SEARCH

SEARCH BY CITATION

Keywords:

  • β2-glycoprotein I;
  • antiphospholipid syndrome;
  • lupus anticoagulants;
  • prothrombin

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Summary.  Lupus anticoagulants (LAC) are a heterogeneous group of autoantibodies that prolong phospholipid-dependent clotting assays. The autoantibodies that cause LAC activity are predominantly directed against β2-glycoprotein I (β2GPI) or prothrombin. In the present study, we describe a method to differentiate between LAC caused by antibodies directed against β2GPI or prothrombin. Monoclonal antibodies, affinity purified patient antibodies, and selected patient samples were used to show that in an aPTT-based clotting assay (PTT-LA; Diagnostica Stago), the use of cardiolipin vesicles in the neutralization procedure discriminates between β2GPI- or prothrombin-dependent LAC activities. Addition of cardiolipin vesicles shortened the prolonged clotting time caused by anti-β2GPI antibodies with LAC activity, whereas this procedure further prolonged clotting times caused by antiprothrombin antibodies with LAC activity. In contrast, addition of phosphatidylcholine/phosphatidylserine vesicles corrected prolonged clotting times caused by either anti-β2GPI or antiprothrombin antibodies with LAC activity. The effects of cardiolipin (CL) on β2GPI-induced LAC activity were specific for contact activation mediated clotting assays. Possible explanations for these findings are the relatively high affinity of β2GPI for cardiolipin, as determined by surface plasmon resonance analysis, and inhibition by anti-β2GPI antibodies of the CL-induced prolongation of the PTT-LA.

Antiphospholipid antibodies (aPL) are traditionally classified as lupus anticoagulants (LAC) and anticardiolipin antibodies (aCL) according to their methods of detection with in vitro clotting tests or ELISA assays [1], respectively. LAC refers to immunoglobulins that prolong in vitro clotting assays [2–4], whereas aCL bind to immobilized cardiolipin (CL) [5,6], a negatively charged phospholipid. In contrast to infection-related aPL, autoimmune aPL are not directed to phospholipids alone, but to lipid-binding (plasma) proteins, notably β2-glycoprotein I (β2GPI) or prothrombin [7–11]. Although β2GPI itself has affinity for phospholipids, its affinity is strongly enhanced in the presence of anti-β2GPI antibodies, due to formation of bivalent complexes [12–15]. It is generally accepted that bivalent complexes compete more strongly with clotting factors for the catalytic phospholipid surfaces than monovalent β2GPI. We have recently described a similar mechanism of enhanced binding of divalent complexes for prothrombin-dependent LAC [16].

It has been shown that the presence of LAC correlates better with a history of thrombotic complications than in the presence of aCL [17,18]. It has been suggested that anti-β2GPI antibodies are more related to thrombotic complications than antiprothrombin-antibodies [19–22]. These observations were based on detection of these antibodies with an ELISA setup. The clinical relevance of anti-β2GPI and antiprothrombin antibodies with a functional activity (viz. induction of LAC activity) is unknown. A simple method to discriminate between LAC activity caused by anti-β2GPI and antiprothrombin antibodies therefore seems relevant for future clinical studies.

Guidelines for the detection of LAC have been published [23] and involve confirmation of the phospholipid-dependent nature of the inhibitory antibody. This confirmation step normally consists of addition of a high concentration of negatively charged phospholipids to abolish the inhibitory effects of the antibodies. The nature of the negatively charged phospholipids has never been defined and it is thought that all negatively charged phospholipids neutralize all LAC activity, irrespective of type of antibodies involved. Here we show that this basic assumption is incorrect as a neutralization procedure with cardiolipin vesicles in an aPTT-based assay only neutralizes LAC activity due to anti-β2GPI antibodies. This observation opens the possibility to discriminate between β2GPI- and prothrombin-dependent LAC activity.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Materials

PTT-LA test was obtained from Diagnostica Stago (Paris, France) and the dRVVT test was obtained from Dade-Behring (Marburg, Germany). Cardiolipin, phosphatidylserine and phosphatidylcholine were obtained from Sigma (St. Louis, MO, USA). Sephadex G-25 and different Sepharoses were obtained from Pharmacia Biotech (Uppsala, Sweden). L1 chip was purchased from Biacore (Uppsala, Sweden).

Patients

We used plasmas from four patients with a combined presence of LAC and anti-β2GPI antibodies, five patients with a combined presence of LAC and antiprothrombin antibodies and three patients with the presence of LAC and the combination of antiprothrombin and anti-β2GPI antibodies (Table 1). The presence of anti-β2GPI and antiprothrombin antibodies was determined as described by Horbach et al. [17]. In the patient samples, LAC activity was determined by a dilute prothrombin time (dPT) and a dilute Russell's viper venom time (dRVVT), as described by Horbach et al. [17], and by a PTT-LA test (Diagnostica Stago).

Table 1.  Prevalence of anti-β2GPI or antiprothrombin antibodies in the 12 selected SLE patients with LAC activity
PatientLACAnti-β2 GPI antibodiesAnti-FII- antibodies
  1. The presence of anti-β2GPI or antiprothrombin antibodies (anti-FII antibodies) in these patients plasmas were determined as described by Horbach et al. [17]. LAC activity was determined by a dilute prothrombin time (dPT) and a dilute Russell's viper venom time (dRVVT) as described by Horbach et al. [17], and by a PTT-LA test as described in Materials and methods.

1++
2++
3++
4++
5++
6++
7++
8++
9++
10+++
11+++
12+++

Proteins

β2GPI was purified from freshly frozen citrated human plasma (obtained from the local blood bank) using DEAE-Sephadex A50, protein G-Sepharose, S-Sepharose and heparin-Sepharose chromatography as described by Horbach et al. [17]. β2GPI appeared as a single band of approximately 42 kDa on sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) under non-reducing conditions.

Prothrombin was purified as described previously by Fujikawa et al. [24]. It appeared as an 80-kDa protein on SDS–PAGE under non-reducing conditions.

Antibodies

Monoclonal antibodies, all described in previous reports [25,26], against β2GPI (22A11, 23H9, 27G7) and against prothrombin (28F4) were used.

Antiprothrombin antibodies were purified from two different LAC positive patient plasmas (patients 1 and 2), which were positive for antiprothrombin antibodies and negative for anti-β2GPI antibodies, as we described previously [16]. In short, patient plasmas were applied to prothrombin-sepharose and the column was washed with Tris-buffered saline (TBS). Bound antibodies were eluted with KSCN (3 mol L−1) in TBS and protein containing fractions were pooled and dialyzed against TBS.

Anti-β2GPI antibodies were isolated from LAC-positive patient plasma 6, which was positive for anti-β2GPI antibodies and negative for antiprothrombin antibodies, by a β2GPI-sepharose column. Purification was further performed as described for the isolation of antiprothrombin antibodies.

Preparation of phospholipid vesicles

Phospholipid vesicles containing 100% phosphatidylcholine (PC) or vesicles containing 20% phosphatidylserine (PS) and 80% PC were prepared as described by Brunner et al. [27], with some modifications. After evaporation of chloroform, phospholipids were dissolved in 10 mmol L−1 Tris, pH 7.2, 100 mmol L−1 NaCl, 0.2 mmol L−1 EDTA, 0.02% (w/v) NaN3, 48 mmol L−1 sodium deoxycholate. Vesicles were obtained by applying this mixture to a PD-10 Sephadex G-25 column. The phospholipid content of the fractions was determined by phosphate analysis [28].

CL vesicles were prepared according to the method described by Pengo et al. [29] with some modifications. In short, CL was dried under a stream of nitrogen. The lipids were resuspended to a concentration of 10 mg mL−1 in TBS by vigorous agitation, using a vortex mixer.

Modified PTT-LA test

We used a PTT-LA test to detect LAC according to the manufacturer instructions, with some modifications. As a screening test, we used 50 µL PTT-LA reagent (cephalin and silica) diluted 1 : 1 with TBS. As a confirm test, we used two different neutralization procedures, namely: (i) neutralization with 25 µL 20% PS/80% PC vesicles (final concentrations in the assay 4.4. 8.8, 17.6 and 35.2 µmol L−1) added to 25 µL PTT-LA reagent (cephalin and silica); and (ii) neutralization with 25 µL CL vesicles (final concentrations in the assays 5.2, 10.4, 20.8, 41.6 µmol L−1) added to 25 µL PTT-LA reagent (cephalin and silica). In short, 50 µL screen PTT-LA reagent (‘screen’) or 50 µL PTT-LA reagents with PS/PC or CL vesicles (‘confirm’) were added to: (i) 50 µL normal pooled plasma (NPP); (ii) 50 µL NPP to which 50 µg mL−1 purified patient antiprothrombin antibodies or 100 µg mL−1 purified patient anti-β2GPI antibodies were added; (iii) 50 µL NPP to which 75 µg mL−1 mAb 28F4 or 100 µg mL−1 mAb 22A11, 23H9 or 27G7 were added; or (iv) 50 µL of different patient plasmas (PP). After 3 min of incubation at 37 °C, coagulation was initiated by adding 50 µL CaCl2 (25 mmol L−1) and clotting time was measured in a KC-10 microcoagulometer (Amelung, Lemgo, Germany).

In the case of NPP, to which purified patient antibodies or mAbs were added, ratios are defined as: clotting time NPP in the presence of antibodies divided by clotting NPP in the absence of antibodies with both clotting times measured at the same phospholipid composition and concentration. In the case of selected patient plasma (PP), the ratio is defined as the clotting time of PP divided by clotting time of NPP, with both clotting times measured at the same phospholipid composition and concentration.

Interaction studies between β2GPI, prothrombin and cardiolipin in the absence or presence of calcium using SPR analysis

Binding studies were performed employing a BIACORE2000 biosensor system (Biacore, Uppsala, Sweden). CL, diluted 1 : 10 in HEPES + EDTA buffer (20 mmol L−1 HEPES, 150 mmol L−1 NaCl, 3 mmol L−1 EDTA, pH 7.4), was immobilized on an L1 chip. A control channel was coated with PC (0.5 µmol L−1 in HEPES + EDTA buffer) and binding to the coated channel was corrected for binding to the control channel. Coating was performed at 25 °C with a flow rate of 2 µL min−1. For quantitative measurements of β2GPI and prothrombin binding to immobilized CL, experiments were performed at three different concentrations. The concentrations were chosen at an appropriate range (around Kd values), and the proteins were passed at 25 °C with a flow rate of 10 µL min−1. When binding of β2GPI and prothrombin was studied in the absence of calcium, HEPES + EDTA buffer was used as a flow buffer, and protein samples were diluted in this buffer. When the binding was studied in the presence of calcium, HEPES + CaCl2 buffer (20 mmol L−1 HEPES, 150 mmol L−1 NaCl, 3 mmol L−1 CaCl2, pH 7.4) was used as flow buffer and protein samples were diluted in this buffer. Elution of β2GPI and prothrombin in the absence of calcium was performed with 10 mmol L−1 NaOH. In the presence of calcium, β2GPI and prothrombin were eluted with 10 mmol L−1 NaOH and 25 mm EDTA, respectively. Elutions were performed at a flow rate of 10 µL min−1 for 30 s.

Analysis of quantitative SPR data

For analysis of the association and dissociation curves of the sensograms, BIA evaluation software was used (Biacore AB, Uppsala, Sweden). Interaction constants were determined by performing non-linear global fitting of data corrected for bulk refractive index changes. Data were fitted according to various models available within the BIA evaluation software. A model describing a 1 : 1 interaction was found to provide the best fit for data regarding the binding of β2GPI and prothrombin to immobilized CL in the absence or presence of calcium.

Results

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Effect of CL and PS/PC vesicles on clotting time of NNP

The clotting time of NPP in the PTT-LA test in the absence of CL vesicles was 38.7 ± 1.2 s. When increasing concentrations CL vesicles were added to NPP, the clotting time lengthened to 51.6 ± 3.8 s at a concentration of 41.6 µmol L−1 CL. In contrast, when increasing concentrations of PS/PC vesicles were added, the clotting time of NPP shortened to 30.1 ± 1.10 s at a concentration of 32 µmol L−1 PS/PC (Fig. 1a). When effects of addition of increasing amounts of CL or PC/PS vesicles on clotting times were evaluated using a dRVVT-or PT based assays, we always found a shortening of the clotting times. This is illustrated in Fig. 1(b) for the dRVVT.

image

Figure 1. Effect of addition of cardiolipin (●) or PS/PC (○) vesicles on PTT-LA (a) and dRVVT (b) of normal plasma.

Download figure to PowerPoint

Effect of cardiolipin on β2GPI- or prothrombin-dependent LAC activity

To measure the influence of CL on the prolongation of the clotting times induced by β2GPI or prothrombin antibodies with LAC activity, 100 µg mL−1 anti-β2GPI-mAb 27G7 or 75 µg mL−1 antiprothrombin-mAb 28F4 were added to NPP. The clotting times of the plasmas tested increased from 38 to 68 and 74 s, respectively. Subsequently, the clotting times were measured in the presence of increasing CL concentrations. As shown in Fig. 2(a), the prolonged clotting time in the presence of anti-β2GPI-mAb 27G7 shortened upon addition of CL. In contrast, in the presence of antiprothrombin-mAb 28F4 the clotting time progressively increased when increasing amounts of CL were added. For anti-β2GPI-mAbs 22A11 and 23H9, and purified patient anti-β2GPI and for antiprothrombin antibodies similar results were obtained (Table 2).

image

Figure 2. Effect of cardiolipin on β2GPI- or prothrombin-dependent LAC activity. Twenty-five microliters of cardiolipin vesicles (26, 52, 104, and 208 µmol L−1) were added to 25 µL PTT-LA reagent and this modified reagent was consequently added to 50 µL plasma samples. After 3 min of incubation at 37 °C, coagulation was initiated with 25 mm CaCl2. (a) Results with NNP spiked with 100 µg/mL anti-β2GPI-mAb 27G7 (●) or 75 µg/mL anti-FII-mAb 28F4 (○). (b) Results of PTT-LA with patient plasma 8 (▪) with β2GPI-dependent LAC activity and patient plasma 5 (□) with prothrombin-dependent LAC activity.

Download figure to PowerPoint

Table 2.  Effect of CL or PS/PC vesicles on clotting time ratios due to addition of various anti-β2GPI and antiprothrombin antibodies
PL (µmol L−1)Anti-β2GPI-mAb  Purified β2GPI antibodyAnti-FII mAbPurified anti-FII antibody
  1. 100 µg mL−1 anti-β2GPI-mAb 22A11, 23H9, 27G7, 100 µg mL−1 purified patient anti-β2GPI antibodies (anti-β2GPI-Ab), 75 µg mL−1 antiprothrombin-mAb (anti-FII-mAb) 28F4, or 50 µg mL−1 purified antiprothrombin antibodies (anti-FII-Ab, patient 1 and 2) were added to NPP. Clotting times were measured in a modified PTT-LA test as described in Materials and methods and the results are expressed as ratios (NPP + antibody/NPP − antibody).

CL22A1123H927G7628F412
 01.452.051.961.392.161.901.65
 5.21.311.611.411.192.281.991.79
 10.41.201.511.291.092.311.961.81
 20.81.141.431.181.062.242.031.89
 41.61.091.391.081.002.062.101.97
PS/PC22A1123H927G7628F412
 01.452.051.961.392.161.901.65
 4.21.532.031.881.302.091.561.48
 8.81.521.841.731.262.061.431.40
 17.61.391.651.601.191.891.301.33
 35.21.271.501.451.121.691.261.33

Figure 2(b) shows the results with 2 LAC-positive patient plasmas selected for the presence of either anti-β2GPI or antiprothrombin antibodies only. Addition of CL vesicles to patient plasma with only anti-β2GPI antibodies shortened the prolonged clotting time, whereas addition of CL vesicles to patient plasmas with prothrombin-dependent LAC activity increased the prolonged clotting time. The results for the nine patient plasmas are presented in Table 3. All patients were LAC positive, three patients had only anti-β2GPI antibodies, three patients had only antiprothrombin antibodies and three patients had both anti-β2GPI antibodies and antiprothrombin antibodies. Note that in one patient (number 10), despite the presence of anti-β2GPI antibodies as detected with an ELISA, no shortening of the clotting times was observed, indicating that the LAC activity in this patient was entirely due to the antiprothrombin antibodies. It should be noted that patient 10 had predominantly IgA-type antibodies in her plasma. The coefficient of variation of the PTT-LA in the presence of 20.8 µmol L−1 was between 3 and 7%. With a dRVVT or a dilute PT, CL vesicles did not discriminate between anti-β2GPI and antiprothrombin antibodies with LAC activity.

Table 3.  Clotting time ratios of different patient plasmas with β2GPI or prothrombin-dependent LAC activity in the absence or presence of increasing concentrations CL or PS/PC vesicles
PL (µmol L−1)Patient plasmaPatient plasma both antibodies  
Only anti-β2GPI antibodiesOnly anti-FII antibodies
  1. Clotting times of patient plasmas 7, 8, and 9 with β2GPI-dependent LAC activity and patient plasmas 3, 4, and 5 with prothrombin-dependent LAC activity were measured in the modified PTT-LA test and the results are expressed as clotting time ratios (patient plasma/NPP, both with the same phospholipid (PL) concentration)

CL789345101112
 02.322.092.791.381.771.852.751.381.40
 5.22.281.762.461.441.891.932.731.311.19
 10.42.141.692.381.401.801.902.971.241.09
 20.82.101.702.331.481.931.942.991.191.06
 41.61.941.662.041.301.771.762.891.091.00
PS/PC789345101112
 02.322.092.791.381.771.852.751.381.40
 4.22.231.952.631.261.351.702.311.291.30
 8.82.021.782.371.271.321.602.171.211.26
 17.61.761.472.151.231.311.491.951.091.12
 35.21.531.311.881.241.201.401.870.951.12

Effect of PS/PC on β2GPI- and prothrombin-dependent LAC activity

Next, we studied the effects of PS/PC vesicles on β2GPI- or prothrombin-dependent LAC activity. Both the prolonged PTT-LA caused by anti-β2GPI-mAb 27G7 or antiprothrombin-mAb 28F4 shortened upon addition of increasing concentrations PS/PC vesicles. Similar results were observed with the other anti-β2GPI-mAbs (22A11 and 23H9), the purified patient anti-β2GPI (patient 6), antiprothrombin antibodies (patients 1 and 2) and with patient samples (Table 2).

Addition of increasing concentrations PS/PC vesicles shortened the prolonged PTT-LA irrespective whether this was caused by β2GPI-dependent LAC or prothrombin-dependent LAC activity (Table 3).

Effect of CL and PS/PC vesicles on LAC activity caused by a combination of anti-β2GPI and antiprothrombin antibodies

To investigate the influence of CL and PS/PC vesicles on the expression of LAC activity caused by a combination of anti-β2GPI and antiprothrombin antibodies, 100 µg mL−1 anti-β2GPI-mAb 27G7 and 75 µg mL−1 antiprothrombin-mAb 28F4 were added together to NPP. After 30-min incubation at 4 °C, clotting times were measured with the PTT-LA test in the presence of different CL and PS/PC concentrations. As depicted in Fig. 3(a), the prolonged clotting time caused by the combination of anti-β2GPI-mAb 27G7 and antiprothrombin-mAb 28F4 was shortened upon addition of 5.2 µmol L−1 CL vesicles. After addition of higher concentrations CL vesicles, the clotting time started to lengthen. By adding increasing concentrations of PS/PC vesicles, clotting times decreased progressively, similar to what was observed when PS/PC vesicles were added to NPP spiked with anti-β2GPI or antiprothrombin mAbs (data not shown).

image

Figure 3. Effect of cardiolipin on a mixture of β2GPI- and prothrombin-dependent LAC activity. (a) NNP was spiked with 100 µg mL−1 anti-β2GPI-mAb 27G7 (□) or 75 µg mL−1 anti-FII-mAb 28F4 (○) or a combination of both mAbs (▪). (b) NNP was spiked with 40 µg mL−1 anti-β2GPI-mAb 22A11 (ratio when added to NNP = 1.19) (▪) or 75 µg mL−1 anti-FII-mAb 28F4 (ratio when added to NNP = 1.60) (○) or a combination of both mAbs (●). Clotting times were measured with a PTT-LA after addition of increasing concentrations of CL. Ratios were calculated by dividing clotting time in the presence of the antibodies by the clotting time in the absence of the antibodies at the indicated cardiolipin concentration. (□) The sum of the ratios that were found when with anti-β2GPI and antiprothrombin antibodies alone.

Download figure to PowerPoint

We also tested whether neutralization with CL vesicles can identify weak LAC activity caused by anti-β2GPI antibodies, when in the same sample antiprothrombin antibodies with strong LAC activity were present. To do so we used a combination of an anti-β2GPI-mAb and an antiprothrombin mAb that when added alone to normal plasma gave LAC ratios (PTT-LA with antibody/PTT-LA without antibody) of 1.19 and 1.60, respectively. Addition of increasing amounts of cardiolipin to plasma with the mixture of antibodies decreased the LAC ratio in the mixture to the same extent as could be predicted on the calculated sum of the ratios that were found when the antibodies were added alone (Fig. 3b).

Binding of β2GPI and prothrombin to immobilized cardiolipin

Binding of β2GPI and prothrombin to CL was investigated by SPR analysis using purified proteins and measuring the association and dissociation rate constants Kon and Koff (Table 4). No binding of β2GPI and prothrombin was observed to the control channel, coated with PC. In the absence of calcium, dose-dependent binding of β2GPI to CL was observed with high affinity (Kd = 1.4 nmol L−1; Table 4). In the presence of calcium, β2GPI still bound to CL with high affinity (Kd = 2.9 nmol L−1; Table 4). Upon replacement of β2GPI by flowbuffer, the resonance signal gradually declined, indicating that β2GPI dissociates from immobilized CL and that binding is reversible.

Table 4.  Kinetics of β2GPI and prothrombin binding to CL in the absence or presence of CaCl2
Injected proteinCaCl2 (µmol L−1)Kon (s−1 mol L−1)Koff (s−1)Kd (nmol L−1)
  1. Association and dissociation constants between immobilized cardiolipin and either β2GPI or prothrombin was determined as described in Materials and methods. The data were analyzed by performing non-linear global fitting of data corrected for bulk refractive index changes to calculate association rate constants (Kon) and dissociation rate constants (Koff). Data represents the mean of two independent experiments.

β2GPI02.0 × 1052.8 × 10−41.4
β2GPI32.3 × 1056.7 × 10−42.9
Prothrombin06.0 × 1043.4 × 10−357.0
Prothrombin39.6 × 1052.2 × 10−220.5

The affinity of prothrombin for CL was significantly lower than that of β2GPI, irrespective of the absence or presence of calcium (Table 4). In the absence of calcium, a dose-dependent binding of prothrombin to CL was observed with relative low affinity (Kd = 57 nmol L−1; Table 4). As expected, in the presence of calcium, the affinity of prothrombin for CL increased (Kd = 20.5 nmol L−1; Table 4). When prothrombin was replaced by flowbuffer, the resonance signal gradually declines, indicating dissociation of prothrombin from immobilized CL and reversible binding.

Discussion

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

The results presented in this paper show that by using a PTT-LA test and two neutralization procedures with either CL or PS/PC vesicles, differentiation is possible between LAC activity caused by anti-β2GPI and antiprothrombin antibodies. Addition of increasing concentrations of CL to NPP spiked with mono- or polyclonal anti-β2GPI antibodies with LAC activity, or to patient plasma with LAC due to anti-β2GPI antibodies shortens the prolonged aPTTs (Fig. 2a,b). Addition of CL to NPP spiked with antiprothrombin antibodies with LAC activity or to patient plasmas with LAC due to antiprothrombin antibodies prolongs the aPTT clotting times (Fig. 2a,b). When the same experiments were done with PS/PC vesicles, we noted neutralization of LAC activity irrespective whether this was due to anti-β2GPI or antiprothrombin antibodies. In many patients, LAC activity is caused by a mixture of both types of antibody [30]. We also performed experiments with NPP spiked with both anti-β2GPI antibodies and antiprothrombin antibodies. Here we found that addition of CL vesicles could reveal even the presence of weak anti-β2GPI antibodies (ratio 1.19) in a mixture of both types of antibodies (Fig. 3).

The basic criteria for the detection of LAC are (i) prolongation of a phospholipid-dependent coagulation test; (ii) evidence of an inhibitor demonstrated by mixing studies with normal plasma; and (iii) confirmation of the phospholipid-dependent nature of the inhibitory antibody by adding extra phospholipids. Commercial tests for the detection of LAC differ with respect to composition and concentration of phospholipids used for detection and conformation assays. These differences probably affect the sensitivity and specificity of the tests. Indeed, a large number of publications have shown that the correlation between different tests for the detection of LAC is unacceptable low [31,32]. Thus far, no guidelines have been proposed on the nature and composition of the phospholipids to be used because of the assumption that all negatively charged phospholipids will neutralize all different types of LAC activity irrespective of the test system used. Here we show for the first time that this assumption is incorrect. We observed major differences for cardiolipin in its ability to neutralize LAC activity. Cardiolipin neutralized anti-β2GPI antibody-dependent LAC in an APTT-based assay while cardiolipin did not neutralize anti-β2GPI antibody-dependent LAC activity in a dRVVT or a dPT. These differences were not found with antiprothrombin antibody dependent LAC activity.

It is noteworthy that when cardiolipin vesicles were added to NPP, we observed a prolongation of the clotting time with PTT-LA reagents (Fig. 1a). This effect was specific for aPTT based clotting tests, and it was not observed for a dPT or dRVVT. We suppose that interference of CL with the contact activation pathway is the cause of the observed increase in clotting time, probably due to the adsorption of high molecular weight kininogen (HMWK) and/or prekallikrein by cardiolipin, thereby depleting the plasma from these essential cofactors of the contact activation (V. Pengo, personal communication). To our knowledge, this interference of cardiolipin with aPTT-based assays has not been described before. Why addition of CL prolongs the aPTT is unclear at the moment, however, this phenomenon enhances the differences in neutralization patterns for β2GPI- and prothrombin-dependent LAC activity.

The effects of the monoclonal antibodies directed against β2GPI on the prolongation of the PTT-LA by cardiolipin are much stronger than the effects of the autoantibodies against β2GPI isolated from patients, especially at higher concentrations of cardiolipin. These differences are due to the higher affinity of the monoclonal antibodies for β2GPI compared to the human antibodies.

The interaction of β2GPI with negatively charged phospholipids was first shown by Shousboe [33]. Later on, by using different techniques, the interaction of β2GPI with negatively charged phospholipids was demonstrated in several other studies [13,14,34–39], but quantitative data on the binding to CL are scarce. Also data on the interaction of prothrombin with negatively charged PS/PC surfaces have been described extensively [30,40–47], but studies on the interaction of prothrombin with CL are lacking. We show here with SPR analysis using purified β2GPI and prothrombin that in the absence or presence of calcium, the affinity of β2GPI for CL is, respectively, 40 and 10 times higher compared to that of prothrombin (Table 4). This relatively high affinity of β2GPI for CL might provide an explanation for the contrasting effects that addition of CL vesicles has on prolonged clotting times caused by either anti-β2GPI or antiprothrombin antibodies with LAC activity. However, as the differences were not seen with clotting assays initiated with tissue factor or Russell's viper venom, competition for catalytic surfaces could not be the only explanation. Addition of CL vesicles to normal plasma by itself caused a prolongation of aPTT-based clotting assays. We presume that besides competition between β2GPI–anti-ß2GPI antibody complexes and clotting factors for the available catalytic surface, the complexes also interfere with the CL-induced prolongation of the PTT-LA.

In conclusion, with a modified PTT-LA test, in which cardiolipin and PS/PC vesicles are used as confirmation reagent, one can discriminate between β2GPI and prothrombin-dependent LAC activities in patient plasmas. This modified PTT-LA test is suitable for use in clinical studies that evaluate the importance of β2GPI- and prothrombin-mediated LAC. Further studies of the specificity of phospholipids for the detection and neutralization of specific LAC inducing antibodies are of major importance to improve our assays used for the laboratory diagnosis of LAC.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

The authors kindly thank Dr P.J. Lenting and E. Westein (Thrombosis and Haemostasis Laboratory, Department of Haematology, University Medical Center, Utrecht, the Netherlands) for technical assistance and helpful discussions.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  • 1
    Kandiah DA, Krilis SA. Laboratory detection of antiphospholipid antibodies. Lupus 1996; 5: 1602.
  • 2
    Shapiro SS, Thiagarajan P. Lupus anticoagulants. Prog Haemost Thromb 1982; 6: 26385.
  • 3
    Exner T, Triplett DA, Taberner D, Machin SJ. Guidlines for testing and revised criteria for lupus anticoagulants. Thromb Haemost 1991; 65: 3202.
  • 4
    Brandt JT, Triplett DA, Alving B, Scharrer I. Criteria for the diagnosis of lupus anticoagulants: An update. Thromb Haemost 1995; 74: 118590.
  • 5
    Harris EN, Gharavi AE, Boey ML, Patel BM, Mackworth-Young CG, Loizou S, Hughes GR. Anticardiolipin antibodies: detection by radioimmunoassay and association with thrombosis in systemic lupus erythematosus. Lancet 1983; ii: 12114.
  • 6
    Loizou S, McCrea JD, Rudge A, Reynolds. R, Boyle CC, Harris EN. Measurement of anti-cardiolipin antibodies by an enzyme-linked immunosorbent assay (ELISA): standardization and quantitation of results. Clin Exp Immunol 1985; 62: 73845.
  • 7
    Bevers EM, Galli M, Barbui T, Comfurius P, Zwaal RFA. Lupus anticoagulant IgG's (LA) are not directed to phospholipids only, but to a complex of lipid-bound human prothrombin. Thromb Haemost 1991; 66: 62932.
  • 8
    Permpikul P, Rao LVM, Rapaport SI. Functional and binding studies of the roles of prothrombin and β2-glycoprotein I in the expression of lupus anticoagulant activity. Blood 1994; 83: 287892.
  • 9
    Roubey RAS, Pratt CW, Buyon JP, Winfield JB. Lupus anticoagulant activity of autoimmune antiphospholipid antibodies is dependent upon beta 2-glycoprotein I. J Clin Invest 1992; 90: 11004.
  • 10
    Oosting JD, Derksen RHWM, Entjes HTI, Bouma BN, De Groot PhG. Lupus anticoagulant activity is frequently dependent on the presence of β2-glycoprotein I. Thromb Haemost 1992; 67: 499502.
  • 11
    Shi W, Chong BH, Hogg PJ, Chesterman CN. Anticardiolipin antibodies block the inhibition by β2-glycoprotein I of the factor Xa generating activity of platelets. Thromb Haemost 1993; 70: 3425.
  • 12
    Roubey RAS, Eisenberg RA, Harper MF, Winfield JB. ‘Anticardiolipin’ autoantibodies recognize β2-glycoprotein I in the absence of phospholipid: importance of Ag density and bivalent binding. J Immunol 1995; 154: 95460.
  • 13
    Arnout J, Wittevrongel C, Vanrusselt M, Hoylaerts M, Vermylen J. Beta-2-glycoprotein I dependent lupus anticoagulants form stable bivalent antibody beta-2-glycoprotein I complexes on phospholipid surfaces. Thromb Haemost 1998; 79: 7986.
  • 14
    Willems GM, Janssen MP, Pelsers MAL, Comfurius P, Galli M, Zwaal RFA, Bevers EM. Role of divalency in the high-affinity binding of anticardiolipin antibody-β2-glycoprotein I complexes to lipid membranes. Biochemistry 1996; 35: 1383342.
  • 15
    Takeya H, Mori T, Gabazza EC, Kuroda K, Deguchi H, Matsuura E, Ichikawa K, Koike T, Suzuki K. Anti-β2-glycoprotein I (β2GPI) monoclonal antibodies with lupus anticoagulant-like activity enhance the β2GPI binding to phospholipids. J Clin Invest 1997; 99: 22608.
  • 16
    Simmelink MJA, Horbach DA, Derksen RHWM, Meijers JCM, Bevers EM, Willems GM, De Groot PG. Complexes of anti-prothrombin antibodies and prothrombin cause lupus anticoagulant activity by competing with the binding of clotting factors for catalytic phospholipid surfaces. Br J Haematol 2001; 113: 6219.
  • 17
    Horbach DA, Van Oort E, Donders RCJM, Derksen RHWM, De Groot PG. Lupus anticoagulant is the strongest risk factor for both venous and arterial thrombosis in patients with systemic lupus erythematosus – comparison between different assays for the detection of antiphospholipid antibodies. Thromb Haemost 1996; 76: 91624.
  • 18
    Wahl DG, Guillemin F, De Maistre E, Perret C, Lecompte T, Thibaut G. Risk for venous thrombosis related to antiphospholipid antibodies in systemic lupus erythematosus – a meta-analysis. Lupus 1997; 6: 46773.
  • 19
    Galli M, Finazzi G, Bevers EM, Barbui T. Kaolin clotting time and dilute Russell's viper venom time distinguish between prothrombin-dependent and β2-glycoprotein I- dependent antiphospholipid antibodies. Blood 1995; 86: 61723.
  • 20
    Swadzba J, Declerck LS, Stevens WJ, Bridts CH, Van Cotthem KA, Musial J, Jankowski M, Szczeklik A. Anticardiolipin, anti-β2-glycoprotein I, antiprothrombin antibodies, and lupus anticoagulants in patients with systemic lupus erythematosus with a history of thrombosis. J Rheumatol 1997; 24: 17105.
  • 21
    D'Angelo A, Safa O, Crippa L, Garlando A, Sabbadini MG, D'Angelo SV. Relationship of lupus anticoagulant, anticardiolipin, anti-β1-GPI and anti-prothrombin autoantibodies with history of thrombosis in patients with the clinical suspicion of APA syndrome. Thromb Haemost 1997; 78: 9678.
  • 22
    Galli M, Barbui T. Antiprothrombin antibodies: detection and clinical significance in the antiphospholipid syndrome. Blood 1999; 93: 214957.
  • 23
    Brandt JT, Triplett DA, Alving B, Scharrer I. Criteria for the diagnosis of lupus anticoagulants: an update, on behalf of the Subcommittee for the Standardization of Lupus Anticoagulants. Thromb Haemost 1995; 74: 118590.
  • 24
    Fujikawa K, Thompson AR, Legaz ME, Meyer RG, Davie EW. Isolation and characterization of bovine factor IX (Christmas factor). Biochemistry 1973; 12: 493845.
  • 25
    Arnout J, Vanrusselt M, Wittevrongel C, Vermylen J. Monoclonal antibodies against beta-2-glycoprotein I. Use as reference material for lupus anticoagulant tests. Thromb Haemost 1998; 79: 9558.
  • 26
    Arnout J, Vanrusselt M, Nevens C, Smans K, Wittevrongel C, Vermylen J. Some murine monoclonal antibodies against human prothrombin have lupus anticoagulant activity. Thromb Haemost 1999; 82(Suppl.): 65.
  • 27
    Brunner J, Skrabal P, Hauser H. Single bilayer vesicles prepared without sonication. Physio-chemical properties. Biochim Biophys Acta 1976; 455: 32231.
  • 28
    Rouser G, Fkeischer S, Yamamoto A. Two dimensional thin layer chromatographic separation of polar lipids and determination of phospholipids by phosphorus analysis of spots. Lipids 1970; 5: 4946.
  • 29
    Pengo V, Thiagarajan P, Shapiro SS, Heine MJ. Immunological specificity and mechanism of action of IgG lupus anticoagulants. Blood 1987; 70: 6976.
  • 30
    Horbach DA, Van Oort E, Tempelman MJ, Derksen RHWM, De Groot PhG. The prevalence of a non-phospholipid-binding form of β2-glycoprotein I in human plasma. Consequences for the development of anti-β2-glycoprotein I-antibodies. Thromb Haemost 1998; 80: 7918.
  • 31
    Arnout J, Meijer P, Vermylen J. Lupus anticoagulant testing in Europe. an analysis of results from the first European Concerted Action on Thrombophilia (ECAT) survey using plasmas spiked with monoclonal antibodies against human beta2-glycoprotein I. Thromb Haemost 1999; 81: 92934.
  • 32
    Jennings I, Kitchen S, Woods TAL, Preston FE, Greaves M. Clinically important inaccuracies in testing for the lupus anticoagulant: an analysis of results from thre surveys of the UK national external quality assessment scheme (NEQAS) for blood coagulation. Thromb Haemost 1997; 77: 9347.
  • 33
    Schousboe I. Purification, characterization and identification of an agglutinin in human serum. Biochim Biophys Acta 1979; 579: 396408.
  • 34
    Brighton TA, Dai YP, Hogg PJ, Chesterman CN. Microheterogeneity of β2-glycoprotein I. implications for binding to anionic phospholipids. Biochem J 1999; 340: 5967.
  • 35
    Comp PC, Nixon RR, Cooper MR, Esmon CT. Familial protein S deficiency is associated with recurrent thrombosis. J Clin Invest 1984; 74: 20828.
  • 36
    Hoshino M, Hagihara Y, Nishii I, Yamazaki T, Kato H, Goto Y. Identification of the phospholipid-binding site of human β2-glycoprotein I domain V by heteronuclear magnetic resonance. J Mol Biol 1998; 304: 92739.
  • 37
    Mehdi H, Naqvi A, Kamboh MI. A hydrophobic sequence at position 313–316 (Leu-Ala-Phe-Trp) in the fifth domain of apolipoprotein H (β2-glycoprotein I) is crucial for cardiolipin binding. Eur J Biochem 2000; 267: 17706.
  • 38
    Arnout J. The role of beta 2-glycoprotein I-dependent lupus anticoagulants in the pathogenesis of the antiphospholipid syndrome. Verh K Acad Geneeskd Belg 2000; 62: 35372.
  • 39
    Hagihara Y, Goto Y, Kato H, Yoshimura T. Role of the N- and C-terminal domains of bovine β2-glycoprotein I in its interaction with cardiolipin. J Biochem (Tokyo) 1995; 118: 12936.
  • 40
    Cuypers PA, Corsel JW, Janssen MP, Kop JM, Hermens WT, Hemker HC. The adsorption of prothrombin to phosphatidylserine multilayers quantitated by ellipsometry. J Biol Chem 1983; 258: 242631.
  • 41
    Lu YF, Nelsestuen GL. Dynamic features of prothrombin interaction with phospholipid vesicles of different size and composition: Implications for protein-membrane contact. Biochemistry 1996; 35: 8193200.
  • 42
    McDonald JF, Shah AM, Schwalbe RA, Kisiel W, Dahlback B, Nelsestuen GL. Comparison of naturally occuring vitamin K-dependent proteins: correlation of amino acid sequences and membrane binding properties suggests a membrane contact site. Biochemistry 1997; 36: 51207.
  • 43
    Ellison EH, Castellino FJ. Adsorption of vitamin K-dependent blood coagulation proteins to spread phospholipid monolayers as determined from combined measurements of the surface pressure and surface protein concentration. Biochemistry 1998; 37: 79978003.
  • 44
    Govers-Riemslag JWP, Johnsen L, Petrovan RJ, Rosing J, Tans G. A kinetic assay to determine prothrombin binding to membranes. Thromb Res 1998; 92: 23947.
  • 45
    Mann KG, Elion J, Butkowski RJ, Downing M, Nesheim ME. Prothrombin. In: Anonymous Blood Clotting. New York, Academic Press Inc., 1981: 286302.
  • 46
    Mayer LD, Nelsestuen GL, Brockman HL. Prothrombin association with phospholipid monolayers. Biochemistry 1983; 22: 31642321.
  • 47
    Mann KG, Lawson JH. The role of the membrane in the expression of the vitamin K-dependent enzymes. Arch Pathol Lab Med 1992; 116: 13306.