We have isolated five monoclonal IgG anti-β2-glycoprotein-1 (anti-β2G-1) and anti-prothrombin Fab from a patient with autoantibodies to oxidized low-density lipoproteins by phage display method. Analysis of their binding specificity revealed that all three β2GP-1-enriched mAbs (B14, B22, B27) reacted with β2GP-1 while both prothrombin-isolated mAbs (P11 and P13) reacted with prothrombin. Intriguingly, mAb P11 reacted with β2GP-1 and prothrombin and showed comparable binding affinity to both Ags, with Kd values of 1.6 × 10−6 M for β2GP-1 vs 3.2 × 10−6 M for prothrombin. This clone may thus, define a hitherto unknown shared epitope between β2GP-1 and prothrombin. Sequence analysis of all five clones showed significant mutations of the expressed genes. One rearranged V-D-J segment was repeatedly employed by three clones (mAbs B22, B27, and P13). However, all three clones used different L chains. Of note, the pairing of VH6-D-J with the L5-Vk1 L chain in mAb P13 resulted in the loss of binding to β2GP-1 and specific reactivity to prothrombin. Together, these data suggest that while the VH6-D-J chain may be important in the binding to β2GP-1, pairing with certain L chains may influence this binding. These data are the first human IgG anti-β2GP-1 and anti-prothrombin sequences reported; both represent the major subsets of antiphospholipid Abs present in antiphospholipid syndrome patients.

Antiphospholipid Abs (aPL),3 which include anticardiolipin Abs (aCL) detected by ELISA and lupus anticoagulant (LAC) Abs detected by in vitro blood clotting assays, are associated with thrombosis, recurrent fetal loss, and thrombocytopenia in patients with antiphospholipid syndrome (APS) (1, 2, 3). It is generally considered that the binding targets of aPLs include negatively charged phospholipids (PLs), various plasma proteins or complexes formed by PLs, and plasma proteins (4, 5, 6, 7). These plasma proteins include β2-glycoprotein-1 (β2GP-1), prothrombin, annexin V, protein C, and its cofactor protein S (5, 8, 9). In 1990, two studies showed that the binding of aPL to cardiolipin (CL) was enhanced by β2GP-1, suggesting that aPLs recognized a complex of CL and β2GP-1 (7, 10). Other studies have reported that aPL reacted with β2GP-1 alone (5, 11, 12). Over the last few years, the consensus is that anti-β2GP-1 Abs make up a significant percentage of aPL found in APS patients (5, 7, 12, 13, 14). In addition, recent studies show that anti-β2GP-1 Abs are more closely associated with APS thrombosis (5, 12, 15).

In contrast, increasing attention is being paid to anti-prothrombin Abs and the role they may play in thrombosis in APS patients. The prevalence of these Abs in patients varies greatly, ranging from 20 to 60% when detected via ELISA using immobilized human prothrombin on activated poly(vinyl chloride) plates (8, 16). Importantly, it was reported that affinity-purified IgG anti-prothrombin Abs bound to immobilized phosphatidylserine (PS) in the presence of Ca2+ and prothrombin, suggesting that IgG anti-prothrombin Abs bound to prothrombin and then is transported onto PS as a “passenger” upon prothrombin binding to PS (17, 18). In this context, it is conceivable that anti-prothrombin IgG may cross-link prothrombin molecules and thus increase the valency of interactions between prothrombin and PS. This implies that the anti-prothrombin IgG may enhance the binding of prothrombin to PS and to PL surface on endothelial cells and thus increase thrombin generation and promote thrombosis.

The oxidative modification of low-density lipoproteins (Ox-LDL) is thought to play an important role in various disease states including atherosclerosis (19, 20). Studies have shown that PLs are structurally similar to LDL and circulating lipoproteins contain various amounts of PLs and β2GP-1 (21). In addition, aPL may be directed against epitopes of oxidized PLs and cross-react with Ox-LDL (22, 23). Together, these data suggest that there is an overlap between aPL and anti-Ox-LDL Abs.

Although significant progress has been made in understanding the binding specificities of aPL, little is known about the structures and genetic basis of these potentially pathogenic autoantibodies. For reasons that are connected with the low efficiency of generating IgG Abs by conventional methods, few IgG aPL have been generated and studied. As a result, structure analysis of the potentially pathogenic Abs in APS has been difficult. Here we describe the isolation of five IgG monoclonal anti-β2GP-1 and anti-prothrombin Abs by phage display method from a patient with anti-Ox-LDL Abs. We report a detailed characterization of their binding specificities and sequence analysis.

The patient is a 72-year-old male who underwent coronary artery bypass surgery despite having low total plasma cholesterol levels for many years. His plasma was screened for the presence of autoantibodies to epitopes of Ox-LDL as part of a study being conducted in patients seen in the Lipid Research Clinic at the University of California at San Diego. Because anti-Ox-LDL Abs overlap substantially with aPL, we analyzed his sera for aCL and reactivities with β2GP-1 and prothrombin. The results showed that he had significant titers of anti-β2GP-1 Abs at one in one hundred (1:100, in borate-buffered saline (BBS), 0.2 M boric acid, 0.15 M NaCl, pH 8.2, containing 0.25% gelatin) and anti-prothrombin Abs (1:50, in calcium buffer, 50 mM Tris-HCl, 150 mM NaCL, 50 mM CaCl2, pH 7.5, containing 0.25% gelatin) (data not shown). Accordingly, his monocytes were isolated and used to prepare a combinatorial library.

The control library was prepared from a normal individual whose plasma did not contain anti-DNA, anti-β2GP-1, and anti-prothrombin Abs (data not shown).

An IgG1 κ and λ libraries were constructed according to published protocols (24, 25). Briefly, lymphocytes from the patient and control subjects were isolated and used as the source of total RNA for the phage library construction. PCR was then used to amplify and clone Fab genes from isolated cells into the phage display vector, pComb3H. Phage Fab clones are then selected based on Ab-binding specificity on β2GP-1 or prothrombin-coated plates. We obtained a library of 108 members with an insert frequency of >80% as determined by restriction endonuclease analysis.

The selection of aPL clones was performed as previously described (24). Briefly, microtiter plates (3690; Costar, Cambridge, MA) were coated overnight at 4°C with either β2GP-1 at 10 μg/ml in BBS or prothrombin (Enzyme Research Laboratories, South Bend, IN) at 10 μg/ml in calcium buffer (50 mM Tris, 150 mM NaCl, and 50 mM calcium chloride, pH 7.5). After washing, β2GP-1-coated plates were blocked with 3% of BSA in BBS while prothrombin-coated plates were blocked with 0.25% gelatin for 1 h at room temperature. Then, freshly prepared phage particles (1012 phage particles) were added and incubated for 2 h at room temperature. Thereafter, wells were washed extensively with TBS containing 0.5% Tween 20 (TBST) and bound phage particles were eluted with 50 μl of 0.1 M HCl/glycine (pH 2.2)/0.1% BSA. Following the third round of panning, phagemid DNA was recovered and used to generate soluble Fab as previously described (24).

Then, each isolated clone was lysed and analyzed for soluble Fab by ELISA. Wells were coated with goat anti-human IgG Fab mAb (Cappel Research Products, Durham, NC) overnight at 4°C and blocked with 0.25% BSA. Bacterial lysates containing soluble Fab were added to wells in duplicates, and the bound Fab were detected with enzyme-labeled goat anti-human IgG. Fab from each positive clone were affinity purified with the goat anti-human Fab column and analyzed for their binding property.

The binding specificity of Fab clones were determined as previously described for anti-β2GP-1 and anti-prothrombin Abs (24). For anti-β2GP-1, microtiter plates were precoated with 10 μg/ml β2GP-1 in BBS overnight at 4°C. After blocking with 0.25% BSA, serial dilutions of purified Fab were distributed to wells in duplicates, and plates were incubated for 2 h at room temperature. The plates were then washed four times with BBS and incubated for 1 h at room temperature with affinity-purified enzyme-labeled goat anti-human IgG.

ELISA for the detection of anti-prothrombin activity was similar to the anti-β2GP-1 ELISA assay with some modification. Briefly, wells were coated with 10 μg/ml prothrombin (Enzyme Research Laboratories) in calcium buffer and blocked with 0.25% gelatin in calcium buffer.

The purified Fab were also used to study the binding specificities of the mAbs to a panel of unrelated Ags, including: chicken OVA, collagen, and ssDNA. All Ags were used at 10 μg/ml, and ELISA was performed as described for β2GP-1 and prothrombin.

Purified mAb Fab and control proteins (150 ng except for mAb P13, which was 100 ng) were loaded into 7% Tris-acetate gel for nonreducing SDS-PAGE analysis (NOVEX, San Diego, CA). After electrophoresis, the gel was analyzed by silver staining for the detection of Fab protein bands according to the manufacturer’s instructions (Pierce, Rockford, IL).

The binding affinity of each mAb was determined by Ag inhibition, with each mAb used at a concentration that gave 50% maximal binding. Diluted mAbs (mAbs P11, 20 μg/ml; B14, 10 μg/ml; B22, 20 μg/ml; B27, 10 μg/ml; P13, 30 μg/ml) were then preincubated for 2 h at room temperature with an equal volume of buffer or increasing concentration of β2GP-1 or prothrombin (100, 200, 400, 600, and 800 μg/ml). The amount of free mAb in the Ab inhibitor mixtures were then measured in an anti-β2GP-1 or anti-prothrombin ELISA using Ag-precoated plates. The average mAb affinity was calculated according to previously described method (26).

Sequencing was performed on purified dsDNA by using previously published sequencing primers (25). Sequence data were compiled and analyzed using the Basic Local Alignment Search Tool (BLAST) (27).

After panning against β2GP-1, three clones were isolated from the κ library while none was isolated from the λ library. For prothrombin, two clones with anti-prothrombin activity were recovered from the κ library alone. Restriction analysis of their DNA revealed that all contained both the H and L chain inserts.

To study the binding property of these clones, affinity-purified Fab were analyzed against β2GP-1, prothrombin, and other unrelated Ag. As shown in Fig. 1, all three β2GP-1-enriched clones (termed mAbs B14, B22, B27) reacted with β2GP-1. Of these, two (mAbs B14 and B22) reacted with CL when complexed with β2GP-1 (data not shown). All three clones did not react with four unrelated Ags, including collagen, OVA, prothrombin, and ssDNA (data not shown). In Fig. 2, both prothrombin-selected clones (termed mAb P13 and P11) reacted with prothrombin. When tested against four unrelated Ags (β2GP-1, collagen, OVA, and ssDNA), P11 reacted strongly with β2GP-1 (Fig. 1), weakly with OVA and ssDNA, but not at all with collagen (data not shown). Fig. 3 shows that each of the mAb Fab displayed the expected 47-kDa Fab band on silver staining of SDS-PAGE gel.

FIGURE 1.

Binding specificity of affinity-purified monoclonal Fab mAbs B14, B22, B27, and P11. mAbs were analyzed against β2GP-1. Bars represent mean net OD ± SD readings of duplicate samples.

FIGURE 1.

Binding specificity of affinity-purified monoclonal Fab mAbs B14, B22, B27, and P11. mAbs were analyzed against β2GP-1. Bars represent mean net OD ± SD readings of duplicate samples.

Close modal
FIGURE 2.

Binding specificity of affinity-purified monoclonal Fab P11 and P13. mAbs were analyzed against prothrombin. Bars represent mean net OD ± SD readings of duplicate samples.

FIGURE 2.

Binding specificity of affinity-purified monoclonal Fab P11 and P13. mAbs were analyzed against prothrombin. Bars represent mean net OD ± SD readings of duplicate samples.

Close modal
FIGURE 3.

SDS-PAGE of mAbs, in silver staining. Lane 1, m.w. marker; lane 2–7, nonreduced mAbs B14, B22, B27, P11, P13, and a control Fab clone isolated from a normal human lymphocyte library; lane 8–9, nonreduced human Fab and IgG1 standard used as positive controls. Arrow indicates the position of the 47-kDa Fab band.

FIGURE 3.

SDS-PAGE of mAbs, in silver staining. Lane 1, m.w. marker; lane 2–7, nonreduced mAbs B14, B22, B27, P11, P13, and a control Fab clone isolated from a normal human lymphocyte library; lane 8–9, nonreduced human Fab and IgG1 standard used as positive controls. Arrow indicates the position of the 47-kDa Fab band.

Close modal

Binding specificity was confirmed by demonstration that soluble β2GP-1 and/or prothrombin inhibited the mAb interactions with immobilized β2GP-1/prothrombin as previously described (26). As shown in Figs. 4 and 5, four Fab clones were specific for β2GP-1 while two clones were reactive with prothrombin. The reactivity of mAb P11 to β2GP-1 and prothrombin was inhibited by soluble Ags, suggesting that mAb P11 may recognize an epitope shared by prothrombin and β2GP-1. The inhibitions range from 23 to 61% and are consistent with the low affinity of anti-β2GP-1 Abs, which remains free in the presence of 200 μg/ml of β2GP-1 in the plasma (28). The binding affinities (Kd) were estimated from the inhibition ELISAs and are shown in Table I.

FIGURE 4.

Competitive inhibition of binding of mAbs B14, B22, and B27 to β2GP-1 and representative mean OD readings of duplicate samples.

FIGURE 4.

Competitive inhibition of binding of mAbs B14, B22, and B27 to β2GP-1 and representative mean OD readings of duplicate samples.

Close modal
Table I.

Ig gene usage of five aPL monoclonal Fab from a patient with Ox-LDL

mAbKdVH germlineDJHVK germlineJK
PutativeHomology (%)PutativeHomology (%)
B14 7.0 × 10−5VH-26 (VH3) 92 21/07 A20 (VK1) 96 
B22 6.0 × 10−56-IG1 (VH6) 94 d5r, 21/10r A20 (VK1) 98 
B27 1.5 × 10−66-IG1 (VH6) 94 d5r, 21/10r L15 (VK1) 96 
P11 1.6 × 10−6 Ma VH4.33 (VH4) 93 23/07, 21/10r,     
 3.2 × 10−6 Mb   22/12, Dxp’1 A30 (VK1) 95 
P13 1.4 × 10−66-IG1 (VH6) 94 d5r, 21/10r     
    dn4 L5 (VK1) 93 
mAbKdVH germlineDJHVK germlineJK
PutativeHomology (%)PutativeHomology (%)
B14 7.0 × 10−5VH-26 (VH3) 92 21/07 A20 (VK1) 96 
B22 6.0 × 10−56-IG1 (VH6) 94 d5r, 21/10r A20 (VK1) 98 
B27 1.5 × 10−66-IG1 (VH6) 94 d5r, 21/10r L15 (VK1) 96 
P11 1.6 × 10−6 Ma VH4.33 (VH4) 93 23/07, 21/10r,     
 3.2 × 10−6 Mb   22/12, Dxp’1 A30 (VK1) 95 
P13 1.4 × 10−66-IG1 (VH6) 94 d5r, 21/10r     
    dn4 L5 (VK1) 93 
a

Kd of anti-β2 GP-1 reactivity of mAb P11.

b

Kd for anti-II response of mAb P11.

Sequence analysis of the H and L chain V regions of all Fab clones (Figs. 6 and 7) revealed that B22, B27, and P13 shared an almost identical rearranged VH6-D-J gene segment termed Humha622. It consists of VH6, D5, and D21/10 in the reverse orientation, and JH4 (Fig. 6,B). The putative D5-encoded segment may derive from DN4. The VH sequences for B22 and B27 are identical and are represented by ha622 (Fig. 6,B); they differed from that of P13, denoted as ha613 in Fig. 6,B, by one silent change in the framework (FR) 1. However, these three clones used different L chains termed Humka122, Humka127, and Humka113, respectively (denoted as ka122 [B], ka127 [C], and ka113 [E] in Fig. 7); each employed different members of the Vk1 family.

FIGURE 6.

The nucleotide and deduced amino acid sequences of Ab H chain V regions B14 (A), B22 and P13 (B), and P11 (C), which are designated Humha314, Humha622, Humha613, and Humha411 and are abbreviated ha314, ha622, ha613, and ha411, respectively. The H chain of B27 is identical with that of B22 and thus is represented by ha622; ha613 differs from ha622 by a single silent base and is given for this region only. The putative corresponding germline gene sequences are included for comparison (293031 ). In each panel, the complete nucleotide and amino acid sequences of a H chain are given, while the corresponding germline sequences (and other related H chain sequence) are given only at the positions where they differ from VH cDNA in the overlapping regions. Dashes denote the identities, the PCR primers are underlined. The CDRs are indicated, and D region and JH genes of all clones are included.

FIGURE 6.

The nucleotide and deduced amino acid sequences of Ab H chain V regions B14 (A), B22 and P13 (B), and P11 (C), which are designated Humha314, Humha622, Humha613, and Humha411 and are abbreviated ha314, ha622, ha613, and ha411, respectively. The H chain of B27 is identical with that of B22 and thus is represented by ha622; ha613 differs from ha622 by a single silent base and is given for this region only. The putative corresponding germline gene sequences are included for comparison (293031 ). In each panel, the complete nucleotide and amino acid sequences of a H chain are given, while the corresponding germline sequences (and other related H chain sequence) are given only at the positions where they differ from VH cDNA in the overlapping regions. Dashes denote the identities, the PCR primers are underlined. The CDRs are indicated, and D region and JH genes of all clones are included.

Close modal
FIGURE 7.

The nucleotide and deduced amino acid sequences of Ab L chain V regions: (A) B14, (B) B22, (C) B27, (D) P11 and (E) P13 which are designated Humka114, Humka122, Humka127, Humka111 and Humka113 and are abbreviated ka114, ka122, ka127, ka111 and ka113. The putative corresponding germline L chain V gene sequences are included for comparison (32333435 ). In each panel, the complete nucleotide and amino acid sequences of a L chain are given, while the germline sequence is given only at the positions where it differs from the L chain in the overlapping regions. Bars denote the identities, the PCR primers are underlined, and the CDRs are indicated.

FIGURE 7.

The nucleotide and deduced amino acid sequences of Ab L chain V regions: (A) B14, (B) B22, (C) B27, (D) P11 and (E) P13 which are designated Humka114, Humka122, Humka127, Humka111 and Humka113 and are abbreviated ka114, ka122, ka127, ka111 and ka113. The putative corresponding germline L chain V gene sequences are included for comparison (32333435 ). In each panel, the complete nucleotide and amino acid sequences of a L chain are given, while the germline sequence is given only at the positions where it differs from the L chain in the overlapping regions. Bars denote the identities, the PCR primers are underlined, and the CDRs are indicated.

Close modal

The mAb B14 employed the VH-26 VH3 gene termed Humha314 and the A20 Vk1 gene termed Humka114 (Figs. 6,A and 7A). Of note, the invariant tryptophan residue that represents the beginning of the fourth FR in ha314 is absent (Fig. 6A). To the best of our knowledge, this is the first VH sequence of a functional Ab that has no invariant tryptophan. Finally, clone P11 used the VH4.33 gene, termed Humha411, and the A30 Vk1 gene, termed Humka111 (Figs. 6 C and 7D).

It was difficult to discern the closest germline D gene used in all five clones because of extensive modification, but there appears to be certain germline genes that could have been employed by these clones as shown in Fig. 6 and Table I. As for the JH gene usage, the mAbs P11 and B14 employed JH1 while others used JH4 gene segments (Fig. 6).

There was no restriction in Jk usage in the clones analyzed (Table I). ka113 used Jk1, ka114 used Jk2, ka122 and ka111 employed Jk3, and ka127 used Jk4. The Jk2 employed by ka114 and the Jk3 in ka111 each contained two mutations (Fig. 7, A and D).

A comparison of the V gene-encoded regions in all five clones with both GenBank and EMBL databases as well as all published sequences revealed that the expressed VH showed a range of 92 to 94% homology with their nearest germline genes for an average of 93.4% (Table I) (29, 30, 31, 32, 33, 34, 35). When compared with these putative germline counterparts, the replacement to silent changes (R/S) in the complementarity-determining regions (CDRs) was 5.0 for mAbs B14, B22, B27, and P13 and <1.0 for mAb P11. In contrast, the R/S ratios in the FRs was 4.0 for mAb B14 and <1.4 for mAbs B22, B27, P11 and P13.

Sequence analysis of the L chains of all five clones revealed significant mutations that range from 6 to 20 nt per L chain V region, which results in a mutation frequencies of 2.1–7.0%. In Fig. 7, the germline nucleotide sequences are included for comparison. When compared with their germline counterparts, the R/S ratios in the CDRs was 2.0 for mAb B22, 4.0 for mAbs B14 and B27. This ratio was 8.0 and 6.0, respectively, for mAbs P11 and P13. In contrast, the R/S ratios in the FRs was 3.0 for mAb P11 and <0.5 for mAbs B14, B22, B27, and P13.

In an attempt to define the structural basis of anti-β2GP-1 and anti-prothrombin activities obtained from a coronary artery bypass patient, we employed the phage display method to isolate and analyze the structural features of five monoclonal IgG aPL Fab. The results show that all three β2GP-1-enriched clones, mAbs B14, B22, and B27, specifically recognized the plasma cofactor, β2GP-1 (Fig. 1). The affinities of these three clones range from 7 × 10−5 to 1.5 × 10−6 M and are comparable to the reported Kd values of serum anti-β2GP-1 Abs ranging from 10−5 to 3.4 10−6 M (14, 36). The mAb B27 had the highest affinity of the three β2GP-1 binding clones and was the most specific Ab.

Both prothrombin-enriched clones (mAbs P11 and P13) reacted with prothrombin (Fig. 2). Intriguingly, mAb P11 also reacted strongly with β2GP-1, suggesting that P11 may recognize an epitope shared by prothrombin and β2GP-1. It is possible that similar dual-reactive autoantibodies may be present in APS patients. In the future, it would be important to study the presence of mAb P11-like aPL in APS patients and the role of such aPL in APS pathogenesis. To this end, it will be required to first define the epitope recognized by mAb P11. The conversion of clone P11 into intact IgG secretor would allow the screening of β2GP-1 and prothrombin peptide libraries to identify the shared epitope. Subsequently, the peptide representing this shared epitope can then be used to screen patients serum samples for the putative P11-like aPL.

The mAbs B22, B27, and P13 shared identical VH6-D-J H chain but different L chains. The first two mAbs recognized β2GP-1, while P13 bound to prothrombin, suggesting that while VH6-D-J gene may have an intrinsic binding affinity for β2GP-1, certain L chain pairings apparently influence that binding. To decipher the role of L chains in binding to β2GP-1, we compared their amino acid sequences. Fig. 8 shows that the two β2GP-1-reactive L chains, ka122 and ka127, are not more similar to each other than to the prothrombin-reactive ka13.

FIGURE 8.

The deduced amino acid sequence of clone P13, B22, and B27 L chain cDNA V regions, designated Humka113, abbreviated ka113; Humka122, abbreviated ka122; and Humka127, abbreviated ka127, respectively. The complete amino acid sequence of ka113 are given, while all others are given only at the positions where they differ from the sequence of ka113. Dashes denote identities, while the CDRs are indicated. The underlined amino acid sequence is coded by the PCR primer.

FIGURE 8.

The deduced amino acid sequence of clone P13, B22, and B27 L chain cDNA V regions, designated Humka113, abbreviated ka113; Humka122, abbreviated ka122; and Humka127, abbreviated ka127, respectively. The complete amino acid sequence of ka113 are given, while all others are given only at the positions where they differ from the sequence of ka113. Dashes denote identities, while the CDRs are indicated. The underlined amino acid sequence is coded by the PCR primer.

Close modal

The H chains of mAbs B22, B27, P13, and B14 derive from VH6/V6–1 and VH26/V3–23 genes segments, respectively. These VH gene segments belong to a set of VH genes that have been shown to be preferentially expressed in ontogeny (37, 38). Previously, analysis of several natural autoantibodies derived from normal individuals revealed usage of the same restricted set of V genes. In this context, the natural IgM autoantibody Kim 13.1, which is encoded by the 51P1 gene in germline configuration, displays anti-CL and rheumatoid factor activity (39). Taken together, these data suggest that some aPL autoantibodies may arise from natural autoantibodies.

The mAb P11 used the A30 Vk1 L chain, which is rarely employed in the functional Ig Vk gene repertoire. In contrast, it was reported recently that in systemic lupus erythematosus (SLE) patients A30 is rearranged to the Jk2 gene and encodes the nephritogenic anti-DNA Ab L chain (35, 40). In addition, several of the presently characterized aPL-associated V genes have been shown to encode diagnostic anti-DNA autoantibodies in SLE (Table II) (40, 41, 42, 43, 44). In particular, VH26 and VH6 encode, respectively, the H chains of the 18/2 and the A10 anti-DNA Ab; A20, A30, and L5 encode the L chain of the III-2R, SC17, and NE-3 anti-DNA autoantibodies, respectively. The meaning of these findings are not clear. However, the extensive overlap of the V gene usage in the aPL with that of characteristic anti-DNA autoantibodies in SLE suggests that some aPL in SLE patients may arise as the byproducts of receptor editing of autoreactive B cells, in which one of the original H and L chain V regions is replaced with a newly rearranged V gene (45, 46). In support of this, sequence analysis of three monoclonal LAC Abs revealed extensive overlap in the Ig V genes with anti-DNA Abs found in lupus patients (47).

Table II.

aPL share Ig V genes with other anti-DNA autoantibodies in SLE

Germline V GeneaaPLPatient-Derived Autoantibodyb
NameHomology (%)NameHomology (%)
VH26 VH3 B14 92 18/2 IgM anti-DNA 100 
4.33 VH4 P11 93 None n/a 
VH6 B22 94 A10 IgM anti-DNA 99 
A20 VK1 B14, B22 96, 98 III-2R IgM anti-DNA 100 
A30 VK1 P11 95 SC17 IgG anti-DNA 98 
L5 VK1 P13 93 NE-13 IgM anti-DNA 100 
L15 VK1 B27 96 None n/a 
Germline V GeneaaPLPatient-Derived Autoantibodyb
NameHomology (%)NameHomology (%)
VH26 VH3 B14 92 18/2 IgM anti-DNA 100 
4.33 VH4 P11 93 None n/a 
VH6 B22 94 A10 IgM anti-DNA 99 
A20 VK1 B14, B22 96, 98 III-2R IgM anti-DNA 100 
A30 VK1 P11 95 SC17 IgG anti-DNA 98 
L5 VK1 P13 93 NE-13 IgM anti-DNA 100 
L15 VK1 B27 96 None n/a 
a

The references for germline Ig V genes are VH26 (29), 4.33 (31), VH6 (30), A20 (34), A30 (35), L5 (33), and L15 (32).

b

The references for patient-derived autoantibodies are 18/2 (41), A10 (42), III-2R (43), SC17 (40), and NE13 (44).

In conclusion, we isolated five mAbs aPL representing two of the major groups of autoantibodies found in APS patients. The structural features of these Abs revealed that certain H and L chain combinations may be important in the development of aPL reactivity.

FIGURE 5.

Competitive inhibition of binding of mAbs P11 and P13 to prothrombin and to β2GP-1 (mAb P11) and representative mean OD readings of duplicate samples.

FIGURE 5.

Competitive inhibition of binding of mAbs P11 and P13 to prothrombin and to β2GP-1 (mAb P11) and representative mean OD readings of duplicate samples.

Close modal
1

This work is supported by National Institutes of Health Grant 7K14HL03523-03.

3

Abbreviations used in this paper: aPL, antiphospholipid Abs; aCL, anti-cardiolipin Abs; APS, antiphospholipid syndrome; LAC, lupus anticoagulant;; SLE, systemic lupus erythematosus; β2GP-1, β2-glycoprotein-1; CL, cardiolipin; LDL, low density lipoprotein; Ox-LDL, oxidized low density lipoprotein; PL, phospholipid; PS, phosphatidylserine; BBS, borate-buffered saline; FR, framework region; R/S, replacement to silent; CDR, complementarity-determining region.

1
Harris, E. N., A. E. Gharavi, M. L. Boey, B. M. Patel, C. G. Mackworth-Young, S. Loizou, G. R. Hughes.
1983
. Anticardiolipin antibodies: detection by radioimmunoassay and association with thrombosis in systemic lupus erythematosus.
Lancet
2
:
1211
2
Alarcon-Segovia, D., M. Deleze, C. V. Oria, J. Sanchez-Guerrero, L. Gomez-Pacheco, J. Cabiedes, L. Fernandez, S. Ponce de Leon.
1989
. Antiphospholipid antibodies and the antiphospholipid syndrome in systemic lupus erythematosus: a prospective analysis of 500 consecutive patients.
Medicine
68
:
353
3
Khamashta, M. A., G. R. V. Hughes.
1994
. Antiphospholipid antibodies: a marker for thrombosis and recurrent abortion.
Clin. Rev. Allergy
12
:
287
4
Roubey, R. A. S..
1994
. Autoantibodies to phospholipid-binding plasma proteins: a new view of lupus anticoagulants and other “antiphospholipid” autoantibodies.
Blood
84
:
2854
5
Galli, M., P. Comfurius, C. Maassen, H. C. Hemker, M. H. de Baets, P. J. C. van Breda-Vriesman, T. Barbui, R. F. A. Zwaal, E. M. Bevers.
1990
. Anticardiolipin antibodies (ACA) directed not to cardiolipin but to a plasma protein cofactor.
Lancet
335
:
1544
6
Matsurra, E., Y. Igarashi, M. Fujimoto, K. Ichikawa, T. Suzuki, T. Sumida, T. Yasuda, T. Koike.
1992
. Heterogeneity of anticardiolipin antibodies defined by the anticardiolipin cofactor.
J. Immunol.
148
:
3885
7
McNeil, H. P., R. J. Simpson, C. N. Chesterman, S. A. Krilis.
1990
. 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. USA
87
:
4120
8
Oosting, J. D., R. H. W. M. Derksen, I. W. G. Bobbink, T. M. Hackeng, B. N. Bouma, P. G. de Groot.
1993
. Antiphospholipid antibodies directed against a combination of phospholipids with prothrombin, protein C, or protein S: an explanation for their pathogenic mechanism?.
Blood
81
:
2618
9
Matsuda, J., M. Gotoh, N. Saitoh, K. Gohchi, M. Tsukamoto, T. Yamamoto.
1994
. Anti-annexin antibody in the sera of patients with habitual fetal loss or preeclampsia.
Thromb. Res.
75
:
105
10
Matsuura, E., Y. Igarashi, M. Fujimoto, K. Ichikawa.
1990
. Anticardiolipin cofactor(s) and differential diagnosis of autoimmune disease.
Lancet
336
:
177
11
Arvieux, J., B. Roussel, M. C. Jacob, M. G. Colomb.
1991
. Measurement of anti-phospholipid antibodies by ELISA using β2-glycoprotein I as an antigen.
J. Immunol. Methods
143
:
223
12
Viard, J. P., Z. Amoura, J. F. Bach.
1992
. Association of anti-β2 glycoprotein I antibodies with lupus-type circulating anticoagulant and thrombosis in systemic lupus erythematosus.
Am. J. Med.
93
:
181
13
Matsuura, E., Y. Igarashi, T. Yasuda, D. A. Triplett, T. Koike.
1994
. Anticardiolipin antibodies recognize β2-glycoprotein I structure altered by interacting with an oxygen modified solid phase surface.
J. Exp. Med.
179
:
457
14
Roubey, R. A. S., R. A. Eisenberg, M. F. Harper, J. B. Winfield.
1995
. “Anticardiolipin” autoantibodies recognize β2-glycoprotein I in the absence of phospholipid: importance of Ag density and bivalent binding.
J. Immunol.
154
:
954
15
Cabiedes, J., A. R. Cabral, D. Alarc-n-Segovia.
1995
. Clinical manifestations of the antiphospholipid syndrome in patients with systemic lupus erythematosus associate more strongly with anti-β2-glycoprotein-I than with antiphospholipid antibodies.
J. Rheumatol.
22
:
1899
16
Arvieux, J., L. Darnige, C. Caron, G. Reber, J. C. Bensa, M. G. Colomb.
1995
. Development of an ELISA for autoantibodies to prothrombin showing their prevalence in patients with lupus anticoagulants.
Thromb. Haemost.
74
:
1120
17
Fleck, R. A., S. I. Rapaport, L. V. Rao.
1988
. Anti-prothrombin antibodies and the lupus anticoagulant.
Blood
72
:
512
18
Permpikul, P., L. V. M. Rao, S. I. Rapaport.
1994
. Functional and binding studies of the roles of prothrombin and β2-glycoprotein I in the expression of lupus anticoagulant activity.
Blood
83
:
2878
19
Witztum, J. L., D. Steinberg.
1991
. Role of oxidized low density lipoprotein in atherogenesis.
J. Clin. Invest.
88
:
1785
20
Aviram, M..
1993
. Modified forms of low density lipoprotein and atherosclerosis.
Atherosclerosis
98
:
1
21
Ploz, E., G. M. Kostner.
1979
. The binding of β2-glycoprotein-I to human serum lipoproteins: distribution among density fractions.
FEBS Lett.
102
:
183
22
H÷rkk÷, S., E. Miller, E. Dudl, P. Reaven, L. K. Curtiss, N. J. Zvaifler, R. Terkeltaub, S. S. Pierangeli, D. W. Branch, W. Palinski, J. L. Witztum.
1996
. Antiphospholipid antibodies are directed against epitopes of oxidized phospholipids: recognition of cardiolipin by monoclonal antibodies to epitopes of oxidized low density lipoprotein.
J. Clin. Invest.
98
:
815
23
Vaarala, O., G. Alfthan, M. Jauhiainen, M. Leirisalo-Repo, K. Aho, T. Palusuo.
1993
. Crossreaction between antibodies to oxidised low-density lipoprotein and to cardiolipin in systemic lupus erythematosus.
Lancet
341
:
923
24
Barbas, C. F., A. S. Kang, R. A. Kerner, S. J. Benkovic.
1991
. Assembly of combinatorial antibody libraries on phage surfaces: the gene III site.
Proc. Natl. Acad. Sci. USA
88
:
7978
25
Ditzel, H. J., K. Itoh, D. R. Burton.
1996
. Determinants of polyreactivity in a large panel of recombinant human antibodies from HIV-1 infection.
J. Immunol.
157
:
739
26
Friguet, B., A. F. Chaffotte, L. Djavadi-Ohaniance, M. E. Goldberg.
1985
. Measurements of the true affinity constant in solution of antigen-antibody complexes by enzyme-linked immunosorbent assay.
J. Immunol. Methods
77
:
305
27
Altschul, S. F., W. Gish, W. Miller, E. W. Myers, D. J. Lipman.
1990
. Basic local alignment search tool.
J. Mol. Biol.
215
:
403
28
Roubey, R. A. S..
1996
. Immunology of the antiphospholipid antibody syndrome.
Arthritis Rheum.
39
:
1444
29
Mathyssens, G., T. H. Rabbits.
1980
. Structure and multiplicity of genes for the human immunoglobulin heavy chain variable region.
Proc. Natl. Acad. Sci. USA
65
:
6561
30
Berman, J. E., S. J. Mellis, R. Pollock, C. L. Smith, H. Suh, B. Heinke, C. Kowal, U. Surti, L. Chess, C. R. Cantor, F. W. Alt.
1988
. Content and organization of the human Ig VH locus: definition of three new VH families and linkage to the Ig CH locus.
EMBO J.
7
:
727
31
Weng, N., J. G. Snyder, L. Yu-Lee, D. M. Marcus.
1992
. Polymorphism of human immunoglobulin VH4 germ-line genes.
Eur. J. Immunol.
22
:
1075
32
Bentley, D. L., T. H. Rabbitts.
1983
. Evolution of immunoglobulin V genes: evidence indicating that recently duplicated human V(k) sequences have diverged by gene conversion.
Cell
32
:
181
33
Pech, M., H. Jaenichen, H. Poblenz, P. S. Neumaler, H. Klobeck, H. G. Zachau.
1984
. Organization and evolution of a gene cluster for human immunoglobulin variable regions of the kappa type.
J. Mol. Biol.
176
:
189
34
Cox, J. P. L., I. M. Tomlinson, G. Winter.
1994
. A directory of human germ-line Vχ segments reveals a strong bias in their usage.
Eur. J. Immunol.
24
:
827
35
Suzuki, N., T. Harada, S. Mihara, T. Sakane.
1996
. Characterization of a germline Vk gene encoding cationic anti-DNA antibody and role of receptor editing for development of the autoantibody in patients with systemic lupus erythematosus.
J. Clin. Invest.
98
:
1843
36
Tincani, A., L. Spatola, E. Prati, F. Allegri, P. Ferremi, R. Cattaneo, P. Meroni, G. Balestrieri.
1996
. The anti-β2-clycoprotein I activity in human anti-phospholipid syndrome sera is due to monoreactive low-affinity autoantibodies directed to epitopes located on native β2-glycoprotein I and preserved during species’ evolution.
J. Immunol.
157
:
5732
37
Schroeder, H. W., Jr, J. L. Hillson, R. M. Perlmutter.
1987
. Early restriction of the human antibody repertoire.
Science
238
:
791
38
Schroeder, H. W., Jr, J. Y. Wang.
1990
. Preferential utilization of conserved immunoglobulin heavy chain variable gene segments during human fetal life.
Proc. Natl. Acad. Sci. USA
87
:
6146
39
Siminovitch, K. A., V. Misener, P. C. Kwong, P.-M. Yang, C. A. Laskin, E. Cairns, D. Bell, L. A. Rubin, P. P. Chen.
1990
. A human anti-cardiolipin autoantibody is encoded by developmentally restricted heavy and light chain variable region genes.
Autoimmunity
8
:
97
40
Harada, T., N. Suzuki, Y. Mizushima, T. Sakane.
1994
. Usage of a novel class of germ-line Ig variable region gene for cationic anti-DNA autoantibodies in human lupus nephritis and its role for the development of the disease.
J. Immunol.
153
:
4806
41
Dersimonian, H., R. S. Schwartz, K. J. Barrett, B. D. Stollar.
1987
. Relationship of human variable region heavy chain germ-line genes to genes encoding anti-DNA autoantibodies.
J. Immunol.
139
:
2496
42
Logtenberg, T., F. M. Young, J. H. Van Es, F. H. J. Gmelig-Meyling, F. W. Alt.
1989
. Autoantibodies encoded by the most Jh-proximal human immunoglobulin heavy chain variable region gene.
J. Exp. Med.
170
:
1347
43
Manheimer-Lory, A., J. B. Katz, M. Pillinger, C. Ghossein, A. Smith, B. Diamond.
1991
. Molecular characteristics of antibodies bearing an anti-DNA-associated idiotype.
J. Exp. Med.
174
:
1639
44
Hirabayashi, Y., Y. Munakata, O. Takai, S. Shibata, T. Sasaki, H. Sano.
1993
. Human B-cell clones expressing lupus nephritis-associated anti-DNA idiotypes are preferentially expanded without somatic mutation.
Scand. J. Immunol.
37
:
533
45
Gay, D., T. Saunders, S. Camper, M. Weigert.
1993
. Receptor editing: an approach by autoreactive B cells to escape tolerance.
J. Exp. Med.
177
:
999
46
Chen, C., M. Z. Radic, J. Erikson, S. A. Camper, S. Litwin, R. R. Hardy, M. Weigert.
1994
. Deletion and editing of B cells that express antibodies to DNA.
J. Immunol.
152
:
1970
47
Lai, C.-J., J. Rauch, C. S. Cho, Y. Zhao, R. U. Chukwuocha, P. P. Chen.
1998
. Immunological and molecular analysis of three monoclonal lupus anticoagulant antibodies from a patient with systemic lupus erythematosus.
J. Autoimmunity
11
:
39