Molecular studies on phospholipid-binding sites and cryptic epitopes appearing on beta 2-glycoprotein I structure recognized by anticardiolipin antibodies

Epitope of antiphospholipid antibodies retrieved from peptide microarray based on R39-R43 of β2-glycoprotein I

Background

Antiphospholipid antibody (aPL) syndrome (APS) is an autoimmune disease characterized by the presence of antiphospholipid antibodies and thromboembolic or pregnancy complications. Although cryptic epitope R39-R43 belonging to beta-2-glycoprotein 1 (β2GP1) has been identified as the main antigenic determinant for aPLs, we have recently demonstrated that the epitope is a motif determined by the polarity, rather than by the sequence or charge of amino acids.

Objective

In the present study, we wanted to identify the association of residues needed to obtain the highest aPL affinity.

Methods

Based on the epitope R39-R43 and our identified motif, we generated a printed peptide microarray of 676 different peptides. These peptides have been then screened for their ability to interact with the plasmas from 11 well-characterized APS patients and confirmed by surface plasma resonance assay.

Results and Conclusions

We identified a peptide that selectively bound immunoglobulin G (IgG) derived from APS patients with 100 times more affinity than β2GP1, Domain I, or epitope R39-R43. This peptide is able to inhibit the activity of IgG derived from APS patients in vitro. We have also generated a monoclonal IgG antibody against this peptide. Using both peptide and monoclonal antibody, we have been able to develop a fully standardized indirect colorimetric immunoassay with highly sensitivity. The identification of the optimized peptide offers a new standardized and accurate tool for diagnostics of APS. Furthermore, having increased affinity for aPL, this peptide could represent a useful tool as prevention strategy for APS and an alternative to the use of anticoagulants.

Cellular immune response to β2-glycoprotein-I valine/leucine247 phenotypes in Mexican patients with primary antiphospholipid syndrome

Homozygote genotype V²⁴⁷ of the β₂-glycoprotein-I (β₂GP-I) gene has been associated with anti-β₂GP-I and thrombosis in patients with primary anti-phospholipid syndrome APS (PAPS). However, the cellular immune response to β₂GP-I²⁴⁷ has been little studied.

OBJECTIVE

To evaluate the immune cellular proliferation in response to native and non-native β₂GP-I²⁴⁷ valine/leucine phenotype from Mexican patients with PAPS.

METHODS

We studied 10 patients with PAPS and 10 healthy control subjects (HC). The polymorphism at position 247 of the β₂GP-I gene was determined by PCR-RFLP and the corresponding β₂GP-I protein was subsequently purified from normal human plasma by affinity chromatography. PBMC purified from patients and controls were stimulated with β₂GP-I under native and in non native (reduced) conditions. We also determined the anti-β₂GP-I production in vitro by B cell clones (EBV) generated in cocultures experiments. Differential Scanning Calorimetry (DSC) was studied to determine the structural differences between the β₂GP-I²⁴⁷ valine/leucine isoforms. Cytokine profile (IL-2, IL-4, IL-6, TNFα, INFγ) was evaluated in culture supernatants.

RESULTS

PAPS and healthy control PBMCs had a higher proliferative response when stimulated with β₂GP-I under reduced cultures conditions compared to non-denatured conditions. PBMCs response from PAPS patients was higher. We observed more cell proliferation in response to β₂GP-I²⁴⁷ valine/leucine or valine isoforms in non-native conditions. In contrast, this response was not significant against β₂GP-I²⁴⁷ leucine. These findings were T CD4⁺-dependent. Similar results were obtained with B cell clones derived from PAPS patients, which showed more pronounced proliferation in non native conditions and higher against β₂GP-I²⁴⁷ valine. No differences were found in anti-β₂GP-I production, but high levels of IL-6 in vitro were identified. The structural analysis of both β₂GP-I²⁴⁷ isoforms by DSC showed a major conformational change due to a single mutation in the β₂GP-I variants.

CONCLUSIONS

PAPS PBMCs had a higher cellular response against β₂GP-I²⁴⁷ in non-native culture conditions preferentially to the β₂GP-I²⁴⁷ valine phenotype. This effect is T CD4⁺ dependent and appears to be driven by tertiary structural changes adopted by β₂GP-I²⁴⁷ polymorphism.

Val247Leu β2-glycoprotein-I allelic variant is associated with antiphospholipid syndrome: systematic review and meta-analysis

INTRODUCTION

Previous studies have suggested that the possession of the Val/Val genotype of the Val247Leu polymorphism of the β(2)-glycoproteinI (β(2)-GPI) gene may be associated with antiphospholipid syndrome (APS), and, among patients with APS, with the production of anti-β(2)-GPI antibodies or the development of thrombosis. Given the controversial results reported, the aim of this work is to combine previous findings by means of a systematic review and a meta-analysis.

METHODS

We retrieved studies analyzing the genotype of the above-mentioned polymorphism among patients with APS by means of electronic database search. A meta-analysis was conducted in a random effects model and calculations of odds ratio (OR) and confidence intervals (CI) were done. Sensitivity analysis and tests for heterogeneity of the results were performed.

RESULTS

Eight previous studies analyzed the association of APS, anti-β(2)-GPI antibodies and/or thrombosis with the Val247Leu polymorphism. After meta-analysis, patients with APS had a significantly higher prevalence of the Val/Val genotype of this genetic variant when compared with controls (OR=2.04; 95% CI: 1.12, 3.73; P=0.02). Among patients with APS, those with anti-β(2)-GPI antibodies had a higher prevalence of this genotype (OR=1.73; 95% CI: 1.04, 2.87; P=0.03). No significant results were found for the presence of arterial or venous thrombosis.

CONCLUSIONS

Val/Val genotype of β(2)-GPI gene is associated with a significant excess risk to suffer from APS and, among patients with APS, to have anti-β(2)-GPI antibodies. No definite conclusions can be made regarding the association of this polymorphism with thrombosis among APS patients.

Endothelium activation in the anti-phospholipid syndrome

Anti-phospholipid syndrome is an autoimmune systemic disease characterized by the persistent presence of anti-phospholipid antibodies and by the occurrence of thrombosis, fetal loss and thrombocytopenia. Anti-phospholipid antibodies are widely accepted as pathogenic antibodies mainly directed against the phospholipid-binding protein beta 2 glycoprotein I. Beta 2 glycoprotein I can be expressed on the endothelial cell membranes of different anatomical localizations and recognized by the autoantibodies. The antibody binding might induce an endothelial activation both in vitro and in vivo experimental models, that was suggested to represent one of the pathogenic mechanisms leading to the prothrombotic state of the syndrome. Beta 2 glycoprotein I endothelial adhesion was found to take place through the interaction of the cationic phospholipid binding site of the molecule with anionic endothelial structures and through annexin II, the endothelial cell receptor for tissue plasminogen activator. Anti-beta 2 glycoprotein I antibodies can directly activate the cells via NF-kB translocation and the signaling cascade triggered by toll like receptors. It has been suggested that beta 2 glycoprotein I might be associated with toll like receptors because of its molecular mimicry with bacterial structures, the natural ligands of toll like receptors. The binding of the antibodies is thought to cross-link beta 2 glycoprotein I and the toll like receptors, eventually switching their signaling pathway.

A systematic analysis of sequences of human antiphospholipid and anti-beta2-glycoprotein I antibodies: the importance of somatic mutations and certain sequence motifs

OBJECTIVE

Previous studies have suggested the importance of somatic mutations and certain residues in the complementarity determining regions (CDRs) of antiphospholipid antibodies (aPL) implicated in the pathogenesis of antiphospholipid antibody syndrome (APS). The authors tested this hypothesis by carrying out a systematic analysis of all published aPL sequences.

METHODS

Each aPL variable region sequence was aligned to the closest germline counterpart in the VBASE Sequence Directory by using DNAPLOT software, allowing analysis of nucleotide homology and distribution of somatic mutations. The probability that this distribution arose as a result of antigen-driven accumulation of replacement mutations in the CDRs was tested statistically.

RESULTS

There was no preferential gene or family use in the 36 aPL sequences identified. Immunoglobulin (Ig) M aPL had few somatic mutations compared with IgG. Of the IgG aPL, 9 of 14 showed evidence of antigen-driven accumulation of replacement mutations in the CDRs. Multinomial analysis allowed a clearer statistical identification of sequences that had been subject to antigen drive. The more specific IgM aPL and some IgG aPL displayed an accumulation of arginine, asparagine, and lysine residues in CDRs.

CONCLUSIONS

High-specificity binding in IgG aPL, but not in more specific IgM aPL, is conferred by antigen-driven somatic mutation. This may in part be caused by an accumulation of arginine, asparagine, and lysine residues in the CDRs, which are germlines encoded in the more specific IgM aPL, but often arise because of somatic mutation in IgG aPL.

RELEVANCE

An understanding of the role of arginine, asparagine, and lysine residues in the binding of pathogenic aPL to phospholipids, and to beta(2)-glycoprotein I, may eventually help in the development of drugs to interfere with those interactions, and thereby improve the treatment of antiphospholipid antibody syndrome.

Valine/valine genotype at position 247 of the beta2-glycoprotein I gene in Mexican patients with primary antiphospholipid syndrome: association with anti-beta2-glycoprotein I antibodies

OBJECTIVE

To determine the polymorphism at position 247 of the beta(2)-glycoprotein I (beta(2)GPI) gene in Mexican patients with antiphospholipid syndrome (APS) and to compare these data in patients with or without antibodies to beta(2)GPI and with the clinical manifestations of APS.

METHODS

We studied 39 patients with primary APS and compared them with 106 clinically healthy subjects. Polymorphism was determined by polymerase chain reaction-restriction fragment length polymorphism. The presence of "true" anticardiolipin (aCL) antibodies, beta(2)GPI-dependent aCL antibodies (IgG and IgM), and phospholipid-free anti-beta(2)GPI antibodies (IgG isotype) were detected by enzyme-linked immunosorbent assay (ELISA) utilizing nonirradiated ELISA plates. Clinical manifestations associated with antiphospholipid antibodies were also evaluated.

RESULTS

We found no significant differences in the genotype expression between the control group and the primary APS patients (13% with VV, 52% with VL, and 35% with LL versus 23% with VV, 51% with VL, and 26% with LL, respectively). In contrast, anti-beta(2)GPI-positive patients had significantly higher frequencies of the VV genotype and V allele expression than the control subjects and the anti-beta(2)GPI-negative patients. These genotype and allele frequencies were also significantly higher in patients with arterial thrombosis than in patients without it. Anti-beta(2)GPI-negative patients without arterial thrombosis did not express the VV genotype. We found no differences in the Val/Leu(247) polymorphism of the beta(2)GPI gene in primary APS patients with or without "true" aCL antibodies or in primary APS patients with or without beta(2)GPI-dependent aCL antibodies.

CONCLUSION

Our results suggest that the VV genotype at position 247 of the beta(2)GPI gene may play a role in the generation of anti-beta(2)GPI antibodies and perhaps in the expression of arterial thrombosis in primary APS.

Antiprothrombin antibodies are associated with pregnancy loss in patients with the antiphospholipid syndrome

OBJECTIVE

To document the clinical association between the history of pregnancy loss in patients with the diagnosis of primary or secondary antiphospholipid syndrome (APS) and the presence of different antiprothrombin antibody subtypes [immunoglobulin G (IgG), IgM and IgA] in a cohort of patients with APS.

METHODS

Records of 170 female patients with primary APS, or APS secondary to systemic lupus erythematosus (SLE) or secondary to other autoimmune diseases were studied.

RESULTS

In female APS patients with IgG antiprothrombin antibodies (n = 105) significant associations to pregnancy loss (p < 0.0001), early pregnancy loss (p < 0.0001) and a negative association to thrombocytopenia (p < 0.01) could be identified. In the group of patients with IgG antiprothrombin antibodies and at least one pregnancy (n = 84) a significant association with pregnancy loss (p < 0.005) and especially with early pregnancy loss (p < 0.0001) was demonstrated. No association with other immunoglobulin subtypes of antiprothrombin antibodies could be documented. In the subgroup of patients with primary APS and at least one pregnancy in the history, pregnancy loss (p < 0.005) and early pregnancy loss (p < 0.0001) were found to be highly associated with the presence of IgG antiprothrombin antibodies. IgG antiprothrombin antibodies represent the highest independent risk factor for pregnancy loss with an odds ratio of 4.5. There was no statistically significant association with venous or arterial thrombosis in all IgG antiprothrombin antibody positive patients.

CONCLUSION

The results of this study document the association of IgG antiprothrombin antibodies with pregnancy loss and in particular early pregnancy loss in a large and well-characterized cohort of patients. We would recommend routine testing for antiprothrombin antibodies in young female patients with APS.

Phospholipid autoantibodies: time for a new immuno-assay?

Antiphospholipid (aPL) antibodies are a wide and heterogeneous group of closely related immunoglobulins that have specificity for a number of phospholipids (PLs), PL-binding proteins, including beta2 glycoprotein I (beta2GP1), prothrombin, and PL-protein complexes. Antiphospholipid antibodies are currently detected by a standard solid-phase immunoassay such as anticardiolipin (aCL) antibodies. Although the association between aCL antibodies and thrombosis is well recognized, their role in stroke pathogenesis, and the possibility that they are an independent stroke risk factors in the general stroke population, remains to be determined. There are other negatively charged PL, largely represented within the cellular membrane and involved in the coagulation. Different studies have demonstrated a positive correlation among other aPL and stroke.

Autoreactive CD4(+) T-cell clones to beta2-glycoprotein I in patients with antiphospholipid syndrome: preferential recognition of the major phospholipid-binding site

Autoreactive CD4(+) T cells to beta2-glycoprotein I (beta2GPI) that promote antiphospholipid antibody production were recently identified in patients with antiphospholipid syndrome (APS). To further examine antigen recognition profiles and T-cell helper activity in beta2GPI-reactive T cells, 14 CD4(+) T-cell clones specific to beta2GPI were generated from 3 patients with APS by repeated stimulation of peripheral blood T cells with recombinant beta2GPI. At least 4 distinct T-cell epitopes were identified, but the majority of the beta2GPI-specific T-cell clones responded to a peptide encompassing amino acid residues 276 to 290 of beta2GPI (KVSFFCKNKEKKCSY; single-letter amino acid codes) that contains the major phospholipid-binding site in the context of the DRB4*0103 allele. Ten of 12 beta2GPI-specific T-cell clones were able to stimulate autologous peripheral blood B cells to promote anti-beta2GPI antibody production in the presence of recombinant beta2GPI. T-cell helper activity was exclusively found in T-cell clones capable of producing interleukin 6 (IL-6). In vitro anti-beta2GPI antibody production induced by T-cell clones was inhibited by anti-IL-6 or anti-CD40 ligand monoclonal antibody. In addition, exogenous IL-6 augmented anti-beta2GPI antibody production in cultures of the T-cell clone lacking IL-6 expression. These results indicate that beta2GPI-specific CD4(+) T cells in patients with APS preferentially recognize the antigenic peptide containing the major phospholipid-binding site and have the capacity to stimulate B cells to produce anti-beta2GPI antibodies through IL-6 expression and CD40-CD40 ligand engagement. These findings are potentially useful for clarifying the pathogenesis of APS and for developing therapeutic strategies that suppress pathogenic antiphospholipid antibody production in these patients.

Genetics of antiphospholipid syndrome

The mechanisms of thrombosis in antiphospholipid syndrome (APS) are highly heterogeneous and multifactorial, and some genetic factors may be involved in its pathophysiology. The genetic variants of representative antigen, beta 2-glycoprotein I (beta 2GPI), have been known, and valine/leucine247 polymorphism is a genetic risk for having anti beta 2GPI antibodies and APS. Congenital beta 2GPI deficiency did not correlate with thrombophilia, thus its responsible gene (beta 2GPI-Sapporo) was not a risk for thrombosis. Many other thrombosis-related genetic factors have been investigated in APS, but no additional risk for thrombosis has been indicated in patients with antiphospholipid antibodies.

Proteolytic cleavage of beta(2)-glycoprotein I: reduction of antigenicity and the structural relationship

Binding of beta(2)-glycoprotein I (beta(2)-GPI)-dependent anticardiolipin antibodies (aCL) derived from antiphospholipid syndrome (APS) is significantly reduced in aCL ELISA due to loss of the phospholipid (PL) binding property of beta(2)-GPI by plasmin treatment. In the present study, the treatment generated a nicked form of beta(2)-GPI and resulted in loss of antigenicity for the autoantibodies detected in ELISA, using an beta(2)-GPI directly adsorbed polyoxygenated carboxylated plate, the assay system of which was not related to PL binding. The nicked form bound to neither Cu(2+)-oxidized low-density lipoprotein (oxLDL) nor to beta(2)-GPI-specific lipid ligands isolated from oxLDL, the result being a complete loss of subsequent binding of anti-beta(2)-GPI autoantibodies. The conformational change in the nicked domain V was predicted from its intact structure determined by an X-ray analysis (implemented in Protein Data Bank: 1C1Z), molecular modeling and epitope mapping of a monoclonal anti-beta(2)-GPI antibody, i.e. Cof-18, which recognizes the related structure. The analysis revealed that novel hydrophobic and electrostatic interactions appeared in domain V after the cleavage, thereby affecting the PL binding of beta(2)-GPI. Such a conformational change may have important implications for exposure of cryptic epitopes located in the domains such as domain IV.

Antiphospholipid antibodies and platelets

The most distinguishing serologic feature of antiphospholipid syndrome (APS) is the moderate to high blood titers of antiphospholipid-binding antibodies (aPL). The pathogenic mechanisms of APS are poorly understood, but may occur as a result of the interaction between anticardiolipin antibodies (aCL), beta-2 glycoprotein-I (beta(2)GP-I) (the aCL cofactor) and blood platelets. However, the relationship between aCL/beta(2)GP-I complexes and platelet aggregation has yet to be clearly elucidated. This article will briefly review aPL, beta(2)GP-I and platelet physiology with respect to recent hypotheses relating aCL/beta(2)GP-I complexes and platelets.

Induction of antiphospholipid antibodies by immunization with synthetic viral and bacterial peptides

We previously induced pathogenic antibodies against anionic phospholipids (PL) in experimental animals by immunization with lipid-free purified human beta2glycoprotein I (beta2GPI). We hypothesized that antiphospholipid antibodies (aPL) are induced by in vivo binding of foreign beta2GPI to self-PL, thus forming an immunogenic complex against which aPL antibodies are produced. If this hypothesis is true, other PL-binding proteins that are products of ubiquitous viral/bacterial agents may also induce aPL. To test this hypothesis, groups of NIH/Swiss mice were immunized with synthetic peptides of viral and bacterial origin that share structural similarity with the putative PL-binding region of beta2GPI. Compared with the control groups, animals immunized with the peptides produced significantly higher levels of aPL and anti-beta2GPI antibodies. These findings demonstrate that some PL-binding viral and bacterial proteins function like beta2GPI in inducing aPL and anti-beta2GPI production, and are consistent with a role for such viral and bacterial proteins in inducing aPL antibody production in humans.

GDKV-Induced Antiphospholipid Antibodies Enhance Thrombosis and Activate Endothelial Cells In Vivo and In Vitro

Antiphospholipid (aPL) Abs are associated with thrombosis, pregnancy loss, and thrombocytopenia in patients with systemic lupus erythematosus or primary antiphospholipid syndrome (APS). β2-Glycoprotein I (β2GPI), a phospholipid-binding serum protein, is involved in aPL binding to phospholipids. aPL can be generated in mice by immunization with β2GPI, and these Abs are thrombogenic and cause pregnancy loss in mice. The objective of this study is to determine whether aPL induced by immunization with the phospholipid-binding site of β2GPI are thrombogenic and whether they activate endothelial cells (EC) in vivo and in vitro. Murine monoclonal aPL were generated from spleen cells of a mouse immunized with GDKV, a synthetic 15-aa peptide spanning Gly274–Cys288 in the fifth domain of human β2GPI, which represents the phospholipid-binding site of β2GPI. The Abs generated had aPL and anti-β2GPI activities. The effect of these Abs on thrombus formation and on EC activation in vivo was determined using a mouse model of thrombosis and microcirculation that enables examination of the adhesion of leukocyte to EC as an indication of EC activation as well as adhesion molecule expression using in vitro ELISA analysis. Mice injected with this monoclonal aPL showed a significant increase in leukocyte sticking and also produced larger thrombi that persisted longer. Exposure to GDKV-induced aPL for 4 h significantly increased surface Ag expression of E-selectin, ICAM-1, and VCAM-1. These data indicate that aPL induced by immunization with the phospholipid binding site of β2GPI are thrombogenic and activate endothelial cells.

Correlation between beta2-glycoprotein I valine/leucine247 polymorphism and anti-beta2-glycoprotein I antibodies in patients with primary antiphospholipid syndrome

OBJECTIVES

Beta2-Glycoprotein I (beta2GPI) exon 7 polymorphism leads to a valine leucine amino acid exchange at position 247 in domain 5 of beta2GPI, between the phospholipid binding site and the cryptic site of the epitopes for anti-beta2GPI antibodies. Therefore, position 247 polymorphism may affect the conformational change of beta2GPI and the exposure of the epitopes for anticardiolipin antibodies (aCL) (= anti-beta2GPI antibodies). In this study we analysed the genetic polymorphism of beta2GPI in a British cohort of well-defined antiphospholipid syndrome (APS) patients.

METHODS

This study comprised 88 Caucasoid patients with APS [57 with primary APS and 31 with APS secondary to systemic lupus erythematosus (SLE)]. Polymorphism assignment was determined by polymerase chain reaction followed by allele-specific restriction enzyme digestion (PCR-RFLP). The presence of anti-beta2GPI antibodies was detected by ELISA utilizing irradiated ELISA plates.

RESULTS AND CONCLUSIONS

Anti-beta2GPI antibodies were present in 28 of 57 primary APS patients (49%) and in 19 of 31 secondary APS patients (61%). The allele containing valine247 was significantly more frequent in primary APS patients with anti-beta2GPI antibodies than in controls (OR = 2.51, 95%, CI 1.03-6.13, P = 0.0396) or in primary APS patients without anti-beta2GPI antibodies (OR = 2.92, 95% CI 1.16-7.39, P = 0.0204). This tendency was not found in the secondary APS group. In conclusion, the beta2GPI polymorphism, valine/leucine247, is correlated with anti-beta2GPI antibody production in patients with primary APS, and valine247 may be important in the formation of beta2GPI antigenicity.

Beta2-glycoprotein I as a 'cofactor' for anti-phospholipid reactivity with endothelial cells

Beta2-glycoprotein I (beta2GPI) is a cofactor for anti-phospholipid (aPL) binding to cardiolipin (CL)-coated plates. Beta2GPI is also able to bind to endothelial cell (EC) membranes as supported by in-vivo as well as by in-vitro studies. The PL-binding site in the fifth domain of the molecule is involved in the adhesion to endothelium. Actually, specific mutations in this molecular portion abolish endothelium binding and a synthetic peptide spanning the sequence Glu274-Cys288 of the CL-binding site displays comparable adhesion to EC monolayers. Heparan sulphate appears to be one of the anionic EC membrane structures with which cationic beta2GPI interacts, as supported by studies with heparitinase-treated EC. Beta2GPI binding to EC might be related to its activity as endothelial growth factor or as a lipid-carrying glycoprotein. Adhesion of beta2GPI to endothelial membranes offers suitable epitopes for circulating aPL that, once bound, can induce cell activation

Epitopes on beta2-GPI recognized by anticardiolipin antibodies

Anticardiolipin antibodies (aCL) found in sera from patients with antiphospholipid syndrome recognize a cryptic epitope that appears on the beta2-glycoprotein I (beta2-GPI) molecule when beta2-GPI interacts with a lipid membrane composed of negatively charged phospholipid or when beta2-GPI is adsorbed on a polyoxygenated polystyrene plate. A homology based model of beta2-GPI was constructed based on the NMR coordinates of sushi domains of human factor H. The conformation was like a cylinder consisting of five domains, its IV and V domains being glued by electrostatic interaction. We used phage-displayed random peptide libraries to search the epitopes of human aCL. Structures similar to consensus sequences selected by a biopanning method was found on domain IV of beta2-GPI.

Antiphospholipid antibody syndrome: new insights on thrombogenic mechanisms

The antiphospholipid antibody syndrome is a thrombophilic condition manifested by vascular thrombosis or recurrent pregnancy loss together with the presence of antibodies against anionic phospholipid protein complexes. These antibodies are detected by their reactivity to the anionic phospholipids (or protein phospholipid complexes) in solid-phase immunoassays or by their property of inhibiting phospholipid-dependent coagulation reactions (the "lupus anticoagulant" effect). The pathophysiologic mechanisms of this syndrome have remained obscure because of the apparent multiplicity of antigenic determinants recognized by the antibodies and also because of the many effects which have been described for them. This article reviews current concepts of the antiphospholipid disease process and evidence for the hypothesis that thrombosis in this syndrome is a result of the displacement of annexin-V, an anionic phospholipid-binding protein with potent anticoagulant activity, from phospholipid surfaces. The authors propose that under physiologic conditions, annexin-V plays a thromboregulatory role at the vascular-blood interface by shielding anionic phospholipids from complexation with coagulation proteins in circulating blood. Thrombosis in the antiphospholipid syndrome is due to disruption of the annexin shield by antiphospholipid (and cofactor) antibodies which results in the increased exposure of thrombogenic phospholipids. Accumulated data are consistent with the hypothesis that the disruption of annexin-V binding to anionic phospholipid surfaces plays an important thrombogenic role in the antiphospholipid syndrome.

Human β2-Glycoprotein I Binds to Endothelial Cells Through a Cluster of Lysine Residues That Are Critical for Anionic Phospholipid Binding and Offers Epitopes for Anti-β2-Glycoprotein I Antibodies

β2-Glycoprotein I (β2GPI) is a phospholipid-binding protein recognized by serum autoantibodies from the anti-phospholipid syndrome both in cardiolipin- and β2GPI-coated plates. We found that: 1) recombinant wild-type β2GPI bound to HUVEC and was recognized by both human monoclonal IgM and affinity-purified polyclonal IgG anti-β2GPI anti-phospholipid syndrome Abs; and 2) a single amino acid change from Lys286 to Glu significantly reduced endothelial adhesion. Double and triple mutants (from Lys284,287 to Glu284,287, from Lys286,287 to Glu286,287, and from Lys284,286,287 to Glu284,286,287) completely abolished endothelial binding. A synthetic peptide (P1) spanning the sequence Glu274–Cys288 of the β2GPI fifth domain still displayed endothelial adhesion. Another peptide (P8), identical with P1 except that Cys281 and Cys288 were substituted with serine residues, did not bind to HUVEC. Anti-β2GPI Abs, once bound to P1 adhered to HUVEC, induced E-selectin expression and up-regulated IL-6 secretion. Control experiments conducted with irrelevant Abs as well as with the P8 peptide did not show any endothelial Ab binding nor E-selectin and IL-6 modulation. Our results suggest that: 1) β2GPI binds to endothelial cells through its fifth domain; 2) the major phospholipid-binding site that mediates the binding to anionic phospholipids is also involved in endothelial binding; 3) HUVEC provide a suitable surface for β2GPI binding comparable to that displayed by anionic phospholipids dried on microtiter wells; and 4) the formation of the complex between β2GPI and the specific Abs leads to endothelial activation in vitro.

Anti-beta2-glycoprotein I (beta2GPI) monoclonal antibodies with lupus anticoagulant-like activity enhance the beta2GPI binding to phospholipids

beta2-Glycoprotein I (beta2GPI), a plasma glycoprotein with phospholipid-binding property, is known to be the actual target antigen for autoimmune type anticardiolipin antibodies (aCLs). Certain groups of aCLs (anti-beta2GPI antibodies) exert lupus anticoagulant (LA) activity and perturb the function of vascular endothelial cells. This investigation aimed at highlighting some insights into the molecular basis by which aCLs exert their biological effects by using anti-beta2GPI mAbs with well-characterized epitopes from mice and from patients with antiphospholipid syndrome. Anti-beta2GPI mAbs directed against the third domain (Cof-20 and Cof-22) and fourth domain (Cof-21, EY1C8, and EY2C9) of beta2GPI inhibited the thrombin generation induced by Russell's viper venom in diluted plasma and that induced by the prothrombinase complex reconstituted with purified clotting factors. This anticoagulant activity was abrogated in the presence of an excess amount of phospholipids, thus resembling the LA activity. In stark contrast, anti-beta2GPI mAbs directed against the fifth domain and the carboxy-terminal region of the fourth domain showed no LA-like activity. These findings suggest that the LA activity of anti-beta2GPI antibodies depends on their epitope specificity. Experiments carried out to clarify the mechanism of the LA activity showed that anti-beta2GPI mAbs with LA-like activity, but not those without this effect, enhance the beta2GPI binding to phospholipids. In addition, the F(ab')2 fragment, but not the Fab' fragment, of the anti-beta2GPI mAbs was found to enhance the LA activity and the beta2GPI binding to phospholipids, suggesting that anti-beta2GPI antibodies induce formation of multiple complexes of beta2GPI on the surface of phospholipids because of their bivalent property. This clustering of beta2GPI molecules induced by anti-beta2GPI antibodies, probably because of their multivalent property and epitope specificity, might hinder the lateral mobility and activation of clotting factors on the surface of phospholipids and thus exert LA activity. Clustering of beta2GPI molecules may also explain the molecular mechanism by which anti-beta2GPI antibodies alter the function of leukocytes and endothelial cells. The well-documented heterogeneous LA activity of aCLs (anti-beta2GPI antibodies) may also be explained by their epitope specificity.

The antiphospholipid/cofactor syndromes: a primary variant with antibodies to beta 2-glycoprotein-I but no antibodies detectable in standard antiphospholipid assays

BACKGROUND

Most systemic lupus erythematosus (SLE) patients with two or more clinical manifestations of the antiphospholipid syndrome (APS) and negative antiphospholipid antibodies (aPL) have antibodies to beta 2-glycoprotein-I (a beta 2 GP-I). Herein we describe a similar set of circumstances, but in patients without evidence of SLE.

PATIENTS AND METHODS

We studied 6 patients with recurrent venous and/or arterial thromboses without aPL as detected by routine assays nor clinical or serological evidence of other autoimmune disease. Immunoglobin (Ig) G and IgM antibodies to bovine and human phospholipid-free beta 2 GP-I were studied by Western blot test and by enzyme-linked immunosorbent assay (ELISA) utilizing radiated and nonirradiated plates. We also tested antibodies to cardiolipin, phosphatidylserine, and phosphatidylethanolamine by ELISA. As controls, 54 normal sera were studied.

RESULTS

All 6 patients had recurrent arterial and/or venous thromboses. Three also had thrombocytopenia, 1 had livedo reticularis, and 2 had valvular heart disease. None of the patients had aPL, but all had serum IgG reactivity against human and bovine beta 2 GP-I (P < 0.001 versus controls for both). Titers of anti-bovine beta 2 GP-I were higher when studied in irradiated plates but were also higher than normal in nonirradiated plates (P < 0.001). These antibodies did not recognize human or bovine beta 2 GP-I bound to cardiolipin in solid phase. We confirmed by Western blot that these autoantibodies recognize human beta 2 GP-I. We found no IgM a beta 2 GP-I.

CONCLUSIONS

We describe a primary condition akin to the antiphospholipid syndrome with negative aPL, but with serum IgG antibodies to human and bovine beta 2 GP-I. These antibodies recognize beta 2 GP-I epitopes that are not accessible when beta 2 GP-I is bound to cardiolipin.

Anti-beta 2-glycoprotein I antibody: specificity and clinical significance

Cardiolipin binding of IgG-class anticardiolipin antibody (aCL) depends on the existence of beta 2-glycoprotein I (beta 2-GPI). We developed an EIA system that enables detection of antibodies against beta 2-GPI, without the presence of cardiolipin. This system involves use of irradiated polystyrene plates, in which oxygen atoms are introduced onto the surfaces of the plates. beta 2-GPI bound to the surface of these plates is assumed to undergo a conformational change that exposes normally cryptic epitopes. Anti-beta 2-GPI antibody measured using this EIA system showed good correlation with aCL measured by conventional EIA methods and may prove useful in evaluating the risk of thrombosis and monitoring the clinical course in patients with SLE. Utilizing this EIA system and beta 2-GPI-deleted mutants, we found that the fourth domain of beta 2-GPI is involved in expression of one of the cryptic epitopes recognized by aCL. We also found that oxidized LDL are sequentially targeted by beta 2-GPI and aCL.

The pathophysiology of alveolar osteonecrosis of the jaw: anticardiolipin antibodies, thrombophilia, and hypofibrinolysis

We studied 55 patients (50 women, 5 men) with severe facial pain and biopsy-proven neuralgia-inducing cavitational osteonecrosis (NICO) of the alveolar bone of the jaws. Our aim was to assess the pathophysiologic contributions to NICO of anticardiolipin antibodies (aCLA), thrombophilia (increased tendency to intravascular thrombi), and hypofibrinolysis (reduced ability to lyse thrombi). Of the 55 patients, 43 (78%) had one or more tests positive for thrombophilia or hypofibrinolysis (or both), and only 12 (22%) were normal. Eighteen of 55 (33%) patients had high aCLA (> 2 SD above mean value for control subjects); immunoglobulin G (IgG) (p = 0.01) and immunoglobulin A (IgA)(p = 0.001) levels were higher in patients than in controls. The distribution of elevated aCLA immunoglobulin classes among patients was as follows: IgG alone, 5 (9%); IgA alone, 7 (13%); and IgM alone, 3 (5%). Three patients (5%) had high levels of both IgG and IgA aCLA. Other defects of the thrombotic or fibrinolytic systems in the 55 patients included high lipoprotein(a) in 36% (vs 20% in control subjects (p = 0.03)), low stimulated tissue plasminogen activator activity (tPA-Fx) in 22% (vs 7% in control subjects (p = 0.08)), high plasminogen activator inhibitor activity (PAI-Fx) in 18% (vs 8% in control subjects (p = 0.03)), resistance to activated protein C in 16% (vs 0% in control subjects (p = 0.007)), low antigenic protein C in 4+ (vs 0% in control subjects (p > 0.2)), and low antigenic protein S in 4% (vs 0% in control subjects (p > 0.2)). Anticardiolipin antibodies and other defects of the thrombotic and fibrinolytic systems appear to be common, potentially reversible pathogenetic risk factors associated with osteonecrosis of the jaw.