- Email: [email protected]
A functional assay of protein C in human plasma
THROMBOSIS RESEARCH 41; 501-508, 1986 0049-3848/86 $3.00 t .OO Printed in the USA. Copyright (c) 1986 Pergamon Press Ltd. All rights reserved.
A FUNCTIONAL ASSAY OF PROTEIN C IN HUMAN PLASMA
Christine M. Hickton, Peter Felding, Kyoichi Ikeda, Gunilla Martinsson and Inga Marie Nilsson Department
for Coagulation Disorders, University of Lund, General Hospital, Malmd, Sweden
(Received 8.11.1985; Accepted in original by Editor F. Brosstad)
form 18.11.1985
ABSTRACT A functional assay for protein C in plasma is described in which barium eluates of plasma are incubated with bovine thrombin and rabbit thrombomodulin to activate protein C. The activated protein C solution is added to an activated partial thromboplastin time (APTT system containing normal plasma and an APTT reagent (Dade Actin R ). The prolongation of coagulation time after recalcification in this system is taken as a measure of the anticoagulant activity of protein C. When expressed as per cent of the value in pooled normal plasma, the results obtained by this method in 34 normal controls and in 3 untreated patients with protein C deficiency were very similar to those obtained by radioimmunoassay of protein C. In 2 patients with protein C deficiency and 23 patients without, all on dicoumarol or warfarin treatment, the anticoagulant activity of protein C was less than its antigen concentration. The day to day analytical coefficient of variation (SD/mean) was 12 % at the 100 % level (n=12), and 10 % at the 25 % level (n=12).
INTRODUCTION Protein C is a vitamin K-dependent serine protease zymogen which, after activation by thrombin bound to an endothelial cofactor, thrombomodulin, acts as an anticoagulant by degrading the activated cofactors, Va and VIIIa (1). The normal plasma concentration of protein C has been estimated to be 2-6 mg/l (1, 2). Hereditary protein C deficiency associated with recurrent venous thromboembolism was first discovered with an electroimmunoassay (3)
Key words:
Protein C, thrombomodulin,
partial thromboplastin
501
time
502
FUNCTIONAL PROTEINC ASSAY
Vol. 41, No. 4
and protein C in plasma is still most commonly assayed by immunological methods. Recently, however, functional assays for protein C in plasma have been described (4-71, and functionally abnormal protein C molecules have been identified (5-7). In the functional assay described by Francis and Patch (41, protein C is separated from its inhibitor (8) by precipitation with BaC12, eluted and then activated by thrombin; thrombin is then inhibited by antithrombin III and heparin, heparin is neutralized by protamine sulfate, and finally the anticoagulant activity of protein C is determined as its ability to prolong the activated partial thromboplastin time (APTT) of normal plasma. The other functional assays (5-7) determine the amidolytic activity of activated protein C towards peptide substrates with (6, 7) or without (5) initial BaC12 or Al(OH)3 precipitation and with (5, 6) or without (7) the use of thrombomodulin for the activation. In the assay described here, protein C in plasma is precipitated with BaCl , eluted with citrate and activated with the thrombin-thrombomodulin camp fex. The ability of the activated protein C solution to prolong the APTT of normal plasma is then measured directly without the use of other inhibitors of the procoagulant activity of thrombin than the thrombomodulin already added (1, 9). The method is now in routine use at our laboratory, and has been used to identify patients with hereditary protein C deficiency.
MATERIALS
AND METHODS
Bovine thrombin, TopostasinR, was obtained from F. Hoffman La Roche & Co AG, Basel. Rabbit thrombomodulin was isolated according to Esmon et al (lo), using 3000 IU TopostasinR/g dry CNBr activated Sepharose 48 (Pharmacia Fine Chemicals AB, Uppsala, Sweden) for preparation of the affinity chromatography gel. Bovine serum albumin, 5 g/l, was added to the pooled thrombomodulin fractions. After dialysis at 4OC for 20 h against 0.05 mol/l Tris/HCl buffer, pH 7.5, containing 0.1 mol/l NaCl and 0.5 % Triton X-100, the thrombomodulin was diluted with this buffer to prolong the APTT to loo-150 set when testing the normal plasma standard in the protein C assay described below (dilution l/10 of the barium citrate eluate). The pool of suitably diluted thrombomodulin was stored in lastic tubes at below -700C until use. The APTT reagent was Dade Actin g Activated Cephaloplastin, containing rabbit brain cephalin and ellagic acid (Dade Diagnostics, Inc, Aguada, Puerto Rico). Plasma samples. Blood was drawn into siliconized glass tubes containing l/IO volume of (1.13 mol/l trisodium citrate. After centrifugation at 2000 g for 10 min, plasma was distributed into plastic tubes and kept at tempera-tures under -200C until analyzed. Individual plasma samples were obtained from 34 apparently healthy hospital staff members (18 women, 16 men, aged 21-50 years), plasma from at least 20 of them being pooled for use as a calibration standard. Plasma was also obtained both from patients with inherited protein C deficiency and from patients on stabilized oral anticoagulant treatment referred to the Department for Coagulation Disorders. Immunoadsorbed, protein C-depleted plasma was prepared by chromatography of normal pooled plasma on a column with goat anti-human protein C IgG coupled to CNBr-activated Sepharose 4B. The protein C against which the antibodies had been raised had been isolated by Dr Walter Kisiel, Seattle, USA. Before use, the column was equilibrated with 0.05 mol/l Tris-HCl, 0.1 mol/l NaCl, pH 7.5. The adsorbed plasma contained no measurable protein C antigen or activity but, of the initial levels of prothrombin, factor VII and factor X, 70-80 % remained.
Vol. 41, No. 4
FUNCTIONAL
CITRATED
Separation
PLASMA
of
BaC12
protein C from its
precipitation
Wash
inhibitor
Citrate
I Ba CITRATE
eluation
ELUATES
Activation of protein
503
PROTEIN C ASSAY
C
Dilution
with
Ca++
Addition
of thrombin-
thrombomodulin
buffer
complex
I ACTIVATED
PROTEIN
C
Determination
Citrated
of anticoagulant
APT
activity
activated
normal
reagent
plasma
+
+
protein
C +
Ca++ ? PROLONGATION
Fig. 1
OF APT
TIME
Functional protein C assay.
In the APTT test system, platelet-poor CPD plasma from blood provided by the Blood Transfusion Service, with about 0.013 mol/l sodium citrate, was used as the test base.
Functional method The principles of the method are shown in Fig. 1. Barium citrate eluates of plasma were prepared as follows: 0.5 ml of plasma was placed into a 1.5 ml capacity Eppendorf plastic centrifuge tubes, and 40 Pl of 1 mol/l BaC12 were added. The samples were gently rocked for 30 min at 4OC, and then centrifuged in an Eppendorf centrifuge (5414s) for one min. The supernatant was discarded, The precipitate was briefly re-suspended with a plastic rod in 100 Pl cold wash buffer composed of 0.05 mol/l Tris HCl, 0.1 mol/l NaCl, pH 7.5, immediately after which the series of samples were centrifuged for one min as above. The supernatant was discarded and the precipitate re-suspended for at least one min with a plastic rod in 0.15 ml/l sodium citrate elution buffer,
504
FUNCTIONAL PROTEIN C ASSAY
Vol. 41, No. 4
pH 7.5, at about 20°C. The suspension was centrifuged for 2 min and the supernatant diluted l/l and l/2 with the elution buffer, and then l/IO with an "activation buffer" composed of 0.05 mol/l Tris HCl, 0.1 mol/l NaCl, 20 mm01 CaC12, 1 g/l bovine serum albumin, pH 7.5. Thrombin-thrombomodulin solution was prepared by mixing one part of thrombin (TopostasinR), 2 NIH U/ml, in activation buffer with 2 parts of thrombomodulin solution at 200C. To activate protein C, 0.06 ml of this solution was added to 0.2 ml of diluted plasma eluate containing protein C. The mixture was incubated at 370C for 2 h in stoppered plastic tubes, which were then kept on ice until assayed (usually within 30 min). To assay the protein C activity the APTT was determined as follows: 0.1 ml of Dade ActinR and 0.1 ml test base were incubated together in a glass tube at 37OC for 3 min, after which 0.1 ml of activated protein C mixture was added to the tube followed by 0.1 ml of 30 mmol/l CaC12 solution, and the 370C clotting time was measured in a light scattering coagulometer (KabiVitrum AB). During the assay all activated protein C samples, the APTT reagent and base plasma were kept on ice, and the CaC12 solution was kept at 37oc. An eluate of normal pooled plasma was diluted l/l, l/Z, l/4, l/8 and l/16 with elution buffer, and then l/10 with activation buffer. These dilutions were designated 100 %, 50 %, 25 %, 12.5 % and 6.25 % of normal pool activity, respectivity. The APTT values of these dilutions were plotted against per cent activity on double logarithmic paper and used to convert the coagulation times obtained for the individual plasma eluates to per cent activity.
Other methods The radioimmunoassay of protein C and the electroimmunoassay of prothrombin (F 1I:Ag) were performed as described by Ikeda and Stenflo (2). In the protein C assay, a polyclonal rabbit IgG with affinity for the activation peptide of protein C was used, and the radio-iodinated activation peptide was used as the tracer. Owren's P&P (prothrombin, factor VII, factor X) and factor VII were analyzed in one-stage clotting assays as described elsewhere (II), with a thromboplastin preparation from human brain. The pool of normal plasma served as the calibration standard in all assays (including the functional assay of protein C), the analytical results for this pool being designated 100 %.
RESULTS Figure 2 illustrates the effect of different components on the APTT in the functional assay of protein C. The results shown in the figure indicate: 1) that protein C was activated to give an anticoagulant effect on the test base plasma only in the presence of both thrombin and thrombomodulin (A compared to B-H); 2) that thrombin (G), and to a lesser extent thrombin + thrombomodulin (E), had a procoagulant effect on the test base plasma; and 3) that thrombomodulin (without thrombin) had a moderate anticoagulant effect on the test base plasma (F). To economize with thrombomodulin, and to avoid too much interference with the APTT by the added thrombin and thrombomodulin, a low amount of thrombin was used, which necessitated an incubation time of about 2 h for protein C to become fully activated (Fig. 3). The concentration of Ca++ in the activation buffer, 20 mmol/l was chosen to obtain maximal activation of protein C without too much interference in
FUNCTIONAL
Vol. 41, No. 4
PROTEIN C
505
ASSAY
Fig. 2
APTT sec. -
APTT in the functional assay of protein C. Dependency of the composition on the solution added as "activated protein C". When not present, PC, T or TM is replaced by the solution buffer for the component. PC is protein C in barium citrate eluates of the normal plasma pool, T is thrombin and TM is thrombomodulin. The plasma eluates or eluate buffer were diluted l/IO with activation buffer before activation. Except for the mentioned replacements the whole assay was run as described.
APTT sec. 160 100%
??
140-
' 50%
??
30 Fig. 3
60
90
120
150
180
25%
210 min.
Dependency of APTT on the time for activation of protein C in 3 dilutions of barium eluates of normal pooled plasma. Except for varying the incubation time, the assay procedure was followed as described in the text.
FUNCTIONAL PROTEIN C ASSAY
506
Vol. 41, No. 4
Table I Mean _+ 1 SD of protein C anticoagulant activity (PC:C), protein C antigen concentration (PC:Ag) and of the ratios of PC:C to prothrombin antigen concentration (F II:Ag), factor VII procoagulant activity (F VII:C) and Owren's P&P in normal controls, patients treated with oral anticoagulants and patients with protein C deficiency
PC:C %
PC:Ag %
PC:C F 1I:Ag
F VII:C
PC:C P&P
1.05+0.20
0.8820.23
1.05_+0.21
0.53_+0.20
0.90_+0.46
1.81_+0.64
PC:C
Controls untreated,
n=34
99_+16
receiving oral anticoagulants, n=23
28211
106220
40_+10
Protein C deficiency untreated
1
38
42
0.39
2
56
49
0.47
3*
28
25
0.32
7
15
8
12
receiving oral anticoagulants 3" 4
0.80
*) the same patient
the APTT assay by the Ca ++ in the activated protein C solution. Table I gives the results of the functional assay compared with those of the radioimmunoassay, and the amounts of prothrombin antigen, factor VII coagulation activity and Owren's P&P for normal controls, patients without in herited protein C deficiency receiving oral anticoagulants, and patients with protein C deficiency receiving or not receiving oral anticoagulants. It can be seen that the patients with protein C deficiency could be identified equally well with either of the two protein C assays. With patients on oral anticoagulant therapy, identification was somewhat more difficult but could be facilitated by comparing the concentration of protein C with that of other vitamin K-dependent proteins. The functional assay gave lower values for protein C than did the radioimmunoassay for patients treated with oral anticoagulants, but the same values for untreated persons. A 1:l mixture of the normal pooled plasma and the protein C depleted plasma had 49 % protein C antigen (n=2) and 51 % protein C activity (n=13). The day to day analytical coefficient of variation (standard deviation/ mean) for the functional protein C assay was 12 % (n=12) for a pool of normal plasma, and 10 % for a pool of plasma from patients receiving oral anticoagu-
Vol. 41, No. 4
FUNCTIONAL
PROTEIN C ASSAY
507
lants (mean protein C level = 22.7 %, n=12).
DICSUSSION The use of the anticoagulant property of activated protein C in determining the p lasma protein C content has hitherto only been described by Francis and Patch (4). In their assay, Ba citrate eluates of plasma are activated in the absence of thrombomodulin and Ca ++ by 355 times more units of thrombin/ml reaction mixture than in our assay. Before using the activated protein C mixture in an APTT system, they neutralized thrombin with antithrombin III and heparin, and excess heparin with protamine sulfate. In our method thrombomodulin is used (as in vivo) to change the thrombin substrate specificity from fibrinogen to protein C (1, 9). Thus low concentrations of thrombin could be used to activate protein C, and APTT could be determined without using antithrombin III, heparin and protamine sulfate. The protein C measured by our method is precipitable with BaC12 (i.e., it contains y -carboxylated glutamic acid residues); it serves as substrate for the thrombin-thrombomodulin complex; and its activated form has anticoagulant properties (i.e., it probably interacts with protein S and phospholipids in the inactivation of factors Va and VIIIa). Thus, in contrast to functional assays of protein C hitherto described (4-71, all the known properties of this protein necessary for its anticoagulant function in vivo are tested. Using both an immunological and a functional assay i-should be possible to detect functionally abnormal protein C molecules. We have demonstrated the abnormal protein C in plasma from patients treated with warfarin (Table I), but so far all the families with hereditary protein C deficiency discovered by us have had an equally decreased level of protein C antigen and function. The functional assay of protein C described here is now in routine use at our laboratory for estimating protein C function in patients with recurrent thromboembolic disease. It is also used for assaying plasma from patients with other disorders, and with it we have found the expected low values in liver disease and disseminated intravascular coagulation. In samples sent to the laboratory for analysis, we have occasionally found very low protein C activity in conjunction with normal antigen concentrations. Using a procedure such as is outlined in Fig. 2 for these samples, it became clear that this phenomenon was due to procoagulant activity, the occurrence of which could be explained by prolonged refrigeration (6 h or overnight) of the whole blood samples before centrifugation and freezing of the plasma. When using Kabi light scattering coagulometers for the APTT determination, the Dade ActinR reagent proved superior to the General Diagnostic automated APTT reagent, whereas both reagents could be used in the Amelung coagulometer KC10 or manually. We have found the method practicable and reliable with both types of coagulometers.
ACKNOWLEDGEMENT The investigation Council (00087).
was supported by grants from the Swedish Medical Research
508
FUNCTIONAL
PROTEIN C ASSAY
Vol. 41, No. 4
REFERENCES 1. DAHLB#CK,B., FERNLUND,P. and STENFLO,J. Protein C. In: New Comprehensive Biochemistry volume on Blood Coagulation. H.C. Hemker and F.A. Zwaal (Eds.). North Holland - Elsevier 1985 (in press). 2. IKEDA,K. and STENFLO,J. A radioimmunoassay 39 297-306, 1985. -Res. __'
for protein C. Thrombosis
3. GRIFFIN,J.H., EVATT,B., ZIMMERMAN,T.S., KLEISS,A.J. and WIDEMAN,C. Deficiency of protein C in congenital thrombotic disease. J. Clin. Invest. 68, 1370-1373, 1981. ~4. FRANCIS Jr, R.B. and PATCH,M.J. A functional assay for protein C in human plasma. Thrombosis Res. -32, 605-613, 1983. 5. COMP,P.C., NIXON,R.R. and ESMON,C.T. Determination of functional levels of protein C, an antithrombotic protein, using thrombin-thrombomodulin complex. -Blood 63, 15-21, 1984. 6. SALA,N., OWEN,W.G. and COLLEN,D. A functional assay of protein C in human plasma. -Blood 63, 671-675, 1984. 7. BERTINA,R.M., BROEKMANS,A.W., KROMMENHOEK-van ES,C. and van WIJNGAARDS, A. The use of a functional and immunologic assay for plasma protein C in the study of the heterogeneity of congenital protein C deficiency. Thromb. Haemostas. (Stuttgart) -51, l-5, 1984, -_ 8. SUZUKI,K. Activated 154-161, 1984.
protein C inhibitor. Sem. Thromb. Hemostas. -' 10
9. ESMON,C.T. and ESMON,N.L. -10, 122-130, 1984.
Protein C activation.
Sem. Thromb. Hemostas.
10. ESMON,N.L., OWEN,W.G. and ESMON,C.T. Isolation of a membrane-bound cofactor for thrombin-catalyzed activation of protein C. J. Biol. Chem. 257, 859-864, 1982. 11. NILSSON,I.M. Hemorrhagic and Thrombotic & Sons 1974, pp 221 and 230.
Diseases.
London - John Wiley
A FUNCTIONAL ASSAY OF PROTEIN C IN HUMAN PLASMA
Christine M. Hickton, Peter Felding, Kyoichi Ikeda, Gunilla Martinsson and Inga Marie Nilsson Department
for Coagulation Disorders, University of Lund, General Hospital, Malmd, Sweden
(Received 8.11.1985; Accepted in original by Editor F. Brosstad)
form 18.11.1985
ABSTRACT A functional assay for protein C in plasma is described in which barium eluates of plasma are incubated with bovine thrombin and rabbit thrombomodulin to activate protein C. The activated protein C solution is added to an activated partial thromboplastin time (APTT system containing normal plasma and an APTT reagent (Dade Actin R ). The prolongation of coagulation time after recalcification in this system is taken as a measure of the anticoagulant activity of protein C. When expressed as per cent of the value in pooled normal plasma, the results obtained by this method in 34 normal controls and in 3 untreated patients with protein C deficiency were very similar to those obtained by radioimmunoassay of protein C. In 2 patients with protein C deficiency and 23 patients without, all on dicoumarol or warfarin treatment, the anticoagulant activity of protein C was less than its antigen concentration. The day to day analytical coefficient of variation (SD/mean) was 12 % at the 100 % level (n=12), and 10 % at the 25 % level (n=12).
INTRODUCTION Protein C is a vitamin K-dependent serine protease zymogen which, after activation by thrombin bound to an endothelial cofactor, thrombomodulin, acts as an anticoagulant by degrading the activated cofactors, Va and VIIIa (1). The normal plasma concentration of protein C has been estimated to be 2-6 mg/l (1, 2). Hereditary protein C deficiency associated with recurrent venous thromboembolism was first discovered with an electroimmunoassay (3)
Key words:
Protein C, thrombomodulin,
partial thromboplastin
501
time
502
FUNCTIONAL PROTEINC ASSAY
Vol. 41, No. 4
and protein C in plasma is still most commonly assayed by immunological methods. Recently, however, functional assays for protein C in plasma have been described (4-71, and functionally abnormal protein C molecules have been identified (5-7). In the functional assay described by Francis and Patch (41, protein C is separated from its inhibitor (8) by precipitation with BaC12, eluted and then activated by thrombin; thrombin is then inhibited by antithrombin III and heparin, heparin is neutralized by protamine sulfate, and finally the anticoagulant activity of protein C is determined as its ability to prolong the activated partial thromboplastin time (APTT) of normal plasma. The other functional assays (5-7) determine the amidolytic activity of activated protein C towards peptide substrates with (6, 7) or without (5) initial BaC12 or Al(OH)3 precipitation and with (5, 6) or without (7) the use of thrombomodulin for the activation. In the assay described here, protein C in plasma is precipitated with BaCl , eluted with citrate and activated with the thrombin-thrombomodulin camp fex. The ability of the activated protein C solution to prolong the APTT of normal plasma is then measured directly without the use of other inhibitors of the procoagulant activity of thrombin than the thrombomodulin already added (1, 9). The method is now in routine use at our laboratory, and has been used to identify patients with hereditary protein C deficiency.
MATERIALS
AND METHODS
Bovine thrombin, TopostasinR, was obtained from F. Hoffman La Roche & Co AG, Basel. Rabbit thrombomodulin was isolated according to Esmon et al (lo), using 3000 IU TopostasinR/g dry CNBr activated Sepharose 48 (Pharmacia Fine Chemicals AB, Uppsala, Sweden) for preparation of the affinity chromatography gel. Bovine serum albumin, 5 g/l, was added to the pooled thrombomodulin fractions. After dialysis at 4OC for 20 h against 0.05 mol/l Tris/HCl buffer, pH 7.5, containing 0.1 mol/l NaCl and 0.5 % Triton X-100, the thrombomodulin was diluted with this buffer to prolong the APTT to loo-150 set when testing the normal plasma standard in the protein C assay described below (dilution l/10 of the barium citrate eluate). The pool of suitably diluted thrombomodulin was stored in lastic tubes at below -700C until use. The APTT reagent was Dade Actin g Activated Cephaloplastin, containing rabbit brain cephalin and ellagic acid (Dade Diagnostics, Inc, Aguada, Puerto Rico). Plasma samples. Blood was drawn into siliconized glass tubes containing l/IO volume of (1.13 mol/l trisodium citrate. After centrifugation at 2000 g for 10 min, plasma was distributed into plastic tubes and kept at tempera-tures under -200C until analyzed. Individual plasma samples were obtained from 34 apparently healthy hospital staff members (18 women, 16 men, aged 21-50 years), plasma from at least 20 of them being pooled for use as a calibration standard. Plasma was also obtained both from patients with inherited protein C deficiency and from patients on stabilized oral anticoagulant treatment referred to the Department for Coagulation Disorders. Immunoadsorbed, protein C-depleted plasma was prepared by chromatography of normal pooled plasma on a column with goat anti-human protein C IgG coupled to CNBr-activated Sepharose 4B. The protein C against which the antibodies had been raised had been isolated by Dr Walter Kisiel, Seattle, USA. Before use, the column was equilibrated with 0.05 mol/l Tris-HCl, 0.1 mol/l NaCl, pH 7.5. The adsorbed plasma contained no measurable protein C antigen or activity but, of the initial levels of prothrombin, factor VII and factor X, 70-80 % remained.
Vol. 41, No. 4
FUNCTIONAL
CITRATED
Separation
PLASMA
of
BaC12
protein C from its
precipitation
Wash
inhibitor
Citrate
I Ba CITRATE
eluation
ELUATES
Activation of protein
503
PROTEIN C ASSAY
C
Dilution
with
Ca++
Addition
of thrombin-
thrombomodulin
buffer
complex
I ACTIVATED
PROTEIN
C
Determination
Citrated
of anticoagulant
APT
activity
activated
normal
reagent
plasma
+
+
protein
C +
Ca++ ? PROLONGATION
Fig. 1
OF APT
TIME
Functional protein C assay.
In the APTT test system, platelet-poor CPD plasma from blood provided by the Blood Transfusion Service, with about 0.013 mol/l sodium citrate, was used as the test base.
Functional method The principles of the method are shown in Fig. 1. Barium citrate eluates of plasma were prepared as follows: 0.5 ml of plasma was placed into a 1.5 ml capacity Eppendorf plastic centrifuge tubes, and 40 Pl of 1 mol/l BaC12 were added. The samples were gently rocked for 30 min at 4OC, and then centrifuged in an Eppendorf centrifuge (5414s) for one min. The supernatant was discarded, The precipitate was briefly re-suspended with a plastic rod in 100 Pl cold wash buffer composed of 0.05 mol/l Tris HCl, 0.1 mol/l NaCl, pH 7.5, immediately after which the series of samples were centrifuged for one min as above. The supernatant was discarded and the precipitate re-suspended for at least one min with a plastic rod in 0.15 ml/l sodium citrate elution buffer,
504
FUNCTIONAL PROTEIN C ASSAY
Vol. 41, No. 4
pH 7.5, at about 20°C. The suspension was centrifuged for 2 min and the supernatant diluted l/l and l/2 with the elution buffer, and then l/IO with an "activation buffer" composed of 0.05 mol/l Tris HCl, 0.1 mol/l NaCl, 20 mm01 CaC12, 1 g/l bovine serum albumin, pH 7.5. Thrombin-thrombomodulin solution was prepared by mixing one part of thrombin (TopostasinR), 2 NIH U/ml, in activation buffer with 2 parts of thrombomodulin solution at 200C. To activate protein C, 0.06 ml of this solution was added to 0.2 ml of diluted plasma eluate containing protein C. The mixture was incubated at 370C for 2 h in stoppered plastic tubes, which were then kept on ice until assayed (usually within 30 min). To assay the protein C activity the APTT was determined as follows: 0.1 ml of Dade ActinR and 0.1 ml test base were incubated together in a glass tube at 37OC for 3 min, after which 0.1 ml of activated protein C mixture was added to the tube followed by 0.1 ml of 30 mmol/l CaC12 solution, and the 370C clotting time was measured in a light scattering coagulometer (KabiVitrum AB). During the assay all activated protein C samples, the APTT reagent and base plasma were kept on ice, and the CaC12 solution was kept at 37oc. An eluate of normal pooled plasma was diluted l/l, l/Z, l/4, l/8 and l/16 with elution buffer, and then l/10 with activation buffer. These dilutions were designated 100 %, 50 %, 25 %, 12.5 % and 6.25 % of normal pool activity, respectivity. The APTT values of these dilutions were plotted against per cent activity on double logarithmic paper and used to convert the coagulation times obtained for the individual plasma eluates to per cent activity.
Other methods The radioimmunoassay of protein C and the electroimmunoassay of prothrombin (F 1I:Ag) were performed as described by Ikeda and Stenflo (2). In the protein C assay, a polyclonal rabbit IgG with affinity for the activation peptide of protein C was used, and the radio-iodinated activation peptide was used as the tracer. Owren's P&P (prothrombin, factor VII, factor X) and factor VII were analyzed in one-stage clotting assays as described elsewhere (II), with a thromboplastin preparation from human brain. The pool of normal plasma served as the calibration standard in all assays (including the functional assay of protein C), the analytical results for this pool being designated 100 %.
RESULTS Figure 2 illustrates the effect of different components on the APTT in the functional assay of protein C. The results shown in the figure indicate: 1) that protein C was activated to give an anticoagulant effect on the test base plasma only in the presence of both thrombin and thrombomodulin (A compared to B-H); 2) that thrombin (G), and to a lesser extent thrombin + thrombomodulin (E), had a procoagulant effect on the test base plasma; and 3) that thrombomodulin (without thrombin) had a moderate anticoagulant effect on the test base plasma (F). To economize with thrombomodulin, and to avoid too much interference with the APTT by the added thrombin and thrombomodulin, a low amount of thrombin was used, which necessitated an incubation time of about 2 h for protein C to become fully activated (Fig. 3). The concentration of Ca++ in the activation buffer, 20 mmol/l was chosen to obtain maximal activation of protein C without too much interference in
FUNCTIONAL
Vol. 41, No. 4
PROTEIN C
505
ASSAY
Fig. 2
APTT sec. -
APTT in the functional assay of protein C. Dependency of the composition on the solution added as "activated protein C". When not present, PC, T or TM is replaced by the solution buffer for the component. PC is protein C in barium citrate eluates of the normal plasma pool, T is thrombin and TM is thrombomodulin. The plasma eluates or eluate buffer were diluted l/IO with activation buffer before activation. Except for the mentioned replacements the whole assay was run as described.
APTT sec. 160 100%
??
140-
' 50%
??
30 Fig. 3
60
90
120
150
180
25%
210 min.
Dependency of APTT on the time for activation of protein C in 3 dilutions of barium eluates of normal pooled plasma. Except for varying the incubation time, the assay procedure was followed as described in the text.
FUNCTIONAL PROTEIN C ASSAY
506
Vol. 41, No. 4
Table I Mean _+ 1 SD of protein C anticoagulant activity (PC:C), protein C antigen concentration (PC:Ag) and of the ratios of PC:C to prothrombin antigen concentration (F II:Ag), factor VII procoagulant activity (F VII:C) and Owren's P&P in normal controls, patients treated with oral anticoagulants and patients with protein C deficiency
PC:C %
PC:Ag %
PC:C F 1I:Ag
F VII:C
PC:C P&P
1.05+0.20
0.8820.23
1.05_+0.21
0.53_+0.20
0.90_+0.46
1.81_+0.64
PC:C
Controls untreated,
n=34
99_+16
receiving oral anticoagulants, n=23
28211
106220
40_+10
Protein C deficiency untreated
1
38
42
0.39
2
56
49
0.47
3*
28
25
0.32
7
15
8
12
receiving oral anticoagulants 3" 4
0.80
*) the same patient
the APTT assay by the Ca ++ in the activated protein C solution. Table I gives the results of the functional assay compared with those of the radioimmunoassay, and the amounts of prothrombin antigen, factor VII coagulation activity and Owren's P&P for normal controls, patients without in herited protein C deficiency receiving oral anticoagulants, and patients with protein C deficiency receiving or not receiving oral anticoagulants. It can be seen that the patients with protein C deficiency could be identified equally well with either of the two protein C assays. With patients on oral anticoagulant therapy, identification was somewhat more difficult but could be facilitated by comparing the concentration of protein C with that of other vitamin K-dependent proteins. The functional assay gave lower values for protein C than did the radioimmunoassay for patients treated with oral anticoagulants, but the same values for untreated persons. A 1:l mixture of the normal pooled plasma and the protein C depleted plasma had 49 % protein C antigen (n=2) and 51 % protein C activity (n=13). The day to day analytical coefficient of variation (standard deviation/ mean) for the functional protein C assay was 12 % (n=12) for a pool of normal plasma, and 10 % for a pool of plasma from patients receiving oral anticoagu-
Vol. 41, No. 4
FUNCTIONAL
PROTEIN C ASSAY
507
lants (mean protein C level = 22.7 %, n=12).
DICSUSSION The use of the anticoagulant property of activated protein C in determining the p lasma protein C content has hitherto only been described by Francis and Patch (4). In their assay, Ba citrate eluates of plasma are activated in the absence of thrombomodulin and Ca ++ by 355 times more units of thrombin/ml reaction mixture than in our assay. Before using the activated protein C mixture in an APTT system, they neutralized thrombin with antithrombin III and heparin, and excess heparin with protamine sulfate. In our method thrombomodulin is used (as in vivo) to change the thrombin substrate specificity from fibrinogen to protein C (1, 9). Thus low concentrations of thrombin could be used to activate protein C, and APTT could be determined without using antithrombin III, heparin and protamine sulfate. The protein C measured by our method is precipitable with BaC12 (i.e., it contains y -carboxylated glutamic acid residues); it serves as substrate for the thrombin-thrombomodulin complex; and its activated form has anticoagulant properties (i.e., it probably interacts with protein S and phospholipids in the inactivation of factors Va and VIIIa). Thus, in contrast to functional assays of protein C hitherto described (4-71, all the known properties of this protein necessary for its anticoagulant function in vivo are tested. Using both an immunological and a functional assay i-should be possible to detect functionally abnormal protein C molecules. We have demonstrated the abnormal protein C in plasma from patients treated with warfarin (Table I), but so far all the families with hereditary protein C deficiency discovered by us have had an equally decreased level of protein C antigen and function. The functional assay of protein C described here is now in routine use at our laboratory for estimating protein C function in patients with recurrent thromboembolic disease. It is also used for assaying plasma from patients with other disorders, and with it we have found the expected low values in liver disease and disseminated intravascular coagulation. In samples sent to the laboratory for analysis, we have occasionally found very low protein C activity in conjunction with normal antigen concentrations. Using a procedure such as is outlined in Fig. 2 for these samples, it became clear that this phenomenon was due to procoagulant activity, the occurrence of which could be explained by prolonged refrigeration (6 h or overnight) of the whole blood samples before centrifugation and freezing of the plasma. When using Kabi light scattering coagulometers for the APTT determination, the Dade ActinR reagent proved superior to the General Diagnostic automated APTT reagent, whereas both reagents could be used in the Amelung coagulometer KC10 or manually. We have found the method practicable and reliable with both types of coagulometers.
ACKNOWLEDGEMENT The investigation Council (00087).
was supported by grants from the Swedish Medical Research
508
FUNCTIONAL
PROTEIN C ASSAY
Vol. 41, No. 4
REFERENCES 1. DAHLB#CK,B., FERNLUND,P. and STENFLO,J. Protein C. In: New Comprehensive Biochemistry volume on Blood Coagulation. H.C. Hemker and F.A. Zwaal (Eds.). North Holland - Elsevier 1985 (in press). 2. IKEDA,K. and STENFLO,J. A radioimmunoassay 39 297-306, 1985. -Res. __'
for protein C. Thrombosis
3. GRIFFIN,J.H., EVATT,B., ZIMMERMAN,T.S., KLEISS,A.J. and WIDEMAN,C. Deficiency of protein C in congenital thrombotic disease. J. Clin. Invest. 68, 1370-1373, 1981. ~4. FRANCIS Jr, R.B. and PATCH,M.J. A functional assay for protein C in human plasma. Thrombosis Res. -32, 605-613, 1983. 5. COMP,P.C., NIXON,R.R. and ESMON,C.T. Determination of functional levels of protein C, an antithrombotic protein, using thrombin-thrombomodulin complex. -Blood 63, 15-21, 1984. 6. SALA,N., OWEN,W.G. and COLLEN,D. A functional assay of protein C in human plasma. -Blood 63, 671-675, 1984. 7. BERTINA,R.M., BROEKMANS,A.W., KROMMENHOEK-van ES,C. and van WIJNGAARDS, A. The use of a functional and immunologic assay for plasma protein C in the study of the heterogeneity of congenital protein C deficiency. Thromb. Haemostas. (Stuttgart) -51, l-5, 1984, -_ 8. SUZUKI,K. Activated 154-161, 1984.
protein C inhibitor. Sem. Thromb. Hemostas. -' 10
9. ESMON,C.T. and ESMON,N.L. -10, 122-130, 1984.
Protein C activation.
Sem. Thromb. Hemostas.
10. ESMON,N.L., OWEN,W.G. and ESMON,C.T. Isolation of a membrane-bound cofactor for thrombin-catalyzed activation of protein C. J. Biol. Chem. 257, 859-864, 1982. 11. NILSSON,I.M. Hemorrhagic and Thrombotic & Sons 1974, pp 221 and 230.
Diseases.
London - John Wiley