THROMBOSIS RESEARCH 65; 593-604.1992 0049-3848/92 $5.00 + .OO Printed in the USA. Copyright (c) 1992 Pergamon Press Ltd. All rights reserved.
CHANGES IN PLASMA THROMBOMODULIN ANTIGEN IN RABBIT DEVELOPING ENDOTOXIN-INDUCED DISSEMINATED INTRAVASCULAR COAGULATION AND THE EFFECT OF HEPARIN Hiroyuki UchiyaMa. Hisaya Ohtani. Sayuri Hiraishi. Shuichi Horie, Hidemi Ishii* and Hutsuyoshi KazaMa Departaent of Clinical Biochemistry, Faculty Teikyo University, 1091-l Suarashi, Sciences, 199-01. Japan. (Received
11.9.1991;
accepted
in revised form 25.12.1991
of Pharmaceutical Tsukui, Kanagawa
by Editor A. Takada)
ABSTRACT Soluble thrombomodulin (TM) antigen level was 1.64 + male0.64 fig/ml (n=18, mean + S.D.) in plasma of normal enzyme immunoassay, and the rabbits as measured by of subspecies of 94. 83 and 51 kd. antigen consisted intravascular coagulation (DIG) was When disseminated into intravenous infusion of endotoxin induced by the TM antigen level in plasma was elevated to rabbits, about 1.5 times of the control value, and an increase in new the 83 kd subspecies as well as the appearance of 76 and 48 kd was observed subspecies of concomitantly with disappearance of the 94 kd subspecies in plasma. Elevation of the antigen level and disappearance of the 94 kd subspecies caused by infusion of endotoxin were reduced by simultaneous infusion of heparin. Addition of with leukocytes stimulated endotoxin plus FMLP to cultured endothelial cells induced release of TM antigen medium accompanying cell injury as measured by ‘PCrthTelease which was prevented by treatment with heparin. It WAS suggested that the increase in plasma TM antigen level in parallel with the generation of DIC reflected endothelial injury of rabbits, and that the elevation of TM antigen and the endothelial cell injury were prevented by heparin treatment. Key Words: DIG. Thrombomodulin. ‘To whom correspondence should
Heparin, Endothelial be addressed.
cell
594
PLASMA TM IN RABBIT DIC MODEL
Vol. 65, Nos. 415
INTRODUCTION Thrombomodulin (TM) is a glycoprotein located on the surface of vascular endothelial cells and functions as an anticoagulant factor by accelerating thrombin-catalyzed activation of protein C (l-3). TM is distributed in large amounts in lung, placenta and and trace amounts are also present in plasma and urine of kidney, humans as a soluble form (4,5). It was recently reported that the soluble TM antigen level was elevated in plasma of patients with diseases such as diabetes mellitus with microangiopathy, disseminated intravascular coagulation (DIG) and systemic lupus and it was suggested that this elevation of erythematosus (6-g), plasma TM antigen was a reflection of damage to endothelial cells (6-8,lO). DIC is a clinical syndrome which is frequently observed in cases of sepsis, solid tumor, acute leukemia, etc., though the mechanism of its development has not been established yet (11). which is released from the cell wall of gram-negative Endotoxin, bacteria, is thought to play a role in the pathogenesis of DIC associated with septicemia. Endotoxin may trigger hypercoagulability in circulating blood by causing endothelial damage and/or by stimulating the procoagulant activity accompanied with down-regulation of TM in endothelial cells (12-14). It has been accepted that endothelial cell damage in endotoxemia is caused by leukocytes activated by endotoxin (15-18). Changes in TM antigen in plasma during the induction of DIG and the influence of therapeutic measures have not previously been investigated. Heparin is an effective therapeutic drug for DIC. In this work, we measured the TM antigen level in plasma of rabbits infused with endotoxin or endotoxin plus heparin, and conducted a vitro studies with cultured endothelial cells in order to cast light on the results obtained & a. MATERIALS AND METHODS Reagents were purchased from Wako Pure Chemical Materials: unless otherwise indicated. Endotoxin Industries, Osaka, Japan, N-formyl-methionyl-leucyl-phenylalanine serotype 012:B8), (LPS, (FMLP) and Lubrol-PX were purchased from Sigma Co., St. Louis, MO. Fibrinogen Determination Set B was purchased from Kokusai Human thrombin was kindly donated by Shiyaku Co., Kobe, Japan. Protein C was Durified from human Green Cross Co., Osaka, Japan. Suzuki et al. plasma by the method described by (19). reagent were Nitrocellulose membrane, Affigel-10 and Enzymobeads purchased from Bio-Rad, Richmond. CA. adjuvant was purchased from Difco Lab., from ICN Biomedicals Inc., Costa Mesa, from Vector Lab. Inc., Burlingame, CA. TM was Purification of rabbit TM and preparation of antibody: ourified from rabbit lung according to the method described by TM extracted from lung by treatment Esmon et al. (20). Briefly. with detergent (0.5 % Lubrol-P,X) was subjected to column chromatography on DIP-thrombin agarose 2 times, and a 94 kd protein was obtained which was detected as a single band by SDS-polyacrylamide gel Coomassie brilliant blue staining after
Vol. 65, Nos. 45
PLASMA TM IN RABBIT DIC MODEL
595
electrophoresis. Cofactor activity of the preparation to activate protein C was 2521.7 U/mg protein as measured by the The purified rabbit TM (1 mg) method previously described (5). was emulsified with an equal volume of Bacto complete Freund’s adjuvant and subcutaneously injected into male sheep. One month later, 0.5 mg of TM in booster emulsion was injected in the same way. The anti-serum was collected at 7 days after the boosting by the method of and sheep anti-rabbit TM IgG was isolated Ishikawa et al. (21). The cofactor activity of 4 fig of purified rabbit TM was neutralized by addition of 15 fig of IgG. For enzyme immunoassay of TM, 1 ml of 4 mg/ml IgG was labeled with 1 ml of 4 mg/ml horseradish peroxidase (HRP, Boehringer Nannheim Yamanouchi Co., Tokyo, Japan) according to the method described previously (5). Male Japanese white rabbits weighing 2.3 - 2.5 Animal treatment: injection of pentobarbital (25 mg/kg kg were anesthetized by i.v. body weight) and then endotoxin dissolved in 2 ml of saline was infused via a marginal ear vein at a dose of 0.1 mg/kg body weight during 1 hr. At 12 hrs after the first infusion, a second infusion of endotoxin was performed in the same way. Heparin at a dose of 2,000 U/kg was infused from the other ear beginning at 30 and given for 1.5 hrs. min before each infusion of endotoxin, Blood was collected into a tube containing 3.8 % sodium citrate (l/10 volume of the expected volume of blood) from a catheter inserted into an ear vein at 30 min before and at 0, 6 and 12 hrs after the first infusion and 0, 2, 4 and 6 hrs after the second infusion. Measurement of platelet count in blood and fibrinogen contents in plasma: Blood collected in a tube containing 3.8% sodium citrate was diluted to l/50,000 with saline, and then counted by using a Sysmex Microcell Counter F-800 (Toa Medical Electronics co. Tokyo, Fibrinogen content in plasma was measured Japan). according to the description in the manual of Fibrinogen Determination Set B. Treatment of endothelial cells: Human umbilical vein endothelial cells were harvested and cultured to confluence in DMEM supplemented 20% fetal serum (FCS, Hyclone with calf Laboratories, Logan, UT.), 72 U/ml penicillin and 0.096 mg/ml streptomycin according to the method described previously (10). Cells at the second or third passage were used for experi ents at 1 to 3 days after confluence. luent cells (2 x 10 $ cells) were incubated with 2 ,KC~ of Na ‘i”Cr04 (56 mCi/mg) in 1 ml of DMEM for 12 hrs at 37-C in a 5% 0 2 atmosphere according to the method of Varani et al. (22). The cells were washed 4 times with DMEM before being used for the experiments. The culture medium was replaced with 1 ml of fresh DMEM as the control medium, or either one of the conditioned media containing unstimulated leukocytes, 100 ng/ml endotoxin plus 1 LLMFMLP or the leukocytes stimulated with 100 ng/ml endotoxin plus 1fiM FMLP, At an appropriate time, the conditioned medium was removed by pipetting from the cell monolayers and the cell monolayers were solubilized with 0.5% Triton X-100 containing 10 mM benzamidine hydrochloride and 50 mM Tris-HCl (PH 7.5) for 1 hr at 4-C. then TM was extracted by centrifugation at 10,000 rpm for 3 min with an MRX150 (Tomy Co., Tokyo, Japan). TM antigens in the conditioned
596
PLASMA TM IN RABBIT DIC MODEL
Vol. 65, Nos. 4l5
medi I# and the cell extract were measured by enzyme immunoassay. Cr content in conditioned medium and cell extracts was The determined by counting with an autowell gamma counter. Enzyme-linked immunoassay (EIA) for TM antigen: For determination of rabbit TM, each well of a microtiter plate was coated with 200 .a1 of 10 fig/ml anti-rabbit TM IgG dissolved in 50 mM carbonate buffer (pH 9.5) overnight at 4-C. The plate was washed with 10 mM phosphate-buffered saline pH 7.4) (PBS, containing 0.05% Tween 20 (washing buffer) and kept in PBS containing 0.2% BSA and 0.05% Tween 20 for 2 hrs at room temperature. After discarding the buffer, 200 ,nl of test specimen diluted with PBS containing 0.2% BSA, 0.05% Tween 20, 0.5% sucrose and 1 mM EDTA was placed in each well and the plate was incubated for 1.5 hrs at room temperature. The plate was washed, HRP-labeled anti-TM IgG (200 ~1, 3.3 ,ug/ml) was placed in each well, and incubation was continued for 1 hr at room temperature. The plate was washed again, then 200 ~1 of substrate solution containing 0.015% H 0 and 1.2 mg/ml ophenylenediamine in 0.5 M citrate buffer was placed in ?Pi? 5.0) each well. The plate was incubated for 1 hr at room temperature and the reaction was terminated with 50 ~1 of 3 M H S04. Absorbance at 490 nm was measured. Purified rabbit lung T& was used as a standard preparation. Determination of human TM was performed according to the method described previously (5). Western blot: Plasma or purified TM from rabbits was subjected to SDS-polyacrylamide gel electrophoresis (PAGE) under reducing conditions and transferred to a nitrocellulose membrane as described previously (10). The membrane was incubated with 25 ml of 20 mM Tris-buffered saline (pH 7.4) containing 60 fig/ml sheep anti-rabbit TM IgG. 1 % skim milk and 0.05% Tween 20 for 2 hrs at room temperature and then washed 4 times with 200 ml of Trisbuffered saline (pH 7.4) containing 0.05% Tween 20. TM-bound IgG was detected using a Vectastain ABC Kit composed of biotinylated anti-sheep IgG and HRP-conjugated streptavidin at room temperature. RESULTS 0.6 r
Evaluation of the enzyme immunoassay for TM of rabbit: The enzyme immunoassay (EIA) for rabbit TM antigen was constructed using anti-rabbit TM IgG. The titsheep ration curves of TM in buffer or in normal plasma were linear within the range from 2 to 14 ng/ml (Fig. that the ability 11, and indicated this EIA to measure TM antigen of in plasma was not interfered with by plasma components. The coefficient of variation at 2 ng/ml TM was 3.4%. With the use of this the normal level of TM in method, plasma of rabbits was found to be + 0.64 fig/ml. n=18. 1.64
0
J;
2 TM
4
6
6
IO
12
14
concentration (nghl)
Titration curves of Fig.1. rabbit thrombomodulin (TM) by enzyme immunoassay. 0 , purified rabbit lung TM in buffer: ??, purified rabbit lung TM in plasma. rabbit
Vol. 65, Nos. 45
PLASMA TM IN RABBIT DIC MODEL
597
TM level in plasma of endotoxin-induced DIC: A single infusion of endotoxin in the range from 0.5 to 1.0 mg/kg/hr induced a marked decrease in platelet count of all of the treated rabbits, though the fibrinogen level in plasma did not decrease. Some rabbits died within 3 hrs after infusion. Therefore, endotoxin was infused twice into rabbits at a dose of 100 .og/kg/hr with a 12 hr interval. The mortality of the animals was reduced to zero by this procedure. Platelet count decreased to 50% after the first infusion, whereas fibrinogen and TM levels in plasma were not influenced within 12 hrs. At 2 - 6 hrs after the second infusion, the platelets and fibrinogen levels decreased to 20% and 75%, respectively, and TM level increased to 150% of the control values (Fig. 2). These results indicated that the TM level in plasma increased in parallel with generation of DIG. On the other hand, continuous infusion of heparin at the dose of 2,000 U/kg/l.5 hrs, beginning from 30 min before endotoxin infusion, completely prevented both the decrease in fibrinogen level and the increase in TM level, and slightly ameliorated the decrease in platelet count (Fig. 2). IA)
(B)
Change In platolot count
Chmp
(Cl in flbrlnogm lovd
Change In plasma
TM level
0 6
12 Tim*
16 thr)
24
0
6
12 Tim6
16 br)
24
6
6
12 Tlmo
16
24
thr)
Fig. 2. Changes in platelet count and plasma fibrinogen and thromboaodulin (TM) levels in rabbits infused with endotoxin (0) or endotoxin plus heparin (0). ( VA) was infused via a Endotoxin marginal ear vein at a dose of 0.1 mg/kg during 1 hr twice with a 12 hr interval. Heparin ( &:I was infused twice via a vein in the other ear at a dose of 2,000 U/kg/l .5 hrs from 30 min before each endotoxin infusion. Citrated blood was collected in the times described in the figure, and platelet count in blood, plasma fibrinogen and TM levels were measured by the methods described in Methods. Each point represents the mean + S.D. from 3 rabbits.
598
PLASMA TM IN RABBIT DIC MODEL
Vol. 65, Nos. 415
The components of plasma TM antigen were investigated in rabbits infused with endotoxin or endotoxin plus heparin by immunoblotting. TM antigen in plasma of untreated control rabbits consisted of 3 subspecies of 94, 83 and 51 kd. An increase in the 83 kd subspecies and the appearance of new subspecies, 76 and 48 kd, were observed concomitantly with a marked decrease in the 94 kd subspecies in plasma of rabbits infused with endotoxin, and the major subspecies were the 83 and 76 kd subspecies. The treatment with heparin markedly reduced the appearance of the new 76 kd subspecies caused by the infusion of endotoxin and sightly reduced the disappearance of the 94 kd subspecies. The major component of plasma TM antigen in rabbits infused with endotoxin plus heparin was the 83 kd subspecies (Fig. 3).
/ -
12
94kD 83kD 76kD
34
Fig. 3. Comparison of thrombomodulin (TM) subspecies in plasma of control, endotoxin-infused or endotoxin plus heparin-infused rabbits. Citrated blood was collected at 6 hrs after the second 1 hr i.v. infusion (12 hr interval) of endotoxin via an ear vein at a dose of 0.1 mg/kg. Heparin was infused from 30 min before each endotoxin infusion, at a dose of 2,000 U/kg/l.5 hrs. Plasma (0.3 ,cl) from rabbits was subjected to SDS-PAGE, and Western blotting was performed as described in Methods. Lane 1, purified lung TM (250 ng); lane 2, control rabbit; lane 3, endotoxin-infused plus heparin-infused rabbit. rabbit; lane 4, endotoxin The effect of Release of TM from cultured endothelial cells: leukocytes stimulated with endotoxin plus FMLP on release of TM cultured endothelial cells was investigated, and antigen fro into the conditioned medium was release of 51 Cr from the cells measured as an indicator of cell injury (Table 1). In the control level and about 8 no significant alteration of TM ep tigen cells, Cr incorporated in the total - 9 % spontaneous release of cells were observed in the conditioned medium during incubation endotoxin plus FMLP Treatment of the cells with for 3 or 6 hrs. induced a decrease in TM antigen level in the ce+...s without Cr in the noticeable increase in either the antigen level or
Vol. 65, Nos. 415
599
PLASMA TM IN RABBIT DIC MODEL
medium. This fact indicated that endotoxin and FMLP induced down-regulation of cellular TM without cell injury. When the cells were treated with a suspension of stimulated leukocytes which had been prekr eated with endotoxin and FMLP, apparent increases of TM and Cr were found in the medium together with a significant decrease in cellular TM. The TM antigen level (0.7 ng of TM at 3 hrs per well) detected in the medium coincided with the difference in the cellular levels between treatment with the endotoxin plus FMLP (4.36 ng of TM at 3 hrs per well) and the stimulated leukocytes (3.64 ng of TM in 3 hrs per well). Furthermore, the amount of TM antigen released into the medium 16.1 % of the total TM in the cells, which was was compatible Table 1. Release of thrombomodulin (TM) antigen endothelial cells into conditioned medium and cell by leukocytes stimulated with endotoxin and FMLP Control Cellular TM’) 3 hrs (% of total)2) 6 hrs (96 of Medium TM 3 hrs (% of total)2) 6 hrs (% of totti)2) 51Cr-release 3 hrs (% cell injury)5) 6 hrs (% cell injury)5)
Endotoxin -FMLP
5.35
+ 0.18
4.36
+ 0.16
5.39
+ 0.21
4.21
+ 0.23
0.02
+ 0.01
0.02
+ 0.02
0.04
+ 0.03
0.04
+ 0.02
7.8
+ 0.2
6.2
+ 1.2
9.2
+ 1.5
8.3
+ 0.6
from cultured injury caused Stimulated leukocytes 3.64 + 0.24 (8319) 3.04 + 0.23 (74,3) 0.70 + 0.10 (16:l) 1.05 + 0.11 (25.7) 25.5 + 3.0 (19T2) 38.2 + 2.4 (31T9)
Endothelial cell monolayers (2 x lo4 cells) incorporating 51Cr were washed and then incubated at 37-C in a 5 % CO atmosphere with unstimulated leukocytes (2 x lo5 cells), enkotoxin (100 ng/ml) plus FMLP (l,~clM) or leukocytes stimulated with endotoxin plus FMLP. After 3 or 6 hrs incubation, the conditioned5piCefi;; and the cells were collected, and TM antigen level and the cells and the conditioned medium were measured. Results represent the means + S.D. of cells from 3 different human umbilical cords. FMIP is the abbreviation of N-formylmethionyl-le cyl-phenylalanine. 1) ng/2 x 10 Y cells 2) % of total = (TM level in the medium or the cells/TM levels in the cells and the medium) x 100 4”; ~g~~o~~:~:os~~~m~~~~:porated in thg cells 5) % injury = [( Cr in the medium r spontaneously in edium)/(total :?Cr incorporated cells the - 51pCr spontaneously released in the x 100
released in the medium)]
PLASMA
600
TM IN RABBIT
DIG MODEL
Vol. 65, Nos. 415
effect of heparin against thrombomodulin (TM) from cultured endothelial cells treated with stimulated leukocyte suspension (A) or the medium separated sf;; stimulated leukocytes (B). Cell monolayers incorporating were prepared according to the method described in “Methods”. Leukocytes stimulated with endotoxin (100 ng/ml) plus FMLP (1 ,aM) were immediately added to the cell monolayers in the presence of various concentrations of heparin in culture medium Leukocytes stimulated with endotoxin plus FMLP for 2 hrs at (A). room temperature were centrifuged and the resulting supernatant was added to the cell monolayers in the presence of heparin in the culture medium. The conditioned medium was collected at 3 of stimulated leukocytes or the medium, ;;s, ;;t~nt,t,h,eena~dW Cr count in the conditioned medium were measured. Each point represents the mean + S.D. of the cells from 3 different umbilical veins. Values of the % injury and % of total were calculated in the same way as in Table 1. ;;;-
‘th 51 Cr
Protective 51. Cr releases
the percentage of the cell injury (19.2 %) calculated from release into the medium and its total amount in the cells. The parallelism between the relative antigen level released into the medium and the relative extent of cell injury was observed These results suggested that regardless of the incubation period. an increase in TM antigen level in the conditioned medium reflected the injury of cultured endothelial cells. The cells treated with unstimulated leukocytes did not exhibit significant changes in TM antigen level in the cells or the medium, and cell Heparin injury was also not detectable (data not shown). treatment of the cells prevented the release of TM antigen into the the conditioned medium as well as the cell injury caused by stimulated leukocytes in a dose-dependent manner in the range between 0.01 and 1.0 U/ml of heparin (Fig. 4A). Also, when the medium separated from cells were treated with the supernatant the stimulated leukocyte suspension, the release of TM antigen The into the medium and cell injury were observed (Fig. 4B). antigen release and the cell injury caused by treatment with the
Vol. 65, Nos. 415
PLASMA TM IN RABBIT DIC MODEL
601
treatment with medium were also prevented by simultaneous heparin in a dose-dependent manner (Fig. 4B). The preventive action of heparin against antigen release and cell injury was more effective on cells treated with the medium as compared with cells treated with the stimulated leukocyte suspension. DISCUSSION TM antigen level in rabbit plasma was 1.64 + 0.64 fig/ml, 47 times higher than that of human plasma (35.2 + 8.3 ng/ml) (5). Therefore, rabbit plasma could be directly subjected to SDS-PAGE to detect molecular subspecies of plasma TM antigen. Human TM antigen in plasma consisted of 6 subspecies (5,101, whereas 3 subspecies were observed in rabbit plasma. Rabbit TM purified from lung in the present work showed an apparent molecular weight of 94 kd under reducing conditions. Bourin et al. (23) reported that heterogeneous TM molecules including 110 - 100 kd species were present in a TM preparation purified from rabbit lung, and the high-molecular subspecies contained chondroitin sulfate-type glycosaminoglycan. Furthermore, it was reported that the molecular weight of human or rabbit thrombomodulin containing glycosaminoglycan was reduced by treatment with chondroitinase It was supposed that the purified 94 kd TM is a ABC (24,25). and that plasma TM antigen degraded form of the native molecule, is composed of various heterogeneous subspecies, with or without glycosaminoglycan and/or with or without degradation of the peptide moiety. Plasma TM levels of rabbits were increased to 1.5 times the control level by infusion of endotoxin and the increase was accompanied with generation of DIG, as confirmed by the decreases in platelet count and plasma fibrinogen level. A marked decrease in the 94 kd subspecies of TM, together with both increase in 83 kd subspecies and the appearance of the new subspecies of 76 and 48 kd, was found in rabbits in which DIC was being induced by the infusion of endotoxin. It is probable that the new subspecies were produced by degradation of high-molecular-weight TM by glycosidases and/or proteases. Similarly, an increase in smaller-molecular forms of TM in plasma in clinical cases of DIC was recently suggested (8). Heparin is an effective drug against the induction of DIC. Heparin treatment of endotoxin-induced DIC rabbits prevented both the increase in plasma TM levels and the degradation of plasma TM subspecies, and produced an improvement of DIG as judged from the platelet count and fibrinogen level. It has been reported that the increase in plasma TM level in clinical cases was reflection of endothelial cell damage (5,8,10), which was also indicated in the present work (Table 1). These facts suggested that heparin not only prevented the development of DIC. but also protected the endothelial cells against injury by leukocytes stimulated with endotoxin. It has been postulated that the therapeutic effectiveness of heparin on DIG is derived from its inhibiting action upon activated coagulation factors in concert with antithrombin III (AT-III). Howevsr. we observed that, when titrated blood was incubated with H-heparin, the labeled heparin was distributed
602
PLASMA TM IN RABBIT DIC MODEL
Vol. 65, Nos. 45
mainly to blood cells (37.7%) and plasma proteins other than ATIII (45.7%). and only 16.6% of the total amount formed a complex with AT-III (unpublished data). Furthermore, it has been reported that heparin was concentrated in endothelium in rats after its injection by the intravenous, intraperitoneal, or subcutaneous route (26.27). These facts suggest that intravenously injected heparin not only binds to AT-III in plasma, but also is distributed to plasma components other than AT-III, and also binds to blood cells and endothelial cells. The current results in vitro indicated that heparin prevented both the TM release into the medium and the cell injury caused by treatment with stimulated leukocytes or the medium separated from the stimulated leukocytes. This preventive effect of heparin was observed in the absence of AT-III. Therefore, it is probable that heparin acts not only as an anticoagulant by formation of a complex with ATIII in plasma during development of DIC, but also directly protects endothelial cells against injury induced by stimulated leukocytes. Endothelial cells are injured by proteases including elastase and active oxygens including superoxide released from endotoxin-stimulated leukocytes (15-18). In order to elucidate the mechanism of its preventive effect, the action of heparin was compared between cells treated with the stimulated leukocyte suspension and those treated with the medium separated from the stimulated leukocytes. When the cells were treated by the former method, they would be injured by both proteases and active oxygens released from the leukocytes, but when the cells were treated by the latter method, they would be injured mainly by proteases, since the half lives of active oxygens are very short and these species should be absent in the medium. Our results showed that the preventive action of heparin against cell injury and antigen release was more effective on the cells treated with the medium than on the cells treated with the stimulated leukocyte suspension. It appears that heparin acted as a proteinase inhibitor rather than as a scavenger for active oxygens. We hypothesize that plasma TM containing degraded subspecies was released from endothelial cells injured by stimulated leukocytes as well as by disturbed microcirculation during the induction of DIC. It is also considered that the preventive effect of heparin against both TM release and cell injury is a consequence of not only its anticoagulant activity but also its direct protection of endothelial cells against the action of leukocyte proteases. REFERENCES 1. 2. 3. 4.
ESMON, C.T. and OWEN, W.G. Identification of an endothelial cell cofactor for thrombin-catalyzed activation of protein Proc.Nat1.Acad.Sci.U.S.A. 78, 2249-2252, 1981. C. ESMON, C.T. The roles of protein C and thrombomodulin in J.Biol.Chem. 264, 4743regulation of blood coagulation. 4746, 1989. Structure and function of DITTMAN, W.A. and MAJERUS, P.W. Blood 75, 329thrombomodulin: A natural anticoagulant. 336, 1990. ISHII, H. and MAJERUS, P.W. Thrombomodulin is present in
Vol. 65, Nos. 415
5.
6. 7.
8.
9.
10. 11. 12. 13.
14.
15.
16.
17. 18.
19.
PLASMA TM IN RABBIT DIC MODEL
603
human plasma and urine. J.Clin.Invest. 76. 2178-2181, 1985. ISHII, H., NAKANO, M., TSUBOUCHI, J., ISHIKAWA.T., UCHIYAMA, H HIRAISHI. S., TAHARA, C., MIYAJIMA, Y. and KAZAMA, M. Esiablishment of enzyme immunoassay of human thrombomodulin in plasma and urine using monoclonal antibodies. Thromb.Haemostas. 63, 157-162. 1990. TANAKA, A., ISHII. H., HIRAISHI, S., KAZAMA, M. and MAEZAWA, H. Increased thrombomodulin values in plasma of diabetic men with microangiopathy. Clin.Chem. 37. 269-272. 1991. IWASHIMA, Y., SATO, T., WATANABE, K., OOSHIMA. E., HIRAISHI, S ., ISHII, H.. KAZAMA, M. and MAKINO, I. Elevation of plasma thrombomodulin level in Diabetic patients with early diabetic nephropathy. Diabetes 39, 983-988, 1990. TAKANO, S., KIMURA, S., OHDAMA, S. and AOKI, N. Plasma thrombomodulin in health and diseases. Blood 76, 2024-2029, 1990. GIDDINGS, J.C., C0LES.P. and WILLIAMS, B.D. Comparison of thrombomodulin and von Willebrand factor antigen in human plasma in various diseases. Thromb.Haemostas. 62, 276 (Abstr.), 1989. ISHII, H.. UCHIYAMA, H. and KAZAMA, M. Soluble thrombomodulin antigen in conditioned medium is increased by damage of endothelial cells. Thromb.Haeaostas. 65, 618-623, 1991. MIZOCK, B. Septic shock. A metabolic perspective. Arch.Intern.Med. 144. 579-585. 1984. MORRISON, D.C.. ULEVITCH, R.J. The effect of bacterial endotoxins on host mediation systems. A review. Am.J.Pathol. 93, 526-617, 1978. COLUCCI. M., BALCONI. G., LORENZET, R., PIETRA. A., LOCATI, D DONATI, M.B. and SEMERARO, N. Cultured human endothelial ceils genarate tissue factor in response to endoxoxin. J.Clin.Invest. 71. 1893-1896, 1983. MOORE, K.L., ANDREOLI, S.P., ESMON, N.L., ESMON, C.T. and BANG, N.U. Endotoxin enhances tissue factor and supresses thrombomodulin expression of human endotherium in vitro. J.Cli.Invest. 79, 124-130, 1987. TILL, G.O., JOHNSON, K.J., KUNKEL. R., and WARD, P.A. Intravascular activation of complement and acute lung injury. Dependency on neutrophils and toxic oxygen metabolites. J.Clin.Invest. 69, 1126-1135, 1982. SACKS, T.. MOLDOW, C.F., CRADDOCK, P.R., BOWERS, T.K., and JACOB, H.S. Oxygen radicals mediate endothelial cell damage by complement-stimulated granulocytes. J.Clin.Invest. 61, 1161-1167, 1978. HARLAN, J.M., KILLEN, P.D., HARKER, L.A., and STRIKER, G.E. Neutrophil-mediated endothelial injury in vitro. J.Clin.Invest. 68. 1394-1403, 1981, SMEDLY, L.A., TONNESEN. M.G.. SANDHAUS, R.A., HASLETT, C., GUTHRIE, L.A., JOHNSTON, R.B., HENSON P.M., and WORTHEN, G.S. Neutrophil-mediated injury to endothelial cells. Enhancement by endotoxin and essential role of neutrophil elastase. J.Clin.Invest. 77, 1233-1243, 1986. SUZUKI, K.. STENFLO, J.. DAHLBACK, B. and TEODORSSON, B. Inactivation of human coagulation factor V by activated
604
20. 21.
22.
23.
24.
25.
26. 27.
PLASMA
TM IN RABBIT
DIC MODEL
Vol. 65, Nos. 415
protein C. J.Biol.Chem. 258, 1914-1920, 1983. OWEN, W.G. and ESMON, C.T. Isolation of a ESMON, N,L., membrane-bound cofactor for thrombin-catalized activation of protein C. J.Biol.Chem. 257, 859-864, 1982. ISHIKAWA, E., IMAGAWA, M., HASHIDA, S., YOSHITAKE, S., HAMAGUCHI, Y. and UENO, T. Enzyme-labeling of antibodies and their fragments for enzyme immunoassay and immunohistochemical staining. J.Immunoassay 4, 209-327, 1983. VARANI, J., BENDELOW, M.J.. SEALEY. D.E., KUNKEL. S.L., factor GANNON, D.E., RYAN, U.S. and WARD, P.A. Tumor necrosis enhances susceptibility of vascular endothelial cells Lab.Invest. 59, 292-295, 1988. neutrophil-mediated killing. BOURIN, M.C., OHLIN, A.K., LANE, D.A., STENFLO, J. and anticoagulant activities LINDAHL, U. Relationship between and properties of rabbit thrombomodulin. polyanionic 263, 8044-8052, 1988. J.Biol.Chem. PARKINSON, J.F., GRINNELL, B.W., MOORE, R.E., HOSKINS, J., VLAHOS, C.J. and BANG, N.U. Stable expression of a secretable deletion mutant of recombinant human thrombomodulin in mammalian cells. J.Biol.Chem. 265, 12602-12610, 1990. BOURIN. M.C. and LINDAHL, U. Functional role of the polyproteoglycan. saccharide component of rabbit thrombomodulin Effects of inactivation of thrombin by antithrombin, cleavage of fibrinogen by thrombin and thrombin-catalized activation of Factor V. Bi0chem.J. 270, 419-425, 1990. HIEBERT, L.M. and JAQUES, L.B. The observation of heparin on endothelium after injection. Thromb.Res. 8, 195-204, 1976. GLIMELIUS, B.. BUSCH, C. and HOOK, M. Binding of heparin on cells. the surf ace of cultured human endothelial Thromb. Res. 12, 773-782, 1978.
CHANGES IN PLASMA THROMBOMODULIN ANTIGEN IN RABBIT DEVELOPING ENDOTOXIN-INDUCED DISSEMINATED INTRAVASCULAR COAGULATION AND THE EFFECT OF HEPARIN Hiroyuki UchiyaMa. Hisaya Ohtani. Sayuri Hiraishi. Shuichi Horie, Hidemi Ishii* and Hutsuyoshi KazaMa Departaent of Clinical Biochemistry, Faculty Teikyo University, 1091-l Suarashi, Sciences, 199-01. Japan. (Received
11.9.1991;
accepted
in revised form 25.12.1991
of Pharmaceutical Tsukui, Kanagawa
by Editor A. Takada)
ABSTRACT Soluble thrombomodulin (TM) antigen level was 1.64 + male0.64 fig/ml (n=18, mean + S.D.) in plasma of normal enzyme immunoassay, and the rabbits as measured by of subspecies of 94. 83 and 51 kd. antigen consisted intravascular coagulation (DIG) was When disseminated into intravenous infusion of endotoxin induced by the TM antigen level in plasma was elevated to rabbits, about 1.5 times of the control value, and an increase in new the 83 kd subspecies as well as the appearance of 76 and 48 kd was observed subspecies of concomitantly with disappearance of the 94 kd subspecies in plasma. Elevation of the antigen level and disappearance of the 94 kd subspecies caused by infusion of endotoxin were reduced by simultaneous infusion of heparin. Addition of with leukocytes stimulated endotoxin plus FMLP to cultured endothelial cells induced release of TM antigen medium accompanying cell injury as measured by ‘PCrthTelease which was prevented by treatment with heparin. It WAS suggested that the increase in plasma TM antigen level in parallel with the generation of DIC reflected endothelial injury of rabbits, and that the elevation of TM antigen and the endothelial cell injury were prevented by heparin treatment. Key Words: DIG. Thrombomodulin. ‘To whom correspondence should
Heparin, Endothelial be addressed.
cell
594
PLASMA TM IN RABBIT DIC MODEL
Vol. 65, Nos. 415
INTRODUCTION Thrombomodulin (TM) is a glycoprotein located on the surface of vascular endothelial cells and functions as an anticoagulant factor by accelerating thrombin-catalyzed activation of protein C (l-3). TM is distributed in large amounts in lung, placenta and and trace amounts are also present in plasma and urine of kidney, humans as a soluble form (4,5). It was recently reported that the soluble TM antigen level was elevated in plasma of patients with diseases such as diabetes mellitus with microangiopathy, disseminated intravascular coagulation (DIG) and systemic lupus and it was suggested that this elevation of erythematosus (6-g), plasma TM antigen was a reflection of damage to endothelial cells (6-8,lO). DIC is a clinical syndrome which is frequently observed in cases of sepsis, solid tumor, acute leukemia, etc., though the mechanism of its development has not been established yet (11). which is released from the cell wall of gram-negative Endotoxin, bacteria, is thought to play a role in the pathogenesis of DIC associated with septicemia. Endotoxin may trigger hypercoagulability in circulating blood by causing endothelial damage and/or by stimulating the procoagulant activity accompanied with down-regulation of TM in endothelial cells (12-14). It has been accepted that endothelial cell damage in endotoxemia is caused by leukocytes activated by endotoxin (15-18). Changes in TM antigen in plasma during the induction of DIG and the influence of therapeutic measures have not previously been investigated. Heparin is an effective therapeutic drug for DIC. In this work, we measured the TM antigen level in plasma of rabbits infused with endotoxin or endotoxin plus heparin, and conducted a vitro studies with cultured endothelial cells in order to cast light on the results obtained & a. MATERIALS AND METHODS Reagents were purchased from Wako Pure Chemical Materials: unless otherwise indicated. Endotoxin Industries, Osaka, Japan, N-formyl-methionyl-leucyl-phenylalanine serotype 012:B8), (LPS, (FMLP) and Lubrol-PX were purchased from Sigma Co., St. Louis, MO. Fibrinogen Determination Set B was purchased from Kokusai Human thrombin was kindly donated by Shiyaku Co., Kobe, Japan. Protein C was Durified from human Green Cross Co., Osaka, Japan. Suzuki et al. plasma by the method described by (19). reagent were Nitrocellulose membrane, Affigel-10 and Enzymobeads purchased from Bio-Rad, Richmond. CA. adjuvant was purchased from Difco Lab., from ICN Biomedicals Inc., Costa Mesa, from Vector Lab. Inc., Burlingame, CA. TM was Purification of rabbit TM and preparation of antibody: ourified from rabbit lung according to the method described by TM extracted from lung by treatment Esmon et al. (20). Briefly. with detergent (0.5 % Lubrol-P,X) was subjected to column chromatography on DIP-thrombin agarose 2 times, and a 94 kd protein was obtained which was detected as a single band by SDS-polyacrylamide gel Coomassie brilliant blue staining after
Vol. 65, Nos. 45
PLASMA TM IN RABBIT DIC MODEL
595
electrophoresis. Cofactor activity of the preparation to activate protein C was 2521.7 U/mg protein as measured by the The purified rabbit TM (1 mg) method previously described (5). was emulsified with an equal volume of Bacto complete Freund’s adjuvant and subcutaneously injected into male sheep. One month later, 0.5 mg of TM in booster emulsion was injected in the same way. The anti-serum was collected at 7 days after the boosting by the method of and sheep anti-rabbit TM IgG was isolated Ishikawa et al. (21). The cofactor activity of 4 fig of purified rabbit TM was neutralized by addition of 15 fig of IgG. For enzyme immunoassay of TM, 1 ml of 4 mg/ml IgG was labeled with 1 ml of 4 mg/ml horseradish peroxidase (HRP, Boehringer Nannheim Yamanouchi Co., Tokyo, Japan) according to the method described previously (5). Male Japanese white rabbits weighing 2.3 - 2.5 Animal treatment: injection of pentobarbital (25 mg/kg kg were anesthetized by i.v. body weight) and then endotoxin dissolved in 2 ml of saline was infused via a marginal ear vein at a dose of 0.1 mg/kg body weight during 1 hr. At 12 hrs after the first infusion, a second infusion of endotoxin was performed in the same way. Heparin at a dose of 2,000 U/kg was infused from the other ear beginning at 30 and given for 1.5 hrs. min before each infusion of endotoxin, Blood was collected into a tube containing 3.8 % sodium citrate (l/10 volume of the expected volume of blood) from a catheter inserted into an ear vein at 30 min before and at 0, 6 and 12 hrs after the first infusion and 0, 2, 4 and 6 hrs after the second infusion. Measurement of platelet count in blood and fibrinogen contents in plasma: Blood collected in a tube containing 3.8% sodium citrate was diluted to l/50,000 with saline, and then counted by using a Sysmex Microcell Counter F-800 (Toa Medical Electronics co. Tokyo, Fibrinogen content in plasma was measured Japan). according to the description in the manual of Fibrinogen Determination Set B. Treatment of endothelial cells: Human umbilical vein endothelial cells were harvested and cultured to confluence in DMEM supplemented 20% fetal serum (FCS, Hyclone with calf Laboratories, Logan, UT.), 72 U/ml penicillin and 0.096 mg/ml streptomycin according to the method described previously (10). Cells at the second or third passage were used for experi ents at 1 to 3 days after confluence. luent cells (2 x 10 $ cells) were incubated with 2 ,KC~ of Na ‘i”Cr04 (56 mCi/mg) in 1 ml of DMEM for 12 hrs at 37-C in a 5% 0 2 atmosphere according to the method of Varani et al. (22). The cells were washed 4 times with DMEM before being used for the experiments. The culture medium was replaced with 1 ml of fresh DMEM as the control medium, or either one of the conditioned media containing unstimulated leukocytes, 100 ng/ml endotoxin plus 1 LLMFMLP or the leukocytes stimulated with 100 ng/ml endotoxin plus 1fiM FMLP, At an appropriate time, the conditioned medium was removed by pipetting from the cell monolayers and the cell monolayers were solubilized with 0.5% Triton X-100 containing 10 mM benzamidine hydrochloride and 50 mM Tris-HCl (PH 7.5) for 1 hr at 4-C. then TM was extracted by centrifugation at 10,000 rpm for 3 min with an MRX150 (Tomy Co., Tokyo, Japan). TM antigens in the conditioned
596
PLASMA TM IN RABBIT DIC MODEL
Vol. 65, Nos. 4l5
medi I# and the cell extract were measured by enzyme immunoassay. Cr content in conditioned medium and cell extracts was The determined by counting with an autowell gamma counter. Enzyme-linked immunoassay (EIA) for TM antigen: For determination of rabbit TM, each well of a microtiter plate was coated with 200 .a1 of 10 fig/ml anti-rabbit TM IgG dissolved in 50 mM carbonate buffer (pH 9.5) overnight at 4-C. The plate was washed with 10 mM phosphate-buffered saline pH 7.4) (PBS, containing 0.05% Tween 20 (washing buffer) and kept in PBS containing 0.2% BSA and 0.05% Tween 20 for 2 hrs at room temperature. After discarding the buffer, 200 ,nl of test specimen diluted with PBS containing 0.2% BSA, 0.05% Tween 20, 0.5% sucrose and 1 mM EDTA was placed in each well and the plate was incubated for 1.5 hrs at room temperature. The plate was washed, HRP-labeled anti-TM IgG (200 ~1, 3.3 ,ug/ml) was placed in each well, and incubation was continued for 1 hr at room temperature. The plate was washed again, then 200 ~1 of substrate solution containing 0.015% H 0 and 1.2 mg/ml ophenylenediamine in 0.5 M citrate buffer was placed in ?Pi? 5.0) each well. The plate was incubated for 1 hr at room temperature and the reaction was terminated with 50 ~1 of 3 M H S04. Absorbance at 490 nm was measured. Purified rabbit lung T& was used as a standard preparation. Determination of human TM was performed according to the method described previously (5). Western blot: Plasma or purified TM from rabbits was subjected to SDS-polyacrylamide gel electrophoresis (PAGE) under reducing conditions and transferred to a nitrocellulose membrane as described previously (10). The membrane was incubated with 25 ml of 20 mM Tris-buffered saline (pH 7.4) containing 60 fig/ml sheep anti-rabbit TM IgG. 1 % skim milk and 0.05% Tween 20 for 2 hrs at room temperature and then washed 4 times with 200 ml of Trisbuffered saline (pH 7.4) containing 0.05% Tween 20. TM-bound IgG was detected using a Vectastain ABC Kit composed of biotinylated anti-sheep IgG and HRP-conjugated streptavidin at room temperature. RESULTS 0.6 r
Evaluation of the enzyme immunoassay for TM of rabbit: The enzyme immunoassay (EIA) for rabbit TM antigen was constructed using anti-rabbit TM IgG. The titsheep ration curves of TM in buffer or in normal plasma were linear within the range from 2 to 14 ng/ml (Fig. that the ability 11, and indicated this EIA to measure TM antigen of in plasma was not interfered with by plasma components. The coefficient of variation at 2 ng/ml TM was 3.4%. With the use of this the normal level of TM in method, plasma of rabbits was found to be + 0.64 fig/ml. n=18. 1.64
0
J;
2 TM
4
6
6
IO
12
14
concentration (nghl)
Titration curves of Fig.1. rabbit thrombomodulin (TM) by enzyme immunoassay. 0 , purified rabbit lung TM in buffer: ??, purified rabbit lung TM in plasma. rabbit
Vol. 65, Nos. 45
PLASMA TM IN RABBIT DIC MODEL
597
TM level in plasma of endotoxin-induced DIC: A single infusion of endotoxin in the range from 0.5 to 1.0 mg/kg/hr induced a marked decrease in platelet count of all of the treated rabbits, though the fibrinogen level in plasma did not decrease. Some rabbits died within 3 hrs after infusion. Therefore, endotoxin was infused twice into rabbits at a dose of 100 .og/kg/hr with a 12 hr interval. The mortality of the animals was reduced to zero by this procedure. Platelet count decreased to 50% after the first infusion, whereas fibrinogen and TM levels in plasma were not influenced within 12 hrs. At 2 - 6 hrs after the second infusion, the platelets and fibrinogen levels decreased to 20% and 75%, respectively, and TM level increased to 150% of the control values (Fig. 2). These results indicated that the TM level in plasma increased in parallel with generation of DIG. On the other hand, continuous infusion of heparin at the dose of 2,000 U/kg/l.5 hrs, beginning from 30 min before endotoxin infusion, completely prevented both the decrease in fibrinogen level and the increase in TM level, and slightly ameliorated the decrease in platelet count (Fig. 2). IA)
(B)
Change In platolot count
Chmp
(Cl in flbrlnogm lovd
Change In plasma
TM level
0 6
12 Tim*
16 thr)
24
0
6
12 Tim6
16 br)
24
6
6
12 Tlmo
16
24
thr)
Fig. 2. Changes in platelet count and plasma fibrinogen and thromboaodulin (TM) levels in rabbits infused with endotoxin (0) or endotoxin plus heparin (0). ( VA) was infused via a Endotoxin marginal ear vein at a dose of 0.1 mg/kg during 1 hr twice with a 12 hr interval. Heparin ( &:I was infused twice via a vein in the other ear at a dose of 2,000 U/kg/l .5 hrs from 30 min before each endotoxin infusion. Citrated blood was collected in the times described in the figure, and platelet count in blood, plasma fibrinogen and TM levels were measured by the methods described in Methods. Each point represents the mean + S.D. from 3 rabbits.
598
PLASMA TM IN RABBIT DIC MODEL
Vol. 65, Nos. 415
The components of plasma TM antigen were investigated in rabbits infused with endotoxin or endotoxin plus heparin by immunoblotting. TM antigen in plasma of untreated control rabbits consisted of 3 subspecies of 94, 83 and 51 kd. An increase in the 83 kd subspecies and the appearance of new subspecies, 76 and 48 kd, were observed concomitantly with a marked decrease in the 94 kd subspecies in plasma of rabbits infused with endotoxin, and the major subspecies were the 83 and 76 kd subspecies. The treatment with heparin markedly reduced the appearance of the new 76 kd subspecies caused by the infusion of endotoxin and sightly reduced the disappearance of the 94 kd subspecies. The major component of plasma TM antigen in rabbits infused with endotoxin plus heparin was the 83 kd subspecies (Fig. 3).
/ -
12
94kD 83kD 76kD
34
Fig. 3. Comparison of thrombomodulin (TM) subspecies in plasma of control, endotoxin-infused or endotoxin plus heparin-infused rabbits. Citrated blood was collected at 6 hrs after the second 1 hr i.v. infusion (12 hr interval) of endotoxin via an ear vein at a dose of 0.1 mg/kg. Heparin was infused from 30 min before each endotoxin infusion, at a dose of 2,000 U/kg/l.5 hrs. Plasma (0.3 ,cl) from rabbits was subjected to SDS-PAGE, and Western blotting was performed as described in Methods. Lane 1, purified lung TM (250 ng); lane 2, control rabbit; lane 3, endotoxin-infused plus heparin-infused rabbit. rabbit; lane 4, endotoxin The effect of Release of TM from cultured endothelial cells: leukocytes stimulated with endotoxin plus FMLP on release of TM cultured endothelial cells was investigated, and antigen fro into the conditioned medium was release of 51 Cr from the cells measured as an indicator of cell injury (Table 1). In the control level and about 8 no significant alteration of TM ep tigen cells, Cr incorporated in the total - 9 % spontaneous release of cells were observed in the conditioned medium during incubation endotoxin plus FMLP Treatment of the cells with for 3 or 6 hrs. induced a decrease in TM antigen level in the ce+...s without Cr in the noticeable increase in either the antigen level or
Vol. 65, Nos. 415
599
PLASMA TM IN RABBIT DIC MODEL
medium. This fact indicated that endotoxin and FMLP induced down-regulation of cellular TM without cell injury. When the cells were treated with a suspension of stimulated leukocytes which had been prekr eated with endotoxin and FMLP, apparent increases of TM and Cr were found in the medium together with a significant decrease in cellular TM. The TM antigen level (0.7 ng of TM at 3 hrs per well) detected in the medium coincided with the difference in the cellular levels between treatment with the endotoxin plus FMLP (4.36 ng of TM at 3 hrs per well) and the stimulated leukocytes (3.64 ng of TM in 3 hrs per well). Furthermore, the amount of TM antigen released into the medium 16.1 % of the total TM in the cells, which was was compatible Table 1. Release of thrombomodulin (TM) antigen endothelial cells into conditioned medium and cell by leukocytes stimulated with endotoxin and FMLP Control Cellular TM’) 3 hrs (% of total)2) 6 hrs (96 of Medium TM 3 hrs (% of total)2) 6 hrs (% of totti)2) 51Cr-release 3 hrs (% cell injury)5) 6 hrs (% cell injury)5)
Endotoxin -FMLP
5.35
+ 0.18
4.36
+ 0.16
5.39
+ 0.21
4.21
+ 0.23
0.02
+ 0.01
0.02
+ 0.02
0.04
+ 0.03
0.04
+ 0.02
7.8
+ 0.2
6.2
+ 1.2
9.2
+ 1.5
8.3
+ 0.6
from cultured injury caused Stimulated leukocytes 3.64 + 0.24 (8319) 3.04 + 0.23 (74,3) 0.70 + 0.10 (16:l) 1.05 + 0.11 (25.7) 25.5 + 3.0 (19T2) 38.2 + 2.4 (31T9)
Endothelial cell monolayers (2 x lo4 cells) incorporating 51Cr were washed and then incubated at 37-C in a 5 % CO atmosphere with unstimulated leukocytes (2 x lo5 cells), enkotoxin (100 ng/ml) plus FMLP (l,~clM) or leukocytes stimulated with endotoxin plus FMLP. After 3 or 6 hrs incubation, the conditioned5piCefi;; and the cells were collected, and TM antigen level and the cells and the conditioned medium were measured. Results represent the means + S.D. of cells from 3 different human umbilical cords. FMIP is the abbreviation of N-formylmethionyl-le cyl-phenylalanine. 1) ng/2 x 10 Y cells 2) % of total = (TM level in the medium or the cells/TM levels in the cells and the medium) x 100 4”; ~g~~o~~:~:os~~~m~~~~:porated in thg cells 5) % injury = [( Cr in the medium r spontaneously in edium)/(total :?Cr incorporated cells the - 51pCr spontaneously released in the x 100
released in the medium)]
PLASMA
600
TM IN RABBIT
DIG MODEL
Vol. 65, Nos. 415
effect of heparin against thrombomodulin (TM) from cultured endothelial cells treated with stimulated leukocyte suspension (A) or the medium separated sf;; stimulated leukocytes (B). Cell monolayers incorporating were prepared according to the method described in “Methods”. Leukocytes stimulated with endotoxin (100 ng/ml) plus FMLP (1 ,aM) were immediately added to the cell monolayers in the presence of various concentrations of heparin in culture medium Leukocytes stimulated with endotoxin plus FMLP for 2 hrs at (A). room temperature were centrifuged and the resulting supernatant was added to the cell monolayers in the presence of heparin in the culture medium. The conditioned medium was collected at 3 of stimulated leukocytes or the medium, ;;s, ;;t~nt,t,h,eena~dW Cr count in the conditioned medium were measured. Each point represents the mean + S.D. of the cells from 3 different umbilical veins. Values of the % injury and % of total were calculated in the same way as in Table 1. ;;;-
‘th 51 Cr
Protective 51. Cr releases
the percentage of the cell injury (19.2 %) calculated from release into the medium and its total amount in the cells. The parallelism between the relative antigen level released into the medium and the relative extent of cell injury was observed These results suggested that regardless of the incubation period. an increase in TM antigen level in the conditioned medium reflected the injury of cultured endothelial cells. The cells treated with unstimulated leukocytes did not exhibit significant changes in TM antigen level in the cells or the medium, and cell Heparin injury was also not detectable (data not shown). treatment of the cells prevented the release of TM antigen into the the conditioned medium as well as the cell injury caused by stimulated leukocytes in a dose-dependent manner in the range between 0.01 and 1.0 U/ml of heparin (Fig. 4A). Also, when the medium separated from cells were treated with the supernatant the stimulated leukocyte suspension, the release of TM antigen The into the medium and cell injury were observed (Fig. 4B). antigen release and the cell injury caused by treatment with the
Vol. 65, Nos. 415
PLASMA TM IN RABBIT DIC MODEL
601
treatment with medium were also prevented by simultaneous heparin in a dose-dependent manner (Fig. 4B). The preventive action of heparin against antigen release and cell injury was more effective on cells treated with the medium as compared with cells treated with the stimulated leukocyte suspension. DISCUSSION TM antigen level in rabbit plasma was 1.64 + 0.64 fig/ml, 47 times higher than that of human plasma (35.2 + 8.3 ng/ml) (5). Therefore, rabbit plasma could be directly subjected to SDS-PAGE to detect molecular subspecies of plasma TM antigen. Human TM antigen in plasma consisted of 6 subspecies (5,101, whereas 3 subspecies were observed in rabbit plasma. Rabbit TM purified from lung in the present work showed an apparent molecular weight of 94 kd under reducing conditions. Bourin et al. (23) reported that heterogeneous TM molecules including 110 - 100 kd species were present in a TM preparation purified from rabbit lung, and the high-molecular subspecies contained chondroitin sulfate-type glycosaminoglycan. Furthermore, it was reported that the molecular weight of human or rabbit thrombomodulin containing glycosaminoglycan was reduced by treatment with chondroitinase It was supposed that the purified 94 kd TM is a ABC (24,25). and that plasma TM antigen degraded form of the native molecule, is composed of various heterogeneous subspecies, with or without glycosaminoglycan and/or with or without degradation of the peptide moiety. Plasma TM levels of rabbits were increased to 1.5 times the control level by infusion of endotoxin and the increase was accompanied with generation of DIG, as confirmed by the decreases in platelet count and plasma fibrinogen level. A marked decrease in the 94 kd subspecies of TM, together with both increase in 83 kd subspecies and the appearance of the new subspecies of 76 and 48 kd, was found in rabbits in which DIC was being induced by the infusion of endotoxin. It is probable that the new subspecies were produced by degradation of high-molecular-weight TM by glycosidases and/or proteases. Similarly, an increase in smaller-molecular forms of TM in plasma in clinical cases of DIC was recently suggested (8). Heparin is an effective drug against the induction of DIC. Heparin treatment of endotoxin-induced DIC rabbits prevented both the increase in plasma TM levels and the degradation of plasma TM subspecies, and produced an improvement of DIG as judged from the platelet count and fibrinogen level. It has been reported that the increase in plasma TM level in clinical cases was reflection of endothelial cell damage (5,8,10), which was also indicated in the present work (Table 1). These facts suggested that heparin not only prevented the development of DIC. but also protected the endothelial cells against injury by leukocytes stimulated with endotoxin. It has been postulated that the therapeutic effectiveness of heparin on DIG is derived from its inhibiting action upon activated coagulation factors in concert with antithrombin III (AT-III). Howevsr. we observed that, when titrated blood was incubated with H-heparin, the labeled heparin was distributed
602
PLASMA TM IN RABBIT DIC MODEL
Vol. 65, Nos. 45
mainly to blood cells (37.7%) and plasma proteins other than ATIII (45.7%). and only 16.6% of the total amount formed a complex with AT-III (unpublished data). Furthermore, it has been reported that heparin was concentrated in endothelium in rats after its injection by the intravenous, intraperitoneal, or subcutaneous route (26.27). These facts suggest that intravenously injected heparin not only binds to AT-III in plasma, but also is distributed to plasma components other than AT-III, and also binds to blood cells and endothelial cells. The current results in vitro indicated that heparin prevented both the TM release into the medium and the cell injury caused by treatment with stimulated leukocytes or the medium separated from the stimulated leukocytes. This preventive effect of heparin was observed in the absence of AT-III. Therefore, it is probable that heparin acts not only as an anticoagulant by formation of a complex with ATIII in plasma during development of DIC, but also directly protects endothelial cells against injury induced by stimulated leukocytes. Endothelial cells are injured by proteases including elastase and active oxygens including superoxide released from endotoxin-stimulated leukocytes (15-18). In order to elucidate the mechanism of its preventive effect, the action of heparin was compared between cells treated with the stimulated leukocyte suspension and those treated with the medium separated from the stimulated leukocytes. When the cells were treated by the former method, they would be injured by both proteases and active oxygens released from the leukocytes, but when the cells were treated by the latter method, they would be injured mainly by proteases, since the half lives of active oxygens are very short and these species should be absent in the medium. Our results showed that the preventive action of heparin against cell injury and antigen release was more effective on the cells treated with the medium than on the cells treated with the stimulated leukocyte suspension. It appears that heparin acted as a proteinase inhibitor rather than as a scavenger for active oxygens. We hypothesize that plasma TM containing degraded subspecies was released from endothelial cells injured by stimulated leukocytes as well as by disturbed microcirculation during the induction of DIC. It is also considered that the preventive effect of heparin against both TM release and cell injury is a consequence of not only its anticoagulant activity but also its direct protection of endothelial cells against the action of leukocyte proteases. REFERENCES 1. 2. 3. 4.
ESMON, C.T. and OWEN, W.G. Identification of an endothelial cell cofactor for thrombin-catalyzed activation of protein Proc.Nat1.Acad.Sci.U.S.A. 78, 2249-2252, 1981. C. ESMON, C.T. The roles of protein C and thrombomodulin in J.Biol.Chem. 264, 4743regulation of blood coagulation. 4746, 1989. Structure and function of DITTMAN, W.A. and MAJERUS, P.W. Blood 75, 329thrombomodulin: A natural anticoagulant. 336, 1990. ISHII, H. and MAJERUS, P.W. Thrombomodulin is present in
Vol. 65, Nos. 415
5.
6. 7.
8.
9.
10. 11. 12. 13.
14.
15.
16.
17. 18.
19.
PLASMA TM IN RABBIT DIC MODEL
603
human plasma and urine. J.Clin.Invest. 76. 2178-2181, 1985. ISHII, H., NAKANO, M., TSUBOUCHI, J., ISHIKAWA.T., UCHIYAMA, H HIRAISHI. S., TAHARA, C., MIYAJIMA, Y. and KAZAMA, M. Esiablishment of enzyme immunoassay of human thrombomodulin in plasma and urine using monoclonal antibodies. Thromb.Haemostas. 63, 157-162. 1990. TANAKA, A., ISHII. H., HIRAISHI, S., KAZAMA, M. and MAEZAWA, H. Increased thrombomodulin values in plasma of diabetic men with microangiopathy. Clin.Chem. 37. 269-272. 1991. IWASHIMA, Y., SATO, T., WATANABE, K., OOSHIMA. E., HIRAISHI, S ., ISHII, H.. KAZAMA, M. and MAKINO, I. Elevation of plasma thrombomodulin level in Diabetic patients with early diabetic nephropathy. Diabetes 39, 983-988, 1990. TAKANO, S., KIMURA, S., OHDAMA, S. and AOKI, N. Plasma thrombomodulin in health and diseases. Blood 76, 2024-2029, 1990. GIDDINGS, J.C., C0LES.P. and WILLIAMS, B.D. Comparison of thrombomodulin and von Willebrand factor antigen in human plasma in various diseases. Thromb.Haemostas. 62, 276 (Abstr.), 1989. ISHII, H.. UCHIYAMA, H. and KAZAMA, M. Soluble thrombomodulin antigen in conditioned medium is increased by damage of endothelial cells. Thromb.Haeaostas. 65, 618-623, 1991. MIZOCK, B. Septic shock. A metabolic perspective. Arch.Intern.Med. 144. 579-585. 1984. MORRISON, D.C.. ULEVITCH, R.J. The effect of bacterial endotoxins on host mediation systems. A review. Am.J.Pathol. 93, 526-617, 1978. COLUCCI. M., BALCONI. G., LORENZET, R., PIETRA. A., LOCATI, D DONATI, M.B. and SEMERARO, N. Cultured human endothelial ceils genarate tissue factor in response to endoxoxin. J.Clin.Invest. 71. 1893-1896, 1983. MOORE, K.L., ANDREOLI, S.P., ESMON, N.L., ESMON, C.T. and BANG, N.U. Endotoxin enhances tissue factor and supresses thrombomodulin expression of human endotherium in vitro. J.Cli.Invest. 79, 124-130, 1987. TILL, G.O., JOHNSON, K.J., KUNKEL. R., and WARD, P.A. Intravascular activation of complement and acute lung injury. Dependency on neutrophils and toxic oxygen metabolites. J.Clin.Invest. 69, 1126-1135, 1982. SACKS, T.. MOLDOW, C.F., CRADDOCK, P.R., BOWERS, T.K., and JACOB, H.S. Oxygen radicals mediate endothelial cell damage by complement-stimulated granulocytes. J.Clin.Invest. 61, 1161-1167, 1978. HARLAN, J.M., KILLEN, P.D., HARKER, L.A., and STRIKER, G.E. Neutrophil-mediated endothelial injury in vitro. J.Clin.Invest. 68. 1394-1403, 1981, SMEDLY, L.A., TONNESEN. M.G.. SANDHAUS, R.A., HASLETT, C., GUTHRIE, L.A., JOHNSTON, R.B., HENSON P.M., and WORTHEN, G.S. Neutrophil-mediated injury to endothelial cells. Enhancement by endotoxin and essential role of neutrophil elastase. J.Clin.Invest. 77, 1233-1243, 1986. SUZUKI, K.. STENFLO, J.. DAHLBACK, B. and TEODORSSON, B. Inactivation of human coagulation factor V by activated
604
20. 21.
22.
23.
24.
25.
26. 27.
PLASMA
TM IN RABBIT
DIC MODEL
Vol. 65, Nos. 415
protein C. J.Biol.Chem. 258, 1914-1920, 1983. OWEN, W.G. and ESMON, C.T. Isolation of a ESMON, N,L., membrane-bound cofactor for thrombin-catalized activation of protein C. J.Biol.Chem. 257, 859-864, 1982. ISHIKAWA, E., IMAGAWA, M., HASHIDA, S., YOSHITAKE, S., HAMAGUCHI, Y. and UENO, T. Enzyme-labeling of antibodies and their fragments for enzyme immunoassay and immunohistochemical staining. J.Immunoassay 4, 209-327, 1983. VARANI, J., BENDELOW, M.J.. SEALEY. D.E., KUNKEL. S.L., factor GANNON, D.E., RYAN, U.S. and WARD, P.A. Tumor necrosis enhances susceptibility of vascular endothelial cells Lab.Invest. 59, 292-295, 1988. neutrophil-mediated killing. BOURIN, M.C., OHLIN, A.K., LANE, D.A., STENFLO, J. and anticoagulant activities LINDAHL, U. Relationship between and properties of rabbit thrombomodulin. polyanionic 263, 8044-8052, 1988. J.Biol.Chem. PARKINSON, J.F., GRINNELL, B.W., MOORE, R.E., HOSKINS, J., VLAHOS, C.J. and BANG, N.U. Stable expression of a secretable deletion mutant of recombinant human thrombomodulin in mammalian cells. J.Biol.Chem. 265, 12602-12610, 1990. BOURIN. M.C. and LINDAHL, U. Functional role of the polyproteoglycan. saccharide component of rabbit thrombomodulin Effects of inactivation of thrombin by antithrombin, cleavage of fibrinogen by thrombin and thrombin-catalized activation of Factor V. Bi0chem.J. 270, 419-425, 1990. HIEBERT, L.M. and JAQUES, L.B. The observation of heparin on endothelium after injection. Thromb.Res. 8, 195-204, 1976. GLIMELIUS, B.. BUSCH, C. and HOOK, M. Binding of heparin on cells. the surf ace of cultured human endothelial Thromb. Res. 12, 773-782, 1978.
