- Email: [email protected]
A Chromogenic Substrate Assay Kit for Factor XII
Thrombosis Research 94 (1999) 103–109
REGULAR ARTICLE
A Chromogenic Substrate Assay Kit for Factor XII: Evaluation and Use for the Measurement of Factor XII Levels in Cardiopulmonary Bypass Patients Michael J. Gallimore1, D.W. Jones1 and Hans P. Wendel2 Kent Haemophilia Centre, Kent and Canterbury Hospital, Ethelbert Road, Canterbury, Kent, United Kingdom; 2Clinical Research Laboratory, Department of Throracic and Cardiovascular Surgery, University of Tuebingen, 72076 Tuebingen, Germany. 1
(Received 30 July 1998 by Editor C.A. Ludlam; revised/accepted 17 November 1998)
Abstract Factor XII levels were determined in plasma samples from 75 patients before undergoing aortocoronary bypass grafting and from 40 healthy age-matched donors by using a microtitre plate adaptation of a new chromogenic peptide substrate assay kit for factor XII. The chromogenic peptide substrate assay values for factor XII correlated well with those obtained in clotting (r50.90; y5 15.81110.8236x) and immunochemical (r50.88; y517.9010.817x) assays in the normal donor samples. Factor XII levels in the patients were significantly lower than those in the normal donors (83.3623.2% versus 103.4623.1; p50.004), and nine patients (12%) had factor XII values below 50% compared with only one of the normal donors (2.5%). Factor XII levels and kallikrein-like activities (a measure of contact system activation) were followed before, during, and one day after cardiopulmonary bypass in 20 patients. Factor XII levels were significantly reduced, and kallikrein-like activities significantly elevated after 5 and 30 minutes cardiopulmonary bypass. One day after cardiopulAbbreviations: FXII, factor XII; DDAVP, desamino D-arginine vasopressin; CPB, cardiopulmonary bypass; FXIIct, factor XII clotting activity; FXIIcs, factor XII determined by chromogenic substrate assay; FXIIag, factor XII antigen. Corresponding author: M.J. Gallimore, Kent Haemophilia Centre, Kent and Canterbury Hospital, Ethelbert Road, Canterbury, Kent, UK. Tel: 44 (0)1227 783 168; Fax: 44 (0) 1227 783 158.
monary bypass both factor XII levels and kallikrein-like activities were significantly lower than preoperation values. 1999 Elsevier Science Ltd. All rights reserved. Key Words: Factor XII; Chromogenic substrate assay; Cardiopulmonary bypass
S
ubsequent to its discovery as a contact coagulation defect in Mr. Hageman, factor XII (FXII) has been shown to be involved in the other plasma defense systems (kallikrein-kinin, complement, and fibrinolysis) and also affects the activities of polymorphonuclear granulocytes, platelets, and other cells [1–3]. Recent interest has focused on the role of FXII in fibrinolysis and thrombosis as it has been shown that FXII-deficient plasma has reduced fibrinolytic activity [4], that FXII-deficient patients have impaired fibrinolytic activity when exposed to desamino D-arginine vasopressin (DDAVP) [5], and that FXII deficiency might be a risk factor for thrombosis [6–11]. Halbmayer et al. [12,13] using clotting and immunochemical assays to measure FXII levels, reported a significantly higher prevalence of FXII deficiency in patients with coronary artery disease (10.3%) than among healthy blood donors (2.3%). We and others have reported that the FXIIrelated contact systems of blood are activated during cardiopulmonary bypass (CPB [14,15]) and that increased fibrinolytic activity occurs as a result
0049-3848/99 $–see front matter 1999 Elsevier Science Ltd. All rights reserved. PII S0049-3848(98)00202-3
104
M.J. Gallimore et al./Thrombosis Research 94 (1999) 103–109
of this activation [16,17], and we have suggested that inhibition of this activation by aprotinin therapy contributes to reduced blood loss in cardiac surgery [18]. Most studies on FXII have been performed with one stage clotting assays by using FXII deficient plasma or immunochemical assays. As these are both expensive and time consuming, both direct and indirect chromogenic peptide substrate assays for FXII have been described [1,19]. These assays, however, are rarely used routinely. A chromogenic substrate assay kit for routine determinations of FXII in clinical samples is now available, and we have evaluated this kit and used it to measure FXII levels in normal healthy subjects and in heart disease patients before, during, and after CPB. The results of our studies are reported in this article.
1. Materials and Methods Ethical committee approval for this study was obtained from the Ethical Committee of Tuebingen University Hospital, Calwerstrass 7, 72076 Tuebingen, Germany. Plasma samples from 75 male patients undergoing aortocoronary bypass grafting (mean age 61.667.55 years, minimum 38, maximum 70 years) and from 40 age-matched healthy male donors (mean age 61.767.43 years, minimum 40, maximum 70 years) were studied. Collection of blood, preparation of plasma, and sampling scheme for the CPB patients were performed as described previously [14,18]. A one-stage clotting assay on an automated coagulometer (Biomatic 2000; Sarstedt, Numbrecht, Germany) using FXII deficient plasma (Immuno Ltd., Vienna, Austria) was used to measure FXII clotting activity levels (FXIIct). FXII antigen (FXIIag) levels were determined by rocket immunoelectrophoresis by using FXII antiserum from Nordic Immunological Laboratories (Tilburg, The Netherlands). Levels of FXII determined by chromogenic substrate assay (FXIIcs) were performed using a microtitre plate technique using a Dynatech MR 5000 reader (Dynatech, Denkendorf, Germany) with FXII chromogenic substrate assay kits (Unitest FXIIe; Unicorn Diagnostics Ltd., London, UK). An evaluation of this kit by using an assay modified
for use on the Cobas Mira Plus analyzer has been reported recently [20]. The microtitre plate assay allows all laboratories with a microtitre plate reader to routinely measure FXII levels in plasma samples. The kit contains all of the reagents needed for the determination of FXII. These include: a standard plasma of known FXII content, an optimized buffer for diluting the plasma, a novel FXII activator (CONT-ACT FXIIt; Channel Diagnostics, Walmer, Kent, UK), an inhibitor cocktail to inhibit proteolytic enzymes other than a and bFXIIa and a chromogenic peptide substrate for a and bFXIIa (2AcOH. H-D-CHT-Gly-Arg-pNA). All reagents are supplied in freeze-dried form with the exception of the buffer. The freeze-dried kit components were reconstituted in sterile distilled water as follows: standard plasma 1 ml; CONT-ACT FXII activator 5 ml; inhibitor cocktail 10 ml; substrate (10 mmol) 10 ml. The buffer (10 ml) was diluted with 90 ml distilled water (buffer A). Inhibitor cocktail solution (2 ml) was added to 48 ml of buffer A (buffer B). The standard plasma and test plasmas have to be treated with organic solvent to destroy inhibitors of kallikrein in the activation stage of the assay and inhibitors of a and bFXIIa after activation. Both acetone or chloroform can be used for this purpose. We used acetone. One volume of acetone was added to 3 volumes of plasma in 1.5 ml polypropylene microcentrifuge tubes, the solutions mixed well, and the tubes were left for 30 minutes at 48C. The samples were then centrifuged (2000 3g for 15 minutes), and the supernatants were removed with plastic pipettes.
1.1. Preparation of the Standard Curve and Assay Method Full details of the preparation of the standard curve and the microtitre plate assay method are given in the kit insert. A brief description is as follows: the acetone treated standard plasma was diluted with buffer A to give values from 25 to 125%. Test plasmas were diluted the same as the 100% standard plasma dilution. Into the wells of a microtitre plate at 378C were pipetted 25 ml plasma dilution or buffer and 25 ml CONT-ACT FXII solution (at 378C). After mixing and incubation for exactly 10 minutes, 75 ml buffer B was added. After mixing and incubation for ex-
M.J. Gallimore et al./Thrombosis Research 94 (1999) 103–109
actly 3 minutes, 50 ml factor XII substrate (at 378C) was added. Finally, after mixing and incubation for exactly 10 minutes, 50 ml acetic acid (50%) was added to stop the reaction. The optical densities were then read at 405 nm in the microtitre plate reader. Blanks were performed substituting buffer A for the CONT-ACT FXII solution.
1.2. Calculation of FXII Level The A405 values for the blanks were subtracted from the values obtained for the standard curve and test samples. The delta A405 values for the standard curve samples were plotted against the percentage standard plasma, and the values for the test plasmas were obtained by plotting their delta A405 values on this standard curve. The values can be expressed either as a percentage or in units per milliliter (U/ml) by applying the formula: FXII (U/ml) % Activity3Potency of the Standard Plasma 5 100 The potency of the standard plasma is given with the kit. Test samples with values higher than 150% were diluted one part of diluted plasma plus one part buffer, reassayed, and the results obtained multiplied by 2. Kallikrein-like activities (thought to be a measure of plasma kallikrein-a2 macroglobulin complexes [21]) were measured using kits supplied by Unicorn Diagnostics Ltd. The method in this kit is essentially as described by Gallimore and Friberger [22] using a more specific substrate for plasma kallikrein (MBz-Pro-Phe-Arg-pNA). Lupus anticoagulant assays were performed on double centrifuged plasma samples by a dilute Russell’s viper venom time technique and platelet correction procedure as recommended by the British Society of Haematology [23], using kits supplied by Unicorn Diagnostics Ltd. All assays were performed in duplicate.
1.3. Statistical Analysis Correlation values and regression analysis were performed using Statistica Mac, Statsoft Ltd., Uckfield, UK.
105
Statistical significance was set at a p value of ,0.05 and calculated using t-test, while correlation between assays was conducted using Pearson correlation.
2. Results The imprecision of the FXIIcs microtitre plate assay was determined by analyzing plasma samples from donors with FXII levels of 114 and 38%. The analyses were performed on five different occasions including five replicates of each plasma. The interassay precision values were 6.15 and 3.8%, and the intraassay precision values were 4.9 and 2.6%, respectively. The assay was able to detect levels of FXII as low as 3% and with suitable dilution levels higher than 200%. Correlation plots for FXIIcs versus FXIIct and FXIIcs versus FXIIag in plasma samples from the 40 normal healthy blood donors are shown in Figures 1a and 1b. Satisfactory correlations were obtained for FXIIcs versus the other two assays (FXIIcs vs. FXIIct50.90, FXIIcs vs. FXIIag50.88). The regression line equation for FXIIcs vs. FXIIct was y5 15.81110.824x and for FXIIcs vs. FXIIag y5 17.9010.817x. Mean values, standard deviations, and ranges for the three assays were FXIIcs5103.4623.1%, 50–182%; FXIIct5106.7624.6%, 53–206%; FXIIag5105.3624.4%, 50–191%. When FXII levels in plasma samples from 75 patients awaiting cardiac surgery were determined using the chromogenic substrate assay, significantly lower values were obtained when compared with the age-matched control group (p50.004, Table 1). All plasma samples from the normal donor and patient groups were lupus anticoagulant negative. In the heart disease patient group, 9 of the 75 patients had low levels of FXII (range 39–50%) as determined in the FXIIcs assay (incidence of 12%) compared with only 1 of the 40 normal healthy donors (incidence of 2.5%). All of the heart disease patients with low FXII levels had normal values for liver function tests (results not shown). Twenty heart disease patients were studied before, during, and after cardiopulmonary bypass surgery. FXII levels (measured by the FXIIcs assay and corrected for haemodilution) fell during CPB
106
M.J. Gallimore et al./Thrombosis Research 94 (1999) 103–109
Fig. 1. (a) Correlation data between factor XII (FXII) levels in 40 healthy blood donors by using the chromogenic substrate and clotting assays. The correlation coefficient (r) for 40 data pairs was 0.90, and the regression line equation was y515.811110.8236x. (b) Correlation data between factor XII (FXII) levels in 40 healthy blood donors by using the chromogenic substrate and antigen assays. The correlation coefficient (r) for 40 data pairs was 0.88, and the regression line equation was y517.9000210.8171x.
and remained reduced throughout and immediately after the operation. One day after operation, FXII levels were significantly lower than preoperation values (Figure 2). In view of the high heparin levels used during CPB, heparin was added to the standard plasma to give a range of heparin levels from 0 to 5.0 U/ml and the samples tested in the FXIIcs assay. Heparin had no effect on the FXIIcs assay, even at plasma levels of 5.0 U/ml (results not shown). Kallikrein-like activities were significantly elevated after 5 and 30 minutes of CPB but had returned to normal immediately after operation. One day after surgery kallikrein-like activities were significantly lower than preoperation values (Figure 3).
samples; however, both of these assays are cumbersome and rather expensive, and the former requires FXII-deficient plasma as an essential component. The chromogenic peptide substrate assay kit for FXII used in our studies allows us to routinely measure FXII without the need for FXII-deficient plasma. The assay is very easy to perform, gives good inter and intra-assay variability, can easily be automated and, unlike the clotting assay, is not affected by phospholipid antibodies [20], and very relevant to the present study, is not affected by heparin. The chromogenic peptide substrate used has been selected because it has good sensitivity to a and bFXIIa and is rather insensitive to plasma kallikrein and plasmin. Specificity for a
3. Discussion There is a growing interest in the FXII molecule as after its activation it can influence both fluid and cellular defense systems. There is also interest in the role of this protein as a possible regulator of fibrinolysis and current debate as to whether FXII deficiency is a risk factor for thrombosis [1,20,24]. Historically, clotting and immunochemical assays have been used to measure FXII levels in plasma
Table 1. Levels of FXII in plasma samples from 40 healthy donors and 75 patients with heart disease determined using the FXIIcs assay Normals (n540) Mean 6SD Range Significance
FXIIcs% 103.4 23.1 50–182
Heart disease (n575)
FXIIcs% 83.3 23.2 39–140 0.004
M.J. Gallimore et al./Thrombosis Research 94 (1999) 103–109
107
Fig. 2. Levels of FXII in plasma samples from 20 patients before, during, and after cardiopulmonary bypass determined using the FXIIcs assay. Results shown as mean6SD. Asterisks indicate significantly different from preoperation values.
Fig. 3. Kallikrein-like activities in plasma samples from 20 patients before, during, and after cardiopulmonary bypass. Results shown as mean6SD. Asterisks indicate significantly different from preoperation values.
and bFXIIa is further ensured by the use of an inhibitor cocktail. When FXII levels in plasma samples from 40 healthy blood donors were determined using the FXIIcs microtitre plate assay, satisfactory correlations with the FXIIct and FXIIag assays were obtained (Figures 1a and 1b). The mean values for the three assays were all slightly above 100%, which might reflect an effect of age on FXII levels (our donors were all older than 40), although our results were very similar to those obtained by others in clotting and immunochemical assays for FXII in large populations of normal healthy donors [8, 13,20,25]. The levels of FXII in 75 patients awaiting cardiopulmonary bypass surgery (83.3%623.2%) were significantly lower (p50.004) than in the agematched control group (Table 1). Also, the incidence of FXII deficiency (FXIIcs value ,50%) in this patient group (9 out of 75 patients incidence 512%) was higher than that for the control group (1 out of 40 incidence 52.5%). These results are very similar to those obtained by Halbmayer et al. [12] who measured FXIIct and FXIIag levels in 426 patients with coronary heart disease awaiting surgery and found an incidence of FXII deficiency of 10.3%. The contact systems of blood are activated dur-
ing cardiopulmonary bypass and because of the high heparin levels used coagulation assays for FXII are unreliable. Because of the possible role of FXII in fibrinolysis and blood loss in CPB [18], we feel it is of interest to determine the functional activity of this contact system protein during CPB. We therefore used the FXIIcs assay to measure FXII levels in 20 patients before, during, and after CPB. The FXII levels fell during CPB and were already significantly lower (p50.001) after 5 minutes CPB (Figure 2). Kallikrein-like activities, a measure of contact system activation [21], were elevated during CPB and were already significantly higher (p50.027) than preoperation values after 5 minutes CPB (Figure 3). Kallikrein-like activities and FXII values were significantly lower than the preoperation values on the day after surgery (p50.008 and 0.004, respectively). Of the 20 patients studied, four (20%) had levels of FXII below 50% 1 day after surgery. Despite the accepted classic role of FXII in intrinsic blood coagulation, none of these patients exhibited enhanced blood loss during or after CPB (results not shown). Our results confirm that FXII activation occurs during CPB and show that plasma levels of this protein that fall during CPB do not recover to preoperation values on the day after surgery but in
108
M.J. Gallimore et al./Thrombosis Research 94 (1999) 103–109
fact fall further, reflecting either continued activation of FXII or the negative acute phase synthesis of FXII recently described [26]. To our knowledge, this is a first report of this finding after CPB and in view of the possibility that reduced FXII levels might constitute a risk for thrombosis further study in this area is obviously warranted. Although FXII has been studied extensively in research laboratories, the lack of simple, reasonably priced, accurate manual and automated assays for this protein has limited the scope of clinical studies. The introduction of chromogenic peptide substrate technology for the determination of this important proenzyme will hopefully rectify this situation in the future.
References 1. Fuhrer G, Gallimore MJ, Heller W, Hoffmeister H-E. FXII. Blut 1990;61:258–66. 2. Saito H. The contact phase of blood coagulation. In: Bloom AL, Forbes CD, Thomas DP, Tuddenham EGD, editors. Haemostasis and Thrombosis, Vol. 1, 3rd ed. Edinburgh: Churchill Livingstone; 1994. p. 289–307. 3. Colman RW, Schmaier AH. Contact System: A vascular biology modulator with anticoagulant, profibrinolytic, antiadhesive and proinflammatory attributes. Blood 1997; 90:3819–43. 4. Kluft C, Dooijewaard GG, Emeis JJ. Role of the contact system in fibrinolysis. Sem Thromb Haemost 1987;13:50–68. 5. Levi M, Hack CE, de Boer JP, Brandjes DPM, Bu¨ller HR, ten Cate JW. Reduction of contact activation related fibrinolytic activity in factor XII deficient patients. J Clin Invest 1991; 88:1155–60. 6. Goodnough LT, Saito H, Ratnoff OD. Thrombosis or myocardial infarction in congenital clotting factors abnormalities and chronic thrombocytopenias: A report of 21 patients and a review of 50 previously reported cases. Medicine 1983;62:248–255. 7. Mannhalter C, Fischer M, Hopmeier P, Deutsch E. Factor XII activity and antigen concentrations in patients suffering from recurrent thrombosis. Fibrinolysis 1987;1: 259–63. 8. La¨mmle B, Wuillemin WA, Huber I, Krauskopf M, Zurcher C, Pflugshaupt R, Furlan M. Thromboembolism and bleeding tendency in
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
congenital factor XII deficiency—a study on 74 subjects from 14 Swiss families. Thromb Haemost 1991;65:117–21. Rodeghiero F, Castaman G, Ruggieri M, Tosetto A. Thrombosis in subjects with homozygous and heterozygous factor XII deficiency. Thromb Haemost 1992;67:590. Halbmayer W-M, Mannhalter C, Feichtinger C, Rubi K, Fischer M. The prevalence of factor XII deficiency in 103 orally anticoagulated patients suffering from venous and/or arterial thromboembolism. Thromb Haemost 1992; 68:285–90. Winter M, Gallimore MJ, Jones DW. Should factor XII assays be included in thrombophilia screening? Lancet 1995;346:52. Halbmayer W-M, Huashofer A, Radek J, Schon R, Deutsch M, Fischer M. Prevalence of factor XII (Hageman factor) deficiency among 426 patients with coronary heart disease awaiting cardiac surgery. Coron Art Dis 1994;5: 451–4. Halbmayer W-M, Haushofer A, Schon R, Mannhalter C, Strohmer E, Baumgarten K, Fischer M. The prevalence of moderate and severe FXII (Hageman factor) deficiency among the normal population: Evaluation of the incidence of FXII deficiency among 300 healthy blood donors. Thromb Haemost 1994; 71:68–72. Fuhrer G, Gallimore MJ, Heller W, Hoffmeister H-E. Studies on components of the plasma kallikrein-kinin system in patients undergoing cardiopulmonary bypass. In: Greenbaum LW, Margolis H, editors. Kinins 1V: Advances in Experimental Medicine and Biology, Vol. 198B. New York: Plenum Press; 1986. p. 385–91. Colman RW, Scott CF, Schmaier AH, Wachtfogel YT, Pixley RA, Edmunds LH Jr. Initiation of blood coagulation at artificial surfaces. Ann NY Acad Sci 1987;516:253–67. Gallimore MJ, Fuhrer G, Heller W, Hoffmeister H-E. TPA levels during cardiopulmonary bypass. Thromb Haemost 1989;62:269. Ray MJ, Marsh NA, Hawson GAT. Relationship of fibrinolysis and platelet function to bleeding after cardiopulmonary bypass. Blood Coag Fibrinol 1994;5:679–85. Fuhrer G, Gallimore MJ, Heller W, Hoffmeister H-E. Aprotinin in cardiopulmonary bypass-
M.J. Gallimore et al./Thrombosis Research 94 (1999) 103–109
19.
20.
21.
22.
effects on Hageman factor (FXII)-kallikrein system and blood loss. Blood Coag Fibrinol 1992;3:99–104. Schousboe I, Rojkaer R, Lintner R. Functional determination of Factor XII in plasma. Clin Appl Thromb Haemost 1996;2:123–9. Jones DW, Gallimore MJ, Winter M. An automated chromogenic peptide substrate assay for coagulation factor XII. Blood Coag Fibrinol 1998;9:183–7. Gallimore MJ, Heller W, Fuhrer G, Wendel H, Klaffschenkel R, Hoffmeister H-E. Contact activation, heparins and cardiopulmonary bypass. Thromb Haemost 1993;69:92–3. Gallimore MJ, Friberger P. Simple chromogenic peptide substrate assays for determining prekallikrein, kallikrein inhibition and kalli-
23. 24.
25.
26.
109
krein “like” activity in human plasma. Thromb Res 1982;25:293–8. The British Society for Haematology. Guidelines on testing for the lupus anticoagulant. J Clin Pathol 1991;44:885–9. Schloesser M, Zeerlander S, Lutze G, Halbmayer W-M, Hoffbert S, Hinney B, Koestering H, La¨mmle B, Pindbur G, Thies K, Kohler M, Engel W. Mutations in the human factor XII gene. Blood 1997;90:3967–77. Von Ka¨nel R, Wuillemin WA, Furlan M, La¨mmle B. Factor XII clotting activity and antigen levels in patients with thromboembolic disease. Blood Coag Fibrinol 1992;3:555–61. Citarella F, Felicia A, Brouwer M, Wagstaff J, Fantoni A, Hack CE. Interleukin-6 downregulates factor XII production by human hepatoma cell line (HepG2). Blood 1997;90:1501–7.
REGULAR ARTICLE
A Chromogenic Substrate Assay Kit for Factor XII: Evaluation and Use for the Measurement of Factor XII Levels in Cardiopulmonary Bypass Patients Michael J. Gallimore1, D.W. Jones1 and Hans P. Wendel2 Kent Haemophilia Centre, Kent and Canterbury Hospital, Ethelbert Road, Canterbury, Kent, United Kingdom; 2Clinical Research Laboratory, Department of Throracic and Cardiovascular Surgery, University of Tuebingen, 72076 Tuebingen, Germany. 1
(Received 30 July 1998 by Editor C.A. Ludlam; revised/accepted 17 November 1998)
Abstract Factor XII levels were determined in plasma samples from 75 patients before undergoing aortocoronary bypass grafting and from 40 healthy age-matched donors by using a microtitre plate adaptation of a new chromogenic peptide substrate assay kit for factor XII. The chromogenic peptide substrate assay values for factor XII correlated well with those obtained in clotting (r50.90; y5 15.81110.8236x) and immunochemical (r50.88; y517.9010.817x) assays in the normal donor samples. Factor XII levels in the patients were significantly lower than those in the normal donors (83.3623.2% versus 103.4623.1; p50.004), and nine patients (12%) had factor XII values below 50% compared with only one of the normal donors (2.5%). Factor XII levels and kallikrein-like activities (a measure of contact system activation) were followed before, during, and one day after cardiopulmonary bypass in 20 patients. Factor XII levels were significantly reduced, and kallikrein-like activities significantly elevated after 5 and 30 minutes cardiopulmonary bypass. One day after cardiopulAbbreviations: FXII, factor XII; DDAVP, desamino D-arginine vasopressin; CPB, cardiopulmonary bypass; FXIIct, factor XII clotting activity; FXIIcs, factor XII determined by chromogenic substrate assay; FXIIag, factor XII antigen. Corresponding author: M.J. Gallimore, Kent Haemophilia Centre, Kent and Canterbury Hospital, Ethelbert Road, Canterbury, Kent, UK. Tel: 44 (0)1227 783 168; Fax: 44 (0) 1227 783 158.
monary bypass both factor XII levels and kallikrein-like activities were significantly lower than preoperation values. 1999 Elsevier Science Ltd. All rights reserved. Key Words: Factor XII; Chromogenic substrate assay; Cardiopulmonary bypass
S
ubsequent to its discovery as a contact coagulation defect in Mr. Hageman, factor XII (FXII) has been shown to be involved in the other plasma defense systems (kallikrein-kinin, complement, and fibrinolysis) and also affects the activities of polymorphonuclear granulocytes, platelets, and other cells [1–3]. Recent interest has focused on the role of FXII in fibrinolysis and thrombosis as it has been shown that FXII-deficient plasma has reduced fibrinolytic activity [4], that FXII-deficient patients have impaired fibrinolytic activity when exposed to desamino D-arginine vasopressin (DDAVP) [5], and that FXII deficiency might be a risk factor for thrombosis [6–11]. Halbmayer et al. [12,13] using clotting and immunochemical assays to measure FXII levels, reported a significantly higher prevalence of FXII deficiency in patients with coronary artery disease (10.3%) than among healthy blood donors (2.3%). We and others have reported that the FXIIrelated contact systems of blood are activated during cardiopulmonary bypass (CPB [14,15]) and that increased fibrinolytic activity occurs as a result
0049-3848/99 $–see front matter 1999 Elsevier Science Ltd. All rights reserved. PII S0049-3848(98)00202-3
104
M.J. Gallimore et al./Thrombosis Research 94 (1999) 103–109
of this activation [16,17], and we have suggested that inhibition of this activation by aprotinin therapy contributes to reduced blood loss in cardiac surgery [18]. Most studies on FXII have been performed with one stage clotting assays by using FXII deficient plasma or immunochemical assays. As these are both expensive and time consuming, both direct and indirect chromogenic peptide substrate assays for FXII have been described [1,19]. These assays, however, are rarely used routinely. A chromogenic substrate assay kit for routine determinations of FXII in clinical samples is now available, and we have evaluated this kit and used it to measure FXII levels in normal healthy subjects and in heart disease patients before, during, and after CPB. The results of our studies are reported in this article.
1. Materials and Methods Ethical committee approval for this study was obtained from the Ethical Committee of Tuebingen University Hospital, Calwerstrass 7, 72076 Tuebingen, Germany. Plasma samples from 75 male patients undergoing aortocoronary bypass grafting (mean age 61.667.55 years, minimum 38, maximum 70 years) and from 40 age-matched healthy male donors (mean age 61.767.43 years, minimum 40, maximum 70 years) were studied. Collection of blood, preparation of plasma, and sampling scheme for the CPB patients were performed as described previously [14,18]. A one-stage clotting assay on an automated coagulometer (Biomatic 2000; Sarstedt, Numbrecht, Germany) using FXII deficient plasma (Immuno Ltd., Vienna, Austria) was used to measure FXII clotting activity levels (FXIIct). FXII antigen (FXIIag) levels were determined by rocket immunoelectrophoresis by using FXII antiserum from Nordic Immunological Laboratories (Tilburg, The Netherlands). Levels of FXII determined by chromogenic substrate assay (FXIIcs) were performed using a microtitre plate technique using a Dynatech MR 5000 reader (Dynatech, Denkendorf, Germany) with FXII chromogenic substrate assay kits (Unitest FXIIe; Unicorn Diagnostics Ltd., London, UK). An evaluation of this kit by using an assay modified
for use on the Cobas Mira Plus analyzer has been reported recently [20]. The microtitre plate assay allows all laboratories with a microtitre plate reader to routinely measure FXII levels in plasma samples. The kit contains all of the reagents needed for the determination of FXII. These include: a standard plasma of known FXII content, an optimized buffer for diluting the plasma, a novel FXII activator (CONT-ACT FXIIt; Channel Diagnostics, Walmer, Kent, UK), an inhibitor cocktail to inhibit proteolytic enzymes other than a and bFXIIa and a chromogenic peptide substrate for a and bFXIIa (2AcOH. H-D-CHT-Gly-Arg-pNA). All reagents are supplied in freeze-dried form with the exception of the buffer. The freeze-dried kit components were reconstituted in sterile distilled water as follows: standard plasma 1 ml; CONT-ACT FXII activator 5 ml; inhibitor cocktail 10 ml; substrate (10 mmol) 10 ml. The buffer (10 ml) was diluted with 90 ml distilled water (buffer A). Inhibitor cocktail solution (2 ml) was added to 48 ml of buffer A (buffer B). The standard plasma and test plasmas have to be treated with organic solvent to destroy inhibitors of kallikrein in the activation stage of the assay and inhibitors of a and bFXIIa after activation. Both acetone or chloroform can be used for this purpose. We used acetone. One volume of acetone was added to 3 volumes of plasma in 1.5 ml polypropylene microcentrifuge tubes, the solutions mixed well, and the tubes were left for 30 minutes at 48C. The samples were then centrifuged (2000 3g for 15 minutes), and the supernatants were removed with plastic pipettes.
1.1. Preparation of the Standard Curve and Assay Method Full details of the preparation of the standard curve and the microtitre plate assay method are given in the kit insert. A brief description is as follows: the acetone treated standard plasma was diluted with buffer A to give values from 25 to 125%. Test plasmas were diluted the same as the 100% standard plasma dilution. Into the wells of a microtitre plate at 378C were pipetted 25 ml plasma dilution or buffer and 25 ml CONT-ACT FXII solution (at 378C). After mixing and incubation for exactly 10 minutes, 75 ml buffer B was added. After mixing and incubation for ex-
M.J. Gallimore et al./Thrombosis Research 94 (1999) 103–109
actly 3 minutes, 50 ml factor XII substrate (at 378C) was added. Finally, after mixing and incubation for exactly 10 minutes, 50 ml acetic acid (50%) was added to stop the reaction. The optical densities were then read at 405 nm in the microtitre plate reader. Blanks were performed substituting buffer A for the CONT-ACT FXII solution.
1.2. Calculation of FXII Level The A405 values for the blanks were subtracted from the values obtained for the standard curve and test samples. The delta A405 values for the standard curve samples were plotted against the percentage standard plasma, and the values for the test plasmas were obtained by plotting their delta A405 values on this standard curve. The values can be expressed either as a percentage or in units per milliliter (U/ml) by applying the formula: FXII (U/ml) % Activity3Potency of the Standard Plasma 5 100 The potency of the standard plasma is given with the kit. Test samples with values higher than 150% were diluted one part of diluted plasma plus one part buffer, reassayed, and the results obtained multiplied by 2. Kallikrein-like activities (thought to be a measure of plasma kallikrein-a2 macroglobulin complexes [21]) were measured using kits supplied by Unicorn Diagnostics Ltd. The method in this kit is essentially as described by Gallimore and Friberger [22] using a more specific substrate for plasma kallikrein (MBz-Pro-Phe-Arg-pNA). Lupus anticoagulant assays were performed on double centrifuged plasma samples by a dilute Russell’s viper venom time technique and platelet correction procedure as recommended by the British Society of Haematology [23], using kits supplied by Unicorn Diagnostics Ltd. All assays were performed in duplicate.
1.3. Statistical Analysis Correlation values and regression analysis were performed using Statistica Mac, Statsoft Ltd., Uckfield, UK.
105
Statistical significance was set at a p value of ,0.05 and calculated using t-test, while correlation between assays was conducted using Pearson correlation.
2. Results The imprecision of the FXIIcs microtitre plate assay was determined by analyzing plasma samples from donors with FXII levels of 114 and 38%. The analyses were performed on five different occasions including five replicates of each plasma. The interassay precision values were 6.15 and 3.8%, and the intraassay precision values were 4.9 and 2.6%, respectively. The assay was able to detect levels of FXII as low as 3% and with suitable dilution levels higher than 200%. Correlation plots for FXIIcs versus FXIIct and FXIIcs versus FXIIag in plasma samples from the 40 normal healthy blood donors are shown in Figures 1a and 1b. Satisfactory correlations were obtained for FXIIcs versus the other two assays (FXIIcs vs. FXIIct50.90, FXIIcs vs. FXIIag50.88). The regression line equation for FXIIcs vs. FXIIct was y5 15.81110.824x and for FXIIcs vs. FXIIag y5 17.9010.817x. Mean values, standard deviations, and ranges for the three assays were FXIIcs5103.4623.1%, 50–182%; FXIIct5106.7624.6%, 53–206%; FXIIag5105.3624.4%, 50–191%. When FXII levels in plasma samples from 75 patients awaiting cardiac surgery were determined using the chromogenic substrate assay, significantly lower values were obtained when compared with the age-matched control group (p50.004, Table 1). All plasma samples from the normal donor and patient groups were lupus anticoagulant negative. In the heart disease patient group, 9 of the 75 patients had low levels of FXII (range 39–50%) as determined in the FXIIcs assay (incidence of 12%) compared with only 1 of the 40 normal healthy donors (incidence of 2.5%). All of the heart disease patients with low FXII levels had normal values for liver function tests (results not shown). Twenty heart disease patients were studied before, during, and after cardiopulmonary bypass surgery. FXII levels (measured by the FXIIcs assay and corrected for haemodilution) fell during CPB
106
M.J. Gallimore et al./Thrombosis Research 94 (1999) 103–109
Fig. 1. (a) Correlation data between factor XII (FXII) levels in 40 healthy blood donors by using the chromogenic substrate and clotting assays. The correlation coefficient (r) for 40 data pairs was 0.90, and the regression line equation was y515.811110.8236x. (b) Correlation data between factor XII (FXII) levels in 40 healthy blood donors by using the chromogenic substrate and antigen assays. The correlation coefficient (r) for 40 data pairs was 0.88, and the regression line equation was y517.9000210.8171x.
and remained reduced throughout and immediately after the operation. One day after operation, FXII levels were significantly lower than preoperation values (Figure 2). In view of the high heparin levels used during CPB, heparin was added to the standard plasma to give a range of heparin levels from 0 to 5.0 U/ml and the samples tested in the FXIIcs assay. Heparin had no effect on the FXIIcs assay, even at plasma levels of 5.0 U/ml (results not shown). Kallikrein-like activities were significantly elevated after 5 and 30 minutes of CPB but had returned to normal immediately after operation. One day after surgery kallikrein-like activities were significantly lower than preoperation values (Figure 3).
samples; however, both of these assays are cumbersome and rather expensive, and the former requires FXII-deficient plasma as an essential component. The chromogenic peptide substrate assay kit for FXII used in our studies allows us to routinely measure FXII without the need for FXII-deficient plasma. The assay is very easy to perform, gives good inter and intra-assay variability, can easily be automated and, unlike the clotting assay, is not affected by phospholipid antibodies [20], and very relevant to the present study, is not affected by heparin. The chromogenic peptide substrate used has been selected because it has good sensitivity to a and bFXIIa and is rather insensitive to plasma kallikrein and plasmin. Specificity for a
3. Discussion There is a growing interest in the FXII molecule as after its activation it can influence both fluid and cellular defense systems. There is also interest in the role of this protein as a possible regulator of fibrinolysis and current debate as to whether FXII deficiency is a risk factor for thrombosis [1,20,24]. Historically, clotting and immunochemical assays have been used to measure FXII levels in plasma
Table 1. Levels of FXII in plasma samples from 40 healthy donors and 75 patients with heart disease determined using the FXIIcs assay Normals (n540) Mean 6SD Range Significance
FXIIcs% 103.4 23.1 50–182
Heart disease (n575)
FXIIcs% 83.3 23.2 39–140 0.004
M.J. Gallimore et al./Thrombosis Research 94 (1999) 103–109
107
Fig. 2. Levels of FXII in plasma samples from 20 patients before, during, and after cardiopulmonary bypass determined using the FXIIcs assay. Results shown as mean6SD. Asterisks indicate significantly different from preoperation values.
Fig. 3. Kallikrein-like activities in plasma samples from 20 patients before, during, and after cardiopulmonary bypass. Results shown as mean6SD. Asterisks indicate significantly different from preoperation values.
and bFXIIa is further ensured by the use of an inhibitor cocktail. When FXII levels in plasma samples from 40 healthy blood donors were determined using the FXIIcs microtitre plate assay, satisfactory correlations with the FXIIct and FXIIag assays were obtained (Figures 1a and 1b). The mean values for the three assays were all slightly above 100%, which might reflect an effect of age on FXII levels (our donors were all older than 40), although our results were very similar to those obtained by others in clotting and immunochemical assays for FXII in large populations of normal healthy donors [8, 13,20,25]. The levels of FXII in 75 patients awaiting cardiopulmonary bypass surgery (83.3%623.2%) were significantly lower (p50.004) than in the agematched control group (Table 1). Also, the incidence of FXII deficiency (FXIIcs value ,50%) in this patient group (9 out of 75 patients incidence 512%) was higher than that for the control group (1 out of 40 incidence 52.5%). These results are very similar to those obtained by Halbmayer et al. [12] who measured FXIIct and FXIIag levels in 426 patients with coronary heart disease awaiting surgery and found an incidence of FXII deficiency of 10.3%. The contact systems of blood are activated dur-
ing cardiopulmonary bypass and because of the high heparin levels used coagulation assays for FXII are unreliable. Because of the possible role of FXII in fibrinolysis and blood loss in CPB [18], we feel it is of interest to determine the functional activity of this contact system protein during CPB. We therefore used the FXIIcs assay to measure FXII levels in 20 patients before, during, and after CPB. The FXII levels fell during CPB and were already significantly lower (p50.001) after 5 minutes CPB (Figure 2). Kallikrein-like activities, a measure of contact system activation [21], were elevated during CPB and were already significantly higher (p50.027) than preoperation values after 5 minutes CPB (Figure 3). Kallikrein-like activities and FXII values were significantly lower than the preoperation values on the day after surgery (p50.008 and 0.004, respectively). Of the 20 patients studied, four (20%) had levels of FXII below 50% 1 day after surgery. Despite the accepted classic role of FXII in intrinsic blood coagulation, none of these patients exhibited enhanced blood loss during or after CPB (results not shown). Our results confirm that FXII activation occurs during CPB and show that plasma levels of this protein that fall during CPB do not recover to preoperation values on the day after surgery but in
108
M.J. Gallimore et al./Thrombosis Research 94 (1999) 103–109
fact fall further, reflecting either continued activation of FXII or the negative acute phase synthesis of FXII recently described [26]. To our knowledge, this is a first report of this finding after CPB and in view of the possibility that reduced FXII levels might constitute a risk for thrombosis further study in this area is obviously warranted. Although FXII has been studied extensively in research laboratories, the lack of simple, reasonably priced, accurate manual and automated assays for this protein has limited the scope of clinical studies. The introduction of chromogenic peptide substrate technology for the determination of this important proenzyme will hopefully rectify this situation in the future.
References 1. Fuhrer G, Gallimore MJ, Heller W, Hoffmeister H-E. FXII. Blut 1990;61:258–66. 2. Saito H. The contact phase of blood coagulation. In: Bloom AL, Forbes CD, Thomas DP, Tuddenham EGD, editors. Haemostasis and Thrombosis, Vol. 1, 3rd ed. Edinburgh: Churchill Livingstone; 1994. p. 289–307. 3. Colman RW, Schmaier AH. Contact System: A vascular biology modulator with anticoagulant, profibrinolytic, antiadhesive and proinflammatory attributes. Blood 1997; 90:3819–43. 4. Kluft C, Dooijewaard GG, Emeis JJ. Role of the contact system in fibrinolysis. Sem Thromb Haemost 1987;13:50–68. 5. Levi M, Hack CE, de Boer JP, Brandjes DPM, Bu¨ller HR, ten Cate JW. Reduction of contact activation related fibrinolytic activity in factor XII deficient patients. J Clin Invest 1991; 88:1155–60. 6. Goodnough LT, Saito H, Ratnoff OD. Thrombosis or myocardial infarction in congenital clotting factors abnormalities and chronic thrombocytopenias: A report of 21 patients and a review of 50 previously reported cases. Medicine 1983;62:248–255. 7. Mannhalter C, Fischer M, Hopmeier P, Deutsch E. Factor XII activity and antigen concentrations in patients suffering from recurrent thrombosis. Fibrinolysis 1987;1: 259–63. 8. La¨mmle B, Wuillemin WA, Huber I, Krauskopf M, Zurcher C, Pflugshaupt R, Furlan M. Thromboembolism and bleeding tendency in
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
congenital factor XII deficiency—a study on 74 subjects from 14 Swiss families. Thromb Haemost 1991;65:117–21. Rodeghiero F, Castaman G, Ruggieri M, Tosetto A. Thrombosis in subjects with homozygous and heterozygous factor XII deficiency. Thromb Haemost 1992;67:590. Halbmayer W-M, Mannhalter C, Feichtinger C, Rubi K, Fischer M. The prevalence of factor XII deficiency in 103 orally anticoagulated patients suffering from venous and/or arterial thromboembolism. Thromb Haemost 1992; 68:285–90. Winter M, Gallimore MJ, Jones DW. Should factor XII assays be included in thrombophilia screening? Lancet 1995;346:52. Halbmayer W-M, Huashofer A, Radek J, Schon R, Deutsch M, Fischer M. Prevalence of factor XII (Hageman factor) deficiency among 426 patients with coronary heart disease awaiting cardiac surgery. Coron Art Dis 1994;5: 451–4. Halbmayer W-M, Haushofer A, Schon R, Mannhalter C, Strohmer E, Baumgarten K, Fischer M. The prevalence of moderate and severe FXII (Hageman factor) deficiency among the normal population: Evaluation of the incidence of FXII deficiency among 300 healthy blood donors. Thromb Haemost 1994; 71:68–72. Fuhrer G, Gallimore MJ, Heller W, Hoffmeister H-E. Studies on components of the plasma kallikrein-kinin system in patients undergoing cardiopulmonary bypass. In: Greenbaum LW, Margolis H, editors. Kinins 1V: Advances in Experimental Medicine and Biology, Vol. 198B. New York: Plenum Press; 1986. p. 385–91. Colman RW, Scott CF, Schmaier AH, Wachtfogel YT, Pixley RA, Edmunds LH Jr. Initiation of blood coagulation at artificial surfaces. Ann NY Acad Sci 1987;516:253–67. Gallimore MJ, Fuhrer G, Heller W, Hoffmeister H-E. TPA levels during cardiopulmonary bypass. Thromb Haemost 1989;62:269. Ray MJ, Marsh NA, Hawson GAT. Relationship of fibrinolysis and platelet function to bleeding after cardiopulmonary bypass. Blood Coag Fibrinol 1994;5:679–85. Fuhrer G, Gallimore MJ, Heller W, Hoffmeister H-E. Aprotinin in cardiopulmonary bypass-
M.J. Gallimore et al./Thrombosis Research 94 (1999) 103–109
19.
20.
21.
22.
effects on Hageman factor (FXII)-kallikrein system and blood loss. Blood Coag Fibrinol 1992;3:99–104. Schousboe I, Rojkaer R, Lintner R. Functional determination of Factor XII in plasma. Clin Appl Thromb Haemost 1996;2:123–9. Jones DW, Gallimore MJ, Winter M. An automated chromogenic peptide substrate assay for coagulation factor XII. Blood Coag Fibrinol 1998;9:183–7. Gallimore MJ, Heller W, Fuhrer G, Wendel H, Klaffschenkel R, Hoffmeister H-E. Contact activation, heparins and cardiopulmonary bypass. Thromb Haemost 1993;69:92–3. Gallimore MJ, Friberger P. Simple chromogenic peptide substrate assays for determining prekallikrein, kallikrein inhibition and kalli-
23. 24.
25.
26.
109
krein “like” activity in human plasma. Thromb Res 1982;25:293–8. The British Society for Haematology. Guidelines on testing for the lupus anticoagulant. J Clin Pathol 1991;44:885–9. Schloesser M, Zeerlander S, Lutze G, Halbmayer W-M, Hoffbert S, Hinney B, Koestering H, La¨mmle B, Pindbur G, Thies K, Kohler M, Engel W. Mutations in the human factor XII gene. Blood 1997;90:3967–77. Von Ka¨nel R, Wuillemin WA, Furlan M, La¨mmle B. Factor XII clotting activity and antigen levels in patients with thromboembolic disease. Blood Coag Fibrinol 1992;3:555–61. Citarella F, Felicia A, Brouwer M, Wagstaff J, Fantoni A, Hack CE. Interleukin-6 downregulates factor XII production by human hepatoma cell line (HepG2). Blood 1997;90:1501–7.