THROMBOSIS RESEARCH 66; 591~598,1992 00494848/92 $5.00 + .OOPrinted in the USA. Copyright (c) 1992 Pergamon Press Ltd. All rights reserved.
ELEVATED LEVELS OF THROMBIN-HEPARIN COFACTOR II COMPLEX IN PLASMA FROM PATIENTS WITH DISSEMINATED INTRAVASCULAR COAGULATION
T.R. Andersson, P. Sie’, H. Pelze?, L-M. Aamodt, K Nustad3, U. Abildgaard. Haematological Research Laboratory, Aker Hospital, 0513 Oslo, Norway, ‘Laboratoire d’Hemostase, Centre de Transfusion Sanguine, Toulouse, France, *Department for Blood Coagulation, Behringwerke AG, Germany. 3Central Laboratory, Radiumhospitale!, 0310 Oslo, Norway.
(Received 16.3.1992; accepted in original form 23.3.1992 by Editor H.A. Vinazzer)
ABSTRACT An ELISA assay for quantitation of the thrombin-heparin cofactor II complex (T-HC II) in plasma was developed. Plasma was incubated with immobilized, specific antibodies to human thrombin. The second, biotinylated antibody was directed against human HC II. The assay was insensitive to thrombinantithrombin complex (TAT) and to uncomplexed HC II. In plasma samples from 31 normal individuals (aged 21 - 68, mean 43.3 years), the T-HC II ranged 0.3 - 6.1 @ml; median 1.5, mean 2.0, and SD 1.6 ng/ml. In plasma samples from 13 patients with disseminated intravascular coagulation (DIC), T-HC II ranged 0.4 - 30.0 (median 13.5) &ml. In plasma samples from 6 patients in which the clinical suspicion of DIC was not verified, T-HC II complex ranged 1.4 - 14.3 (median 3.6) @ml. In plasma samples with elevated T-HC II levels, TAT was usually elevated, and on the average more than was T-HC II. These results indicate that HC II contributes significantly to the inactivation of in viva generated thrombin.
INTRODUCTION The physiological role of heparin cofactor II (HC II) is not evident. In vitro, this inhibitor inhibits thrombin very slowly by the formation of an inactive complex. The reaction is accelerated by heparin but more specifically by dermatan sulfate (DS) (1). When blood clots in the presence of DS, the thrombin generated is preferentially complexed by HC II (2). The sequence in DS catalyzing the thrombin-HC II reaction has recently been identified (3).
Key words:
Thrombin, heparin cofactor II, disseminated intravascular coagulation, ELISA assay, protease-inhibitor complex. 591
592
THROMBIN-HC
II COMPLEXES IN DIC
Vol. 66, No. 5
Heterozygous deficiency of HC II carries a low (4) or moderate (5) risk for venous thrombosis, in contrast to the strong tendency to thrombosis seen in deficiency of antithrombin (AT). Circumstantial evidence suggesting, however, that HC II may be involved in the control of blood coagulation in vivo may be summarized as follows: In disseminated intravascular coagulation (DIG) the level of HC II is usually quite low, often as low as that of the main inhibitor of thrombin, AT. The low AT level results, at least in part from the formation of inactive thrombin-AT (TAT) complexes (6,7). By analogy, the low HC II level in DIC might also result from consumption during inactivation of thrombin. Glycosaminoglycans on the cellular surface catalyze the inactivation of thrombin. The formation of a thrombin-HC II complex is accelerated by fibroblasts and vascular smooth muscle cells, but apparently not by human umbilical vein endothelial cells (8). The latter cells harbour mainly heparan sulfate, which catalyzes the thrombin - AT reaction. We eluted small amounts of DS-like material by perfusion of the lungs from newborn calves (9). These findings suggested to us that DS and HC II may be involved in the control of thrombin activity during DIC. In order to test this hypothesis, we have developed a quantitative enzyme immuno-assay for thrombin-HC II complex in human plasma.
MATERIALS AND METHODS
Citrated nlasma. Venous blood was collected in Vacutainer tubes containing l/10 volume of 0.13 M sodium citrate, centrifuged for 20 minutes at 2000 g. Plasma was stored at -70 “C. Patient material. Blood samples were obtained from A) 31 healthy persons not on medication, 17 females and 14 males, mean age 43.3 years B) 19 patients, 9 females and 10 males, aged l-68 years (mean 37.4 years), consecutively referred for evaluation of coagulation because DIC was suspected on clinical grounds. These patients had severe infection (n = 17), or preeclampsia in pregnancy complicated by the HELLP syndrome (n = 2). Depending on laboratory findings, these were classified as DIC if D-dimer test was positive (above 500 @ml) and TAT complex was 10 @ml or higher. The DIC group was divided in patients with thrombocytopenia ($100 x 109/1) (n = 6), DIC without thrombocytopenia (n = 7). In 6 plasma samples, DIC was not verified by these criteria (“controls”). Hemostasis tests: TAT complex was quantitated with the commercial ELISA kit from Behringwerke, Germany. D-dimer was measured with the semiquantitative latex kit test from Stago, France. AT and HC II were determined with chromogenic substrate assays (2). Proteins: Human thrombin, Sigma thrombin T 6759, 3050 NIH U/mg was dissolved in distilled water to 100 U/ml (32.7 ,@ml), (and stored at - 20 “C for a maximum of three months. On SDSPAGE a single peak (35 kD) appeared. Human HC II was prepared by chromatography as previously reported (10). The product was free from contamination with antithrombin by activity analysis and immunoelectrophoresis, and showed one main band (- 65 kD) and a faint low molecular weight contamination (probably degraded protein) at SDS PAGE. Human antithrombin, laboratory grade, was from Kabi, Sweden. Antiserum against human HC II was raised in rabbits as previously described (11). Buffer: 015 M NaCl, 0.05M Phosphate buffer, pH 7.4 (PBS). For dilution of standards, test plasma, biotinylated antibodies, and streptavidin, PBS was used with the following additions: 0.5 % BSA, 0.5 % Tween 20, 0.1 M EDTA and pH adjusted to 7.4. Washing buffer was PBS with
Vol. 66, No. 5
THROMBIN-HC
II COMPLEXES IN DIC
593
0.05 % Tween 20. Substrate buffer: 0.1 M Citrate-phosphate buffer, pH 5.0. Dermatan sulfate (DS) was obtained as chondroitin sulfate, type B from porcine skin, from Sigma; 100 mg/vial. Heparin contamination was removed as described previously (12) and the DS solution stored frozen. Biotin-XX-NHS was obtained from Calbiochem, USA (cat. no. 203114). Streptavidine-Peroxidase was from Dakopatts, Denmark. Substrate: orto-phenylenediamine, free base (OPD), Sigma. One tablets (20 mg) was diluted in 50 ml substrate buffer, and 20 ~1 H,O, (30 %) was added just before use. Methods: Pre-coated micro-well plates: These were coated with specific antibodies against human thrombin, as described for the Thrombin-Antithrombin (TAT) assay (7) and identical to those used in the TAT kit from Behringwerke AG, Germany. The plates were packed into foil-bags, with a drying agent, and kept in the refrigerator. Thrombin activity was tested with chromogenic substrate S-2238 from Kabi, Sweden; undiluted reaction mixture was mixed with 9 volumes substrate and incubated at pH 8.0, tp 37 “C, 30 minutes. Absorbance was then read at 405 nm. SDS-Polvacrvlamide gel electrophoresis (SDS-PAGE) was performed according to Laemmli (13) Molecular weight standards included myosin (205 kD), I3-galactosidase (116 kD), phosphorylase B (94 kD), bovine serum albumin (68 kD), and ovalbumin (43 kD). Preparation of thrombin-HC II complex and standards Principle: An exact amount of human thrombin was incubated with a moderate molar excess of HC II in presence of dermatan sulfate which accelerates the formation of the bimolecular complex. A fixed amount of thrombin was first incubated with varying amounts of HC II. The reaction mixture was analyzed for remaining thrombin activity and run on SDS PAGE. With the amount of HC II chosen, no thrombin activity was present after incubation. SDS PAGE electrophoresis showed a prominent band at - 95000D and a faint band at - 65000, representing T-HC II complex and a slight amount of uncomplexed HC II. Procedure: A mixture was prepared of human thrombin (500 ~1 containing 16.4 pg), human HC II (200 ~1 containing about 60 pg), and dermatan sulfate (DS) (100 ~1 containing 240 ,r@. After incubation at 37 “C (pH 7.4,O.lO M NaCl, 0.05 M Tris) for 180 minutes, the tube was immersed in ice water and aliquotes stored in plastic tubes at -70 “C. These were analyzed to check absence of thrombin activity and SDS-PAGE electrophoresis (Cf above) and used for preparation of standards. The amount of T-HC II complex was calculated from the amount of thrombin (16.4 pg) in the reaction mixture and the molecular weights of thrombin (36 kD) and HC II (66 kD) assuming that all thrombin molecules took part in formation of a bimolecular complex (102 kD). Standards. The stem solution, containing 57.8 &ml was diluted l/100, l/400, and l/600 in dilution buffer. One volume of these dilutions, and of buffer, was mixed with nine volumes of a normal plasma. This plasma contained 1.3 Icglml of “intrinsic” T-HC II (Cf results). The plasma standards thus contained 59.0, 15.6, 4.8 and 1.17 ng/ml of T-HC II complex. Thrombin-antithrombin (TAT) complex was prepared similar to T-HC II complex, by incubating human thrombin with human AT (1:1.2 molar concentration) without glycosaminoglycan, for 2 hours at 37 “C.
HC II antibodies: Polyvalent IgG fraction from HC II rabbit antiserum was purified by Protein A Sepharose CL-4B (Pharmacia, Sweden). The antibody fraction (4.7 mg or 0.29 umol IgG in a volume of 500 ul) was biotinylated. Biotin-XX-NHS (Calbiochem cat. No 203114), 130 ,ug
THROMBIN-HC II COMPLEXES IN DIC
594
Vol. 66, No. 5
(10 times molar excess of the antibody) was dissolved in 50 ~1 DMSO and mixed with the antibody solution. 100 ,ulO.5 M borate buffer pH 8.0 were added and the mixture was incubated at 4 “C overnight. After the incubation 50 ~1 1M ethanolamine pH 8.0 were added. Free biotin was removed by gel filtration using Biogen P30 (Biorad) with a column size of 1.5 x 15 cm and PBS as buffer. The pooled fractions containing biotinylated HC II antibodies were stabilized by adding BSA to a final 1 % concentration. Aliquotes were stored at -70 “C. A l:lO,OOO dilution was used in the ELISA assay. ELISA assay of T-HC II complex: A sandwich technique using two different antibodies which bind selectively to the corresponding antigens of the complex was used. Plasma was first incubated with immobilized antibodies to thrombin. After washing, a biotinylated antibody towards HC II was used to detect bound T-HC II complex. Procedure: Plasma (standards and tests) were diluted 1:2 with dilution buffer. 100 ~1 diluted plasma was pipetted in each well, incubated in humidity-chamber with stirring for 2 hours at 20 - 25 “C. After three washings, 100 ~1 of biotinylated anti-HC II (diluted l:lO,OOO) was added and incubated over night at room temperature humidity-chamber with stirring). After three washings, 100 ,~l Streptavidin, (diluted l:lO,OOO) was added and incubated for 1 hour at room temperature (humidity-chamber with stirring). After washing, freshly made substrate was added (200 ul) and incubated in the dark for approximately 30 minutes (depending on the development of colour); 50 ul of 1 M H,SO, was added and absorbance read at 492 nm and 630 nm (blank). Each sample was run in duplicate and the mean value used for calculations.
A 492
nm
1.5
1 .o
0.5
0.1 1
T-HCII
I
I
10
100
CONC.,
ng/ml
Figure 1. Standard curve for the ELISA assay of T-HC II in plasma. To a normal plasma containing 1.3 ng/ml of T-HC II was added l/10 volume of purified T-HC II complex.
Vol. 66, No. 5
THROMBIN-HC II COMPLEXES IN DIG
595
RESULTS
There was a linear correlation in a log plot between absorbance and the amount of T-HC II added to a normal plasma from one person (Fig 1). To assess the intrinsic T-HC II concentration in this particular plasma, it was adsorbed with immobilized, specific antibodies to human thrombin. As this procedure reduced the measured T-HC II concentration by 1.3 &ml, this value was taken to reflect its intrinsic T-HC II concentration. This unabsorbed plasma was always used for preparing standards. In order to determine the T-HC II concentration in the other plasma samples from healthy persons, a (non-linear) standard curve also based on the buffer value (zero T-HC II) was used. The addition of purified HC II (200 @ml) or TAT complex (100 &ml) did not influence the measured T-HC II value.
TABLE 1. Median and Range Values in the Various Groups of Normal and Patients. AT = antithrombin, plt = platelet counts, n. d. = not determined.
n.d.
CONTROLS
95 78-113
2.6 1.7-14.8
99 77-150
1.5 0.3-6.1
55 51-100
53 30-78
25 10.5-57.0
49.5 22-60
11.5 3.8-30.0
304 135-603
82 53-116
15.5 10.0-60.0
76 40-112
14.5 0.4-28.5
186 97-313
80 53-110
6.5 3.8-7.8
73 59-115
3.8 1.4-14.3
The highest T-HC II concentration in plasma from healthy persons was 6.1 @ml; the median value was 1.5 @ml. The mean value was 2.0 @ml with an SD of 1.6 &ml. The values were thus not normally distributed, as is also evident from Fig 2. The intra-assay coefficient of variation was 7 %. In plasma from patients with clinically suspected DIC and elevated levels of fibrin-derived Ddimer and TAT complex, T-HC II values ranged 0.4 - 30.0 ng/ml with a median value of 13.5 @ml. Similar T-HC II values were found in the two DIC subgroups, with or without thrombocytopenia (Table 1). In contrast, normal or near normal T-HC II values (median 3.8 &ml) were found in plasma samples from septic patients in which DIC was excluded by normal TAT and D-Dimer values (Table 1, Fig 2). Plasma samples with elevated T-HC II (above 3.2 ng/ml) also contained TAT above normal reference, the correlation was of borderline significance (p = 0.07, n = 21).
596
Vol. 66, No. 5
THROMBIN-HC II COMPLEXES IN DIC
T-HCII (4-d 60
60
40
40 0
0 20
20
0 15
-
_a -
0
15
0 0 10
-
10 0 0 0
0
5-
u 0.0
OJ
e
NORMAL
0
0.
5
: 0 0 CTR.
DIC
DIC
P’t 3
plt N
Figure 2. Individual and median values of T-HC II in plasma in patient groups and healthy perS0ll.S.
DISCUSSION Our findings show that formation of a complex between thrombin and HC II occurs at a low rate in normal individuals, and at a higher rate in septic patients with laboratory signs of DIC. The results strongly suggest a role for HC II in the control of coagulation in viva. The low levels of HC II regularly found in severe DIC (14) is at least in part a result of consumption, as previously suggested (2). It has been assumed that HC II is primarily an extravascular thrombin inhibitor (3). It is possible that the T-HC II complexed are generated in the extravascular space and transported to the blood. Intravascular generation of thrombin is a key event in DIC, and it may seem more likely that the T-HC II complexes actually are formed intravascularly. Future studies are required to establish where the T-HC II in plasma originate.
Vol. 66, No. 5
THROMBIN-HC
II COMPLEXES IN DIC
597
Traditionally, the diagnosis of DIC has been based on prolonged clotting times, and test results supporting consumption, elevated FDP and thrombocytopenia. By these criteria, consumption coagulopathy may be hard to distinguish from liver failure (15). Therefore, we based the diagnosis of DIC on elevated levels of the molecular markers, TAT and D-dimer, signifying increased in viva generation of thrombin. The assay used presented here is designed as the TAT assay (7) and the same procedure is used for catching the T-inhibitor complex. The T-HC II level found in plasma from normal individuals is similar to those found for TAT (7). In DIC, TAT levels are markedly increased, often above the T-HC II values reported here. Surprisingly, occasional plasma samples show much higher T-HC II values than TAT values. Comparing mean values for the group of DIC patients, the T-HC II level was about half of the TAT level. The inhibitor level measured is a function both of the rate of formation, and their rate of elimination, which at present is not known. Endothelial glycosaminoglycans are probably involved in the function of several coagulation inhibitors. Heparan sulfate (HS) may be the main intravascular catalyst of the thrombin-AT reaction and may well be the receptor for Tissue factor pathway inhibitor (TFPI). Dermatan sulfate (DS) appears to be the specific catalyst of the thrombin-HC II reaction. We eluted anticoagulant GAGS from the pulmonary circulation of newborn calves; with a HS:DS ratio of about 3: 1. Although this is a higher DS fraction than would be surmised from studies on cultured cells (8), the DS-like activity was distinct. It is possible that the T-HC II complexes we have demonstrated are formed with endothelial DS as catalyst.
REFERENCES 1.
TOLLEFSEN, D.M., PESTKA, C.A., MONFAO, W.J. Activation of heparin cofactor II by dermatan sulfate. J Biol Chem 258. 6713-6716, 1983.
2.
ANDERSSON, T.R. Consumption of heparin cofactor II (Dermatan sulfate Cofactor) and antithrombin during coagulation. Thromb Res 46, 355-362, 1987.
3.
TOLLEFSEN,
4.
BERTINA, R.M., VAN DER LINDEN, I.K., ENGESSER, L., MijLLER, H.P., BROMMER, E.J.P. Hereditary heparin cofactor II deficiency and the risk of development of thrombosis. Thromb Haemost 57, 196-200, 1987.
5.
VINAZZER, H., STOCKER, K. Heparin cofactor II: Experimental approach to a new assay and clinical results. Thromb Res 61, 235-241, 1991.
6.
LAU, H.K., ROSENBERG, R.D. The isolation and characterization of a specific antibody population directed against the thrombin antithrombin complex. J Biol Chem 225, 58855893, 1980.
D.M. Heparin cofactor II. Thromb Haemost 65, 136, 1991 (Abstract).
598
THROMBIN-HC
II COMPLEXES IN DIC
Vol. 66, No. 5
7.
PELZER, H., SCHWARZ, A., HEIMBURGER, N. Determination of human thrombinantithrombin III complex in plasma with an enzyme-linked immunosorbent assay. Thromb Haemost 59 (1). 101-106, 1988.
8.
McGUIRE, E.A., TOLLEFSEN, M. Activation of Heparin cofactor II by fibroblasts and vascular smooth muscle cells. J Biol Chem 262, 169-175, 1987.
9.
HALE, G., ABILDGAARD, U., ANDERSSON T.R., LARSEN, M.L., SANDSET, P.M., LINDBOE, C.F., SKJ(dRTEN, F. Anticoagulant glycosaminoglycans (GAG) extracted from pulmonary vascular surface. Thromb Haemost 65, 951, 1991 (Abstract).
10. SIE, P., DUPOUY, D., PICHON, J., BONEU, B. Turnover study of heparin cofactor II in healthy man. Thromb Haemost 54, 635-638, 1985. 11. SIE, P., DUPOUY, D., PICHON, J., BONEU, B. Constitutional heparin cofactor deficiency associated with recurrent thrombosis. Lancet 1985, II, 414-416.
II
12. TEIEN, A.N., ABILDGAARD, U., HOOK, M. The anticoagulant effect of heparan sulfate and dermatan sulfate. Thromb Res 8, 859-867, 1976. 13. LAEMMLI, U.K. Cleavage of structural proteins during the assemble of the head of bacteriophage t 4. Nature 227, 680-685, 1970. 14. ABILDGAARD, U., LARSEN, M.L. Assay of dermatan sulfate cofactor (Heparin Cofactor II) activity in human plasma. Thromb Res 35, 1984, 257-266. 15. ANKER, E, ABILDGAARD, U., ANDERSEN, R., FAGERHOL, M., BJUNE, G. Low antithrombin in severe disease: Consumption or decreased synthesis? Stand J Haematol 30, 1983, 59-63.
ELEVATED LEVELS OF THROMBIN-HEPARIN COFACTOR II COMPLEX IN PLASMA FROM PATIENTS WITH DISSEMINATED INTRAVASCULAR COAGULATION
T.R. Andersson, P. Sie’, H. Pelze?, L-M. Aamodt, K Nustad3, U. Abildgaard. Haematological Research Laboratory, Aker Hospital, 0513 Oslo, Norway, ‘Laboratoire d’Hemostase, Centre de Transfusion Sanguine, Toulouse, France, *Department for Blood Coagulation, Behringwerke AG, Germany. 3Central Laboratory, Radiumhospitale!, 0310 Oslo, Norway.
(Received 16.3.1992; accepted in original form 23.3.1992 by Editor H.A. Vinazzer)
ABSTRACT An ELISA assay for quantitation of the thrombin-heparin cofactor II complex (T-HC II) in plasma was developed. Plasma was incubated with immobilized, specific antibodies to human thrombin. The second, biotinylated antibody was directed against human HC II. The assay was insensitive to thrombinantithrombin complex (TAT) and to uncomplexed HC II. In plasma samples from 31 normal individuals (aged 21 - 68, mean 43.3 years), the T-HC II ranged 0.3 - 6.1 @ml; median 1.5, mean 2.0, and SD 1.6 ng/ml. In plasma samples from 13 patients with disseminated intravascular coagulation (DIC), T-HC II ranged 0.4 - 30.0 (median 13.5) &ml. In plasma samples from 6 patients in which the clinical suspicion of DIC was not verified, T-HC II complex ranged 1.4 - 14.3 (median 3.6) @ml. In plasma samples with elevated T-HC II levels, TAT was usually elevated, and on the average more than was T-HC II. These results indicate that HC II contributes significantly to the inactivation of in viva generated thrombin.
INTRODUCTION The physiological role of heparin cofactor II (HC II) is not evident. In vitro, this inhibitor inhibits thrombin very slowly by the formation of an inactive complex. The reaction is accelerated by heparin but more specifically by dermatan sulfate (DS) (1). When blood clots in the presence of DS, the thrombin generated is preferentially complexed by HC II (2). The sequence in DS catalyzing the thrombin-HC II reaction has recently been identified (3).
Key words:
Thrombin, heparin cofactor II, disseminated intravascular coagulation, ELISA assay, protease-inhibitor complex. 591
592
THROMBIN-HC
II COMPLEXES IN DIC
Vol. 66, No. 5
Heterozygous deficiency of HC II carries a low (4) or moderate (5) risk for venous thrombosis, in contrast to the strong tendency to thrombosis seen in deficiency of antithrombin (AT). Circumstantial evidence suggesting, however, that HC II may be involved in the control of blood coagulation in vivo may be summarized as follows: In disseminated intravascular coagulation (DIG) the level of HC II is usually quite low, often as low as that of the main inhibitor of thrombin, AT. The low AT level results, at least in part from the formation of inactive thrombin-AT (TAT) complexes (6,7). By analogy, the low HC II level in DIC might also result from consumption during inactivation of thrombin. Glycosaminoglycans on the cellular surface catalyze the inactivation of thrombin. The formation of a thrombin-HC II complex is accelerated by fibroblasts and vascular smooth muscle cells, but apparently not by human umbilical vein endothelial cells (8). The latter cells harbour mainly heparan sulfate, which catalyzes the thrombin - AT reaction. We eluted small amounts of DS-like material by perfusion of the lungs from newborn calves (9). These findings suggested to us that DS and HC II may be involved in the control of thrombin activity during DIC. In order to test this hypothesis, we have developed a quantitative enzyme immuno-assay for thrombin-HC II complex in human plasma.
MATERIALS AND METHODS
Citrated nlasma. Venous blood was collected in Vacutainer tubes containing l/10 volume of 0.13 M sodium citrate, centrifuged for 20 minutes at 2000 g. Plasma was stored at -70 “C. Patient material. Blood samples were obtained from A) 31 healthy persons not on medication, 17 females and 14 males, mean age 43.3 years B) 19 patients, 9 females and 10 males, aged l-68 years (mean 37.4 years), consecutively referred for evaluation of coagulation because DIC was suspected on clinical grounds. These patients had severe infection (n = 17), or preeclampsia in pregnancy complicated by the HELLP syndrome (n = 2). Depending on laboratory findings, these were classified as DIC if D-dimer test was positive (above 500 @ml) and TAT complex was 10 @ml or higher. The DIC group was divided in patients with thrombocytopenia ($100 x 109/1) (n = 6), DIC without thrombocytopenia (n = 7). In 6 plasma samples, DIC was not verified by these criteria (“controls”). Hemostasis tests: TAT complex was quantitated with the commercial ELISA kit from Behringwerke, Germany. D-dimer was measured with the semiquantitative latex kit test from Stago, France. AT and HC II were determined with chromogenic substrate assays (2). Proteins: Human thrombin, Sigma thrombin T 6759, 3050 NIH U/mg was dissolved in distilled water to 100 U/ml (32.7 ,@ml), (and stored at - 20 “C for a maximum of three months. On SDSPAGE a single peak (35 kD) appeared. Human HC II was prepared by chromatography as previously reported (10). The product was free from contamination with antithrombin by activity analysis and immunoelectrophoresis, and showed one main band (- 65 kD) and a faint low molecular weight contamination (probably degraded protein) at SDS PAGE. Human antithrombin, laboratory grade, was from Kabi, Sweden. Antiserum against human HC II was raised in rabbits as previously described (11). Buffer: 015 M NaCl, 0.05M Phosphate buffer, pH 7.4 (PBS). For dilution of standards, test plasma, biotinylated antibodies, and streptavidin, PBS was used with the following additions: 0.5 % BSA, 0.5 % Tween 20, 0.1 M EDTA and pH adjusted to 7.4. Washing buffer was PBS with
Vol. 66, No. 5
THROMBIN-HC
II COMPLEXES IN DIC
593
0.05 % Tween 20. Substrate buffer: 0.1 M Citrate-phosphate buffer, pH 5.0. Dermatan sulfate (DS) was obtained as chondroitin sulfate, type B from porcine skin, from Sigma; 100 mg/vial. Heparin contamination was removed as described previously (12) and the DS solution stored frozen. Biotin-XX-NHS was obtained from Calbiochem, USA (cat. no. 203114). Streptavidine-Peroxidase was from Dakopatts, Denmark. Substrate: orto-phenylenediamine, free base (OPD), Sigma. One tablets (20 mg) was diluted in 50 ml substrate buffer, and 20 ~1 H,O, (30 %) was added just before use. Methods: Pre-coated micro-well plates: These were coated with specific antibodies against human thrombin, as described for the Thrombin-Antithrombin (TAT) assay (7) and identical to those used in the TAT kit from Behringwerke AG, Germany. The plates were packed into foil-bags, with a drying agent, and kept in the refrigerator. Thrombin activity was tested with chromogenic substrate S-2238 from Kabi, Sweden; undiluted reaction mixture was mixed with 9 volumes substrate and incubated at pH 8.0, tp 37 “C, 30 minutes. Absorbance was then read at 405 nm. SDS-Polvacrvlamide gel electrophoresis (SDS-PAGE) was performed according to Laemmli (13) Molecular weight standards included myosin (205 kD), I3-galactosidase (116 kD), phosphorylase B (94 kD), bovine serum albumin (68 kD), and ovalbumin (43 kD). Preparation of thrombin-HC II complex and standards Principle: An exact amount of human thrombin was incubated with a moderate molar excess of HC II in presence of dermatan sulfate which accelerates the formation of the bimolecular complex. A fixed amount of thrombin was first incubated with varying amounts of HC II. The reaction mixture was analyzed for remaining thrombin activity and run on SDS PAGE. With the amount of HC II chosen, no thrombin activity was present after incubation. SDS PAGE electrophoresis showed a prominent band at - 95000D and a faint band at - 65000, representing T-HC II complex and a slight amount of uncomplexed HC II. Procedure: A mixture was prepared of human thrombin (500 ~1 containing 16.4 pg), human HC II (200 ~1 containing about 60 pg), and dermatan sulfate (DS) (100 ~1 containing 240 ,r@. After incubation at 37 “C (pH 7.4,O.lO M NaCl, 0.05 M Tris) for 180 minutes, the tube was immersed in ice water and aliquotes stored in plastic tubes at -70 “C. These were analyzed to check absence of thrombin activity and SDS-PAGE electrophoresis (Cf above) and used for preparation of standards. The amount of T-HC II complex was calculated from the amount of thrombin (16.4 pg) in the reaction mixture and the molecular weights of thrombin (36 kD) and HC II (66 kD) assuming that all thrombin molecules took part in formation of a bimolecular complex (102 kD). Standards. The stem solution, containing 57.8 &ml was diluted l/100, l/400, and l/600 in dilution buffer. One volume of these dilutions, and of buffer, was mixed with nine volumes of a normal plasma. This plasma contained 1.3 Icglml of “intrinsic” T-HC II (Cf results). The plasma standards thus contained 59.0, 15.6, 4.8 and 1.17 ng/ml of T-HC II complex. Thrombin-antithrombin (TAT) complex was prepared similar to T-HC II complex, by incubating human thrombin with human AT (1:1.2 molar concentration) without glycosaminoglycan, for 2 hours at 37 “C.
HC II antibodies: Polyvalent IgG fraction from HC II rabbit antiserum was purified by Protein A Sepharose CL-4B (Pharmacia, Sweden). The antibody fraction (4.7 mg or 0.29 umol IgG in a volume of 500 ul) was biotinylated. Biotin-XX-NHS (Calbiochem cat. No 203114), 130 ,ug
THROMBIN-HC II COMPLEXES IN DIC
594
Vol. 66, No. 5
(10 times molar excess of the antibody) was dissolved in 50 ~1 DMSO and mixed with the antibody solution. 100 ,ulO.5 M borate buffer pH 8.0 were added and the mixture was incubated at 4 “C overnight. After the incubation 50 ~1 1M ethanolamine pH 8.0 were added. Free biotin was removed by gel filtration using Biogen P30 (Biorad) with a column size of 1.5 x 15 cm and PBS as buffer. The pooled fractions containing biotinylated HC II antibodies were stabilized by adding BSA to a final 1 % concentration. Aliquotes were stored at -70 “C. A l:lO,OOO dilution was used in the ELISA assay. ELISA assay of T-HC II complex: A sandwich technique using two different antibodies which bind selectively to the corresponding antigens of the complex was used. Plasma was first incubated with immobilized antibodies to thrombin. After washing, a biotinylated antibody towards HC II was used to detect bound T-HC II complex. Procedure: Plasma (standards and tests) were diluted 1:2 with dilution buffer. 100 ~1 diluted plasma was pipetted in each well, incubated in humidity-chamber with stirring for 2 hours at 20 - 25 “C. After three washings, 100 ~1 of biotinylated anti-HC II (diluted l:lO,OOO) was added and incubated over night at room temperature humidity-chamber with stirring). After three washings, 100 ,~l Streptavidin, (diluted l:lO,OOO) was added and incubated for 1 hour at room temperature (humidity-chamber with stirring). After washing, freshly made substrate was added (200 ul) and incubated in the dark for approximately 30 minutes (depending on the development of colour); 50 ul of 1 M H,SO, was added and absorbance read at 492 nm and 630 nm (blank). Each sample was run in duplicate and the mean value used for calculations.
A 492
nm
1.5
1 .o
0.5
0.1 1
T-HCII
I
I
10
100
CONC.,
ng/ml
Figure 1. Standard curve for the ELISA assay of T-HC II in plasma. To a normal plasma containing 1.3 ng/ml of T-HC II was added l/10 volume of purified T-HC II complex.
Vol. 66, No. 5
THROMBIN-HC II COMPLEXES IN DIG
595
RESULTS
There was a linear correlation in a log plot between absorbance and the amount of T-HC II added to a normal plasma from one person (Fig 1). To assess the intrinsic T-HC II concentration in this particular plasma, it was adsorbed with immobilized, specific antibodies to human thrombin. As this procedure reduced the measured T-HC II concentration by 1.3 &ml, this value was taken to reflect its intrinsic T-HC II concentration. This unabsorbed plasma was always used for preparing standards. In order to determine the T-HC II concentration in the other plasma samples from healthy persons, a (non-linear) standard curve also based on the buffer value (zero T-HC II) was used. The addition of purified HC II (200 @ml) or TAT complex (100 &ml) did not influence the measured T-HC II value.
TABLE 1. Median and Range Values in the Various Groups of Normal and Patients. AT = antithrombin, plt = platelet counts, n. d. = not determined.
n.d.
CONTROLS
95 78-113
2.6 1.7-14.8
99 77-150
1.5 0.3-6.1
55 51-100
53 30-78
25 10.5-57.0
49.5 22-60
11.5 3.8-30.0
304 135-603
82 53-116
15.5 10.0-60.0
76 40-112
14.5 0.4-28.5
186 97-313
80 53-110
6.5 3.8-7.8
73 59-115
3.8 1.4-14.3
The highest T-HC II concentration in plasma from healthy persons was 6.1 @ml; the median value was 1.5 @ml. The mean value was 2.0 @ml with an SD of 1.6 &ml. The values were thus not normally distributed, as is also evident from Fig 2. The intra-assay coefficient of variation was 7 %. In plasma from patients with clinically suspected DIC and elevated levels of fibrin-derived Ddimer and TAT complex, T-HC II values ranged 0.4 - 30.0 ng/ml with a median value of 13.5 @ml. Similar T-HC II values were found in the two DIC subgroups, with or without thrombocytopenia (Table 1). In contrast, normal or near normal T-HC II values (median 3.8 &ml) were found in plasma samples from septic patients in which DIC was excluded by normal TAT and D-Dimer values (Table 1, Fig 2). Plasma samples with elevated T-HC II (above 3.2 ng/ml) also contained TAT above normal reference, the correlation was of borderline significance (p = 0.07, n = 21).
596
Vol. 66, No. 5
THROMBIN-HC II COMPLEXES IN DIC
T-HCII (4-d 60
60
40
40 0
0 20
20
0 15
-
_a -
0
15
0 0 10
-
10 0 0 0
0
5-
u 0.0
OJ
e
NORMAL
0
0.
5
: 0 0 CTR.
DIC
DIC
P’t 3
plt N
Figure 2. Individual and median values of T-HC II in plasma in patient groups and healthy perS0ll.S.
DISCUSSION Our findings show that formation of a complex between thrombin and HC II occurs at a low rate in normal individuals, and at a higher rate in septic patients with laboratory signs of DIC. The results strongly suggest a role for HC II in the control of coagulation in viva. The low levels of HC II regularly found in severe DIC (14) is at least in part a result of consumption, as previously suggested (2). It has been assumed that HC II is primarily an extravascular thrombin inhibitor (3). It is possible that the T-HC II complexed are generated in the extravascular space and transported to the blood. Intravascular generation of thrombin is a key event in DIC, and it may seem more likely that the T-HC II complexes actually are formed intravascularly. Future studies are required to establish where the T-HC II in plasma originate.
Vol. 66, No. 5
THROMBIN-HC
II COMPLEXES IN DIC
597
Traditionally, the diagnosis of DIC has been based on prolonged clotting times, and test results supporting consumption, elevated FDP and thrombocytopenia. By these criteria, consumption coagulopathy may be hard to distinguish from liver failure (15). Therefore, we based the diagnosis of DIC on elevated levels of the molecular markers, TAT and D-dimer, signifying increased in viva generation of thrombin. The assay used presented here is designed as the TAT assay (7) and the same procedure is used for catching the T-inhibitor complex. The T-HC II level found in plasma from normal individuals is similar to those found for TAT (7). In DIC, TAT levels are markedly increased, often above the T-HC II values reported here. Surprisingly, occasional plasma samples show much higher T-HC II values than TAT values. Comparing mean values for the group of DIC patients, the T-HC II level was about half of the TAT level. The inhibitor level measured is a function both of the rate of formation, and their rate of elimination, which at present is not known. Endothelial glycosaminoglycans are probably involved in the function of several coagulation inhibitors. Heparan sulfate (HS) may be the main intravascular catalyst of the thrombin-AT reaction and may well be the receptor for Tissue factor pathway inhibitor (TFPI). Dermatan sulfate (DS) appears to be the specific catalyst of the thrombin-HC II reaction. We eluted anticoagulant GAGS from the pulmonary circulation of newborn calves; with a HS:DS ratio of about 3: 1. Although this is a higher DS fraction than would be surmised from studies on cultured cells (8), the DS-like activity was distinct. It is possible that the T-HC II complexes we have demonstrated are formed with endothelial DS as catalyst.
REFERENCES 1.
TOLLEFSEN, D.M., PESTKA, C.A., MONFAO, W.J. Activation of heparin cofactor II by dermatan sulfate. J Biol Chem 258. 6713-6716, 1983.
2.
ANDERSSON, T.R. Consumption of heparin cofactor II (Dermatan sulfate Cofactor) and antithrombin during coagulation. Thromb Res 46, 355-362, 1987.
3.
TOLLEFSEN,
4.
BERTINA, R.M., VAN DER LINDEN, I.K., ENGESSER, L., MijLLER, H.P., BROMMER, E.J.P. Hereditary heparin cofactor II deficiency and the risk of development of thrombosis. Thromb Haemost 57, 196-200, 1987.
5.
VINAZZER, H., STOCKER, K. Heparin cofactor II: Experimental approach to a new assay and clinical results. Thromb Res 61, 235-241, 1991.
6.
LAU, H.K., ROSENBERG, R.D. The isolation and characterization of a specific antibody population directed against the thrombin antithrombin complex. J Biol Chem 225, 58855893, 1980.
D.M. Heparin cofactor II. Thromb Haemost 65, 136, 1991 (Abstract).
598
THROMBIN-HC
II COMPLEXES IN DIC
Vol. 66, No. 5
7.
PELZER, H., SCHWARZ, A., HEIMBURGER, N. Determination of human thrombinantithrombin III complex in plasma with an enzyme-linked immunosorbent assay. Thromb Haemost 59 (1). 101-106, 1988.
8.
McGUIRE, E.A., TOLLEFSEN, M. Activation of Heparin cofactor II by fibroblasts and vascular smooth muscle cells. J Biol Chem 262, 169-175, 1987.
9.
HALE, G., ABILDGAARD, U., ANDERSSON T.R., LARSEN, M.L., SANDSET, P.M., LINDBOE, C.F., SKJ(dRTEN, F. Anticoagulant glycosaminoglycans (GAG) extracted from pulmonary vascular surface. Thromb Haemost 65, 951, 1991 (Abstract).
10. SIE, P., DUPOUY, D., PICHON, J., BONEU, B. Turnover study of heparin cofactor II in healthy man. Thromb Haemost 54, 635-638, 1985. 11. SIE, P., DUPOUY, D., PICHON, J., BONEU, B. Constitutional heparin cofactor deficiency associated with recurrent thrombosis. Lancet 1985, II, 414-416.
II
12. TEIEN, A.N., ABILDGAARD, U., HOOK, M. The anticoagulant effect of heparan sulfate and dermatan sulfate. Thromb Res 8, 859-867, 1976. 13. LAEMMLI, U.K. Cleavage of structural proteins during the assemble of the head of bacteriophage t 4. Nature 227, 680-685, 1970. 14. ABILDGAARD, U., LARSEN, M.L. Assay of dermatan sulfate cofactor (Heparin Cofactor II) activity in human plasma. Thromb Res 35, 1984, 257-266. 15. ANKER, E, ABILDGAARD, U., ANDERSEN, R., FAGERHOL, M., BJUNE, G. Low antithrombin in severe disease: Consumption or decreased synthesis? Stand J Haematol 30, 1983, 59-63.
