THROMBOSIS RESEARCH 65; 253-262,1992 0049-3646J92 $5.00 + .OO Printed in the USA. Copyright (c) 1992 Pergamon Press pk. All rights reserved.
FUNCTIONAL PROTEIN S IN WOMEN WITH LUPUS ANTICOAGULANT INHIBITOR
E. Rossi, L. Gatti, D. Guarneri, E. Finotto*, A. Lombardi*, L. Preda* Centro Trasfusionale e di Immunoematologia, I.C.P. and * R.& D., Instrumentation Laboratory Srl, MilanoJtaly (Received 19.3.1991; accepted in revised form 16.10.1991 by Editor M.B. Donati) (Received in final form by Executive Editorial Office 26.11.1991)
ABSTRACT Protein C (PC) and protein S (PS) are components of a potent, natural anticoagulant system. A deficiency of one of these two inhibitors is associated with thrombotic events in young people. A significant reduction in functional PS activity has been observed during normal pregnancy, and recurrent fetal loss may occur in women with lupus anticoagulant (LA) inhibitor. We measured functional PS activity and free PS antigen in 16 non pregnant patients with LA inhibitor and in 17 normal women as controls. A significant difference was observed between patients and controls in functional PS activity (65&23% vs 87+15%, p=O.O2) but not in free PS antigen (H&17% vs 93&17%). Functional PS activity decreased only in six patients (37%). Removal of IgG from plasma reduced the difference in functional PS activity between patients and controls. Immunologic IgG levels did not correlate with antiphospholipid antibodies (APA) activities, activated partial thromboplastin time/kaolin clotting time (aP’IT/KCT) data or functional PS activity.
INTRODUCTION Protein C (PC) and protein S (PS) are potent, natural anticoagulant plasma proteins (l-3). The former is cleaved by thrombin and converted to a serine protease, activated protein C (aPC), Key words: protein S, protein C, lupus anticoagulant, fetal loss, thrombosis
253
254
PROTEIN S IN WOMEN WITH LA INHIBITOR
Vol. 65, No. 2
that inactivates factors Va (4) and VIIIa (5,6); the activation rate is physiologically enhanced by the presence of thrombomodulin, an endothelial glycoprotein (7,s). The inactivation rate of factors Va and VIIIa is increased by PS. Its function is to bind aPC to the surface of phospholipids, which is where the proteolytic inactivation of the two factors occurs. The risk of tbromboembolic events is increased in patients with a hereditary deficiency of one of the two proteins (9-11). Functional and immunologic levels of PS are also reduced during normal pregnancy (12). Lupus anticoagulant (LA) is an auto-antibody of the IgG or IgM class of immunoglobulins; its activity is directed against phospholipids. In vitro, LA interacts with cephalin and thromboplastin, quenching the pro-coagulant activity of the test system and prolonging the clotting time. The presence of LA is frequently associated with a predisposition to thrombotic events and fetal loss (13). The pathogenesis of thrombosis in patients with LA is still unclear, although several hypotheses have been proposed to explain this association (14-23). In this study we investigated 16 patients with LA inhibitor and a history of venous or arterial thrombosis and/or recurrent fetal loss. Functional activity of PS was measured in these patients and in a control group of normal women in order to seek a correlation between thromboembolic events, fetal loss and decreased PS activity.
MATERIALS. METHODS and PATIENTS Materials, Blood samples were collected in one-tenth volume of 129 n-M (38 g/L) trisodium citrate. Platelet poor plasma was obtained by double centrifugation at 2,100 g for 15 minutes at room temperature; plasma supernatant was then quickly frozen at -40 “C and stored at the same temperature. Blood samples from 20 healthy donors (10 men and 10 women) were collected and the plasma mixed to obtain a normal plasma pool (NPP). This plasma was also quickly frozen, stored at -40 “C and used as PS standard (100%) to construct the calibration curve. Protein ASepharose CL 4B, used to adsorb IgG of patients’ and controls’ plasma, as well as CNBractivated Sepharose 4B and prepacked columns of Sephadex G-25M (PD-10) were supplied by Pharmacia (Cologno M., Milano, Italy). NOR-Partigen IgG immunoplates (Istituto Behring, Scoppito, l’Aquila, Italy) were used to measure the quantity of immunoglobulins removed from plasma. Asseraplate Protein S kit (Stago, Asnieres, France) was used to measure the level of total and free PS antigen, by electroimmuno assay method. The free form was measured on the supernatant of the plasma treated with PEG 8000, at the final concentration of 3.6% (24). The mixture (300 l.tL plasma + 50 pL PEG) remained initially at 37°C for 5 minutes, and then for 30 minutes at 4°C; to separate the precipitate, the mixture was centrifugated at 3500 rpm for 10 minutes. Asserachrome APA (Stago, Asnieres, France) was employed as enzyme immunoassay of anti-phospholipid antibodies (APA). APA levels < 5 PL units were considered normal, 5-15 PL units borderline, 15-30 PL units low-positive, 30-100 PL units moderate-positive and >lOO PL units high-positive, as suggested by the manufacturer of the kit. PC activator (Protac) was purchased from Pentapharm (Basel, Switzerland). Reagents for determination of aPIT, KCT, bovine thromboplastin and PS deficient plasma were supplied by Instrumentation Laboratory (Milano, Italy). PS deficient plasma was prepared by depletion of a fresh normal plasma on a column of anti-PS mouse IgG (25) coupled to CNBr-activated Sepharose (8 mg/mL). After depletion, the plasma was lyophilized in 1-mL vials. All salts used to prepare buffer solutions were RPE grade and supplied by Carlo Erba (Milano, Italy). ACL 300/R coagulometer (ACL) (26,27) was used for all coagulation tests.
Vol. 65, No. 2
PROTEIN S IN WOMEN WITH LA INHIBITOR
255
Functional PS assay (28): PS cofactor activity was measured as degree of prolongation of prothrombin time (PT) of a PS deficient plasma after addition of diluted test plasma. The assay was performed as follows: two volumes of PS deficient plasma were preincubated for 20 minutes at 37 “C with one volume of Protac (1 U/mL) to fully activate PC. Samples and standards were diluted 1:lO with saline solution (NaCl, 150 mM). Using an extrinsic single factor cycle of ACL, we performed an automatic 3-point calibration curve (loo50-25 % PS activity) employing NPP as PS 100% standard and PS deficient plasma as 0% standard. By Research Program (27) the data were put on a personal computer, and the clotting time was calculated. The PS activity values were calculated using the function Y=q + m*X2 , where Y is the PS(%) activity and X the clotting time (seconds). Kaolin clotting time assay (29): KCT was performed automatically on ACL with Research Program (27), using an Exner assay modified as follows: 60 pL, sample volume; 40 pL, kaolin suspension (2 mg/mL); 60 pL, CaCl, 25 mM; 240 set, activation time; 360 see, acquisition time; 600 rpm, rotor speed. ZgG adsorption: 2 mL of protein A-Sepharose gel was put into a column ($ = 15 mm) and equilibrated with phosphate buffer (0.1 M, pH=7.0). After addition of 2.5 mL of plasma, the two column ends were closed and the plasma was mixed with gel suspension for 30 minutes in an end-over-end or similar mixer. Plasma was eluted completely with 1.5 mL of saline solution and the gel washed with about 10 volumes of phosphate buffer to remove unbound contaminants. IgG were eluted from protein A-Sepharose by adding 4.0 mL of glycine-HCl buffer (0.1 M, pH=2.6) and recovering the first 3.0 mL eluted. Finally, IgG solution was desalted on PD-10 columns equilibrated with saline solution.
Methods.
Table 1 Patient Characteristics Patient Age Fetal loss 1 35 yes 2 26 yes 3 30 yes 4 35 yes 5 33 yes 6 40 no 7 32 yes 8 28 no 9 36 yes 10 27 yes 11 34 yes 12 27 yes 13 50 no 14 26 yes 15 32 yes 16 30 yes + : single event or mild disease ++ : several events or severe disease
Thrombosis ++ (A)
Thrombocytopenia + +
+ + ++ ++ + + + +
(W (A) (A) (A) (V) (V) (V) (V)
+
+ (A) + (V) ++ A/v : arterial/venous
256
PROTEIN S IN WOMEN WITH LA INHIBITOR
Vol. 65, No. 2
Inhibition test: We tested 450 pL of IgG saline solution (patients and controls) to which were added 50 pL of NPP as source of coagulation factors and PS. Subsequently, the residual factor (VIII, V, X) and PS activities were measured on the ACL using a single factor cycle and functional PS assay respectively. m. We obtained plasma from 17 normal women (mean age 31 years, range 19-40) and from 16 patients with LA inhibitor (mean age 32 years, range 25-50) with a history of recurrent fetal loss and/or thrombosis. The patients’ histories are summarized in Table 1. stical analvsig. All values are expressed as mean f S.E.M. Significant differences between and within the groups were evaluated by a non parametric test (unpaired and paired Wilcoxon test respectively).
RESULTS Confnnation of presence of LA inhibitor. Figure 1 shows the aPTT (columns A and C) and KCI (column B) ratios. In the patients, the aP’IT and KCT ratios were 1.67k0.55 and 2.66kO.62 respectively, versus 0.98+0.08 and 1.20f0.26 in the controls. Unpaired Wilcoxon analysis revealed a highly significant difference between the two groups (p
FUNCTIONAL PROTEIN S IN WOMEN WITH LUPUS ANTICOAGULANT INHIBITOR
E. Rossi, L. Gatti, D. Guarneri, E. Finotto*, A. Lombardi*, L. Preda* Centro Trasfusionale e di Immunoematologia, I.C.P. and * R.& D., Instrumentation Laboratory Srl, MilanoJtaly (Received 19.3.1991; accepted in revised form 16.10.1991 by Editor M.B. Donati) (Received in final form by Executive Editorial Office 26.11.1991)
ABSTRACT Protein C (PC) and protein S (PS) are components of a potent, natural anticoagulant system. A deficiency of one of these two inhibitors is associated with thrombotic events in young people. A significant reduction in functional PS activity has been observed during normal pregnancy, and recurrent fetal loss may occur in women with lupus anticoagulant (LA) inhibitor. We measured functional PS activity and free PS antigen in 16 non pregnant patients with LA inhibitor and in 17 normal women as controls. A significant difference was observed between patients and controls in functional PS activity (65&23% vs 87+15%, p=O.O2) but not in free PS antigen (H&17% vs 93&17%). Functional PS activity decreased only in six patients (37%). Removal of IgG from plasma reduced the difference in functional PS activity between patients and controls. Immunologic IgG levels did not correlate with antiphospholipid antibodies (APA) activities, activated partial thromboplastin time/kaolin clotting time (aP’IT/KCT) data or functional PS activity.
INTRODUCTION Protein C (PC) and protein S (PS) are potent, natural anticoagulant plasma proteins (l-3). The former is cleaved by thrombin and converted to a serine protease, activated protein C (aPC), Key words: protein S, protein C, lupus anticoagulant, fetal loss, thrombosis
253
254
PROTEIN S IN WOMEN WITH LA INHIBITOR
Vol. 65, No. 2
that inactivates factors Va (4) and VIIIa (5,6); the activation rate is physiologically enhanced by the presence of thrombomodulin, an endothelial glycoprotein (7,s). The inactivation rate of factors Va and VIIIa is increased by PS. Its function is to bind aPC to the surface of phospholipids, which is where the proteolytic inactivation of the two factors occurs. The risk of tbromboembolic events is increased in patients with a hereditary deficiency of one of the two proteins (9-11). Functional and immunologic levels of PS are also reduced during normal pregnancy (12). Lupus anticoagulant (LA) is an auto-antibody of the IgG or IgM class of immunoglobulins; its activity is directed against phospholipids. In vitro, LA interacts with cephalin and thromboplastin, quenching the pro-coagulant activity of the test system and prolonging the clotting time. The presence of LA is frequently associated with a predisposition to thrombotic events and fetal loss (13). The pathogenesis of thrombosis in patients with LA is still unclear, although several hypotheses have been proposed to explain this association (14-23). In this study we investigated 16 patients with LA inhibitor and a history of venous or arterial thrombosis and/or recurrent fetal loss. Functional activity of PS was measured in these patients and in a control group of normal women in order to seek a correlation between thromboembolic events, fetal loss and decreased PS activity.
MATERIALS. METHODS and PATIENTS Materials, Blood samples were collected in one-tenth volume of 129 n-M (38 g/L) trisodium citrate. Platelet poor plasma was obtained by double centrifugation at 2,100 g for 15 minutes at room temperature; plasma supernatant was then quickly frozen at -40 “C and stored at the same temperature. Blood samples from 20 healthy donors (10 men and 10 women) were collected and the plasma mixed to obtain a normal plasma pool (NPP). This plasma was also quickly frozen, stored at -40 “C and used as PS standard (100%) to construct the calibration curve. Protein ASepharose CL 4B, used to adsorb IgG of patients’ and controls’ plasma, as well as CNBractivated Sepharose 4B and prepacked columns of Sephadex G-25M (PD-10) were supplied by Pharmacia (Cologno M., Milano, Italy). NOR-Partigen IgG immunoplates (Istituto Behring, Scoppito, l’Aquila, Italy) were used to measure the quantity of immunoglobulins removed from plasma. Asseraplate Protein S kit (Stago, Asnieres, France) was used to measure the level of total and free PS antigen, by electroimmuno assay method. The free form was measured on the supernatant of the plasma treated with PEG 8000, at the final concentration of 3.6% (24). The mixture (300 l.tL plasma + 50 pL PEG) remained initially at 37°C for 5 minutes, and then for 30 minutes at 4°C; to separate the precipitate, the mixture was centrifugated at 3500 rpm for 10 minutes. Asserachrome APA (Stago, Asnieres, France) was employed as enzyme immunoassay of anti-phospholipid antibodies (APA). APA levels < 5 PL units were considered normal, 5-15 PL units borderline, 15-30 PL units low-positive, 30-100 PL units moderate-positive and >lOO PL units high-positive, as suggested by the manufacturer of the kit. PC activator (Protac) was purchased from Pentapharm (Basel, Switzerland). Reagents for determination of aPIT, KCT, bovine thromboplastin and PS deficient plasma were supplied by Instrumentation Laboratory (Milano, Italy). PS deficient plasma was prepared by depletion of a fresh normal plasma on a column of anti-PS mouse IgG (25) coupled to CNBr-activated Sepharose (8 mg/mL). After depletion, the plasma was lyophilized in 1-mL vials. All salts used to prepare buffer solutions were RPE grade and supplied by Carlo Erba (Milano, Italy). ACL 300/R coagulometer (ACL) (26,27) was used for all coagulation tests.
Vol. 65, No. 2
PROTEIN S IN WOMEN WITH LA INHIBITOR
255
Functional PS assay (28): PS cofactor activity was measured as degree of prolongation of prothrombin time (PT) of a PS deficient plasma after addition of diluted test plasma. The assay was performed as follows: two volumes of PS deficient plasma were preincubated for 20 minutes at 37 “C with one volume of Protac (1 U/mL) to fully activate PC. Samples and standards were diluted 1:lO with saline solution (NaCl, 150 mM). Using an extrinsic single factor cycle of ACL, we performed an automatic 3-point calibration curve (loo50-25 % PS activity) employing NPP as PS 100% standard and PS deficient plasma as 0% standard. By Research Program (27) the data were put on a personal computer, and the clotting time was calculated. The PS activity values were calculated using the function Y=q + m*X2 , where Y is the PS(%) activity and X the clotting time (seconds). Kaolin clotting time assay (29): KCT was performed automatically on ACL with Research Program (27), using an Exner assay modified as follows: 60 pL, sample volume; 40 pL, kaolin suspension (2 mg/mL); 60 pL, CaCl, 25 mM; 240 set, activation time; 360 see, acquisition time; 600 rpm, rotor speed. ZgG adsorption: 2 mL of protein A-Sepharose gel was put into a column ($ = 15 mm) and equilibrated with phosphate buffer (0.1 M, pH=7.0). After addition of 2.5 mL of plasma, the two column ends were closed and the plasma was mixed with gel suspension for 30 minutes in an end-over-end or similar mixer. Plasma was eluted completely with 1.5 mL of saline solution and the gel washed with about 10 volumes of phosphate buffer to remove unbound contaminants. IgG were eluted from protein A-Sepharose by adding 4.0 mL of glycine-HCl buffer (0.1 M, pH=2.6) and recovering the first 3.0 mL eluted. Finally, IgG solution was desalted on PD-10 columns equilibrated with saline solution.
Methods.
Table 1 Patient Characteristics Patient Age Fetal loss 1 35 yes 2 26 yes 3 30 yes 4 35 yes 5 33 yes 6 40 no 7 32 yes 8 28 no 9 36 yes 10 27 yes 11 34 yes 12 27 yes 13 50 no 14 26 yes 15 32 yes 16 30 yes + : single event or mild disease ++ : several events or severe disease
Thrombosis ++ (A)
Thrombocytopenia + +
+ + ++ ++ + + + +
(W (A) (A) (A) (V) (V) (V) (V)
+
+ (A) + (V) ++ A/v : arterial/venous
256
PROTEIN S IN WOMEN WITH LA INHIBITOR
Vol. 65, No. 2
Inhibition test: We tested 450 pL of IgG saline solution (patients and controls) to which were added 50 pL of NPP as source of coagulation factors and PS. Subsequently, the residual factor (VIII, V, X) and PS activities were measured on the ACL using a single factor cycle and functional PS assay respectively. m. We obtained plasma from 17 normal women (mean age 31 years, range 19-40) and from 16 patients with LA inhibitor (mean age 32 years, range 25-50) with a history of recurrent fetal loss and/or thrombosis. The patients’ histories are summarized in Table 1. stical analvsig. All values are expressed as mean f S.E.M. Significant differences between and within the groups were evaluated by a non parametric test (unpaired and paired Wilcoxon test respectively).
RESULTS Confnnation of presence of LA inhibitor. Figure 1 shows the aPTT (columns A and C) and KCI (column B) ratios. In the patients, the aP’IT and KCT ratios were 1.67k0.55 and 2.66kO.62 respectively, versus 0.98+0.08 and 1.20f0.26 in the controls. Unpaired Wilcoxon analysis revealed a highly significant difference between the two groups (p
