Complexing of Tissue Plasminogen Activator With PAI-1, a,-Macroglobulin, and C1-Inhibitor: Studies in Patients With Defibrination and a Fibrinolytic State After

Electroshock or Complicated Labor

By Bruce Bennett, Alison Croll, Kathleen Ferguson, and Nuala A. Booth

Release of tissue plasminogen activator (t-PA) and its interaction with plasma protease inhibitors were studied in two patients with massive defibrination, one after elec- troshock and soft tissue injury and the other after compli- cated labor; both had very severe hemorrhage. Large quantities of free t-PA were present in the circulation for several hours. Complexes of t-PA with plasminogen activa- tor inhibitor 1 (PAI-11, a,-macroglobulin and C1 -inhibitor were also observed. PAI-1 antigen rose dramatically in both patients, and complexes of t-PA with PAL1 rose rapidly during the period of observation. In contrast, the complexes of t-PA with a,-macroglobulin and C1-inhibitor, present initially, persisted for short periods only and dis-

WO IMMUNOLOGICALLY distinct types of plas- T minogen activator (PA) occur in human blood. Tissue plasminogen activator (t-PA) is the principal activator and occurs in endothelial cells, from which it can be released by stimuli such as physiologic stress. Urokinase (u-PA) circu- lates principally in an inactive precursor form, single chain

Both t-PA and u-PA can be inhibited by the circulating plasma inhibitor PAI- 1, present in small quantities in cell- free plasma and in higher amounts in platelets.’s2 The sequence of interactions between PA, PAI- 1, and the other inhibitors that secure control of physiologic and pathologic fibrinolysis is not defined. Since pathologic fibrinolytic bleed- ing is rare, this control is sufficiently powerful to prevent plasmin generation in the circulation under normal circum- stances. Certainly, plasmin is only generated in the circula- tion of normal individuals under extreme stress3 The manner in which released plasminogen activator is cleared from the circulation is uncertain, but it involves both interaction with plasma inhibitors3s4 and clearance from the circulation by the liver. 5*6

This report examines interaction of endogenous t-PA, released in large amounts over a very short period of time, with plasma protease inhibitors; records the time for which free and complexed t-PA persists in the circulation; and notes the consequences of these events in terms of plasmin genera- tion and depletion of plasma protease inhibitors.

U-PA (XU-PA).

CASE REPORT

Patient 1, a 58-year-old man, purchased a disk-grinder (a tool with a rotating head used to sharpen certain farm implements) in April 1987. Under the influence of alcohol, he inserted the bare wires into a wall socket. He was seen to fall to the ground immediately, and was unconscious. The grinder functioned briefly, inflicting severe injury to his left thigh and severing the femoral artery and vein. He was pulseless when medical help arrived, but was vigorously resusci- tated with external cardiac massage and the infusion of large volumes of intravenous fluid. These measures restored heartbeat and circulation, and he was transferred to Aberdeen Royal Infirmary, where he arrived semi-conscious but grossly anemic and hemor- rhagic. Table 1 lists the initial findings on admission. He was transferred directly to theater and an attempt was made to repair the injured blood vessels. On return to the intensive care unit after

appeared when free t-PA disappeared from the circulation. Plasmin was generated initially, as indicated by the pres- ence of plasmin-a,-antiplasmin complexes. Plasma concen- trations of a,-macroglobulin, C1 -inhibitor, antithrombin 111, and a,-antiplasmin were severely depleted initially, but rapidly returned to normal. The observations demonstrate that there is a major release of t-PA in such defibrinating patients, that there is a role for protease inhibitors other than PAI-1 in the regulation of,endogenous t-PA. and indi- cate the great rapidity with which such free t-PA is complexed and cleared. 0 1990 by The American Society of Hematology.

surgery, severe generalized bleeding occurred and he was returned to theater. Further surgical reconstruction of the major bleeding sites in the femoral vessels was attempted but was unsuccessful, and amputation of the left leg was undertaken; thereafter massive bleeding ceased but generalized oozing continued for many hours. Before admission to Aberdeen Royal Infirmary, this patient had received massive but unquantified infusion of crystalloid. In the 12 hours after admission he received 13.5 L whole blood, 4.5 L packed red cells, 4.5 L fresh frozen plasma, 6.4 L purified plasma protein solution, 4.0 L synthetic volume expanders, 2.5 L crystalloid, 1.0 L platelet-rich plasma, and 12 U platelet concentrate.

Patient 2 was a 37-year-old woman who had three previous miscarriages. Her first completed pregnancy ended at term in delivery of a healthy child, but labor was complicated by fetal distress necessitating forceps delivery, with the child in the occipito- posterior position. Placenta and membranes were delivered appar- ently intact. One hour after delivery, severe vaginal bleeding and cardiovascular collapse occurred. The initial coagulation data indi- cated in Table 2 were recorded. Resuscitation was accomplished with blood transfusion (8.5 L, part as whole blood and part as packed red cells), cryoprecipitate, and fresh frozen plasma, and ergo- metrine, syntocinon, and prostaglandin were administered during this period. Cardiovascular stability was re-established, and from 5 hours after the delivery recovery was uncomplicated.

MATERIALS AND METHODS

Plasminogen activators in circulating blood were demonstrated by sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-

~~ ~~

From the Department of Medicine and Therapeutics, University of Aberdeen: and Intensive Care Unit. Aberdeen Royal Injirmary. Scotland.

Submitted December 28, 1988: accepted October 4, 1989. Supported by Grants from The British Heart Foundation, Gram-

pian Health Board, and Scottish Hospitals Endowment Research Trust.

Address reprint requests to B. Bennett, MD, Department of Medi- cine & Therapeutics, University of Aberdeen, Foresterhill, Aber- deen AB9 ZZD, Scotland. UK.

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. section 1734 solely to indicate this fact.

0 1990 by The American Society of Hematology. 0006-4971/90/7503-0028$3.00/0

Blood, Vol 75, No 3 (February 1). 1990: pp 671-676 67 1

672 BENNElT ET AL

Table 1. Patient 1

Plasminogen Hours Platelet Fibrin- Activator: After Count TCT APlT PT own ATlll Fibrin Plate Plasminogen a,AP CI-inh a-,M PAI-1 t-PA t-PA-PAI-1

Admission ( x 109/L) (s) (SI (SI (g/L) (% normal) (mm diameter) (% normal) (% normal) (% normal) (% normal) (ng/mL) (ng/mL) (% normal)

0 139 18 180 88 0.70 53 24 46 15 46 17 26 22 50 7 116 19 95 34 1.40 61 16 85 29 49 39 110 43 1,Ooo

1 1 64 17 71 26 2.00 89 0 85 57 80 55 550 32 700 17 133 16 63 27 2.40 98 0 95 89 100 75 1,756 51 1,OOO 19 151 14 58 23 2.00 89 0 95 98 100 77 1,900 54 800 21 63 15 55 24 1.90 89 0 100 65 100 66 1.950 43 800

Normal

values 200-400 13 40-50 12-17 1.5-4.0 80-120 0 100 100 100 100 8-35 1.0-6.6 1 0 0

PAGE) with zymography.' This involves separation by SDS-PAGE of plasma proteins according to their molecular size, with subsequent identification of plasminogen activators by applying an agarose gel containing fibrin and plasminogen to the SDS gel. Plasminogen activators diffuse from the SDS gel into the detector gel, where they convert plasminogen to plasmin, producing a clear zone of lysis in the opaque fibrin detector gel.

Identification of bands of plasminogen activator activity as com- plexes with a,-macroglobulin (a,-M) or C1-inhibitor (Cl-inh) was made by incubating plasma samples with immunoglobulins (Ig) specific for these inhibitors (Dako Ltd, High Wycombe, UK). The treated samples were then centrifuged at 100.000 x g to remove immune precipitates, and the supernatants were applied to SDS gels for zymography as described above.

PAI- 1 was quantified by enzyme-linked immunosorbent assay (ELISA),' as was &PA8; the latter assay was kindly performed by Dr Ian R. MacGregor. These ELISAs do not discriminate between free and complexed forms of t-PA8 or of PAI-1.9 T-PA-PAI-1 com- plexes were quantified by a two-site ELISA, using rabbit anti-t-PA (Organon-Teknika, Cambridge, UK) as the capture antibody and biotin-conjugated anti-PAI-19 as the second antibody. The quanti- ties of complex detected were expressed as a percentage of that in pooled plasma from 20 healthy resting normal subjects aged 20 to 40 years.

Purified t-PA and a,-macroglobulin were from Organon Teknika (Freiburg, FRG). In the experiments described in the text, 20 pL t-PA (4 U/mL) was incubated at 37OC for 30 minutes with 10 pL a,-macroglobulin (10 mg/mL) before study by SDS-PAGE with zymography, as described above. As a control, t-PA was similarly incubated in the absence of a,-macroglobulin.

Plasmin-a,-antiplasmin complexes were demonstrated by two- dimensional immunoelectrophoresis (2DIEP) against antiserum to a,-antiplasmin.lo Antithrombin 111 (ATIII) complexes were demon-

strated similarly." Plasminogen, a , -M, C1-inh, ATIII, and a2- antiplasmin (a,-AP) were quantified by rocket immunoelectrophore- sis against specific antisera." Fibrinogen was quantified by a functional assay based on that of Ratnoff and Menzies." Coagula- tion screening techniques, prothrombin time (PT), activated partial thromboplastin time (APTT), and thrombin time (TT) were per- formed by standard techniques. Overall plasma fibrinolytic activity was assessed by applying untreated citrated plasma to plasminogen- rich fibrin plates and measuring the diameter of the lysed zone after 24 hours at 37OC."

RESULTS

On admission to Aberdeen Royal Infirmary, patient 1 showed a gross reduction of the hemoglobin, and marked prolongation of PT, APTT, and TT was observed (Table 1). These abnormalities could have been due either to a defi- brinating process or to severe hemodilution, as massive amounts of fluid had been infused before hospital admission. Evidence of activation of both coagulation and fibrinolytic pathways was present because thrombin-AT111 and plasmin a,-AP complexes could both be demonstrated in the admis- sion blood sample (Fig 1). The former disappeared almost immediately and fibrinogen and ATIII levels rose rapidly toward normal after resuscitation and transfusion (Table 1). Plasmin-a,-AP complexes persisted for many hours and traces remained detectable 19 hours but had disappeared by 2 1 hours after admission (Fig 1 ) .

The explanation for the generation of plasmin in this patient was evident from the studies of plasma plasminogen activator. Figure 2 shows clearly that on admission to the hospital, massive amounts of free plasminogen activator were

Table 2. Patient 2

Hours Post-

partum

Platelet Count TCT APlT

( X 1 0 ~ 1 ~ ) (s) (SI

PT (S)

Plasminogen Activator:

Fibrin- Fibrin Plate ogen ATlll (mm Plasminogen a,AP (g/L) (% normal) diameter) (% normal) (% normal)

CI-inh (% normal)

1 2 5 6 9

12 27

101 66 163 87 27 71 82 19 56 60 17 54 48 17 51 48 15 47

103 13 47

34 28 18 20 19 19 18

0.35 51 21 1.05 60 4 2.07 72 4 - 66 0

2.31 72 0 - 76 0

2.13 69 . 0

100 81 86 79 79 72 69

57 55 66 66 69 69 69

37 51 60 54 57 60 89

a-,M PAI-1 (% normal) (ng/mL)

24 49.5 44 189.9 58 613.1 63 m . 7 63 645.1 72 230.4 70 95.7

1-PA t-PA-PAI-1 (ng/mL) (% normal)

105 225 14 450 16 480 25 610 43 1.270 40 1,120 15 460

Normal values 200-400 13 40-50 12-17 1.5-4.0 80-120 0 1 0 0 100 100 1 0 0 8-35 1.0-6.6 1 0 0

T-PA COMPLEXES WITH PAI-1, a,M, CI-INH IN DIC 673

I

Oh

7h

-7 :-------- - - 7

21h

Fig 1. Two-dlmensionel immunoelectrophoresis of plasma agsinst antiserum to ATlll (right) and to u,-AP (left) of samples from petient 1 et intervals after admission to the hospital as indi- cated in hours. The slow-moving peaks represent inhibitor com- plexed with active enzyme."

detectable in this patient's plasma at an apparent molecular mass of approximately 65 Kd, characteristic of free t-PA. A band of plasminogen activator activity was detected at 110 Kd, as is the case in the plasma of normal individuals; this represents t-PA complexed to PAI-I.' Additional new high molecular mass bands of PA activity, not seen in normal plasma, were detected. One had an estimated molecular mass of 180 Kd, and the second had a very high mass but migrated so little from the top of the gel that its size could not realistically be estimated. All these bands represented t-PA, all being removed when antiserum monospecific for t-PA was included in the detector gel (Fig 3). The two bands of large molecular size represented t-PA complexed with a,-M and CI-inh. as demonstrated in Fig 3; antisera specific for a,-M removed the highest molecular mass band and antibody to C1-inh removed the 180-Kd band. A band of PA activity with identical mobility to the slow-moving band was gener- ated when purified t-PA was incubated with purified a,-M (Fig 4). We previously showed that the band of activity of 180 Kd is generated when purified t-PA is incubated with purified C1-inh.' All three types of t-PA-inhibitor complex, namely with PAI-I, a,-M and C1-inh, appeared in the euglobulin fraction of plasma (not shown).

The free t-PA at 65 Kd and the high molecular mass complexes of t-PA with a,-M and C1-inh also disappeared rapidly from the circulation, being present in the 7-hour blood sample but absent from the 1 I-hour sample (Fig 5) .

Immunoassay of PAL1 showed normal plasma levels of the protein on admission to the hospital and its rapid increase during the following hours. Table 1 and Figs 1 and 5 summarize the changes in plasmin, plasmin a,-AP, plasmino-

gen activator, and inhibitors during the first 24 hours after admission. Rapid disappearance of free t-PA and of t-PA complexes with a,-M and C1-inh, and slightly slower disap pearance of plasmin a,-AP complexes, occurred while the plasma concentrations of a,-AP. a,-M and C1-inh were returning toward normal and plasma PAL1 was increasing rapidly.

The sequential changes in complexes of t-PA with PAI-1, quantified by ELISA, are shown in Table 1. It is evident that PAI-I rose massively and remained elevated throughout the period of observation and that t-PA-PAL1 complex showed a similar trend. It should be noted here that in the days after this period of observation, this patient showed evidence of gross cardiac, hepatic, and renal damage, the last requiring hemodialysis.

Patient 2 also defibrinated when first studied, immediately after complicated labor (Table 2). Like patient 1, her hemo- static profile was suggestive of disseminated intravascular coagulation. She also showed the features of massive sys- temic activation of the fibrinolytic system, with evidence of circulating plasmin-a,-AP complexes and severe transient depletion of inhibitors of fibrinolysis, a,-AP, C1-inh, and a,-M. Her plasma also contained free t-PA (Fig 2). with additional bands of t-PA activity identifiable as complexes with PAI-1, a , - M , and C1-inh (data not shown). Free t-PA

180kh

110 kDa

65kDa

Fig 2. SDS-PAGE with zymography to show bands of plasmino- gen activator (clear lysed zones in the opaque fibrinagerose gel). (A) Normal plasma: (e) petient 1 edmission sample: (C) petient 2, l hour postpertum.

674 BENNETT ET AL

A B C D 1 I

180kDe

6MtDa

Fig 3. SDS-PAGE with zymography of admission sample from patient 1. (A) After treetment with nonimmune rabbit IgG: (9) aiter treatment with IgG specific for a2-M: (C) &or treatment with IgG specific for C1-inh; (D) in the presence of IgG specific for t-PA. The same obaervations were made on the plasme of petient 2 (not shown).

and t-PA complexed with a,M and CI-inh disappeared rapidly from her plasma within 5 hours of their appearance, confirming the pattern observed in patient 1.

This patient had massive quantities of t-PA antigen present in her initial blood sample, and showed a dramatic rise in levels of PAI-I and of t-PA-PAL1 complexes there- after (Table 2). Her starting levels of PAI-1 were, of course, elevated by pregnancy itself.“ The sequential changes, with a rise and subsequent fall of PAI-I and t-PA-PAL1 com- plexes, differ from those of patient 1, in that the levels were clearly returning toward normal at the end of the period of study. Some of these changes may reflect her response to the initial fibrinolytic event, but this pattern may have been mod- ified by the normal sequence of changes after labor and de- livery. In notable contrast to patient l , this patient’s plasma PAI-I concentration peaked and was falling at the end of the study period. This seems likely to reflect the fact that she made a totally uncomplicated recovery from her episode of severe hemorrhage and did not go on to show any evidence of the extensive organ damage sustained by patient 1.

DISCUSSION

It is evident that the trauma suffered by these patients, either electroshock with tissue injury or complicated labor, resulted in defibrination with systemic activation of the fibrinolytic system. Free t-PA was released into the circula- tion in both patients. This resulted in the formation of complexes of t-PA, not only with the principal inhibitor of plasminogen activator, PAI-I, but also with a,-macroglobu- lin and CI-inh. C1-inh was previously found to inhibit t-PA released after exercise.) Inhibition of endogenous (post- exercise) t-PA by a,-antitrypsin and a,-antiplasmin has also

been reported: High concentrations of exogenous t-PA added to human plasma resulted in complex formation with a,-AP and a2-M.” Thus our observations demonstrate that a,-M participates in vivo in the control of t-PA as well as that of plasminI6 and u-PA.”

However, the formation of t-PA complexes with PAI-1, a, -M, and C1-inh in our patients was insufficient to remove free t-PA activity from the circulation, where it remained detectable for several hours. These events were accompanied by severe depletion of many of the protease inhibitors operating in the fibrinolytic system, namely a,-AP, a2-M, and C1-inh. Clearly the circulating plasma inhibitory mech- anisms were overwhelmed by the enormous release of t-PA into the circulation, such that plasmin generation resulted, as indicated by the presence of plasmin-a,-AP complexes. Initially thrombin was also generated, as indicated by the presence of thrombin-AT111 complexes, but this phenome- non was more short-lived and these complexes disappeared rapidly, while the fibrinolytic abnormalities persisted for many hours. PAI-1 rose progressively and dramatically in both patients to very high levels. This contrasts the findings relating to the other inhibitors, a,AP. a , M . CI-inh, and ATIII, and indicates that the level of this protein is under different control from the other inhibitors, consonant with evidence that it is an acute phase

C -

A B

180kDa

1lOkDa

65kDa

Fig 4. SDS-PAGE with zymography of (A) purified t-PA: (B) purified t-PA after incubation with puritied a2-M; (C) plasma from patient 1 (admission sample).

T-PA COMPLEXES WITH PAI-1, a,M, C~INH IN DIC 675

oh 7h l l h 21h

18OkDa

iiokDa mkDa

65kDa

Fig 6. SDS-PAGE with rymogrephy on the series of semples from petient 1 et the indicated intervals efter edmission, to show diseppearance of free t-PA end its complexes with a,-M and C1-inh.

The plasma concentration of t-PA-PAL1 complex rose and remained elevated over the entire period of observation in patient I . who developed major organ damage after the episode of defibrination. In patient 2, who recovered rapidly from the hemorrhagic episode, PAI-I and t-PA-PAI-I rose

strikingly, but the concentrations were returning to normal after about 24 hours. Complexes of t-PA with a, -M and C1-inh were, in contrast, cleared very rapidly from the circulation, disappearing between 7 and 11 hours. Free t-PA was also cleared rapidly, becoming undetectable between the 7- and 1 1-hour observations, consistent with its short half-life in humans' as well as in other species.,@,,

In a previous study that showed release of large amounts of free t-PA after exercise, complexing of t-PA with PAI-I and C1-inh wasdemonstrated, but complexes with a, -M were not observed. Interestingly, in that study only traces of plasmin were generated, in striking contrast to the situation in the patients studied here. Possibly plasmin generation in the circulation occurs only if thecapacity of all three inhibitors is overwhelmed.

In summary, this study indicates that massive release of t-PA may occur in certain circumstances; a part of this t-PA complexes with PAI-I and, in the extreme circumstances described here, with a,-M and C1-inh. However, such complexing was insufficient to prevent plasmin formation as indicated by the generation of plasmin-a,-AP complexes. The released plasminogen activator, whether free or com- plexed with a, -M or C1-inh, was cleared rapidly from the circulation. t-PA complexes with PAL1 remained in the circulation for considerably longer than the three other forms of t-PA. The rapidity of clearance was such that definite conclusions as to whether clearance of free t-PA occurs or whether clearance requires its complexing with inhibitor proteins cannot be drawn; this area requires further study.

ACKNOWLEDGMENT

We thank Dr Ian MacGregor. Scottish National Blood Transfu- sion Service, Edinburgh, for performing the t-PA antigen determina- tions.

REFERENCES

1. Erickson LA, Ginsberg MH, Loskutoff D J Detection and partial characterization of an inhibitor of plasminogen activator in human platelets. J Clin Invest 74:1465.1984

2. Booth NA. Simpson AJ. Croll A, Bennett B, Madjrcgor I R Plasminogen activator inhibitor (PAI-I) in plasma and platelets. Br J Haematol70327.1988

3. Booth NA, Walker E. Maughan R, Bennett B: Plasminogen activator in normal subjects after exercise and venous occlusion: t-PA circulates as complexes with CI inhibitor and PAI-I. Blood 69:1600. 1987

4. Rijken DC, Juhan-Vague I, Collen D Complexes between tissue-type plasminogen activator and proteinase inhibitors in hu- man plasma, identified with an immunoradiometric assay. J Lab Clin Med 101:285, 1983

5. Nilsson T, Wallen P, Mellbring G: In vivo metabolism of human tissue-type plasminogen activator. S a n d J Haematol33:49, 1984

6. Fletcher AP, Biederman 0. Moore D, Alkjaersig N, Sherry S: Abnormal plasminogen-plasmin system activity (fibrinolysis) in patients with hepatic cirrhosis; its cause and consequences. J Clin Invest 43:681,1964

7. Booth NA, Anderson JA, Bennett B: Plasminogen activators in alcoholic cirrhosis; demonstration of increased tissue type and urokinase type activator. J Clin Pathol37:772, 1984

8. MacGregor IR, MacDonald S, Dawes J. Micklem LR, James

K. A monoclonal antibody enzyme linked immunoabsorbent assay (ELISA) directed towards a fibrin binding region of tissue-type plasminogen activator. Fibrinolysis 1:247, 1987

9. MacGregor IR, Booth NA: An enzyme linked immunoabsor- bent assay (ELISA) used to study the cellular secretion of endothe- lial plasminogen activator inhibitor (PAI-1). Thromb Haemostas 5968.1988

10. Booth NA, Bennett B, Wijngaards G, Grieve JHK: A new life-long haemorrhagic disorder due to ex- plasminogen activator. Blood 61:267,1983

1 1. Booth NA, Bennett B Plasmin a,-antiplasmin complexes in bleeding disorders characteriscd by primary or secondary fibrinoly- sis. Br J Haematol56:545, 1984

12. Laurel1 C-B: Quantitative estimation of proteins by electro- phoresis in agarose gel containing antibodies. Anal Biochem 15:45, 1966

13. Ratnoff OD, Menzies C A new method for the determination of fibrinogen in small samples of plasma. J Lab Clin Med 32316, 1951

14. Kruithof EKO, "hang CT, Gudinchet A, Huaert J, Nicoloso G. Genton C, Welti H , Bachmann F Fibrinolysis in pregnancy: A study of plasminogen activator inhibitors. Blood 69:460, 1987

15. Kominger C, Collen D Neutralization of human extrinsic (tissue-type) plasminogen activator in human plasma: No evidence for a specific inhibitor. Thromb Haemostas 46:662, 1981

676 BENNET ET AL

16. Harpel PC: a,-Plasmin inhibitor and a,-macroglobulin- plasmin complexes in plasma. J Clin Invest 68:46,1981

17. Straight DL, Hassett MA, McKee PA Structural and functional characterization of the inhibition of urokinase by a2- macroglobulin. Biochemistry 24390 1, 1985

18. Juhan-Vague I, Aillaud MF, de Cock F, Philip-Joet C, Arnaud C, Seradimigni A, Collen D: The fast-acting inhibitor of tissue-type plasminogen activator is an acute-phase reactant protein, in Davidson JF, Donati MB, Coccheri S (eds): Progress in Fibrinoly- sis VII. London, UK, Churchill Livingstone, 1985, p 146

19. Kluft C, Verheijen JH, Jie AFH, Rijken DC, Preston FE, Sue-Ling HM, Jespersen J, Aason 0: The postoperative fibrinolytic

shutdown: Rapidly reverting acute-phase pattern for the fast-acting inhibitor of tissue-type plasminogen activator after trauma. Scand J Clin Invest 45:605, 1985

20. Korninger C, Stassen JM, Collen D: Turnover of human extrinsic (tissue-type) plasminogen activator in rabbits. Thromb Haemostas 46:658, 1981

21. Emeis JJ, van den Hoogen CM, Jense D: Hepatic clearance of tissue-type plasminogen activator in rats. Thromb Haemostas 54: 661,1985

22. Fuchs HE, Berger H, Pizzo SV: Catabolism of human tissue plasminogen activator in mice. Blood 65:539, 1985