Avid Bioservices Inc (CDMO)

[goodJohnhunting]

Jake,

and the 200 to 240 range... how limiting is that range with respect to patient populations...

Here are some limitations:

1. Age (This is the #1 factor for "free floating, unbound B2GP1).
2. Smoker/Non
3. Race
4. Arteriosclerotic Vascular Disease
5. Weight
6. History of APS
7. History of coagulants
8. Overall metabolic health
9. Diabetes
10. Inflammatory disease

Ab2gp1 is only expressed when bound/exposed/expressed to lipid by-layer domain V 1,2.

B2gp1 is "free floating" and readable in plasma. Which by the way exist in two separate conformations.

I'm assuming "free floating" samples were taken pre-administration in Sunrise trial. The data mining was not for relevant antibodies directed towards b2gp1, IMO.

See confirmation from your study link below>>

To measure relevant antibodies, considered to
be those of the IgG isotype directed towards b2GP1 and
particularly those directed to Domain 1 (Dm1) of the molecule.
Patients/methods: In this cross-sectional study we
measured IgG ab2GP1-Dm1 by a chemiluminescent immunoassay
in a group of individuals initially positive for IgG
ab2GP1 and classified as triple (LAC+, IgG aCL+, IgG
ab2GP1+, n = 32), double (LAC–, IgG aCL+, IgG
ab2GP1+, n = 23) or single positive (LA-, IgG aCL-, IgG
ab2GP1+, n= 10)http://onlinelibrary.wiley.com/doi/10.1111/jth.12865/pdf

I've searched the best possible abstract to enlighten anyone interested in knowing what and how B2GP1 functions.

BTW, there is confirmation in the abstract to my statin/bavi theory..

Read below:

http://onlinelibrary.wiley.com/doi/10.1111/j.1538-7836.2011.04327.x/full

Philip G. de Groot, Department of Hematology (G03.550), University Medical Center, Heidelberglaan 100, 3584 CX Utrecht, the Netherlands.
Tel.: [phone]; fax: [phone].
E-mail: [e-mail]
Abstract

Summary. ß2-Glycoprotein I (ß2-GPI) is a protein that circulates in blood at high concentrations. The function of ß2-GPI has long been an enigma. More than 20 years ago, it was discovered that ß2-GPI is the major antigen for the circulating antibodies in the antiphospholipid syndrome. However, this knowledge has not advanced our understanding of the physiologic role of the protein. In recent years, new insights have suggested an important function of this protein in innate immunity. ß2-GPI was found to scavenge lipopolysaccharide and was able to clear unwanted anionic cellular remnants such as microparticles from the circulation. The function of ß2-GPI seems to depend on the structural conformation of the protein, and it has been established that ß2-GPI can exist in at least two conformations. In this review, we will highlight and summarize the current knowledge on this protein.
Introduction

ß2-Glycoprotein I (ß2-GPI), also known as apolipoprotein H, is a 50-kDa protein that was described for the first time in 1961, and in 1968 the first apparently healthy person deficient in this protein was identified [1,2]. As no function could be attributed to ß2-GPI, the protein did not receive much attention and, on average, a single publication per year appeared in the literature. ß2-GPI’s alternative name, apolipoprotein H, suggested a function in lipid metabolism, but this second name turned out to be a misnomer. In plasma, ß2-GPI is not incorporated into or associated with lipoprotein particles [3]. From 1990 onwards, the interest in this apparently obsolete protein increased significantly, when ß2-GPI was identified as the most important antigen in an autoimmune disease called the antiphospholipid syndrome (APS) [4,5]. APS is an autoimmune disease characterized by thrombotic complications in both arteries and veins, as well as by pregnancy-related complications in combination with the presence of so-called antiphospholipid antibodies in the plasma [6]. It is now generally accepted that these autoantibodies are not directed against negatively charged phospholipids, but towards proteins bound to these phospholipids. Animal studies have shown that the most prominent antigen in APS is ß2-GPI, a protein with (relatively low) affinity for anionic phospholipids [7]. The importance of antibodies against ß2-GPI was demonstrated by injection of these antibodies into mice; this resulted in increased thrombus formation when the mice were challenged, and they showed increased resorption of fetuses when pregnant [8–11]. Despite the obvious importance of ß2-GPI in the pathophysiology of APS, these in vivo experiments did not reveal a physiologic function for this protein. To explain the role of ß2-GPI in the clinical manifestations observed in APS, it was thought that the autoantibodies are gain-of-function antibodies; the autoantibodies induce a new function for ß2-GPI. Studies performed with Fab and F(ab')2 fragments showed that dimerization of ß2-GPI resulted in 1000-fold increased affinity for anionic phospholipids [7]. Indeed, in vivo experiments demonstrated that F(ab')2, but not Fab, fragments of isolated patient IgG injected into hamsters induced increased thrombus formation after vascular injury, supporting the importance of bivalency of the interaction of ß2-GPI with anionic phospholipids [12]. However, accumulating data from recent publications has challenged the hypothesis that dimerization of ß2-GPI resulting from antibody binding can explain all of the in vitro observations made on the function of ß2-GPI. In this review, we will discuss the recent insights into the structural changes that can occur within ß2-GPI, and the consequences of these changes for the function of ß2-GPI in health and disease.
Structure of ß2-GPI

ß2-GPI is an anionic phospholipid-binding glycoprotein that belongs to the complement control protein (CCP) superfamily. The members of this family are composed of repeating stretches of about 60 amino acids with two fully conserved disulfide bonds. More than 50 mammalian proteins, mainly members of the complement system, belong to this family [13]. The CCP domain functions as a protein–protein interaction module in many different proteins. ß2-GPI consists of 326 amino acids organized in five CCP domains [14]. The first four domains have the regular, conserved sequences, but the fifth domain is aberrant. This domain contains a six-residue insertion, a 19-residue C-terminal extension, and an additional disulfide bond that includes a C-terminal cysteine. These additional amino acids in domain V are responsible for a unique feature of this CCP domain, because they constitute a large positively charged patch that determines the affinity for anionic phospholipids. Human ß2-GPI contains four N-glycosylation sites (Arg143, Arg164, Arg174, and Arg234), localized in the third and fourth domains, and one O-linked sugar on Thr130. The glycans account for approximately 20% w/w of the total molecular mass [15].

The crystal structure of ß2-GPI was solved in 1999 by two groups [16,17]. The crystal structure of the protein reveals a stretched arrangement of the first four domains, with domain V at a right angle to the other domains, like a hockey stick. The phospholipid-binding site is located at the bottom side of domain V. This site consists of two major parts, a large positive patch of 14 charged amino acids, and a flexible and hydrophobic loop. The flexible loop contains a classic Trp-Lys sequence at its interfacial region, giving it the potential to insert into the membrane [18]. The crystal structure predicts that when ß2-GPI is bound to a lipid layer, domains I–IV point away from the lipid layer, and the potential binding site for autoantibodies against ß2-GPI that has been located in domain I is fully exposed [19].

Small-angle X-ray scattering experiments suggested that, in solution, ß2-GPI adopts an S-shaped conformation with an additional buckle between domains II and III [20]. This different spatial orientation of ß2-GPI in solution could explain the observation that autoantibodies against ß2-GPI only recognize ß2-GPI when it is coated on a surface, and not when it is in solution [19]. Owing to the twist between domains II and III, the carbohydrate residues are positioned in such a way that they could shield the epitope for the autoantibodies in domain I. Binding of ß2-GPI to anionic surfaces results in a change in conformation and exposure of the epitope for the autoantibodies [20,21].

Recently, electron microscopy (EM) studies have further complicated the story [22]. Direct visualization of plasma purified ß2-GPI in the absence and presence of antibodies showed that it can exist in two completely different conformations (Fig. 1A,B). In plasma, it is present as a circular protein in which domain I interacts with domain V. After interaction with anionic surfaces, the protein opens up and adopts the hockey stick-like conformation, which resembles the crystal structure [22]. The harsh crystallization conditions (high salt and pH) might have resulted in the opening of ß2-GPI. The S-shaped configuration was never observed with EM; this might be explained by the loss of interaction between ß2-GPI and its liquid environment. In contrast to the proposed S-shaped structure, in which the autoantibody binding site was shielded by a carbohydrate side chain, in a circular conformation the epitope for the autoantibodies is shielded via interaction with domain V of the same molecule (Fig. 2). The circular conformation also predicts the presence of shielded epitopes within domain V [22]. This latter prediction is supported by the observations that ß2-GPI binds to cellular receptors via domain V and that this binding is strongly enhanced in the presence of antibodies [23]. From all of these studies, it is obvious that ß2-GPI is a flexible molecule, and that the protein is not constrained to a single, specific conformation, but that its conformation depends on interactions with its surroundings. Factor H, a complement factor composed of 20 CCP domains, can also adopt different domain orientations in solution, with consequences for its functional activity [24,25]. Apparently, proteins consisting of CCP domains can vary their conformation, which probably depends on the length and flexibility of the linker sequences between the CCP domains.

Figure 1.
Figure 1. Open in figure viewerDownload Powerpoint slide
ß2-Glycoprotein I (ß2-GPI) can exist in two conformations. (A, B) Electron microscopy images of the closed circular conformation of plasma ß2-GPI (A) and the open, hockey stick-like conformation when ß2-GPI is complexed with an antibody directed against domain I (B). Photographs courtesy of M. Mörgelin, University of Lund, Sweden.
Figure 2.
Figure 2. Open in figure viewerDownload Powerpoint slide
Model of the conversion of ß2-glycoprotein I from a circular conformation into an open, hockey-stick like conformation. Note that the antibody-binding site is not accessible to the autoantibodies in the circular conformation. LPS, lipopolysaccharide.
The potential of ß2-GPI to adopt different conformations is a critical factor to consider in studies performed with purified ß2-GPI. Perchloric acid precipitation is a popular first step in many purification protocols. The exposure of ß2-GPI to an extreme pH may result in refolding of the protein [26,27]. Conflicting results obtained in studies on the function of ß2-GPI from different laboratories could be explained by differences in conformation caused by the purification method used.
ß2-GPI in evolution

To our knowledge, the evolutionary conservation of ß2-GPI has not been studied in detail. We have investigated the conservation of the ß2-GPI gene in 40 currently published genomes, and found its presence in all mammals, but also in birds, fish, and reptiles (C. Agar, P. G. de Groot, J. C. M. Meijers, unpublished data). Important sites in the molecule, such as the disulfide bridges in the CCP domains, the binding site for phospholipids, and the amino acids in domain I to which the antiphospholipid antibodies are directed (see below), were very well and, in most cases, completely conserved. There was even 14% amino acid identity with the protein from the fruit fly Drosophila melanogaster and 17% amino acid identity with that from the roundworm Caenorhabditis elegans, the most primitive organisms in which ß2-GPI could be identified. This suggests that the function of ß2-GPI may be more important than expected, given the evolutionary conservation across the animal kingdom.
Oxidation status of ß2-GPI

Recently, the presence of a unique structural property of ß2-GPI, conferred by a C-terminal cysteine, has been discussed [28]. This amino acid is part of a disulfide bridge in the fifth domain, and is exposed on the surface of the molecule. It has been shown that ß2-GPI can be reduced in a thioredoxin-1-dependent or a protein disulfide isomerase-dependent mechanism. These thiol-oxidoreductases predominantly reduce the Cys288–Cys326 disulfide bond within ß2-GPI. An assay was developed to measure the reduced state of ß2-GPI. It was suggested that ß2-GPI circulates in healthy volunteers predominantly in the reduced state, and that, in patients with APS, ß2-GPI circulates in a relatively oxidized and nitrated state [29,30]. However, the crystal structure suggested that all cysteines of ß2-GPI are oxidized [16,17]. Furthermore, nitrosylation of cysteines was not observed. It therefore remains questionable whether ß2-GPI circulates in a reduced state. Nevertheless, it is possible that ß2-GPI in the circulation can rapidly switch between an oxidized and reduced state, and that the rather harsh conditions used in the assay to detect reduced ß2-GPI shift the balance between the reduced and the oxidized form towards reduction. Thiol-exchange reactions play a central role in both primary and secondary hemostasis [31]. Further experiments are necessary to answer the question of whether ß2-GPI can act as an intermediate in these reactions.
Plasma levels of ß2-GPI

ß2-GPI is predominantly synthesized in hepatocytes, and it circulates in blood at variable levels (50–500 µg mL-1, 1–10 µm) [32]. ß2-GPI levels increase with age (Fig. 3), and are reduced in pregnant women and in patients with stroke and myocardial infarction [33]. Interestingly, ß2-GPI levels seem to be slightly higher in patients with APS [34].

Figure 3.
Figure 3. Open in figure viewerDownload Powerpoint slide
Plasma levels of ß2-glycoprotein I (ß2-GPI) in 500 healthy volunteers. The ß2-GPI level increases with age (12.6 µg mL-1 for every 10-year increase) [89]. Samples courtesy of C. J. M. Doggen, University of Twente, Enschede, The Netherlands.
Plasma levels of ß2-GPI are partially regulated by polymorphisms in the promotor region of the gene. Of the many single-nucleotide polymorphisms (SNPs) in the promoter region of the ß2-GPI gene, only two have been identified (-32C>A and -700C>A) that correlate with a significant reduction in plasma levels of ß2-GPI [35,36]. The –32C>A SNP is part of a promotor region that serves as a binding site for transcription factor IID, whereas the -700C>A SNP is not part of a known binding motif for a transcription factor. However, these SNPs only explain a proportion of the variance in plasma ß2-GPI levels. Besides SNPs in the promoter region, additional SNPs have been identified in the coding region. An interesting polymorphism is Cys306Gly, which disrupts the phospholipid-binding site within ß2-GPI. One study showed that this SNP was correlated with plasma levels of ß2-GPI [37]. Also, theTrp316Ser polymorphism, a variation within the flexible hydrophobic loop that also affects phospholipid binding, has been found to be associated with plasma levels of ß2-GPI [38]. Apparently, amino acid changes in the two regions involved in binding to phospholipid membranes influence plasma levels of ß2-GPI. Mutations in the phospholipid-binding site could influence the conformation, resulting in altered clearance of ß2-GPI from the circulation. Additional studies are warranted to better delineate the genetic and molecular bases of plasma ß2-GPI levels.
Autoantibodies against ß2-GPI

Three different assays are available for detection of the autoantibodies that cause APS: two ELISAs with cardiolipin or ß2-GPI as antigen, and a clotting assay that detects lupus anticoagulant [6]. Although the detection principles that these assays use to identify the presence of the autoantibodies are different, the common denominator of these three assays is that a positive result depends on the presence of ß2-GPI [6]. Further analysis of the specificity of the autoantibodies has identified an epitope on domain I of ß2-GPI to which the antibodies are directed [39–41]. This discontinuous epitope includes Arg39 and Arg43. An ELISA that detects autoantibodies against domain I of ß2-GPI correlates better with thrombosis than the classic anti-ß2-GPI antibody ELISA, in which all domains are available for autoantibody binding [42]. Autoantibodies against ß2-GPI constitute a heterogeneous population of antibodies, and only a subset of these seem to be related to the clinical manifestations.

As discussed before, ß2-GPI circulates in blood in a circular conformation in which the epitope for these antibodies is shielded from plasma. After interaction with anionic surfaces, ß2-GPI opens up and the epitope is exposed. The observation that autoantibodies are directed against cryptic epitopes within ß2-GPI could be a clue to how these autoantibodies can break tolerance. CD4-positive T cells isolated from patients with APS respond to ß2-GPI fragments and chemically modified ß2-GPI, but fail to respond to native ß2-GPI [43]. This suggests that expression of cryptic epitopes is a prerequisite for induction of the formation of autoantibodies against ß2-GPI. To further study this, we injected mice with domain I or domains II to V of murine ß2-GPI [44]. The individual domain I expresses the cryptic epitope for the autoantibodies. We found that only the mice injected with domain I developed antibodies against murine ß2-GPI; their plasmas expressed lupus anticoagulant activity, and contained elevated levels of thrombin–antithrombin complexes. Mice injected with domains II to V did not develop these autoantibodies. These observations suggest that exposure of the cryptic epitope for ß2-GPI in mice is sufficient for the development of autoantibodies with all of the characteristics of the autoantibodies that characterize APS.

What are the pathophysiologic conditions that can lead to the development of these antibodies? Many studies have shown that the presence of antiphospholipid antibodies is associated with a history of infections [45]. The prevailing theory to explain this correlation is molecular mimicry [46]. Sequence similarities between foreign proteins and self-proteins are sufficient to induce a loss of immune tolerance, resulting in the formation of autoantibodies. An alternative theory is that infectious agents can serve as adjuvants [47]. An adjuvant enhances the recipient’s immune response by unfolding the injected protein, resulting in exposure of antigenic epitopes that are normally shielded. To further support this theory as an explanation for the development of autoantibodies against ß2-GPI, we injected mice with different surface proteins isolated from Streptococcus pyogenes. Four S. pyogenes surface proteins (M1 protein, protein H, SclA, and SclB) were found to interact with ß2-GPI (G. M. van Os, J. C. Meijers, C. Agar, M. Valls Seron, J. A. Marquart, P. Akesson, R. T. Urbanus, R. H. Derksen, H. Herwald, M. Morgelin, P. G. de Groot, unpublished data). Only binding to protein H induced a conformational change in ß2-GPI, thereby exposing a cryptic epitope for APS-related autoantibodies. When mice were injected with the four proteins, only mice injected with protein H developed antibodies against mouse ß2-GPI, and these antibodies were directed against an epitope in domain I. These studies support the suggestion that a conformational change in ß2-GPI is sufficient to induce autoantibodies (G. M. van Os, J. C. Meijers, C. Agar, M. Valls Seron, J. A. Marquart, P. Akesson, R. T. Urbanus, R. H. Derksen, H. Herwald, M. Morgelin, P. G. de Groot, unpublished data). We hypothesize that other (bacterial, viral, or parasitic) proteins or anionic membrane fragments could also induce this conformational change. Levine et al. [48] showed that when mice were immunized with ß2-GPI, addition of lipopolysaccharide (LPS) to the injected ß2-GPI was enough to break tolerance. We have recently shown that binding to LPS is sufficient to convert ß2-GPI from a circular to a hockey stick-like conformation [49]. Therefore, we propose that the conformational change in ß2-GPI is the common denominator in the development of autoantibodies against ß2-GPI.

It has been shown that artificially oxidized ß2-GPI induces dendritic cell maturation and T-cell proliferation, and that addition of an antioxidant prevents the maturation of dendritic cells [50]. Analysis of ß2-GPI after oxidation showed the presence of aggregates, suggesting that the exposure of cryptic epitopes resulting from denaturation, and not a change in oxidation state, could be responsible for the cellular responses. Nevertheless, it is possible that oxidized proteins participate in the initiation of an immune response [51].

Initial studies suggested that the Val247Leu polymorphism was correlated with the presence of autoantibodies against ß2-GPI [52], but this assumption has been questioned, and needs further confirmation [53]. It is possible that the Val247Leu polymorphism influences the stability of the circular conformation of ß2-GPI.
Other possible roles and functions of ß2-GPI

Individuals deficient in ß2-GPI seem to be healthy, and mice deficient in ß2-GPI do not express a clear phenotype, indicating that the presence of ß2-GPI is not essential for life. In contrast, a very important observation was that the percentage of null offspring born was significant lower in these mice than expected on the basis of Mendelian ratios [54], indicating that ß2-GPI might play a role in embryonic development. However, no further experiments have been performed during the last 10 years to elucidate the underlying mechanism.

Despite the observation that ß2-GPI-deficient humans and mice do not bleed or show an increased risk of thrombosis, many functions in the regulation of hemostasis have been attributed to ß2-GPI, such as inhibition of ADP-induced platelet aggregation and regulation of contact activation [55,56]. Although it is possible that, in purified systems, ß2-GPI interferes with platelet aggregation and activates the intrinsic coagulation pathway, there are no indications that these interactions take place in vivo.

ß2-GPI has been identified in atherosclerotic plaques [57], and a number of studies have suggested that the presence of anti-ß2-GPI antibodies results in accelerated atherosclerosis [58,59]. However, further studies have shown that the increased atherogenesis was not the result of the antiphospholipid antibodies, but more a general consequence of autoimmune diseases [60,61]. Circulating oxidized LDL–ß2-GPI complexes have also been detected in patients with both systemic lupus erythematosus and APS. The presence of circulating oxidized LDL–ß2-GPI complexes could point to an active role of these complexes in autoimmune-mediated atherothrombosis [62].

The first publication suggesting a role of ß2-GPI in angiogenesis showed that clipped ß2-GPI was able to inhibit bladder cancer progression in mice [63]. Clipped ß2-GPI arises after plasmin cleavage of the Lys317–Thr318 bond in ß2-GPI [64]. ß2-GPI-deficient mice show increased microvessel formation in comparison with ß2-GPI-reconstituted controls when injected with growth factor-free matrigel implants [65]. It was suggested that clipped ß2-GPI is able to inhibit endothelial cell proliferation [66].
ß2-GPI and the risk of thrombosis in patients with APS

Animal studies in which anti-ß2-GPI antibodies isolated from patients with APS were injected into mice suggest that there is a direct link between these autoantibodies and thrombosis [12,67,68]. Purified autoantibodies increased thrombus size in a dose-dependent manner [68]. The mechanism by which these autoantibodies lead to thromboembolic events is unknown. Many different hypotheses have been proposed, but none of these has been convincingly proven [69]. The most popular theory to explain why APS patients have an increased thrombotic risk is that the autoantibodies can cause activation of different cell types involved in regulation of the hemostatic balance, such as endothelial cells, platelets, neutrophils, fibroblasts, trophoblasts, and monocytes [70–75]. Normally, ß2-GPI in plasma circulates in a circular conformation with a relatively low affinity for anionic surfaces [22]. When ß2-GPI encounters cells that (temporarily) expose anionic phospholipids on their surface, ß2-GPI will bind to these phospholipids, resulting in a conformational change. This change will expose the epitope for the autoantibodies, which is normally shielded from recognition [22]. The autoantibodies will bind and stabilize ß2-GPI in its hockey stick-like conformation. The binding of the autoantibodies can also result in the formation of bivalent complexes. The antibody–ß2-GPI complexes have a much stronger affinity for anionic phospholipids, and also for different receptors that are expressed on these cells. Many different receptors have been described that bind ß2-GPI–antibody complexes: Toll-like receptor (TLR)2, TLR4 [70,71], annexin A2 [9], GPIba [76,77], and a member of the LDL-receptor family, LRP8 [75]. Interestingly, functional or absolute deficiencies of all of these receptors lead to decreased thrombus formation in murine models of APS, suggesting that they all play a role in the syndrome, which seems unlikely. Clearly, ß2-GPI in its hockey stick-like conformation is a very adhesive protein, and binds to different receptors on cells. Additional experiments in which the affinities of the different receptors for ß2-GPI are determined are absolutely necessary to solve this puzzle. Interestingly, three groups have shown that a fragment of the LRP8 receptor can be used as an inhibitor of ß2-GPI–antibody complexes in different models, suggesting an important role for LRP8 in the pathophysiology of APS [10,11,78]. ß2-GPI binds readily to members of the LDL-receptor family, which are involved in the uptake of different components by cells (see below). In contrast to the other members of its family, LRP8 is not a clearance receptor, but is predominantly involved in signal transduction.
ß2-GPI and infectious diseases

The phospholipid-binding site in domain V of ß2-GPI contains a number of positively charged amino acids. Peptides with these residues are known to interact with and target bacteria [79]. We tested the effects of domain V-derived peptides against a number of bacteria [80]. ß2-GPI by itself was not able to directly kill bacteria, but peptides from domain V had potent antibacterial activities against Gram-positive and Gram-negative bacteria [80]. The peptides inserted in the bacterial membrane, and caused leakage of the cytosol and death of the bacteria. ß2-GPI can be cleaved by plasmin and neutrophil proteases [64,80]. Indeed, it was shown that neutrophil protease-treated ß2-GPI possessed antibacterial activity [80]. ß2-GPI can therefore be considered to be involved in innate immunity. Some bacteria, however, have evolved to circumvent the antibacterial action of cleaved ß2-GPI. S. pyogenes expresses different proteins, such as protein H and M1 protein, that bind ß2-GPI. By binding to ß2-GPI, the proteins prevent cleavage by neutrophil proteases. Furthermore, neutrophil proteases cleave protein H and M1 protein from the bacterial surface, and these soluble proteins can counteract the effects of antibacterial peptides formed from ß2-GPI [80].

Surprisingly, a direct interaction between ß2-GPI and LPS was recently discovered [49]. LPS is one of the constituents of the outer membrane of Gram-negative bacteria, and plays a major role in activating the host immune response by binding to monocytes and other cells [81]. There are a number of proteins known to neutralize the effects of LPS [82–84]. Besides a direct binding affinity of LPS for ß2-GPI, it was shown that ß2-GPI inhibited LPS-induced tissue factor and interleukin-6 expression from monocytes and endothelial cells [49]. The interaction between ß2-GPI and LPS may be of clinical relevance. It was shown that administration of LPS to healthy volunteers elicited an immediate drop in the ß2-GPI level of approximately 25%. This is a very different response from those of to other coagulation, fibrinolytic or LPS-scavenging proteins, and suggests an immediate interaction between LPS and ß2-GPI. Furthermore, the levels of ß2-GPI before LPS injection were inversely correlated with the expression of inflammatory markers and with the temperature rise observed in the volunteers after the LPS injection. Binding of LPS to ß2-GPI caused a conformational change within ß2-GPI that facilitated binding of the LPS–ß2-GPI complex to monocytes and its internalization [49]. Binding to monocytes was inhibited by receptor-associated protein (RAP), a universal inhibitor of the members of the LDL-receptor family. RAP also reduced the inhibitory effect of ß2-GPI on LPS-induced tissue factor expression on monocytes [49]. This suggests that binding of LPS to ß2-GPI alone is not sufficient for the neutralization of LPS, but that a second step, such as uptake of the complex via one of the members of the LDL-receptor family, is necessary.
ß2-GPI and von Willebrand factor (VWF)

VWF) is a multimeric plasma protein involved in platelet adhesion to injured vessel walls at high shear rates [85]. VWF acts as a molecular bridge between subendothelial collagens and the GPIb–IX–V receptor complex on platelets. Normally, VWF in the circulation does not interact with platelets. VWF undergoes a conformational change when it binds to collagen, resulting in exposure of the GPIb-binding site within its A1-domain. We have found that ß2-GPI binds with low affinity to the A1-domain, thereby inhibiting VWF-mediated platelet adhesion and agglutination [86]. Autoantibodies against ß2-GPI neutralize this inhibitory activity, pointing to a role of ß2-GPI–VWF interactions in the onset of thrombosis. In this respect, it is interesting to note that ultralarge VWF multimers have been reported to be present in patients with APS [87]. Ultralarge VWF multimers are in a GPIb-binding conformation, and their presence has been implicated in the pathology of many microangiopathic disorders [88]. Apparently, ß2-GPI could also have a role in the clearance of VWF from the circulation. In one study, lower plasma levels of ß2-GPI were correlated with an increased risk of myocardial infarction in elderly men [89].

ß2-GPI only binds to VWF when the protein is in its GPIb-binding conformation. This suggests an important role of ß2-GPI in pathologic conditions such as thrombotic thrombocytopenic purpura (TTP), in which ultralarge VWF multimers circulate. Indeed, the levels of ß2-GPI in plasma of patients with TTP is much lower than in healthy controls, again suggesting a role for ß2-GPI in the neutralization and clearance of ultralarge VWF from the circulation (B. de Laat, J. A. Kremer Hovinga, G. van Os, I. Dienava, B. Lammle, J. Wollersheim, R. Fijnheer, P. G. de Groot, unpublished data).

Recently, the concept that ß2-GPI interferes with VWF activity by simply binding to its A1-domain, thereby blocking its interaction with GPIb, has been challenged [90]. VWF has a tendency to self-associate and form fibrils, a process that might be supported by a rearrangement of disulfide bridges within the molecule [91]. This rearrangement of disulfide bridges coincides with an increased affinity for platelets. ADAMTS-13 has been proposed as one of the proteins that regulate this thiol-disulfide exchange in VWF [92,93]. As mentioned before, ß2-GPI can participate in thiol-exchange reactions. It has been suggested that ß2-GPI forms disulfide bridges with VWF [90]. ß2-GPI might then prevent ADAMTS-13 from reducing disulfide bridges that are necessary for VWF fibril formation. The prevention of binding of ß2-GPI to VWF by the autoantibodies would neutralize the inhibition of ADAMTS-13 reductase activity by ß2-GPI and, consequently, result in a decrease in the number of ultralarge VWF multimers and VWF activity. In this respect, it is of interest that ß2-GPI can also interact with GPIb, the receptor for VWF on platelets [76,77]. Whether the binding of ß2-GPI to GPIb influences the affinity of GPIb for VWF is unknown.
ß2-GPI and phagocytosis

One of the mechanisms by which antibodies could be formed against ß2-GPI is the repeated exposure of negatively charged phospholipids. Repeated surface exposure of, mainly, phosphatidylserine (PS) on cells in the vasculature could lead to binding of ß2-GPI to PS-exposing cells, a conformational change in ß2-GPI, and the formation of antibodies [94,95]. Although PS is generally located in the cytoplasmic leaflet of the plasma membrane of cells, cellular activation, apoptosis or certain pathologic conditions may cause a reorganization of the transbilayer lipid distribution, resulting in exposure of PS on the cell surface [96]. Normally, PS-exposing cells are removed rapidly by phagocytosis. A malfunction in this process, however, could give rise to increased binding of plasma proteins with affinity for anionic phospholipids. ß2-GPI has been shown to facilitate phagocytosis, by binding to PS-containing vesicles or apoptotic cells, which promotes engulfment by phagocytes [97,98]. Strikingly, this reaction was shown to be mediated by one of the members of the LDL-receptor family, although the specific member still needs to be identified [99]. Other studies suggested Ro60 as a receptor for ß2-GPI [100]. Also, platelet microvesicles bind ß2-GPI, allowing phagocytosis by macrophages [101]. Addition of antiphospholipid antibodies inhibited this process, and could therefore induce a procoagulant state [102]. Interestingly, all of these studies were performed with ß2-GPI that was purified with perchloric acid. It would be interesting to repeat these experiments with ß2-GPI in different conformations.
Conclusions and future perspectives

ß2-GPI must play an important role in biology, given its abundance in the (human) circulation, and its conservation during evolution. The homology with other proteins involved in innate immunity and the recent discovery that ß2-GPI can act as a scavenger for LPS suggest that ß2-GPI plays a role in host defense against bacteria (and, perhaps, also viruses and parasites). Moreover, increasing evidence suggests that not only is ß2-GPI involved in the clearance of microparticles, but also that it seems to act as a scavenger in the circulation that removes protein and cellular waste. The elucidation of the physiologic role(s) of ß2-GPI may also shed light on the events that cause the formation of antibodies against the protein, occasionally leading to APS. The observation that the amino acids that constitute the epitope for the antiphospholipid antibodies are conserved during evolution demonstrates that they are essential for the biological role of ß2-GPI. The coming years will no doubt be exciting and provide novel important and interesting findings on this fascinating protein.
Acknowledgements

We thank R. Urbanus, B. de Laat, C. Agar and G. van Os for critically reviewing this manuscript. We thank H. Herwald and M. Mörgelin, University of Lund, Sweden, for valuable discussions on this subject. This work was supported by a grant of the Netherlands Organization for Scientific Research (ZonMW 91207002).

All the best,
John