- Email: [email protected]
Dynamic aspects of thrombus formation in mutant analbuminemic rats
THROMBOSIS RESEARCH 58; 6X3-643,1990 0049-3848/90 $3.00 + .OOPrinted in the USA. Copyright (c) 1990 Pergamon Press plc. All rights reserved.
DYNAMIC ASPECTS OF THROMBUS
FORMATION
IN MUTANT ANALBUMINEMIC
RATS.
Shin Kogal ,4Kenji Okajima:, Masayasu Inoue3,1Hiroaki Okabe', Haruo Araki , Sumi Nagase ,Kiyoshi Takatsuki Departments of Internal Medicinel, Laboratory Medicine2, Biochemistry3, Pharmacologyb, Kumamoto University Medical School, Kumamoto 860. Department of Chemistry, Sasaki Institute5, Tokyo 101, Japan.
(Received 25.1.1990; accepted in revised form 2.4.1990 by Editor H. Yamazaki)
ABSTRACT To elucidate the mechanism for thrombus formation in hypoalbuminemic and hyperlipidemic subjects, the experimental thrombus formation was analyzed in Nagase analbuminemic rats (MAR). The mutant rat reveals similar biochemical findings to those of NS, such as hyperlipidemia. When thrombus was induced in a small branch of the mesenteric artery by inserting a glass micropipette, thrombus formation was observed within 4 min in NAR and Sprague-Dawley rat (SDR). In SDR, the thrombus gradually grew up in size and detached from the inserted glass micropipette within 4 min after its formation. In ccntrast, the thrombus formed in NAR did not dissociate from the micropipette until lo-13 min after its formation. The maximum size of the thrombus formed in NAR was about 4 times larger than that in SDR. In the presence of fibrin, the plasma samples from NAR inhibited the activity of tissue-plasminogen activator (t-PA) 1.7 times more potently than did SDR plasma. Plasma .A%-plasmin inhibitor activity was significantly higher than that in SDR. Albumin significantly enhanced the t-PA-catalyzed activation of plasminogen, suggesting that the serum albumin might contribute, at least in part, to the plasma fibrinolytic activity. Thus, the higher thrombogenic potential in MAR than in SDR might be due to the reduced thrombolytic activity in NAR.
INTRODUCTION It is well documented that patients with nephrotic syndrome (NS) frequently associate with thrombosis (l-4). Concerning the mechanism for
Correspondence should be addressed to Dr. Kenji Okajima, Department of Laboratory Medicine, Kumamoto University Medical School, Honjo l-l-l, Kumamoto 860 Japan. Key Words: Nephrotic syndrome, Analbuminemic rat, Hypoalbuminemia, Hyperlipidemia, Thrombosis, Fibrinolysis
633
634
THROMBUS FORMATION IN NA RATS
Vol. 58, No. 6
thrombus formation in NS, some Critical changes in coagulation and fibrinolysis might seem to underlie the mechanism for their thrombotic tendency (5-10). The most characteristic change in coagulation system in NS is the decreased levels of antithrombin III (AT-III) due to its urinary loss, which might predispose patients with NS to thrombosis (11-12). Moreover, platelet activation by hypoalbuminemia and immune complex deposited at the glomerular lesion might contribute to the occurrence of thrombosis in NS (13-15). Since the most common and characteristic findings in NS are hypoalbuminemia and hyperlipidemia, these conditions might possibly correlate with thrombosis in NS (16-17). However, whether hypoalbuminemia and hyperlipidemia contribute to the hypercoagulable state in NS remains to be elucidated. The present study was undertaken to examine whether hypoalbuminemia and hyperlipidemia affect the potential for thrombosis in NS. The profile of thrombus formation in normal rats was compared to that in Nagase Analbuminemic Rats (NAR). Like hypoalbuminemic human subjects, the mutant NAR also shows hyperlipidemia (18-19).
MATERIALS AND METHODS Materials. Human serum albumin (faction V) and rat serum albumin were purchased from Sigma Chemical Co. (St. Louis, MO, U.S.A.). Tissue-type plasminogen activator (t-PA), plasminogen, fibrin and CBS 10.65 (synthetic substrate for plasmin) were obtained from Diagnostica-Stago (Asnieres, France). Chromogenic substrates, S-2251 (H-D-Val-Leu-Lys-pNA) and S-2238 (H-D-Phe-Pip-Arg-pNA)were obtained from Kabi-Vitrum (Stockholm, Sweden). Other reagents used were of analytical grade. Thrombus formation and thrombolysis in NAR and SDR. Female NAR and SDR. 220-240 p;. were intraveritoneallv administered 40 mg/kg pentobarbital. After a small-midline incision on abdominal wall, the terminal portion of the jejunum was pulled out carefully to spread over a transparent warm stage. Under dissecting microscope, fat and connective tissues surrounding the mesenteric artery of about 300 yrn in inner diameter were removed carefully. The preparation was mounted on a biological triocular equipped with a color (Nikon, Optiphot, Tokyo, Japan) microscope videorecording system (Hitachi, model GP 5J, Tokyo, Japan). The surface of mesenteric artery was continuously perfused at 37'C with physiological saline solution at a rate of 1 ml/min. The magnification of microscope was 100 times. The thrombus was induced by inserting a glass micropipette (10 Frn in diameter) using a micromanipulator. The glass micropipette was made by a puller (Narishige, model PD-5, Tokyo, Japan) and its inside was filled with physiological saline solution. The tip of the glass micropipette was advanced to one third of the vascular lumen as described previously (20). The maximal diameter of thrombus was measured by tracing their image on a played back display. Statistical analysis was performed by unpaired Student's t-test for between-group comparison. Values were expressed as means + S.E.M.. Assays. Under pentobarbital anesthesia, blood was collected from the abdominal aorta by a titrated plastic syringe; final concentration of sodium citrate was 0.38%. Blood samples were centrifuged at 3,000 x g for 10 min at 4°C. Antithrombin-III (AT-III) activity was determined by measuring the inhibition of thrombin activity in the presence of heparin using the chromogenic substrate S-2238 according to the method as described in (21). Plasminogen activity was determined by measuring plasmin activity after activation by
Vol. 58, No. 6
THROMBUS FORMATION IN NA RATS
635
streptokinase using the chromogenic substrate S-2251 (22). Fibrinogen was measured according to the method of Clauss et al (23). t-PA activity was measured spectrophotometrically as described by Verheijen et al (24). Plasma 3(2-plasmin inhibitor 6(2-PI) activity was determined by measuring the remaining plasmin activity after incubation of plasma with plasmin as described previously (25). Platelet Aggregation. Blood samples were collected from the abdominal aorta into titrated plastic syringes. Platelet rich plasma (PRP) was prepared by centrifuging the blood samples at 800 rpm for 15 min. Platelet poor plasma (PPP) was prepared by re-centrifuging the remaining blood samples at 3,000 rpm for 10 min at room temperature. Platelet aggregation was measured spectrophotometrically at 37°C in a platelet aggregometer {Chrono-log Co., Havertown, PA, U.S.A.) using 450 ~1 PRP adjusted to 3.0 x 10 platelets/,ul. The aggregometer was adjusted to 0% transmission with PRP and to 100% with PPP. Unlike human platelet, rat platelet could hardly aggregate in PRP. Therefore, we added CaC12 in reaction cuvette to a final concentration of 6.2 mM. Collagen (Chrono-log co., Havertown, PA, U.S.A.) and ADP (Chrono-log Co., Havertown, PA, U.S.A.) were used for triggering aggregation. Measurement of t-PA activity in the presence of rat and human albumin. The reaction mixture contained in a final volume of 1.0 ml. 0.02 IlJ/ml t-PA, 0.05 U/ml plasminogen, 75 yg/ml fibrin, 0.4 mM CBS 10.65 (synthetic substrate), 50 mM phosphate buffer (pH 7.5) and various concentrations of rat and human albumin. Formation of plasmin was determined spectrophotometrically at 405 nm. The method employed herein is essentially the same as described in the method for measurement of t-PA activity (24).
RESULTS
1. Thrombus formation in NAR and SDR.
When a micropipette was inserted into a small branch of the mesenteric artery, thrombus was formed after about 40 set both in SDR and NAR (FIG. 1-A). No significant difference in the time for initiating thrombus formation was found between the two animal groups (TABLE 1). In SDR, the thrombus grew up in size and dissociate from the micropipette about 4 min after its insertion The thrombus was formed again on the surface of micropipette 7 (FIG. i-c). min after insertion (FIG. 1-D). The second thrombus also flowed away into the circulation 5-6 min after its formation. On the other hand, the thrombus formed in NAR did not dissociate from the micropipette at 4 min of insertion, grew up to occupy the vascular lumen and, finally, dissociated from the FIG. 2 depicts the summary of micropipette after 12-13 min of insertion. thrombus formation in the present study. Just before the dissociation, the maximum size of thrombus formed in NAR was about 4 times larger than that in SDR (TABLE 1). In SDR, thrombus was formed repeatedly, within 30-40 set of the dissociation. The time required for the dissociation of thrombus from the micropipette was about 3-4 min. These findings suggest that thrombotic tendency might be stronger in NAR than in SDR presumably due to lower thombolytic activity in the former than in the latter.
THROMBUS FORMATION IN NA RATS
636
Vol. 58, No. 6
NAR
40 set
B
C
D
E
Imm FIG.
1
637
THROMBUS FORMATION IN NA RATS
Vol. 58, No. 6
FIG.
1 (on previous
page)
Thrombus formation in SDR and NAR. Photographs taken from TV display show thrombus formation and thrombus dissociation. Arrows in the figure indicate the formed thrombus. A horizontal bar indicates 1 mm. A: Thrombus started to occur about 40 set after insertion of a micropipette in both animal groups. B: Thrombus was formed and it gradually grew up around the inserted micropipette. C: Thrombus formed in SDR (left panel) dissociated from the micropipette and vascular wall and flowed away into the blood circulation. The thrombus initially formed in NAR (right panel) did not flow away by this time and further grew up. D: The second thrombus was formed in SDR, while the first thrombus formed in NAR still attached to the micropipette. E: The third thrombus formed in SDR was going to flow away into the circulation, while the first thrombus formed in NAR still continued to grow in size.
mesenteric artery
SDR
dissociation
new thrombus
Fglass micropipette $:99&m
/
/ +
@ : 400Um
: 400pm (Max)
NAR
F9 : BOllIn
Time after insertion of the micropipette
40sec
4min
9.5min
12min
T:Thrombus FIG.
2
Summary of thrombus formation and thrombolysis in SDR and NAR.
638
THROMBUS FORMATION IN NA RATS
Vol. 58, No. 6
TABLE 1. Difference in the profiles of thrombus formation and thrombolysis between SDR NAR.
and
Animal group
Time taken for
StimulationThrombus formationThrombus formation Dissociation (min) (set) SDR (n=18) 32.1 + 0.93 3.75 + 0.24 NAR (n=14) 33.3 7 2.46 10.30 T - 1.72* Data are expressed as mean + S.E.M. number of animals examined. *: P(O.04, **: P(O.01
Maximum size of the formed thrombus (diameter, pm) 430 + 18 1530 z 136**
Figures in the parentheses indicate the
2. Coagulation and fibrinolysis of blood samples from SDR and NAR. To elucidate the mechanism by which thrombolysis was impaired in NAR, coagulation and fibrinolysis of blood samples from NAR were compared with those from SDR. Chemical analysis revealed that AT-III activity in NAR plasma (158 23.4%) was about 1.4 times higher than that in SDR plasma (116 + 5.8X, P
TABLE 2 Effect of SDR and NAR plasma on t-PA activity -in vitro. Control t-PA activity in the presence of fibrin (A 405)
2.223
% Inhibition
0
SDR
NAR
1.174
0.417
47.1
81.2
THROMBUS
Vol. 58. No. 6 P<
0.01
FORMATION
639
IN NA RATS
P
I’i
A_ ” 0
a,
-o-
i?
“8”
z
Iam
I ci 50
SDR
(n-a)
NAR
cn=u
b=lv
(A)
ol=6)
(B)
h=l)
@=6)
(Cl
FIG. 3 Plasma levels of AT-III, Fibrinogen and z(2-PI in SDR and NAR. Plasma levels of AT-III (A), Fibrinogen (B) and 3(2-PI (C) were determined in SDR and NAR according to the methods as described in text. Figures in parentheses indicate the number of animals examined. 3. Effect of rat and human albumin on t-PA activity. To konw the relationship between hypoalbuminemia and decreased fibrinolysis, the effect of albumin was examined on t-PA activity.. As expected, rat albumin significantly enhanced the t-PA activity in a dose denendent fashion (FIG. 4). Similar enhancement of t-PA activity was also observed with humanalbumin (FIG. 4). The reaction mixture contained in a final volume of 1.0 ml, 0.02 IU/ml t-PA, 0.05 U/ml plasminogen, 75 pg/ml fibrin, 0.4 mM CBS 10.65 (synthetic substrate for plasmin), 50 mM phosphate buffer (pH 7.5) and various concentrations of rat and human albumin. Fonna,tion of plasmin was determined spectrophotometrically at 405 nm. The method employed herein is essentially the same as described in the method for determinig of t-PA activity.
0
0.25 Concentration
of albumin
u.3
(mM)
FIG. 4 Effect of rat and human albumin on t-PA activity.
640
THROMBUS FORMATION IN NA RATS
Vol. 58. No. 6
DISCUSSION The present study demonstrates that the potential for thrombosis is more evident in NAR than in SDR as judged from the profiles of experimental thrombus formation. No significant difference in lag-time required for the initiation of thrombus formation was observed between the two animal groups, suggesting that the activity to form thrombus might not be different between the two groups. Although plasma levels of fibrinogen and other coagulation factors are higher in NAR than in SDR, no significant difference in the incidence of thrombosis was found between the two animal groups (26). The increased plasma level.of fibrinogen in NAR might suggest the high potential of NAR to form thrombus. Although the mechanism for the increase in plasma AT-III level in NAR is not known, this increase might seem to contribute to decrease the thrombogenic potential in NAR. The most outstanding difference in the profiles of experimental thrombus formation between the two animal groups was the delayed dissociation of the thrombus from the micropipette in This phenomenon in NAR might be correlated with low thrombolytic NAR. activity. Consistent with this notion is the fact that plasma levels ofs(2-PI activity and the inhibitory activity of plasma samples against t-PA activity were significantly higher in NAR than in SDR. Emori et al reported that plasma levels of 0(2-PI, d2-macroglobulin (d2-MG) and &-antitrypsin (&-AT), inhibitors of plasmin, and fibrinogen increased significantly in NAR (27). The reason why the plasma levels of such acute phase reactants are increased in NAR remained obscure (28, 29). Du et al reported that plasma levels of 3(2-PI significantly increased in NS patients who associated with thrombosis and suggested the possibility that high antiplasmin activity might increase Concerning the the incidence of thrombosis in patients with NS (30). relationship between fibrinolytic activity and hyperlipidemia, Padro et al reported that the activity of antiplasmin and plasminogen activator inhibitor remarkable (31). A increased significantly in hyperlipidemic rats hyperlipidemia was demonstrated in NAR by Ando et al (19, 32); the serum levels of both total cholesterol and triglyceride were significantly higher in NAR (171 2 14 mgfdl and 169 + 20 mg/dl, respectively) than those in SDR (78 2 7 mg/dl and 101 + 22 mg/dl, respectively). The type of hyperlipidemia in NAR was quite similar to IIb type of hyperlipidemia frequently observed in NS. Recently, a remarkably high homology was demonstrated between the structure of plasminogen and apolipoprotein (a), suggesting that apolipoprotein (a) in LDL-cholesterol and plasminogen might compete the sites where plasminogen activation would take place (33). Thus, the low fibrinolytic activity and hyperlipidemia in NAR might possibly be correlated with each other. These findings suggested that the potent inhibitory activity of NAR plasma against t-PA activity might be explained by the increase in the levels of c(2-PI, d2-MG, $,l-ATand lipoprotein. The mechanism by which albumin enhanced the activity of t-PA is not known at present. Preliminary experiments in this laboratory revealed that albumin enhanced t-PA activity even in the absence of -fibrin. However, such an enhancement by albumin was not observed when urokinase-type plasminogen activator was employed. Thus, albumin might possibly increase the t-PA activity by enhancing the interaction between t-PA and plasminogen. Preliminary experiment also demonstrated that infusion of albumin to NAR and SDR inhibited the increase in the size of thrombus and accelerated the dissociation of the formed thrombus from the micropipette. These findings suggested that serum albumin might contribute, at least in part, to firinolysis. The precise mechanism for the enhancement of t-PA by albumin is under our current investigation. The present findings suggested that the' thrombotic potential might be higher in NAR than in SDR presumably due to low thrombolytic activity in the
Vol. 58, No. 6
THROMBUS FORMATION IN NA RATS
641
former that shows analbuminemia and hyperlipidemia. It has been reported that the decrease in plasma albumin levels also enhanced the aggregation of Thus, hypoalbuminemia and hyperlipidemia might be platelets (13, 14). However, some important factors for thrombosis in patients with NS. compensatory mechanisms including the increased AT-III levels might contribute to decrease the incidence of thrombosis in NAR. In case of NS, the decrease in AT-III levels by its urinary loss and the activation of platelets by primary glomerular lesion might increase the incidence of thrombosis synergistically with the two risk factors described above (15).
REFERENCES 1.
ROSENMANN, E., POLLAK, V.E. and PIRANI, C.L. Renal vein thrombosis in adults: A clinical and pathologic study based on renal biopsies. Medicine. 47, 269, 1968.
2.
WAGONER, R.D., STANSON, A.W., HOLLEY, K.E. and WINTER, C.S. Renal vein thrombosis in idiopathic membranous glomerulopathy and nephrotic syndrome: Incidence and significance. Kidney Int. 23, 368-374, 1983.
3.
LLACH, F., PAPPER, S. and MASSRY, S.G. The clinical spectrum of renal vein thrombosis: Acute and chronic. Am. J. Med. 69, 819-827, 1980.
4.
KENDALL, A.G., LOHMANN, R.E. and DOSSETOR, J.B. Nephrotic syndrome: A hypercoagulable state. Arch. Intern. Med. 127, 1021-1027, 1971.
5.
THOMPSON, C., FORBES, C.D., PRENTICE, C.R.M. and KENNEDY, A.C. Changes in the blood coagulation and fibrinolysis in the nephrotic syndrome. Quart. J. Med. 43, 399-407. 1974.
6.
LANGE, L.G. III, CARVALHO, A., BAGDASARIAN, A., LAUIRI, B. and COLLMAN, R.W. Activation of Hageman factor in the nephrotic syndrome. Am. J. Med. 56, 564-568, 1974.
7.
BANG, N.U., TRYGSTAD, C.W., SCHROEDER, J.E., HEINDENREICH, R.O., CSISCKO, B.M. Enhanced platelet function in glomerular renal disease. J. Lab. Clin. Med. 81, 651-660, 1973.
8.
PANICUCCI, F., SAGRIPANTI, A., VISPI, M., PINORI, E., LECCHINI, L., BARSOTTI, G. and GIOVANETTI, S. Comprehensive study of hemostasis in nephrotic syndrome. Nephron 33, 9-13, 1983.
9.
WINTER, C.S., WAGONER, R.D., BOWIE, E.J. and OWEN, G.A.Jr. Dysfibrinogenemia and hypercoagulability in patients with membranous glomerulopathy. Thromb. Haemostas. 42, 66-67, 1979.
10. KALLEN, R.J. and LEE, S.K. A study of the plasma kinin-generating system in children with the minimal lesion, idiopathic nephrotic syndrome. Ped. Res. 9, 705-709, 1975. 11.
KAUFFMANN, R.H., VELTKAMP, J.J., VAN TILBURG, N.H. and VAN ES, L.A. Acquired antithrombin III deficiency and thrombosis in the nephrotic: syndrome. Am. J. Med. 65, 607-613, 1978.
12.
VZIRI, N.D., PAULE, P., TOOHEY, J., HUNG, E.. ALIKHANI, S., DARWISH, R. and PAHL, M. Acquired deficiency and urinary excretion of antithrombin III in nephrotic syndrome. Arch. Intern. Med. 144, 1802-1803, 1984.
642
THROMBUS FORMATION IN NA RATS
vOl.58,NO. 6
13. YOSHIDA, N. and AOKI, N. Release of arachidonic acid from human platelets. A key role for the potentiation of platelet aggregability in normal subjects as well as in those with nephrotic syndrome. Blood 52, 969-977, 1978. 14.
REMLJZZI,G., MECCA, G., MARCHESI, D., LIVIO, M., DE GAETANO, G., DONATI, M.B. and SIVER, M.J. Platelet hyperaggregability and the nephrotic syndrome. Thromb. Res. 16, 345-354, 1979.
15.
CAMERON, J.S. Coagulation and thromboembolic complications in the nephrotic syndrome, Year Book Medical Publishers, Inc. 75-114, 1984. 16.
16.
LLACH, F. Hypercoagulability and thrombotic complications of nephrotic syndrome. Kidney International 28, 429-439, 1985.
17.
GANDRILLE, S., JOWIN, M.H., TOULON, P., PEMY, P., FIESSINGER, J.N., RONCATOM, M., MOATTI, N. and AIACH M. A study of fibrinogen and fibrinolysis in 10 adults with nephrotic syndrome. Thromb. Haemost. 53, 445-454, 1988.
18.
NAGASE, S., SHIMAMUNE, K., SEIGO S. Science 205, 590-591, 1979.
19.
ANDO, S., KON, K., TANAKA, Y., NAGASE, S. and NAGAI, Y. Characterization of hyperlipidemia in Nagase analbuminemia rat (NAR). J. Biochem. 87, 1859-1862, 1980.
20.
ARAKI, H. and NISHI, K. Effect of vascular constriction on occlusive thrombus formation of rat mesenteric artery. Thromb. Res. 44, 39-45, 1986.
21.
ABILDGAARD, U., LIE, M., ODEGARD 0. R. Antithrombin (heparin cofactor) assay with new chromogenic substrates (S-2238 and Chromozym TH). Thromb. Res. 11, 549-553, 1977.
22.
FRIBERGER, P., K. MARTA, GUSTAVSSON, S., AURELL L. and CLAESON G. Methods for determination of plasmin, antiplasmin and plasminogen by means of substrate S-2251. Haemostasis 7, 138-145, 1978.
23.
CLAUSS, A. Gerinnungsphysiologiesche Schnellmethode zur Bestimmung des Fibrinogens. Acta Haematol,(Basel), 17, 237-246, 1957.
24.
VERHEIJEN, J.H., MULLAART, E., CHANG, G.T.G., KLUFT, C. and WIJINGAARDS, G. A simple sensitive spectrophotometric assay for extrinsic (tissue-type) plasminogen activator applicalbe to measurement in plasma. Thromb. Haemost.48, 266-269, 1982.
25.
VERHEIJEN, J.H., CHANG G.T.G. and KLUFT, B. Evidence for the occurrence of a fast acting inhibitor of tissue type plasminogen activator in human plasma. Thromb. Haemost. 51, 392-395, 1984.
26.
MATSUDA, T., SEKI, T., MIURA, R., OGAWARA, M., and NAGASE, S. Coagulability of blood fibrinolysis and platelet function in congenital analbuminemic rats. Blood and Vessel 14, 231-233, 1983.
27.
EMORI, T., TAKAHASHI, PI.,SUGIYAMA, K., SHUMIYA, S. and NAGASE, S. Age-related changes in plasma proteins of analbuminemic rats. Exp. Anim. 32. 123-134. 1983.
Albumin-deficient rat mutant,
Vol. 58, No. 6
THROMBUS FORMATION IN NA RATS
643
28.
MATSUDA, M., WAKABAYASHI, K., AOKI, N. and MORIOKA, Y. o(22-Plasmin inhibitor is among acute phase reactants. Thromb. Res. 17, 527-532, 1980.
29.
PEPYS, M.B. and BALTZ, M.L. Acute phase proteins with special reference to C-reactive protein and related proteins (pentaxins) and serum amyloid protein. Advances in Immunology, 141-212, 1983, 34.
30.
DU, X.H., GLAS-GREENWALT, P., KANT, K.S., ALLEN, C.M., HAYES, S. and POLLAK, V.E. Nephrotic syndrome with renal vein thrombosis: pathogenetic importance of a plasmin inhibitor (3(2-antiplasmin). Clin .Nephrol. 24, 4, 186-191, 1985.
31.
PADRO, T., VILLAVERDE, C.A. and CANOVAS, M. Hyperlipemia, fibrinolysis and arteriosclerosis. Thromb. Res. 49, 519-530, 1988.
32.
TAKAHASHI, M., KUSUMI, K., SHUMIYA, S. and NAGASE, S. Age-related changes in plasma lipid concentrations and enzyme activities in Nagase analbuminemia rats (NAR). Exp. Anim. 32, 39-46, 1983.
33.
MCLEAN, J.W. TOMLINON, J.E. KUANG, W-J., EATON D.L., CHEN, E.Y., ELESS, G.M., SCANU, A.M. and LAWN, R.M. 'DNA sequence of human apolipoprotein (a) is homologous to plasminogen. Nature 300, 132-137, 1987.
DYNAMIC ASPECTS OF THROMBUS
FORMATION
IN MUTANT ANALBUMINEMIC
RATS.
Shin Kogal ,4Kenji Okajima:, Masayasu Inoue3,1Hiroaki Okabe', Haruo Araki , Sumi Nagase ,Kiyoshi Takatsuki Departments of Internal Medicinel, Laboratory Medicine2, Biochemistry3, Pharmacologyb, Kumamoto University Medical School, Kumamoto 860. Department of Chemistry, Sasaki Institute5, Tokyo 101, Japan.
(Received 25.1.1990; accepted in revised form 2.4.1990 by Editor H. Yamazaki)
ABSTRACT To elucidate the mechanism for thrombus formation in hypoalbuminemic and hyperlipidemic subjects, the experimental thrombus formation was analyzed in Nagase analbuminemic rats (MAR). The mutant rat reveals similar biochemical findings to those of NS, such as hyperlipidemia. When thrombus was induced in a small branch of the mesenteric artery by inserting a glass micropipette, thrombus formation was observed within 4 min in NAR and Sprague-Dawley rat (SDR). In SDR, the thrombus gradually grew up in size and detached from the inserted glass micropipette within 4 min after its formation. In ccntrast, the thrombus formed in NAR did not dissociate from the micropipette until lo-13 min after its formation. The maximum size of the thrombus formed in NAR was about 4 times larger than that in SDR. In the presence of fibrin, the plasma samples from NAR inhibited the activity of tissue-plasminogen activator (t-PA) 1.7 times more potently than did SDR plasma. Plasma .A%-plasmin inhibitor activity was significantly higher than that in SDR. Albumin significantly enhanced the t-PA-catalyzed activation of plasminogen, suggesting that the serum albumin might contribute, at least in part, to the plasma fibrinolytic activity. Thus, the higher thrombogenic potential in MAR than in SDR might be due to the reduced thrombolytic activity in NAR.
INTRODUCTION It is well documented that patients with nephrotic syndrome (NS) frequently associate with thrombosis (l-4). Concerning the mechanism for
Correspondence should be addressed to Dr. Kenji Okajima, Department of Laboratory Medicine, Kumamoto University Medical School, Honjo l-l-l, Kumamoto 860 Japan. Key Words: Nephrotic syndrome, Analbuminemic rat, Hypoalbuminemia, Hyperlipidemia, Thrombosis, Fibrinolysis
633
634
THROMBUS FORMATION IN NA RATS
Vol. 58, No. 6
thrombus formation in NS, some Critical changes in coagulation and fibrinolysis might seem to underlie the mechanism for their thrombotic tendency (5-10). The most characteristic change in coagulation system in NS is the decreased levels of antithrombin III (AT-III) due to its urinary loss, which might predispose patients with NS to thrombosis (11-12). Moreover, platelet activation by hypoalbuminemia and immune complex deposited at the glomerular lesion might contribute to the occurrence of thrombosis in NS (13-15). Since the most common and characteristic findings in NS are hypoalbuminemia and hyperlipidemia, these conditions might possibly correlate with thrombosis in NS (16-17). However, whether hypoalbuminemia and hyperlipidemia contribute to the hypercoagulable state in NS remains to be elucidated. The present study was undertaken to examine whether hypoalbuminemia and hyperlipidemia affect the potential for thrombosis in NS. The profile of thrombus formation in normal rats was compared to that in Nagase Analbuminemic Rats (NAR). Like hypoalbuminemic human subjects, the mutant NAR also shows hyperlipidemia (18-19).
MATERIALS AND METHODS Materials. Human serum albumin (faction V) and rat serum albumin were purchased from Sigma Chemical Co. (St. Louis, MO, U.S.A.). Tissue-type plasminogen activator (t-PA), plasminogen, fibrin and CBS 10.65 (synthetic substrate for plasmin) were obtained from Diagnostica-Stago (Asnieres, France). Chromogenic substrates, S-2251 (H-D-Val-Leu-Lys-pNA) and S-2238 (H-D-Phe-Pip-Arg-pNA)were obtained from Kabi-Vitrum (Stockholm, Sweden). Other reagents used were of analytical grade. Thrombus formation and thrombolysis in NAR and SDR. Female NAR and SDR. 220-240 p;. were intraveritoneallv administered 40 mg/kg pentobarbital. After a small-midline incision on abdominal wall, the terminal portion of the jejunum was pulled out carefully to spread over a transparent warm stage. Under dissecting microscope, fat and connective tissues surrounding the mesenteric artery of about 300 yrn in inner diameter were removed carefully. The preparation was mounted on a biological triocular equipped with a color (Nikon, Optiphot, Tokyo, Japan) microscope videorecording system (Hitachi, model GP 5J, Tokyo, Japan). The surface of mesenteric artery was continuously perfused at 37'C with physiological saline solution at a rate of 1 ml/min. The magnification of microscope was 100 times. The thrombus was induced by inserting a glass micropipette (10 Frn in diameter) using a micromanipulator. The glass micropipette was made by a puller (Narishige, model PD-5, Tokyo, Japan) and its inside was filled with physiological saline solution. The tip of the glass micropipette was advanced to one third of the vascular lumen as described previously (20). The maximal diameter of thrombus was measured by tracing their image on a played back display. Statistical analysis was performed by unpaired Student's t-test for between-group comparison. Values were expressed as means + S.E.M.. Assays. Under pentobarbital anesthesia, blood was collected from the abdominal aorta by a titrated plastic syringe; final concentration of sodium citrate was 0.38%. Blood samples were centrifuged at 3,000 x g for 10 min at 4°C. Antithrombin-III (AT-III) activity was determined by measuring the inhibition of thrombin activity in the presence of heparin using the chromogenic substrate S-2238 according to the method as described in (21). Plasminogen activity was determined by measuring plasmin activity after activation by
Vol. 58, No. 6
THROMBUS FORMATION IN NA RATS
635
streptokinase using the chromogenic substrate S-2251 (22). Fibrinogen was measured according to the method of Clauss et al (23). t-PA activity was measured spectrophotometrically as described by Verheijen et al (24). Plasma 3(2-plasmin inhibitor 6(2-PI) activity was determined by measuring the remaining plasmin activity after incubation of plasma with plasmin as described previously (25). Platelet Aggregation. Blood samples were collected from the abdominal aorta into titrated plastic syringes. Platelet rich plasma (PRP) was prepared by centrifuging the blood samples at 800 rpm for 15 min. Platelet poor plasma (PPP) was prepared by re-centrifuging the remaining blood samples at 3,000 rpm for 10 min at room temperature. Platelet aggregation was measured spectrophotometrically at 37°C in a platelet aggregometer {Chrono-log Co., Havertown, PA, U.S.A.) using 450 ~1 PRP adjusted to 3.0 x 10 platelets/,ul. The aggregometer was adjusted to 0% transmission with PRP and to 100% with PPP. Unlike human platelet, rat platelet could hardly aggregate in PRP. Therefore, we added CaC12 in reaction cuvette to a final concentration of 6.2 mM. Collagen (Chrono-log co., Havertown, PA, U.S.A.) and ADP (Chrono-log Co., Havertown, PA, U.S.A.) were used for triggering aggregation. Measurement of t-PA activity in the presence of rat and human albumin. The reaction mixture contained in a final volume of 1.0 ml. 0.02 IlJ/ml t-PA, 0.05 U/ml plasminogen, 75 yg/ml fibrin, 0.4 mM CBS 10.65 (synthetic substrate), 50 mM phosphate buffer (pH 7.5) and various concentrations of rat and human albumin. Formation of plasmin was determined spectrophotometrically at 405 nm. The method employed herein is essentially the same as described in the method for measurement of t-PA activity (24).
RESULTS
1. Thrombus formation in NAR and SDR.
When a micropipette was inserted into a small branch of the mesenteric artery, thrombus was formed after about 40 set both in SDR and NAR (FIG. 1-A). No significant difference in the time for initiating thrombus formation was found between the two animal groups (TABLE 1). In SDR, the thrombus grew up in size and dissociate from the micropipette about 4 min after its insertion The thrombus was formed again on the surface of micropipette 7 (FIG. i-c). min after insertion (FIG. 1-D). The second thrombus also flowed away into the circulation 5-6 min after its formation. On the other hand, the thrombus formed in NAR did not dissociate from the micropipette at 4 min of insertion, grew up to occupy the vascular lumen and, finally, dissociated from the FIG. 2 depicts the summary of micropipette after 12-13 min of insertion. thrombus formation in the present study. Just before the dissociation, the maximum size of thrombus formed in NAR was about 4 times larger than that in SDR (TABLE 1). In SDR, thrombus was formed repeatedly, within 30-40 set of the dissociation. The time required for the dissociation of thrombus from the micropipette was about 3-4 min. These findings suggest that thrombotic tendency might be stronger in NAR than in SDR presumably due to lower thombolytic activity in the former than in the latter.
THROMBUS FORMATION IN NA RATS
636
Vol. 58, No. 6
NAR
40 set
B
C
D
E
Imm FIG.
1
637
THROMBUS FORMATION IN NA RATS
Vol. 58, No. 6
FIG.
1 (on previous
page)
Thrombus formation in SDR and NAR. Photographs taken from TV display show thrombus formation and thrombus dissociation. Arrows in the figure indicate the formed thrombus. A horizontal bar indicates 1 mm. A: Thrombus started to occur about 40 set after insertion of a micropipette in both animal groups. B: Thrombus was formed and it gradually grew up around the inserted micropipette. C: Thrombus formed in SDR (left panel) dissociated from the micropipette and vascular wall and flowed away into the blood circulation. The thrombus initially formed in NAR (right panel) did not flow away by this time and further grew up. D: The second thrombus was formed in SDR, while the first thrombus formed in NAR still attached to the micropipette. E: The third thrombus formed in SDR was going to flow away into the circulation, while the first thrombus formed in NAR still continued to grow in size.
mesenteric artery
SDR
dissociation
new thrombus
Fglass micropipette $:99&m
/
/ +
@ : 400Um
: 400pm (Max)
NAR
F9 : BOllIn
Time after insertion of the micropipette
40sec
4min
9.5min
12min
T:Thrombus FIG.
2
Summary of thrombus formation and thrombolysis in SDR and NAR.
638
THROMBUS FORMATION IN NA RATS
Vol. 58, No. 6
TABLE 1. Difference in the profiles of thrombus formation and thrombolysis between SDR NAR.
and
Animal group
Time taken for
StimulationThrombus formationThrombus formation Dissociation (min) (set) SDR (n=18) 32.1 + 0.93 3.75 + 0.24 NAR (n=14) 33.3 7 2.46 10.30 T - 1.72* Data are expressed as mean + S.E.M. number of animals examined. *: P(O.04, **: P(O.01
Maximum size of the formed thrombus (diameter, pm) 430 + 18 1530 z 136**
Figures in the parentheses indicate the
2. Coagulation and fibrinolysis of blood samples from SDR and NAR. To elucidate the mechanism by which thrombolysis was impaired in NAR, coagulation and fibrinolysis of blood samples from NAR were compared with those from SDR. Chemical analysis revealed that AT-III activity in NAR plasma (158 23.4%) was about 1.4 times higher than that in SDR plasma (116 + 5.8X, P
TABLE 2 Effect of SDR and NAR plasma on t-PA activity -in vitro. Control t-PA activity in the presence of fibrin (A 405)
2.223
% Inhibition
0
SDR
NAR
1.174
0.417
47.1
81.2
THROMBUS
Vol. 58. No. 6 P<
0.01
FORMATION
639
IN NA RATS
P
I’i
A_ ” 0
a,
-o-
i?
“8”
z
Iam
I ci 50
SDR
(n-a)
NAR
cn=u
b=lv
(A)
ol=6)
(B)
h=l)
@=6)
(Cl
FIG. 3 Plasma levels of AT-III, Fibrinogen and z(2-PI in SDR and NAR. Plasma levels of AT-III (A), Fibrinogen (B) and 3(2-PI (C) were determined in SDR and NAR according to the methods as described in text. Figures in parentheses indicate the number of animals examined. 3. Effect of rat and human albumin on t-PA activity. To konw the relationship between hypoalbuminemia and decreased fibrinolysis, the effect of albumin was examined on t-PA activity.. As expected, rat albumin significantly enhanced the t-PA activity in a dose denendent fashion (FIG. 4). Similar enhancement of t-PA activity was also observed with humanalbumin (FIG. 4). The reaction mixture contained in a final volume of 1.0 ml, 0.02 IU/ml t-PA, 0.05 U/ml plasminogen, 75 pg/ml fibrin, 0.4 mM CBS 10.65 (synthetic substrate for plasmin), 50 mM phosphate buffer (pH 7.5) and various concentrations of rat and human albumin. Fonna,tion of plasmin was determined spectrophotometrically at 405 nm. The method employed herein is essentially the same as described in the method for determinig of t-PA activity.
0
0.25 Concentration
of albumin
u.3
(mM)
FIG. 4 Effect of rat and human albumin on t-PA activity.
640
THROMBUS FORMATION IN NA RATS
Vol. 58. No. 6
DISCUSSION The present study demonstrates that the potential for thrombosis is more evident in NAR than in SDR as judged from the profiles of experimental thrombus formation. No significant difference in lag-time required for the initiation of thrombus formation was observed between the two animal groups, suggesting that the activity to form thrombus might not be different between the two groups. Although plasma levels of fibrinogen and other coagulation factors are higher in NAR than in SDR, no significant difference in the incidence of thrombosis was found between the two animal groups (26). The increased plasma level.of fibrinogen in NAR might suggest the high potential of NAR to form thrombus. Although the mechanism for the increase in plasma AT-III level in NAR is not known, this increase might seem to contribute to decrease the thrombogenic potential in NAR. The most outstanding difference in the profiles of experimental thrombus formation between the two animal groups was the delayed dissociation of the thrombus from the micropipette in This phenomenon in NAR might be correlated with low thrombolytic NAR. activity. Consistent with this notion is the fact that plasma levels ofs(2-PI activity and the inhibitory activity of plasma samples against t-PA activity were significantly higher in NAR than in SDR. Emori et al reported that plasma levels of 0(2-PI, d2-macroglobulin (d2-MG) and &-antitrypsin (&-AT), inhibitors of plasmin, and fibrinogen increased significantly in NAR (27). The reason why the plasma levels of such acute phase reactants are increased in NAR remained obscure (28, 29). Du et al reported that plasma levels of 3(2-PI significantly increased in NS patients who associated with thrombosis and suggested the possibility that high antiplasmin activity might increase Concerning the the incidence of thrombosis in patients with NS (30). relationship between fibrinolytic activity and hyperlipidemia, Padro et al reported that the activity of antiplasmin and plasminogen activator inhibitor remarkable (31). A increased significantly in hyperlipidemic rats hyperlipidemia was demonstrated in NAR by Ando et al (19, 32); the serum levels of both total cholesterol and triglyceride were significantly higher in NAR (171 2 14 mgfdl and 169 + 20 mg/dl, respectively) than those in SDR (78 2 7 mg/dl and 101 + 22 mg/dl, respectively). The type of hyperlipidemia in NAR was quite similar to IIb type of hyperlipidemia frequently observed in NS. Recently, a remarkably high homology was demonstrated between the structure of plasminogen and apolipoprotein (a), suggesting that apolipoprotein (a) in LDL-cholesterol and plasminogen might compete the sites where plasminogen activation would take place (33). Thus, the low fibrinolytic activity and hyperlipidemia in NAR might possibly be correlated with each other. These findings suggested that the potent inhibitory activity of NAR plasma against t-PA activity might be explained by the increase in the levels of c(2-PI, d2-MG, $,l-ATand lipoprotein. The mechanism by which albumin enhanced the activity of t-PA is not known at present. Preliminary experiments in this laboratory revealed that albumin enhanced t-PA activity even in the absence of -fibrin. However, such an enhancement by albumin was not observed when urokinase-type plasminogen activator was employed. Thus, albumin might possibly increase the t-PA activity by enhancing the interaction between t-PA and plasminogen. Preliminary experiment also demonstrated that infusion of albumin to NAR and SDR inhibited the increase in the size of thrombus and accelerated the dissociation of the formed thrombus from the micropipette. These findings suggested that serum albumin might contribute, at least in part, to firinolysis. The precise mechanism for the enhancement of t-PA by albumin is under our current investigation. The present findings suggested that the' thrombotic potential might be higher in NAR than in SDR presumably due to low thrombolytic activity in the
Vol. 58, No. 6
THROMBUS FORMATION IN NA RATS
641
former that shows analbuminemia and hyperlipidemia. It has been reported that the decrease in plasma albumin levels also enhanced the aggregation of Thus, hypoalbuminemia and hyperlipidemia might be platelets (13, 14). However, some important factors for thrombosis in patients with NS. compensatory mechanisms including the increased AT-III levels might contribute to decrease the incidence of thrombosis in NAR. In case of NS, the decrease in AT-III levels by its urinary loss and the activation of platelets by primary glomerular lesion might increase the incidence of thrombosis synergistically with the two risk factors described above (15).
REFERENCES 1.
ROSENMANN, E., POLLAK, V.E. and PIRANI, C.L. Renal vein thrombosis in adults: A clinical and pathologic study based on renal biopsies. Medicine. 47, 269, 1968.
2.
WAGONER, R.D., STANSON, A.W., HOLLEY, K.E. and WINTER, C.S. Renal vein thrombosis in idiopathic membranous glomerulopathy and nephrotic syndrome: Incidence and significance. Kidney Int. 23, 368-374, 1983.
3.
LLACH, F., PAPPER, S. and MASSRY, S.G. The clinical spectrum of renal vein thrombosis: Acute and chronic. Am. J. Med. 69, 819-827, 1980.
4.
KENDALL, A.G., LOHMANN, R.E. and DOSSETOR, J.B. Nephrotic syndrome: A hypercoagulable state. Arch. Intern. Med. 127, 1021-1027, 1971.
5.
THOMPSON, C., FORBES, C.D., PRENTICE, C.R.M. and KENNEDY, A.C. Changes in the blood coagulation and fibrinolysis in the nephrotic syndrome. Quart. J. Med. 43, 399-407. 1974.
6.
LANGE, L.G. III, CARVALHO, A., BAGDASARIAN, A., LAUIRI, B. and COLLMAN, R.W. Activation of Hageman factor in the nephrotic syndrome. Am. J. Med. 56, 564-568, 1974.
7.
BANG, N.U., TRYGSTAD, C.W., SCHROEDER, J.E., HEINDENREICH, R.O., CSISCKO, B.M. Enhanced platelet function in glomerular renal disease. J. Lab. Clin. Med. 81, 651-660, 1973.
8.
PANICUCCI, F., SAGRIPANTI, A., VISPI, M., PINORI, E., LECCHINI, L., BARSOTTI, G. and GIOVANETTI, S. Comprehensive study of hemostasis in nephrotic syndrome. Nephron 33, 9-13, 1983.
9.
WINTER, C.S., WAGONER, R.D., BOWIE, E.J. and OWEN, G.A.Jr. Dysfibrinogenemia and hypercoagulability in patients with membranous glomerulopathy. Thromb. Haemostas. 42, 66-67, 1979.
10. KALLEN, R.J. and LEE, S.K. A study of the plasma kinin-generating system in children with the minimal lesion, idiopathic nephrotic syndrome. Ped. Res. 9, 705-709, 1975. 11.
KAUFFMANN, R.H., VELTKAMP, J.J., VAN TILBURG, N.H. and VAN ES, L.A. Acquired antithrombin III deficiency and thrombosis in the nephrotic: syndrome. Am. J. Med. 65, 607-613, 1978.
12.
VZIRI, N.D., PAULE, P., TOOHEY, J., HUNG, E.. ALIKHANI, S., DARWISH, R. and PAHL, M. Acquired deficiency and urinary excretion of antithrombin III in nephrotic syndrome. Arch. Intern. Med. 144, 1802-1803, 1984.
642
THROMBUS FORMATION IN NA RATS
vOl.58,NO. 6
13. YOSHIDA, N. and AOKI, N. Release of arachidonic acid from human platelets. A key role for the potentiation of platelet aggregability in normal subjects as well as in those with nephrotic syndrome. Blood 52, 969-977, 1978. 14.
REMLJZZI,G., MECCA, G., MARCHESI, D., LIVIO, M., DE GAETANO, G., DONATI, M.B. and SIVER, M.J. Platelet hyperaggregability and the nephrotic syndrome. Thromb. Res. 16, 345-354, 1979.
15.
CAMERON, J.S. Coagulation and thromboembolic complications in the nephrotic syndrome, Year Book Medical Publishers, Inc. 75-114, 1984. 16.
16.
LLACH, F. Hypercoagulability and thrombotic complications of nephrotic syndrome. Kidney International 28, 429-439, 1985.
17.
GANDRILLE, S., JOWIN, M.H., TOULON, P., PEMY, P., FIESSINGER, J.N., RONCATOM, M., MOATTI, N. and AIACH M. A study of fibrinogen and fibrinolysis in 10 adults with nephrotic syndrome. Thromb. Haemost. 53, 445-454, 1988.
18.
NAGASE, S., SHIMAMUNE, K., SEIGO S. Science 205, 590-591, 1979.
19.
ANDO, S., KON, K., TANAKA, Y., NAGASE, S. and NAGAI, Y. Characterization of hyperlipidemia in Nagase analbuminemia rat (NAR). J. Biochem. 87, 1859-1862, 1980.
20.
ARAKI, H. and NISHI, K. Effect of vascular constriction on occlusive thrombus formation of rat mesenteric artery. Thromb. Res. 44, 39-45, 1986.
21.
ABILDGAARD, U., LIE, M., ODEGARD 0. R. Antithrombin (heparin cofactor) assay with new chromogenic substrates (S-2238 and Chromozym TH). Thromb. Res. 11, 549-553, 1977.
22.
FRIBERGER, P., K. MARTA, GUSTAVSSON, S., AURELL L. and CLAESON G. Methods for determination of plasmin, antiplasmin and plasminogen by means of substrate S-2251. Haemostasis 7, 138-145, 1978.
23.
CLAUSS, A. Gerinnungsphysiologiesche Schnellmethode zur Bestimmung des Fibrinogens. Acta Haematol,(Basel), 17, 237-246, 1957.
24.
VERHEIJEN, J.H., MULLAART, E., CHANG, G.T.G., KLUFT, C. and WIJINGAARDS, G. A simple sensitive spectrophotometric assay for extrinsic (tissue-type) plasminogen activator applicalbe to measurement in plasma. Thromb. Haemost.48, 266-269, 1982.
25.
VERHEIJEN, J.H., CHANG G.T.G. and KLUFT, B. Evidence for the occurrence of a fast acting inhibitor of tissue type plasminogen activator in human plasma. Thromb. Haemost. 51, 392-395, 1984.
26.
MATSUDA, T., SEKI, T., MIURA, R., OGAWARA, M., and NAGASE, S. Coagulability of blood fibrinolysis and platelet function in congenital analbuminemic rats. Blood and Vessel 14, 231-233, 1983.
27.
EMORI, T., TAKAHASHI, PI.,SUGIYAMA, K., SHUMIYA, S. and NAGASE, S. Age-related changes in plasma proteins of analbuminemic rats. Exp. Anim. 32. 123-134. 1983.
Albumin-deficient rat mutant,
Vol. 58, No. 6
THROMBUS FORMATION IN NA RATS
643
28.
MATSUDA, M., WAKABAYASHI, K., AOKI, N. and MORIOKA, Y. o(22-Plasmin inhibitor is among acute phase reactants. Thromb. Res. 17, 527-532, 1980.
29.
PEPYS, M.B. and BALTZ, M.L. Acute phase proteins with special reference to C-reactive protein and related proteins (pentaxins) and serum amyloid protein. Advances in Immunology, 141-212, 1983, 34.
30.
DU, X.H., GLAS-GREENWALT, P., KANT, K.S., ALLEN, C.M., HAYES, S. and POLLAK, V.E. Nephrotic syndrome with renal vein thrombosis: pathogenetic importance of a plasmin inhibitor (3(2-antiplasmin). Clin .Nephrol. 24, 4, 186-191, 1985.
31.
PADRO, T., VILLAVERDE, C.A. and CANOVAS, M. Hyperlipemia, fibrinolysis and arteriosclerosis. Thromb. Res. 49, 519-530, 1988.
32.
TAKAHASHI, M., KUSUMI, K., SHUMIYA, S. and NAGASE, S. Age-related changes in plasma lipid concentrations and enzyme activities in Nagase analbuminemia rats (NAR). Exp. Anim. 32, 39-46, 1983.
33.
MCLEAN, J.W. TOMLINON, J.E. KUANG, W-J., EATON D.L., CHEN, E.Y., ELESS, G.M., SCANU, A.M. and LAWN, R.M. 'DNA sequence of human apolipoprotein (a) is homologous to plasminogen. Nature 300, 132-137, 1987.