tech_banner
Beta2‐glycoprotein I binds thrombin via exosite I and...
Beta2‐glycoprotein I binds thrombin via exosite I and exosite II: Anti–β2‐glycoprotein I antibodies potentiate the inhibitory effect of β2‐glycoprotein I on thrombin‐mediated factor XIa generation - Rahgozar - 2007 - Arthritis amp; Rheumatism - Wiley Online Library Soheila Rahgozar, St. George Hospital, University of New South Wales, Sydney, New South Wales, Australia Drs. Rahgozar and Yang contributed equally to this work.Search for more papers by this authorQiuxia Yang, St. George Hospital, University of New South Wales, Sydney, New South Wales, Australia Drs. Rahgozar and Yang contributed equally to this work.Search for more papers by this authorBill Giannakopoulos, St. George Hospital, University of New South Wales, Sydney, New South Wales, AustraliaSearch for more papers by this authorXiaokai Yan, St. George Hospital, University of New South Wales, Sydney, New South Wales, AustraliaSearch for more papers by this authorSpiros Miyakis, St. George Hospital, University of New South Wales, Sydney, New South Wales, AustraliaSearch for more papers by this authorSteven A. Krilis, Corresponding Author s.krilis@unsw.edu.au St. George Hospital, University of New South Wales, Sydney, New South Wales, AustraliaDepartment of Immunology, Allergy and Infectious Diseases, University of New South Wales, St. George Hospital, 2 South Street, Sydney, New South Wales 2217, AustraliaSearch for more papers by this author Soheila Rahgozar, St. George Hospital, University of New South Wales, Sydney, New South Wales, Australia Drs. Rahgozar and Yang contributed equally to this work.Search for more papers by this authorQiuxia Yang, St. George Hospital, University of New South Wales, Sydney, New South Wales, Australia Drs. Rahgozar and Yang contributed equally to this work.Search for more papers by this authorBill Giannakopoulos, St. George Hospital, University of New South Wales, Sydney, New South Wales, AustraliaSearch for more papers by this authorXiaokai Yan, St. George Hospital, University of New South Wales, Sydney, New South Wales, AustraliaSearch for more papers by this authorSpiros Miyakis, St. George Hospital, University of New South Wales, Sydney, New South Wales, AustraliaSearch for more papers by this authorSteven A. Krilis, Corresponding Author s.krilis@unsw.edu.au St. George Hospital, University of New South Wales, Sydney, New South Wales, AustraliaDepartment of Immunology, Allergy and Infectious Diseases, University of New South Wales, St. George Hospital, 2 South Street, Sydney, New South Wales 2217, AustraliaSearch for more papers by this author Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URLShare a linkShare onEmailFacebookTwitterLinked InRedditWechat Abstract Objective Beta2-glycoprotein I (β2GPI) is a dominant antigenic target in antiphospholipid syndrome (APS). Beta2-glycoprotein I may bind to factor XI and serve a physiologic function as a regulator of factor XI activation by thrombin. We undertook this study to investigate the possible interactions of β2GPI with thrombin in β2GPI-regulated factor XI activation by thrombin and to evaluate the effect of anti-β2GPI antibodies on this system.Methods The β2GPI interaction with thrombin was investigated in direct and competitive assays using β2GPI domain mutants and thrombin-binding exosite oligonucleotides. Beta2-glycoprotein I inhibition of thrombin-mediated factor XI activation was assessed in the presence of 8 anti-β2GPI monoclonal antibodies (mAb) directed against domain I.Results Domain V of β2GPI was involved in direct binding to thrombin, and exosite I and exosite II on thrombin took part in this interaction. Anti-β2GPI mAb produced a 70% inhibition of thrombin-mediated factor XI activation in the presence of β2GPI.Conclusion We demonstrate that β2GPI interacts with thrombin exosites I and II. This novel finding necessitates a reinterpretation of previous studies from which the detection of anti-human thrombin antibodies in APS has been reported. We also show that anti-β2GPI antibodies potentiate the inhibitory effect of β2GPI on thrombin-mediated factor XIa generation.Antiphospholipid syndrome (APS) is a condition characterized by an increased predisposition to arterial and venous thromboses, and recurrent fetal loss, in association with the presence of antiphospholipid antibodies (aPL) (1, 2). APS is now the most common cause of acquired hypercoagulability in the general population, and it is a major cause of morbidity in pregnancy (3). Beta2-glycoprotein I (β2GPI) is the primary target antigen recognized by autoantibodies in patients with APS (4-6). In vivo studies using animal models suggest that anti-β2GPI antibodies are able to directly promote thrombus propagation (7, 8). In view of the prime importance of the β2GPI molecule not only as a substrate in the detection of β2GPI-dependent aPL (9), but also potentially as a mediator in a complex with anti-β2GPI antibodies, detailed studies have been undertaken to delineate its structure (9, 10). The key features will be reviewed, for a basic understanding is required to appreciate the experiments described in this report (for a comprehensive review, see refs.9 and10). Beta2-glycoprotein I circulates in plasma in a free and bound form associated with lipoproteins. It is a single-chain glycoprotein that comprises 5 domains. Each domain is characterized by disulfide bridges joining the first-to-third and second-to-fourth cysteine residues. Although the first 4 domains (DI, DII, DIII, and DIV) are similar, the fifth domain (DV) has an extra internal disulfide bond and a highly conserved C-terminal loop, encompassing residues Cys306–Cys326. The C-terminus is unusual in that it is disulfide linked and does not have a free carboxyl residue (9, 10). The C-terminal loop is surface exposed and susceptible to proteolytic cleavage between Lys317 and Thr318 by factor XIa and plasmin, thus abolishing β2GPI binding to anionic phospholipids (11-13). Earlier studies suggested that β2GPI may function as a mild natural anticoagulant, and the proposed mechanism was that it binds the anionic phospholipid surface on platelets, thus inhibiting the initiation of the contact pathway (14). A novel mechanism by which the β2GPI molecule may regulate the contact pathway has been suggested recently by our group. The crucial insights into how this may occur were provided by the discovery that β2GPI is able to directly bind factor XI and to inhibit factor XI activation by factor XIIa and thrombin (15). Once thrombin is generated (depending on the milieu and the substrate with which it interacts), it may induce either a procoagulant activity contributing to the thrombin burst and the stabilizing of clot, or an anticoagulant effect by activating protein C in the presence of thrombomodulin (16). To delineate the mechanism whereby β2GPI inhibits factor XI activation by thrombin, we undertook studies looking at the possible interaction of β2GPI with thrombin. We demonstrated the novel finding that domain V of β2GPI is involved in direct binding to thrombin and that exosite I and exosite II take part in this interaction. In the context of our previous studies, we propose the existence of a quaternary complex on the surface of activated platelets in which β2GPI may colocalize with factor XI and thrombin to regulate thrombin generation. The β2GPI interaction with thrombin has significant implications for the interpretation of assays that purport to detect anti-human thrombin antibodies in patients with APS (17, 18) and may necessitate a reinterpretation of their significance in light of this finding. We also investigated the influence of anti-β2GPI antibodies on the interaction between factor XI, β2GPI, and thrombin. Surprisingly, we found that anti-β2GPI antibodies potentiate the β2GPI-mediated inhibition of factor XI activation by thrombin. To explain the pathophysiology of APS, other mechanisms may need to be examined. Human plasma–derived factor XI, α-thrombin (factor IIa), and plasmin were purchased from Calbiochem-Novabiochem (La Jolla, CA). Biotin-D-Phe-Pro-Arg-chloromethylketone–blocked human α-thrombin (BFPRck–thrombin) was from Hematologic Technologies (Essex Junction, VT). S2238 (H-D-Phe-Pip-Arg-p-nitroanilide) and S2366 (pyroGlu-Pro-Arg-p-nitroanilide) were purchased from Chromogenix Instrumentation Laboratory (Milan, Italy). MaxiSorp Lockwell plates were from Nunc (Roskilde, Denmark). Streptavidin–horseradish peroxidase (HRP) conjugate and tetramethylbenzidine (TMB) peroxidase substrate were purchased from Rockland (Gilbertsville, PA). Hirudin, human serum albumin, bovine serum albumin (BSA), plasmin, Dextran sulfate, and p-nitrophenyl phosphate were purchased from Sigma (St. Louis, MO). Baculovirus vector pBakPac6 and oligonucleotide-directed mutagenesis kit were purchased from Clontech Laboratories (Franklin Lakes, NJ). Spodoptera frugiperda Sf9 cells and serum-free medium Sf-900II were purchased from Invitrogen (Carlsbad, CA). Recombinant human β2GPI and domain deletion mutants of recombinant human β2GPI were generated as previously described (19). Domain deletion mutants of recombinant human β2GPI were described as follows: domain I (DI), domain I–II (DI–II), domain I–III (DI–III), domain I–IV (DI–IV), domain II–V (DII–V), domain III–V (DIII–V), and domain IV–V (DIV–V). Plasma-derived native β2GPI was purified in our laboratory by a sequential protocol with perchloric acid precipitation, cation-exchange chromatography, heparin affinity chromatography, and gel filtration with Sephacryl S-300. Native β2GPI proteolytically clipped by plasmin at Lys317–Thr318 was generated as previously described (20). Eight murine monoclonal antibodies (mAb) directed against β2GPI domain I were used in this study. The mAb 3G11F9 and 4B2E7F4 (designated mAb 1 and 7, respectively) and the mAb FC1 were generated in our laboratory as previously described (21) and purified using affinity chromatography with protein A–agarose. The mAb 6A4, 6F3, 10D12, 8C3, and 9GI were kindly provided by Dr. J. Arvieux (Grenoble, France). Sheep anti-human factor XI, rabbit anti–factor IIa, and anti-rabbit IgG were obtained from Affinity Biologicals (Ancaster, Ontario, Canada). The 2 single-stranded DNA oligonucleotides HD22 and HD1 (AGTCCGTGGTAGGGCAGGTTGGGGTGACT and GGTTGGTGTGGTTGG), directed toward exosite II and exosite I, respectively, of thrombin, and their nonbinding analogs AGTCCGTAATAAAGCAGGTTAAAATGACT and GGTGGTGGTTGTGGT, referred to as HD4 and HD3, respectively, were synthesized by Invitrogen. Binding of thrombin to immobilized recombinant human β2GPI was performed using MaxiSorp Lockwell plates, the wells of which were coated with 100 μl recombinant human β2GPI or BSA (10 μg/ml) in 50 mM carbonate–bicarbonate buffer, pH 9.6, by incubation overnight at 4°C. The plates were washed 5 times with phosphate buffered saline (PBS)–0.1% Tween 20, pH 7.5 (PBST) using a microplate washer (Beckman Coulter, Fullerton, CA). The wells were blocked with 2% BSA/PBST for 2 hours at 25°C and then washed 5 times with PBST. One hundred microliters of thrombin (10 μg/ml) in PBS was applied to individual wells in triplicate and incubated for various times at 25°C. Plates were washed at the indicated time points, and 100 μl of thrombin substrate (S2238, 1.6 mM) was then added and incubated for 3 hours at 37°C. Optical density (OD) was measured at 405 nm using a Microplate Scanning Spectrophotometer (Bio-Tek Instruments, Winooski, VT). The amount of thrombin bound was derived from a standard curve constructed with known concentrations of thrombin. The cleavage of S2238 was plotted as a function of time, from 0 to 3 hours. Specific saturation binding of thrombin to immobilized recombinant human β2GPI was performed using a MaxiSorp Lockwell plate coated with 100 μl of recombinant human β2GPI (200 nM) as indicated above. The total binding was assessed by adding 100 μl of various concentrations of thrombin (23–680 nM) in PBS to the wells and incubating the plate for 2 hours at 25°C, since this was the optimal time for maximal binding (results not shown). The nonspecific binding was identified by adding a mixture of 50 μl of BFPRck–thrombin (2.3–6.8 μM) and 50 μl of thrombin (23–680 nM) to the wells and incubating the plate as described above. The wells were washed 5 times with PBST, and 100 μl of S2238 (1.6 mM) was then added and incubated for 3 hours at 37°C. OD was measured at 405 nm. Specific binding of thrombin to recombinant human β2GPI was derived by subtraction of the nonspecific binding from the total binding using GraphPad Prism software, version 3.03 (GraphPad Software, San Diego, CA). The amount of thrombin bound was measured and converted to pmoles bound from a standard curve with known concentrations of thrombin. Reversibility of binding of thrombin to β2GPI was assessed using 100 μl of 200 nM recombinant human β2GPI coated on the plate. The wells were washed with PBST and blocked with 2% BSA/PBST for 2 hours. One hundred microliters of 700 nM thrombin was added to individual wells and incubated for 2 hours at 25°C. The plate was then washed, and 100 μl of BFPRck–thrombin or BSA (35 μM) was added to individual wells. At defined time points (from 10 minutes to 2 hours), the wells were washed 5 times with PBST, and 100 μl of S2238 (1.6 mM) was added and then incubated for 3 hours at 37°C. The amount of thrombin bound was measured and converted to pmoles. Competitive inhibition of thrombin with active site–blocked thrombin for binding to recombinant human β2GPI. A MaxiSorp Lockwell plate was coated overnight at 4°C with 100 μl of recombinant human β2GPI at 227 nM in 50 mM carbonate–bicarbonate coating buffer, pH 9.6, then washed 5 times with PBST, pH 7.4, and blocked for 2 hours at 25°C with 300 μl 2% BSA/PBST. The wells were then washed, and mixtures of 50 μl of human α-thrombin (227 nM) and 50 μl of various concentrations of BFPRck–thrombin (227–4,540 nM) or BSA were added to individual wells and incubated for 2 hours at 25°C. Following 3 washes, 100 μl of S2238 in prewarmed 50 mM Tris, 150 mM NaCl (Tris buffered saline [TBS]), pH 7.6, at 37°C was dispensed to the wells. After 18 hours of incubation at 25°C, generation of p-nitroaniline was monitored by measuring the OD at 405 nm. The 50% inhibition concentration (IC50) was calculated by nonlinear regression using GraphPad Prism software, version 3.03. Following the competitive inhibition experiments, binding of BFPRck–thrombin to recombinant human β2GPI was confirmed after washing the plate 5 times with PBST and adding 100 μl 1/10,000 streptavidin–HRP in 2% BSA/PBST to the wells. The plate was incubated for 30 minutes with shaking at room temperature and washed as described above. One hundred microliters of TMB was added and incubated for 10 minutes, and after the addition of 0.18M H2SO4 stop solution, the plate was read at 450 nm in a Microplate Scanning Spectrophotometer. Binding of thrombin to various preparations of immobilized β2GPI was performed using MaxiSorp Lockwell plates. One hundred microliters of recombinant human β2GPI, DI, DI–II, DI–III, DI–IV, DII–V, DIII–V, DIV–V, and DV or BSA (10 μg/ml) in 50 mM carbonate–bicarbonate coating buffer, pH 9.6, was incubated overnight at 4°C. The plate was treated as indicated above, and 100 μl of thrombin (10 μg/ml) in PBS was then added to the individual wells and incubated for 2 hours at 25°C. The wells were then washed 5 times with PBST, and S2238 (1.6 mM) was then added and incubated for 3 hours at 37°C. The amount of thrombin bound was derived as described above. Effect of single-stranded DNA oligonucleotides HD22 and HD1 on the thrombin–β2GPI interaction. The effect of HD22 and HD1 on the thrombin–β2GPI interaction was studied using a solid-phase binding assay. One hundred microliters of 200 nM recombinant human β2GPI was absorbed to the wells of a MaxiSorp Lockwell plate by incubating in 50 mM carbonate–bicarbonate buffer, pH 9.60, overnight at 4°C. Wells were washed 5 times with 300 μl of PBST (pH 7.5) and were then blocked with 300 μl of 2% BSA/PBST for 2 hours at 25°C. After washing in the same manner, the wells were incubated with 50 μl of HD22, HD4, HD1, and HD3 at final concentrations of 24–768 nM and 50 μl of thrombin at 500 nM for 2 hours at 25°C. The wells were washed, and 100 μl of S2238 (1.6 mM) was then added and incubated for 3 hours at 37°C. The OD was measured at 405 nm. Factor XI activation by thrombin in the presence of β2GPI and anti-β2GPI antibodies. Activation of factor XI by thrombin was carried out in TBSA (50 mM Tris HCl, 150 mM NaCl, 0.1% BSA, pH 7.6) as described previously with some modifications (15). Factor XI activation was assessed with thrombin in the presence of recombinant human β2GPI, or BSA in the presence or absence of 8 different mouse mAb against domain I of β2GPI, or isotype control mAb. Factor XI (60 nM) was mixed with 1 μg/ml of Dextran sulfate, 0.5 μM of different preparations of β2GPI, and/or 0.5 μM different mAb in TBSA, as mentioned above. The mixtures were then incubated for 5 minutes at 37°C, followed by the addition of 2 nM of thrombin in TBSA and incubation for a further 5 minutes at 37°C. Reactions were stopped by adding 25 units/ml of hirudin and diluting the reaction mixture 1:5 with TBS. Factor XIa activity was measured by adding 100 μl of 1.2 mM S2366 to 100-μl aliquots of each reaction mixture in individual microtiter wells. The OD was measured at 405 nm. Data were analyzed by GraphPad Prism software, version 3.03. Data are expressed as the mean ± SEM. Differences between groups were evaluated by using Student\'s t-test and one-way unpaired analysis of variance (Tukey\'s multiple comparison test). The interaction of thrombin with β2GPI was assessed in direct binding experiments. We investigated the binding of thrombin to β2GPI by measuring the thrombin enzymatic activity on the chromogenic substrate S2238. A mean ± SD of 0.108 ± 0.007 pmoles of thrombin (P 0.0001 versus control) (n = 6 experiments) was bound when assayed with S2238 (Figure 1A). The amidolytic activity of thrombin on S2238 was first detected after 5 minutes of incubation and increased linearly with time (Figure 1B). The values for the kd and maximum binding (BMAX) of thrombin bound to β2GPI were derived using GraphPad Prism software version 3.03. The analysis incorporated the data demonstrated in the specific saturation binding experiments (Figure 2A). The kd was assessed as 0.486 × 10−6M, and the BMAX was assessed as 0.154 × 10−12 moles. In this system, the nonspecific binding was equivalent to specific binding. A, Direct binding of thrombin (FIIa) to recombinant human β2-glycoprotein I (β2GPI). Plates were coated with recombinant human β2GPI or bovine serum albumin (BSA) and blocked with 2% BSA/phosphate buffered saline–Tween 20. Thrombin was applied to the wells and incubated for 2 hours at 25°C. S2238 was added for 3 hours at 37°C, and absorbance was assessed at 405 nm. The amount of thrombin bound was derived from a standard curve. B, Cleavage rate of S2238 in 3-hour incubation with recombinant human β2GPI–bound thrombin. The amount of hydrolysis of S2238 during a 3-hour incubation with recombinant human β2GPI–bound thrombin was monitored over time (0–3 hours). Data were plotted using GraphPad Prism software, version 3.03. Solid squares represent thrombin; open squares represent BSA. Values are the mean ± SEM of triplicate points. OD = optical density. A, Saturation binding of thrombin to recombinant human β2GPI. The dissociation constant of thrombin-specific binding to recombinant human β2GPI was derived from saturation binding experiments in which thrombin at various concentrations was incubated in microtiter plates coated with recombinant human β2GPI. A 100-fold molar excess of biotin-D-Phe-Pro-Arg-chloromethylketone–blocked human α-thrombin (BFPRck–thrombin) was used to measure the nonspecific binding. Solid squares represent total binding; open squares represent nonspecific binding, solid circles represent specific binding. B, Reversibility of binding of thrombin to β2GPI. One hundred microliters of thrombin (700 nM) was incubated with recombinant human β2GPI coated on microtiter wells for 2 hours at 25°C. The wells were washed and a 50-fold molar excess of BFPRck–thrombin was added, and at defined time points, wells were washed and bound thrombin was assessed with 1.6 mM S2238. Solid squares represent BFPRck–thrombin; open squares represent BSA. Values are the mean ± SEM of triplicate points. See Figure 1 for other definitions. The reversibility of thrombin binding to β2GPI was investigated by adding a 50-fold molar excess of BFPRck–thrombin to the wells equilibrated with thrombin in a direct binding β2GPI assay. A dramatic decrease of bound thrombin was observed at the 10-minute time point, and this reached a plateau at 25 minutes (Figure 2B). Binding of thrombin to immobilized recombinant human β2GPI inhibited by BFPRck–thrombin. Binding of thrombin to recombinant human β2GPI was competitively inhibited in a dose-dependent manner by BFPRck–thrombin, with an IC50 value of 232 nM (Figure 3A). Saturation binding of BFPRck–thrombin to recombinant human β2GPI was confirmed by using streptavidin–HRP after extensive washing of the microtiter wells (Figure 3B). However, it is not clear whether inhibition of thrombin amidolytic activity by 50% is equivalent to an IC50. A, Inhibitory effect of biotin-D-Phe-Pro-Arg-chloromethylketone–blocked human α-thrombin (BFPRck–thrombin) on the binding of thrombin to recombinant human β2-glycoprotein I (β2GPI). A fixed concentration of thrombin was incubated with serial dilutions of BFPRck–thrombin in wells of microtiter plates coated with recombinant human β2GPI. The amount of thrombin bound was obtained by using substrate S2238 and determining absorbance at 405 nm. B, Confirmation of BFPRck–thrombin binding to recombinant human β2GPI. Biotinylated active site–inhibited thrombin bound in a dose-dependent manner to coated recombinant human β2GPI, as detected with streptavidin–horseradish peroxidase (see Materials and Methods). Values are the mean ± SEM of triplicate points. OD = optical density. To investigate the thrombin binding site on β2GPI, we used a number of domain deletion mutants of β2GPI. Direct binding experiments demonstrated that thrombin binds to DI–V, DII–V, DIII–V, DIV–V, and DV, but not to DI, DI–II, DI–III, DI–IV, or BSA (Figure 4), confirming that DV of β2GPI is the critical domain for binding to thrombin. Previous studies have shown that domain V is necessary for binding to phospholipids, heparin, factor XI, plasminogen, and apolipoprotein E receptor 2, indicating that this region may mediate multiple binding functions in vivo (12, 20, 22-24). Direct binding of thrombin (FIIa) to various preparations of recombinant human β2-glycoprotein I (β2GPI). Thrombin bound to recombinant human β2GPI (DI–V) and to domain deletion mutants (DII–V, DIII–V, DIV–V, DV) of recombinant human β2GPI, but not to DI, DI–II, DI–III, DI–IV, or bovine serum albumin (BSA), confirming that thrombin binds to domain V of recombinant human β2GPI. The amount of thrombin bound to immobilized β2GPI mutants or BSA was determined after the addition of S2238 and reading the absorbance at 405 nm. Values are the mean and SEM. The competitive inhibition of BFPRck–thrombin for binding of thrombin to immobilized recombinant human β2GPI demonstrated that the active site of thrombin is not involved in the interaction of thrombin with β2GPI. Subsequently, we applied a competitive inhibition assay, using 2 single-stranded DNA oligonucleotides, HD22 and HD1, to identify the possible role of exosite II and exosite I in the thrombin–β2GPI interaction. HD22 is directed against heparin binding exosite (exosite II) on human thrombin, and HD1 is directed to fibrin binding exosite (exosite I). Both HD22 and HD1 were demonstrated to competitively inhibit thrombin binding to the immobilized recombinant human β2GPI with IC50 values of 92.7 nM and 116 nM, respectively. HD4 and HD3, which are nearly identical to HD22 and HD1 but which differ in only 8 substitutions of G to A for HD4 and 6 substitutions between T and G for HD3 (which has a net negative charge equal to that of HD1), did not show any inhibitory effect on the thrombin–β2GPI interaction (Figure 5), implying that this is a specific interaction and not greatly influenced by the negative charge. Thrombin amidolytic activity on its chromogenic substrate S2238 was not influenced by native β2GPI in the presence or absence of anti-β2GPI antibodies (data not shown). A, Inhibitory effect of HD22 on thrombin interaction with recombinant human β2-glycoprotein I (β2GPI). A fixed concentration of thrombin was incubated with serial dilutions of HD22 and HD4 in the wells of microtiter plates coated with recombinant human β2GPI. The amount of thrombin bound was obtained by using substrate S2238 and determining absorbance at 405 nm. Solid squares represent HD22; open squares represent HD4. B, Inhibitory effect of HD1 on the thrombin–recombinant human β2GPI interaction. The experiment was performed as described in A with serial dilutions of HD1 and HD3. Solid squares represent HD1; open squares represent HD3. Values are the mean ± SEM. Anti-β2GPI antibodies potentiate the inhibitory effect of β2GPI on the activation of factor XI by factor IIa. The observation that β2GPI inhibited factor XI activation by thrombin (15) led us to evaluate the effect of monoclonal anti-β2GPI antibodies on this process in the presence of β2GPI. We used a panel of anti-β2GPI mAb that have been previously characterized. Beta2-glycoprotein I inhibited factor XI activation by thrombin with an IC50 of 1.4 μM. The 8 anti-β2GPI mAb alone did not cause any significant inhibition of the activation of factor XI by thrombin. The inhibitory effect of recombinant human β2GPI on factor XI activation by thrombin was increased in the presence of anti-β2GPI antibodies. This increase was 200% and was statistically significant compared with the isotype control antibodies (P 0.001) (Figure 6). Inhibition of factor XI (FXI) activation by thrombin in the presence of recombinant human β2GPI (rhβ2GPI) and anti-β2GPI antibodies. Factor XI and Dextran sulfate were incubated with recombinant human β2GPI, bovine serum albumin (BSA), or various monoclonal antibodies (Ab) alone or with a combination of β2GPI and the monoclonal antibodies, for 5 minutes at 37°C. Thrombin was then added, and the reaction was stopped with hirudin after 5 minutes. The reaction mixture was diluted 1:5, and factor XI activation was assessed using S2366 and determining absorbance at 405 nm. Values are the mean and SEM of triplicate points. AbC = antibody isotype control. Thrombin is not only a key enzyme in the coagulation pathway, but it is also an important factor in the immune system. The interaction of β2GPI with this critical protein may interfere with the wide-ranging activities in which thrombin takes part. While being generated on the surface of platelets, thrombin is engaged in a spectrum of different biologic functions in the body. Thrombin may act as a procoagulant factor by converting fibrinogen into an insoluble fibrin clot, which anchors platelets to the site of injury and initiates the process of wound healing (25). Furthermore, thrombin may function prothrombotically by interacting with protease-activated receptors and GPIb-IX-V, the process which may lead to platelet aggregation and activation (26). In an entirely opposite function, thrombin may activate protein C on the membrane of endothelial cells via thrombomodulin, and act as an anticoagulant (27). Additional regulation is achieved when thrombin is irreversibly inhibited by the stoichiometric serine protease inhibitors (serpins) antithrombin III and heparin cofactor II (28). On the other hand, thrombin may play a significant role in the immune response. It can promote tumor growth and metastasis, angiogenesis, atherosclerosis, and inflammation (29). It has been noted previously that thrombin interaction with its macromolecular substrates, inhibitors, and effectors via either of its 2 electropositive sites, exosites I and II, may direct thrombin to the relevant substrate and allosterically regulate its activities. In the current study, we demonstrate that β2GPI is able to bind exosites I and II on thrombin and that domain V is the crucial domain of β2GPI for this interaction. Independent binding of both exosites I and II was previously shown for platelet GPIbα (30) and thrombomodulin (31, 32). The physiologic significance of β2GPI binding to thrombin is yet to be identified. Beta2-glycoprotein I may potentially play an as yet to be determined modulatory role in the extensive biologic functions of thrombin. This might happen by blockade of thrombin exosites or by inhibiting its interactions with substrates and ligands. The abundant β2GPI molecule potentially binds factors XI and XIa and mediates a regulatory role in factor XI activation, both via factor XIIa and thrombin as discussed initially in the introduction (12, 15). The activation of factor XI by thrombin represents an important amplification loop, leading to additional thrombin generation on the surface of platelets (33), where thrombin and factor XI bind to GPIb-IX-V. This binding allows factor XI and thrombin to colocalize and thus leads to efficient factor XIa generation (34, 35). Our group has recently demonstrated that β2GPI binds with high affinity to the GPIbα subunit of the GPIb-IX-V receptor via domains II, III, V, and possibly IV (36). Pennings et al, using a different experimental design, have demonstrated that artificially constructed dimers of β2GPI immobilized on a plate are able to support platelet adhesion via GPIbα under conditions of flow (37). In the context of the aforementioned studies, we speculate that β2GPI may colocalize with factor XI, thrombin, and GPIbα in a quaternary complex to regulate factor XI activation by thrombin. GPIbα binds thrombin via exosite II and provides a proper orientation of the thrombin molecule to bind factor XI. Thrombin in turn binds apple 1 on factor XI via exosite I and activates factor XI by cleaving the scissile bond at Arg369–Ile370 (38). In the suggested quaternary complex, β2GPI may bind exosite I and exosite II on thrombin and interfere with the aforementioned interactions. This disturbance may lead to inhibition of factor XI activation by thrombin. Anti-human thrombin antibodies are not frequent and are usually observed with bleeding disorders (39, 40). However, it has recently been reported that in a cohort of 120 patients with APS, 43% displayed elevated levels of anti-human thrombin antibodies (17). Furthermore, it was noted that the presence of anti-human thrombin antibodies was closely related to the presence of anti-β2GPI antibodies (96%) (17). The pertinent point to note is that in that study an enzyme-linked immunosorbent assay (ELISA) was used in which thrombin was coated on a microtiter plate, followed by the addition of the patients\' sera (17). In view of our finding that β2GPI can bind to thrombin, we suggest that it is feasible that β2GPI present in the patients\' sera may have bound to thrombin coated on the microtiter plate, thereby contaminating the assay and making interpretation impossible regarding the presence or absence of anti-human thrombin antibodies in this cohort of patients. A similar case can be made for other studies which have reported detecting anti-human thrombin antibodies in patients with APS using a similar ELISA-based method (18). Although multiple mechanisms have been reported, the pathophysiologic mechanism by which anti-β2GPI antibodies exert their adverse effects is still open to debate. To delineate the effect of anti-β2GPI antibodies on β2GPI inhibition of factor XI activation by thrombin, we undertook in vitro studies using anti-β2GPI antibodies directed against domain I and examined their effect on factor XI activation by thrombin in the presence of β2GPI. The rationale for utilizing antibodies against domain I of β2GPI was that the β2GPI binding site for antibodies appears to be quite distinct from the β2GPI binding site for factor XI, anionic phospholipids, GPIbα, and thrombin. Anti-β2GPI antibodies in APS are directed to a restricted epitope spanning amino acids 40–43 on domain I (19, 41, 42). As a point of contrast, anti-β2GPI antibodies not associated with a prothrombotic tendency, such as those detected in association with atopic dermatitis in children (43) and in patients with leprosy (44), tend to be directed against domain V. The current study demonstrates that the addition of anti-β2GPI mAb to a system containing factor XI, intact β2GPI, and thrombin produces a mean ± SEM 74.4 ± 2.6% inhibition of thrombin-mediated factor XI activation, which is greater than when β2GPI alone acts in this system (mean ± SEM 45.4 ± 1.7%). Anti-β2GPI antibodies from patients with APS bind with lower affinity to β2GPI compared with murine mAb used in this study (45, 46). Further experiments using autoantibodies from APS patients are required to determine the significance of these findings in these individuals. The functional effect of anti-β2GPI antibodies demonstrated in this study is characterized by a decreased generation of thrombin. This represents a paradox analogous to the lupus anticoagulant (LAC) effect, in that clinically this effect would be expected to lead to a bleeding diathesis. It is unlikely that this interaction explains the prothrombotic tendency in APS; however, it may contribute to the disturbance of primary hemostasis and prolonged bleeding time demonstrated in APS patients with LAC positivity (47). In conclusion, the current study shows that β2GPI binds thrombin via exosite I and exosite II. The physiologic significance of this finding waits to be determined. This discovery necessitates a reevaluation of the methodology used in previous studies to establish the presence of anti-human thrombin antibodies in patients with APS. We also demonstrate that anti-β2GPI antibodies in the presence of β2GPI may promote inhibition of factor XI activation, potentially interfering with the coordinated generation of thrombin. Dr. Krilis had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study design. Dr. Krilis. Acquisition of data. Drs. Rahgozar, Yang, and Yan. Analysis and interpretation of data. Drs. Rahgozar, Giannakopoulos, Miyakis, and Krilis. Manuscript preparation. Drs. Rahgozar, Yang, Giannakopoulos, Miyakis, and Krilis. Statistical analysis. Dr. Rahgozar. Immunology and clinical importance of antiphospholipid antibodies. In: FJ Dixon, editor. Advances in immunology. San Diego: Academic Press;1991. p. 193– 280.Cervera R, et al. International consensus statement on an update of the classification criteria for definite antiphospholipid syndrome (APS).J Thromb Haemost 2006; 295– 306. Wiley Online Library Euro-Phospholipid Project Group. Antiphospholipid syndrome: clinical and immunologic manifestations and patterns of disease expression in a cohort of 1,000 patients.Arthritis Rheum 2002; 1019– 27. Wiley Online LibraryKrilis SA. Anti-phospholipid antibodies are directed against a complex antigen that includes a lipid-binding inhibitor of coagulation: β2-glycoprotein I (apolipoprotein H).Proc Natl Acad Sci U S A 1990; 4120– 4.van Breda-Vriesman PJ, et al. Anticardiolipin antibodies (ACA) directed not to cardiolipin but to a plasma protein cofactor.Lancet 1990; 335: 1544– 7.Hoylaerts MF, et al. Thrombogenicity of β2-glycoprotein I-dependent antiphospholipid antibodies in a photochemically induced thrombosis model in the hamster.Blood 2003; 101: 157– 62.Bulla R, et al. Thrombus formation induced by antibodies to β2-glycoprotein I is complement dependent and requires a priming factor.Blood 2005; 106: 2340– 6.Krilis SA. Identification of a region of β2-glycoprotein I critical for lipid binding and anti-cardiolipin antibody cofactor activity.Proc Natl Acad Sci U S A 1993; 2141– 5.Krilis SA. Domain V of β 2 glycoprotein I binds factor XI/XIa and is cleaved at Lys317-Thr318.J Biol Chem 2005; 280: 907– 12.Kato H. Plasmin can reduce the function of human β2 glycoprotein I by cleaving domain V into a nicked form.Blood 1998; 4173– 9. Schousboe I. β2-Glycoprotein I: a plasma inhibitor of the contact activation of the intrinsic blood coagulation pathway.Blood 1985; 1086– 91.Konecny P, et al. β2-Glycoprotein I binds factor XI and inhibits its activation by thrombin and factor XIIa: loss of inhibition by clipped β2-glycoprotein I.Proc Natl Acad Sci U S A 2004; 101: 3939– 44.Scharrer I. Identification of thrombin antibodies in patients with antiphospholipid syndrome.Ann N Y Acad Sci 2005; 1050: 250– 6. Wiley Online LibraryLe DT, et al. Identification of anti-thrombin antibodies in the antiphospholipid syndrome that interfere with the inactivation of thrombin by antithrombin.J Immunol 2001; 167: 7192– 8.Krilis SA. Epitope studies with anti-β2-glycoprotein I antibodies from autoantibody and immunized sources.J Autoimmun 2000; 91– 6.Krilis SA. Heparin inhibits the binding of β2-glycoprotein I to phospholipids and promotes the plasmin-mediated inactivation of this blood protein.J Biol Chem 2002; 277: 2644– 9.Koike T, et al. Detection of ‘antiphospholipid’ antibodies: a single chromogenic assay of thrombin generation sensitively detects lupus anticoagulants, anticardiolipin antibodies, plus antibodies binding β2-glycoprotein I and prothrombin.Clin Exp Immunol 2001; 124: 502– 8. Wiley Online LibraryKrilis SA. Site-directed mutagenesis of recombinant human β2-glycoprotein I identifies a cluster of lysine residues that are critical for phospholipid binding and anti-cardiolipin antibody activity.J Immunol 1996; 157: 3744– 51.de Groot PG. Dimers of β2-glycoprotein I increase platelet deposition to collagen via interaction with phospholipids and the apolipoprotein E receptor 2′.J Biol Chem 2003; 278: 33831– 8.Inagaki J, et al. Nicked β2-glycoprotein I: a marker of cerebral infarct and a novel role in the negative feedback pathway of extrinsic fibrinolysis.Blood 2004; 103: 3766– 72.Varughese KI, et al. Modulation of α-thrombin function by distinct interactions with platelet glycoprotein Ibα.Science 2003; 301: 218– 21.Johnson AE. The fifth and sixth growth factor-like domains of thrombomodulin bind to the anion-binding exosite of thrombin and alter its specificity.J Biol Chem 1992; 267: 11023– 8.Esmon CT. Occupancy of anion binding exosite 2 on thrombin determines Ca2+ dependence of protein C activation.J Biol Chem 1994; 269: 11807– 12.Walsh PN. Factor XI binding to the platelet glycoprotein Ib-IX-V complex promotes factor XI activation by thrombin.J Biol Chem 2002; 277: 1662– 8.Lopez JA, et al. Factor XI interacts with the leucine-rich repeats of glycoprotein Ibα on the activated platelet.J Biol Chem 2004; 279: 49323– 9.Andrews RC, et al. Anti-β2-glycoprotein I antibodies in complex with β2-glycoprotein I can activate platelets in a dysregulated manner via glycoprotein Ib-IX-V.Arthritis Rheum 2006; 2558– 67. Wiley Online Libraryde Groot PG. Platelet adhesion to dimeric β2-glycoprotein-I under conditions of flow is mediated by glycoprotein Ibα [abstract].J Thromb Haemost 2005; Suppl 1:Abstract OR256.Walsh PN, et al. Thrombin activation of factor XI on activated platelets requires the interaction of factor XI and platelet glycoprotein Ib α with thrombin anion-binding exosites I and II, respectively.J Biol Chem 2003; 278: 48112– 9.Corash L. Development of antithrombin antibodies following surgery in patients with prosthetic cardiac valves.Blood 1988; 1375– 80.Leung LL. Development of antibodies to thrombin and factor V with recurrent bleeding in a patient exposed to topical bovine thrombin.Blood 1990; 2011– 6.de Groot PG. IgG antibodies that recognize epitope Gly40-Arg43 in domain I of β2-glycoprotein I cause LAC, and their presence correlates strongly with thrombosis.Blood 2005; 105: 1540– 5.Marti-Renom MA, et al. Use of single point mutations in domain I of β2-glycoprotein I to determine fine antigenic specificity of antiphospholipid autoantibodies.J Immunol 2002; 169: 7097– 103.Hojnik M, et al. Anti-β2-glycoprotein I antibodies in children with atopic dermatitis.Int Immunol 2002; 823– 30.Youinou P. Distinguishing features of anti-β2 glycoprotein I antibodies between patients with leprosy and the antiphospholipid syndrome.Thromb Haemost 2002; 599– 605.Krilis SA. Anti-β 2-glycoprotein I autoantibodies from patients with the \"antiphospholipid” syndrome bind to β2-glycoprotein I with low affinity: dimerization of β2-glycoprotein I induces a significant increase in anti-β2-glycoprotein I antibody affinity.J Immunol 1998; 161: 2038– 43.Derksen RH. Prolonged bleeding time and lupus anticoagulant: a second paradox in the antiphospholipid syndrome.Arthritis Rheum 2004; 3605– 9. Wiley Online Library The full text of this article hosted at iucr.org is unavailable due to technical difficulties. Please check your email for instructions on resetting your password. If you do not receive an email within 10 minutes, your email address may not be registered, and you may need to create a new Wiley Online Library account. Can\'t sign in? Forgot your username? Enter your email address below and we will send you your username If the address matches an existing account you will receive an email with instructions to retrieve your username