normal, with similar half-lives for scu-PA and active TPA but a longer half-life for total TPA, consistent with slower clearance of TPA/PAI-1 and TPA/C1-inhibitor complexes.19 Changes in clearance are likely to be the most important cause of increased TPA and scu-PA during sustained submaximal exercise. Results from the study by de Boer et al29 were used to develop and validate the exercise clearance model in our kinetic simulation.

In contrast to submaximal exercise, graded exercise to exhaustion initially produced little change in plasma TPA or scu-PA levels, followed by an exponential increase in TPA and scu-PA that peaked at the point of exhaustion.19 26 27 28 Kinetic modeling of the exercise to exhaustion protocol in our study indicated that while reduced hepatic clearance did occur (Fig 2Up), it was insufficient to explain the rapid exponential rise in plasminogen activators. Instead, increased secretion of both TPA and scu-PA was predicted to be the most important cause of increased levels of these proteins during graded exercise to exhaustion. This is supported by our previous finding that TPA secretion was predicted to be directly proportional to the epinephrine concentration in blood during exercise and epinephrine infusion.27 28 Exercise to exhaustion produces an exponential increase in plasma epinephrine levels which, based on increases in TPA secretion, accurately predicts the changes seen in plasma TPA levels.

The highest levels of active TPA and greatest percentage of TPA in the active form occurred at the point of maximum exercise. The average specific molar activity of uncomplexed TPA at maximum exercise, defined as TPA activity (IU/mL) divided by total TPA antigen minus TPA/PAI-1 complex minus TPA/C1-inhibitor complex, was 4.5x1013 IU/mol. This was the same as the value of 4.48x1013 IU/mol from prior work used to calculate the molar concentration of active TPA and active PAI-1.35 It is somewhat surprising that the specific activities of the two studies were this similar. The prior study used a less specific assay for TPA activity and based the calculation of free TPA on total TPA minus TPA/PAI-1 complex only. It is likely that compensating errors in the older study contributed to the similarity of the results. More work is still needed on the specificity and standardization of assays for all forms of TPA and PAI-1.

Study Limitations
It is important to point out the limitations of this model. First, the site of PAI-1 synthesis is still controversial. While we chose to model liver PAI-1 secretion, endothelial PAI-1 secretion produced similar results for TPA and only slightly different forearm vein results for PAI-1. The major conclusions in the study on release and clearance of TPA were not affected by the site of PAI-1 secretion. Second, it is likely that different endothelial beds secrete TPA (and possibly PAI-1) at different rates.34 59 Third, we used estimated total and regional blood volumes and regional blood flows. Errors in total blood volume may affect the absolute level of predicted TPA and PAI-1 secretion (see Table 4Up) but did not have an effect on the relative changes in secretion found during exercise and the adrenergic infusions. To better understand regional differences in TPA and PAI-1 secretion, more data are needed on regional TPA and PAI-1 release, blood volumes, and blood flows. Finally, better measurements of hepatic clearance rates as a function of hepatic blood flow for active TPA, TPA/PAI-1 complex, TPA/C1-inhibitor complex, active PAI-1, and latent PAI-1 are needed to improve the clearance model.

In summary, the most important findings in this study were (1) the ß-adrenergic agonist isoproterenol increased plasma TPA and scu-PA levels by stimulating increased secretion, (2) the {alpha}-adrenergic agonist phenylephrine increased TPA and scu-PA levels in plasma by reducing hepatic blood flow and thus clearance, (3) graded exercise to exhaustion increased plasma TPA and scu-PA levels through a combination of increased secretion and reduced clearance, (4) the time course of scu-PA release was similar to that of TPA release, but the magnitude of the responses differed, (5) in vivo, the model predicted that C1-inhibitor bound to TPA at a rate of 553 mol-1 · s-1, 200-fold higher than measured in vitro,25 71 and (6) C1-inhibitor contributed equally with PAI-1 to TPA inhibition when active PAI-1 levels were low (20 to 50 pmol/L) but was less important when active PAI-1 levels were high. We conclude that secretion, inhibition, clearance, and regional blood flow effects must all be taken into account when evaluating changes in plasminogen activator levels.


*    Selected Abbreviations and Acronyms
 
PAI-1 = plasminogen activator inhibitor
scu-PA = inactive single-chain zymogen
TPA = tissue plasminogen activator
UPA = urokinase plasminogen activator


*    Acknowledgments
 
This study was supported by VA Medical Research Service and the American Heart Association, Washington Affiliate, Seattle, Wash.

Received March 16, 1995; revision received July 5, 1995; accepted July 7, 1995.


*    References
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
up arrowResults
up arrowDiscussion
*References
 

  1. Hamsten A, Walldius G, Szamosi A, Blombaeck M, De Faire U, Dahlen G, Landou C, Wiman B. Plasminogen activator inhibitor in plasma: risk factor for recurrent myocardial infarction. Lancet. 1987;2:3-9. [Medline] [Order article via Infotrieve]
  2. Ridker PM, Vaughan DE, Stampfer MJ, Manson JE, Hennekens CH. Endogenous tissue-type plasminogen activator and risk of myocardial infarction. Lancet. 1993;341:1165-1168. [Medline] [Order article via Infotrieve]
  3. Munkvad S, Gram J, Jespersen J. A depression of active tissue plasminogen activator in plasma characterizes patients with unstable angina pectoris who develop myocardial infarction. Eur Heart J. 1990;11:525-528. [Abstract/Free Full Text]
  4. Jansson JH, Olofsson BO, Nilsson TK. Predictive value of tissue plasminogen activator mass concentration on long-term mortality in patients with coronary artery disease. Circulation. 1993;88:2030-2034. [Abstract/Free Full Text]
  5. Gram J, Jespersen J. A selective depression of tissue plasminogen activator (t-PA) activity in euglobulin characterizes a risk group among survivors of acute myocardial infarction. Thromb Haemost. 1987;61:289-293.
  6. Zalewski A, Shi Y, Nardone D, Bravette B, Weinstock P, Fischman D, Wilson P, Goldberg S, Levin DC, Bjornsson TD. Evidence for reduced fibrinolytic activity in unstable angina at rest: clinical, biochemical, and angiographic correlates. Circulation. 1991;83:1685-1691. [Abstract/Free Full Text]
  7. Cortellaro M, Cofrancesco E, Boschetti C, Mussoni L, Donati MB, Cardillo M, Catalano M, Gabrielli L, Lombardi B, Specchia G, Tavazzi L, Tremoli E, Pozzoli E, Turri M, Group ftP. Increased fibrin turnover and high PAI-1 activity as predictors of ischemic events in atherosclerotic patients. Arterioscler Thromb. 1993;13:1412-1417. [Abstract/Free Full Text]
  8. Margaglione M, Di Minno G, Grandone E, Vecchione G, Celentano E, Cappucci G, Grilli M, Simone P, Panico S, Mancini M. Abnormally high circulation levels of tissue plasminogen activator and plasminogen activator inhibitor-1 in patients with a history of ischemic stroke. Arterioscler Thromb. 1994;14:1741-1745. [Abstract/Free Full Text]
  9. Kluft C. Constitutive synthesis of tissue-type plasminogen activator (t-PA) and plasminogen activator inhibitor type 1 (PAI-1): conditions and therapeutic targets. Fibrinolysis. 1994;8(suppl 2):1-7.
  10. Emeis JJ. Mechanisms involved in short-term changes in blood levels of t-PA. In: Kluft C, ed. Tissue-type Plasminogen Activator (t-PA): Physiological and Clinical Aspects. Boca Raton, Fla: CRC Press; 1988:21-36.
  11. Greenway CV. Mechanisms and quantitative assessment of drug effects on cardiac output with a new model of the circulation. Pharmacol Rev. 1982;33:213-251. [Medline] [Order article via Infotrieve]
  12. Gader AMA, da Costa J, Cash JD. The effect of propranolol, alprenolol and practolol on the fibrinolytic and factor VIII responses to adrenaline and salbutamol in man. Thromb Res. 1974;4:25-33. [Medline] [Order article via Infotrieve]
  13. Gader AMA, Clarkson AR, Cash JD. The plasminogen activator and coagulation factor VII responses to adrenaline, noradrenaline, isoprenaline and salbutamol in man. Thromb Res. 1973;2:9-16.
  14. Cash JD, Woodfield DG, Allan AGE. Adrenergic mechanisms in the systemic plasminogen activator response to adrenaline in man. Br J Haematol. 1970;18:487-494. [Medline] [Order article via Infotrieve]
  15. Rosing DR, Redwood DR, Brakman P, Astrup T, Epstein SE. The fibrinolytic response of man to vasoactive drugs measured in arterial blood. Thromb Res. 1978;13:419-428. [Medline] [Order article via Infotrieve]
  16. Gurewich V. Pro-urokinase: physiochemical properties and promotion of its fibrinolytic activity by urokinase and by tissue plasminogen activator with which it has a complementary mechanism of action. Semin Thromb Hemost. 1988;14:110-115. [Medline] [Order article via Infotrieve]
  17. Lijnen HR, Van Hoef B, de Cock F, Collen D. The mechanism of plasminogen activation and fibrin dissolution by single chain urokinase-type plasminogen activator in a plasma milieu in vitro. Blood. 1989;73:1864-1872. [Abstract/Free Full Text]
  18. Pannell R, Gurewich V. A comparison of the rates of clot lysis in a plasma milieu induced by tissue plasminogen activator (t-PA) and rec-pro-urokinase: evidence that t-PA has a more restricted mode of action. Fibrinolysis. 1992;6:1-5.
  19. Dooijewaard G, de Boer A, Turion PNC, Cohen AF, Breimer DD, Kluft C. Physical exercise induces enhancement of urokinase-type plasminogen activator (u-PA) levels in plasma. Thromb Haemost. 1991;65:82-86. [Medline] [Order article via Infotrieve]
  20. Imura Y, Stassen JM, Kurokawa T, Iwasa S, Lijnen HR, Collen D. Thrombolytic and pharmacokinetic properties of an immunoconjugate of single-chain urokinase-type plasminogen activator (u-PA) and a bispecific monoclonal antibody against fibrin and against u-PA in baboons. Blood. 1992;79:2322-2329. [Abstract/Free Full Text]
  21. Van De Werf F, Nobuhara M, Collen D. Coronary thrombolysis with human single-chain, urokinase type plasminogen activator (pro-urokinase) in patients with acute myocardial infarction. Ann Intern Med. 1986;104:345-348.
  22. Dooijewaard G. The post-thrombolytic era: is it possible to prevent thrombosis by stimulating indigenous fibrinolysis. Fibrinolysis. 1993;7(suppl 1):42-43.
  23. Booth NA, Walker E, Maughan R, Bennett B. Plasminogen activator in normal subjects after exercise and venous occlusion: t-PA circulates as complexes with C1-inhibitor and PAI-1. Blood. 1987;69:1600-1604. [Abstract/Free Full Text]
  24. Bennett B, Croll A, Ferguson K, Booth NA. Complexing of tissue plasminogen activator with PAI-1, {alpha}2-macroglobulin, and C1-inhibitor: studies in patients with defibrination and a fibrinolytic state after electroshock or complicated labor. Blood. 1990;75:671-676. [Abstract/Free Full Text]
  25. Huisman LGM, van Griensven JMT, Kluft C. On the role of C1-inhibitor as inhibitor of tissue-type plasminogen activator in human plasma. Thromb Haemost. 1995;73:466-471.[Medline] [Order article via Infotrieve]
  26. van den Burg PJM, Dooijewaard G, van Vliet M, Mosterd WL, Kluft C, Huisveld IA. Differences in u-PA and t-PA increase during acute exercise: relation with exercise parameters. Thromb Haemost. 1994;71:236-239. [Medline] [Order article via Infotrieve]
  27. Chandler WL, Levy WC, Veith RC, Stratton JR. A kinetic model of the circulatory regulation of tissue plasminogen activator during exercise, epinephrine infusion, and endurance training. Blood. 1993;81:3293-3302. [Abstract/Free Full Text]
  28. Chandler WL, Veith RC, Fellingham GW, Levy WC, Schwartz RS, Cerqueira MD, Kahn SE, Larson VG, Cain KC, Beard JC, Abrass IB, Stratton JS. Fibrinolytic response during exercise and epinephrine infusion in the same subjects. J Am Coll Cardiol. 1992;19:1412-1420. [Abstract]
  29. de Boer A, Kluft C, Kroon JM, Kasper FJ, Schoemaker HC, Pruis J, Breimer DD, Soons PA, Emeis JJ, Cohen AF. Liver blood flow as a major determinant of the clearance of recombinant human tissue-type plasminogen activator. Thromb Haemost. 1992;67:83-87. [Medline] [Order article via Infotrieve]
  30. Emeis JJ. Regulation of the acute release of tissue-type plasminogen activator from the endothelium by coagulation activation products. Ann N Y Acad Sci. 1992;667:249-258. [Medline] [Order article via Infotrieve]
  31. Chandler WL. A kinetic model of the circulatory regulation of tissue plasminogen activator. Thromb Haemost. 1991;66:321-322. [Medline] [Order article via Infotrieve]
  32. Chandler WL, Schmer G, Stratton JR. Optimum conditions for the stabilization and measurement of tissue plasminogen activator activity in human plasma. J Lab Clin Med. 1989;113:362-371. [Medline] [Order article via Infotrieve]
  33. Levy WC, Cerqueira MD, Matsuoka DT, Harp GD, Sheehan FN, Stratton JR. Four radionuclide methods of left ventricular volume determination: comparison of a manual and an automated technique. J Nucl Med. 1992;33:763-770. [Abstract/Free Full Text]
  34. Chandler WL, Loo SC, Mornin D. Adrenergic stimulation of regional plasminogen activator release in rabbits. Thromb Haemost. 1992;68:545-549. [Medline] [Order article via Infotrieve]
  35. Chandler WL, Trimble SL, Loo SC, Mornin D. Effect of PAI-1 levels on the molar concentrations of active tissue plasminogen activator (t-PA) and t-PA/PAI-1 complex in plasma. Blood. 1990;76:930-937. [Abstract/Free Full Text]
  36. Wojta J, Binder BR, Huber K, Hoover RL. Evaluation of fibrinolytic capacity in plasma during thrombolytic therapy with single (scu-PA) or two-chain urokinase type plasminogen activator (tcu-PA) by a combined assay system for urokinase type plasminogen activator antigen and function. Thromb Haemost. 1989;61:289-293. [Medline] [Order article via Infotrieve]
  37. Holvoet P, Cleemput H, Collen D. Assay of human tissue-type plasminogen activator (t-PA) with an enzyme-linked immunosorbent assay (ELISA) based on three murine monoclonal antibodies to t-PA. Thromb Haemost. 1985;54:684-687. [Medline] [Order article via Infotrieve]
  38. Declerck PJ, Verstreken M, Collen D. An immunofunctional assay for active plasminogen activator inhibitor-1 (PAI-1). Fibrinolysis. 1988;2(suppl 2):17-18.
  39. Amiral J, Plassart V, Grosley M, Mimilla F, Constant G, Guyader AM. Measurement of tPA - PAI-1 complexes by ELISA, using monoclonal antibodies: clinical relevance. Thromb Res. 1988;(suppl VIII):99-113.
  40. Wieczorek I, Ludlam CA, MacGregor I. Venous occlusion does not release von Willebrand factor, factor VIII, or PAI-1 from endothelial cells: the importance of consensus on the use of correction factors for haemoconcentration. Thromb Haemost. 1993;69:91-93. [Medline] [Order article via Infotrieve]
  41. Andreasen PA, Nielsen LS, Kristensen P, Grøndahl-Hansen J, Skriver L, Danø K. Plasminogen activator inhibitor from human fibrosarcoma cells binds urokinase-type plasminogen activator but not its pro-enzyme. J Biol Chem. 1986;261:7644-7651. [Abstract/Free Full Text]
  42. Mollison PL. Blood transfusion in clinical medicine. Boston, Mass: Blackwell Scientific; 1983.
  43. Witzleb E. Functions of the vascular system. In: Schmidt RF, Thews G, eds. Human Physiology. New York, NY: Springer-Verlag; 1983:397-455.
  44. Keber D, Blinc A, Fettich J. Increase of tissue plasminogen activator in limbs during venous occlusion: a simple haemodynamic model. Thromb Haemost. 1990;64:433-437. [Medline] [Order article via Infotrieve]
  45. Rowell LB. Human circulation, regulation during physical stress. New York, NY: Oxford University Press; 1986.
  46. Immink FGA, Beijer HJM, Charbon GA. {alpha}- and ß-Receptor blockade of isoproterenol- and norepinephrine-induced effects on regional blood flow and blood flow acceleration. Eur J Pharmacol. 1978;50:159-173. [Medline] [Order article via Infotrieve]
  47. Rutlen DL, Supple EW, Powell WJ. The role of the liver in the adrenergic regulation of blood flow from the splanchnic to the central circulation. Yale J Biol Med. 1979;52:99-106. [Medline] [Order article via Infotrieve]
  48. Pringle TH, Riddell JG, Shanks RG. Characterization of the beta-adrenoreceptors which mediate the isoprenaline-induced changes in finger tremor and cardiovascular function in man. Eur J Clin Pharmacol. 1988;35:507-514. [Medline] [Order article via Infotrieve]
  49. Gasic S, Eichler HG, Korn A. Nifedipine antagonizes alpha-adrenoreceptor mediated splanchnic and systemic vasocontriction in man. Int J Clin Pharm Ther Tox. 1987;25:382-388. [Medline] [Order article via Infotrieve]
  50. Johnson JM. Responses of forearm blood flow to graded leg exercise in man. J Appl Physiol. 1979;46:457-462. [Abstract/Free Full Text]
  51. van Hinsbergh VWM. Synthesis and secretion of plasminogen activators and plasminogen activator inhibitor by endothelial cells. In: Kluft C, ed. Tissue-Type Plasminogen Activator (t-PA): Physiological and Clinical Aspects. Boca Raton, Fla: CRC Press; 1988:3-20.
  52. Hajjar KA, Nachman RL. The human endothelial cell plasmin-generating system. In: Colman RW, Hirsh J, Marder VJ, Salzman EW, eds. Hemostasis and Thrombosis: Basic Principles and Clinical Practice. Philadelphia, Pa: JB Lippincott; 1994:823-836.
  53. Jaffe EA. Biochemistry, immunology, and cell biology of endothelium. In: Colman RW, Hirsh J, Marder VJ, Salzman EW, eds. Hemostasis and Thrombosis: Basic Principles and Clinical Practice. Philadelphia, Pa: JB Lippincott; 1994:718-744.
  54. Koh SCL, Yuen R, Viegas OAC, Chua SE, Ng BL, Sen DK, Ratnam SS. Plasminogen activators t-PA, u-PA, and its inhibitor (PAI) in normal males and females. Thromb Haemost. 1991;66:581-585. [Medline] [Order article via Infotrieve]
  55. Nicoloso G, Hauert J, Kruithof EKO, Melle GV, Bachmann F. Fibrinolysis in normal subjects, comparison between plasminogen activator inhibitor and other components of the fibrinolytic system. Thromb Haemost. 1988;59:299-303.[Medline] [Order article via Infotrieve]
  56. Brommer EJP, Derkx FHM, Schalekamp MADH, Dooijewaard G, Klaauw MMvd. Renal and hepatic handling of endogenous tissue-type plasminogen activator and its inhibitor in man. Thromb Haemost. 1988;59:404-411. [Medline] [Order article via Infotrieve]
  57. Alessi MC, Juhan-Vague I, Declerck PJ, Anfosso F, Gueunoun E, Collen D. Correlations between t-PA and PAI-1 antigen and activity and t-PA/PAI-1 complexes in plasma of control subjects and of patients with increased t-PA or PAI-1 levels. Thromb Res. 1990;60:509-516. [Medline] [Order article via Infotrieve]
  58. Huber K, Kirchheimer C,