| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Division of Rheumatology, Department of Medicine, Research Institute of the McGill University Health Center, McGill University, Montreal, Quebec, Canada; and
Section of Nephrology, Department of Medicine, University of Chicago, Chicago, IL 60637
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2-glycoprotein I (2GPI) or prothrombin (PT), two PL-binding proteins. We have previously shown that 2GPI and 2GPI-dependent aPL bind specifically to apoptotic, but not viable, thymocytes. In this study, we demonstrate that PT, like 2GPI, binds selectively to the surface of apoptotic, but not viable, Jurkat cells. Furthermore, PT supports the binding of systemic lupus erythematosus-derived polyclonal and murine monoclonal LA Ab to apoptotic cells. Two LA mAb, which differed dramatically in their relative affinities for PT, were studied. Although one mAb (29J3-62) had a high affinity for PT alone, the other (29I4-24) showed minimal reactivity with PT alone and required PL for elevated binding. Monovalent fragments of 29I4-24 reacted with PL-bound PT with high affinity, suggesting that this mAb recognizes a PL-dependent epitope. Despite these differences, PT-dependent binding of both mAb to apoptotic cells was 30-fold greater than that to viable cells. Moreover, binding of PT to apoptotic cells was, itself, increased in the presence of bivalent, but not monovalent, forms of either mAb. In summary, our data demonstrate the following: 1) specific binding of PT to apoptotic cells, an effect enhanced by PT-dependent LA Ab; 2) heterogeneity of PT-dependent LA Ab; and 3) potential pathogenicity of Ab of either low or high affinity for PT. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2-glycoprotein I (2GPI), annexin V, and prothrombin (PT) (3, 4, 5, 6, 7). Recent work has shown that certain aPL recognize these PL-binding proteins alone, whereas other aPL require the presence of PL for this recognition (8, 9, 10). Lupus anticoagulant (LA) Ab are a subset of aPL. Although LA Ab are defined by their ability to prolong clotting times in in vitro coagulation assays, these Ab are associated with thrombosis in vivo (11). LA Ab fall into two major groups: 2GPI-dependent (12) and PT-dependent (6, 10, 13). Each of these groups of LA Ab is thought to affect the interaction of the relevant protein with PL in vitro and, more importantly, in vivo. Despite these advances in our understanding, the identity of the natural target(s) and/or immunogen(s) for aPL and the sequence of events underlying their induction remain unclear. Increasing evidence suggests that apoptotic cells may be involved in the initiation and/or maintenance of autoimmunity (14, 15, 16). Apoptotic cells have been shown to express autoantigens that are specifically targeted by autoantibodies found in SLE and in APS (17, 18, 19, 20). Indeed, anionic PL, such as phosphatidylserine (PS) and cardiolipin, which are not normally present on the surface of viable cells, are translocated to the external surface of the plasma membrane of cells undergoing apoptosis (21, 22, 23). We have previously demonstrated that 2GPI binds to the surface of apoptotic, but not viable, cells, and that the interaction of 2GPI with the surface of apoptotic cells generates epitopes that are both antigenic for patient-derived aPL (24) and immunogenic in normal mice (25).
In the present study, we determine whether the paradigm for recognition of apoptotic cells by aPL can be extended from 2GPI-dependent aPL to other types of aPL. Specifically, we ask whether apoptotic cells are also antigenic targets for PT-dependent aPL. We demonstrate that PT, a PL-binding protein known to interact with PS, binds to the surface of apoptotic, but not viable, Jurkat cells, and is specifically recognized by human SLE-derived polyclonal and murine monoclonal LA Ab. Combined with our previously published results on 2GPI-dependent LA Ab, these data demonstrate that both major groups of LA Ab recognize PL-binding proteins bound to the surface of apoptotic cells. Furthermore, we show that the binding of PT to apoptotic cells is enhanced in the presence of PT-dependent LA Ab, suggesting that LA Ab can modulate the binding of PT to apoptotic cells. The finding that apoptotic cell-bound PT constitutes a major antigenic target of aPL is consistent with the concept that apoptotic cells are important in the initiation and pathogenesis of APS.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Unless stated otherwise, all chemicals were obtained commercially and used without further purification. Purified human apolipoprotein H (2GPI) was obtained from Crystal Chem (Chicago, IL). 1,2-Dioleoyl-sn-glycero-3-phosphocholine (DOPC), 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC), 1,2-dioleoyl-sn-glycero-3-[phospho-L-serine] (DOPS), bovine heart cardiolipin (CL), and egg phosphatidylethanolamine were obtained from Avanti Polar Lipids (Alabaster, AL). Purified human PT was generously provided by Dr. K. Mann (University of Vermont, Burlington, VT) or purchased from Hematologic Technologies (Essex Junction, VT), and recombinant human annexin V was generously provided by Dr. J. Tait (University of Washington, Seattle, WA).
IgG isolated from SLE patient plasma
IgG was isolated from plasma (3 ml) of individual patients (coded 1–6) with SLE and confirmed LA activity, as previously described (10). Characteristics of these patients and IgG fractions have been described elsewhere (10). All patients fulfilled the American College of Rheumatology criteria for the classification of SLE (26). Control IgG was isolated in the same manner from freshly reconstituted Verify 1 (normal human coagulation control plasma; Organon Teknika, Durham, NC). IgG fractions were assayed for reactivity to PL-bound PT by ELISA on plates coated with phosphatidylethanolamine, as previously described (10).
Murine mAb
Murine hybridoma mAb 29J3-62.2.2 and 29I4-24.3.2 (referred to hereafter as 29J3-62 and 29I4-24, respectively) were generated from specific pathogen-free female BALB/c mice (Harlan Sprague-Dawley, Indianapolis, IN) immunized i.v. with human PT in combination with either DOPS (29J3-62) or DOPS/DOPC (29I4-24) in HEPES buffer (10 mM HEPES and 150 mM NaCl (pH 7.2)), containing 1.25 mM CaCl2. PL for immunization was prepared as previously described (27), and mice were given four biweekly injections of 50 µg of PL mixed with 10 µg of human PT. Mice producing high serum levels of anti-PT and LA Ab were boosted i.v. with Ag 3–5 days before cell fusion, and their splenocytes were fused with NSO murine myeloma cells, as previously described (28). The hybridomas obtained were screened for reactivity to PL (DOPS and DOPC) and PT individually by ELISA (25), and for LA activity, as described in LA activity. 29J3-62 and 29I4-24 were selected and paired for study because of their strong LA activity, but different reactivities with PT. A negative control mAb, 27D2-83.14 (referred to hereafter as 27D2-83), was derived from a mouse immunized with human PT, combined with DOPS/DOPC in the absence of CaCl2. 27D2-83 was selected as a random negative control mAb, because it was negative in all relevant assays. All three mAb were of the IgG1 subclass and were purified from culture supernatant using protein A affinity columns from Amersham Pharmacia Biotech (Uppsala, Sweden) (29J3-62) or RepliGen (Needham, MA) (29I4-24 and 27D2-83), according to the manufacturer’s instructions. The IgG was eluted from the column with 0.1 M (protein A column; Amersham Pharmacia Biotech) or 0.2 M (protein A column; RepliGen) glycine (pH 2.3). Following neutralization with 1 M Tris, the eluate was dialyzed against 0.01 M PBS (pH 7.3) for 72 h at 4°C. Purified IgG1 mAb was quantitated using the Micro BCA protein assay reagent kit (Pierce, Rockford, IL), and the purity of mAb preparations was assessed by SDS-PAGE.
F(ab')2 of the mAb were prepared using the ImmunoPure IgG1 Fab and F(ab')2 preparation kit (Pierce). Optimal enzymatic digestion times for each mAb were 46 h for 29J3-62 and 29I4-24, and 24 h for 27D2-83, both at 37°C. Monovalent Fab' were derived from F(ab')2 by reduction with 0.01 M L-cysteine (free base; Sigma-Aldrich, St. Louis, MO) for 2 h at 37°C, followed by alkylation with 0.15 M iodoacetamide (Sigma-Aldrich) for 30 min at 25°C (29). IgG that was both reduced and alkylated (IgG') was also prepared from each mAb to control for the effect of reduction and alkylation on Ab reactivity. The Ab and fragments were then dialyzed against PBS and stored at 4°C. All fragments were free of undigested IgG, Fc fragments, and F(ab')2 (for Fab' fractions), as determined by reducing and nonreducing SDS-PAGE.
Labeling of human annexin V and PT
Recombinant annexin V was biotinylated, using a modification of a previously described protocol (30). Briefly, 500 µl of annexin V (0.5 mg/ml in PBS) were mixed with 50 µl of NaHCO3 (1.0 M) and 50 µl of biotin-amidocaproate-N-hydroxysuccinimide ester (Sigma-Aldrich) (1.4 mM in DMSO), and incubated for 2 h at 25°C in the dark. The reaction was stopped by adding 50 µl of NH4Cl (1.0 M) and incubating for 10 min. The mixture was dialyzed against 0.01 M TBS (pH 7.4) for 72 h at 4°C.
Purified human PT, 2GPI, or human serum albumin (HSA) was labeled with FITC at a 13:1 FITC:protein molar ratio, as follows. Eight microliters of FITC Isomer I (Sigma-Aldrich) in DMSO (3.62 mg/ml) was added to 200 µl of PT (2 mg/ml in 0.05 M boric acid buffer (pH 9.2) containing 0.2 M NaCl). The mixture was incubated for 2 h at 25°C in the dark, followed by dialysis against PBS for 72 h at 4°C in the dark.
Cell culture and induction of apoptosis
The Jurkat human T cell leukemia line (ATCC TIB-152; American Type Culture Collection, Manassas, VA) was maintained in RPMI 1640 (Life Technologies, Rockville, MD), containing 10% FBS (Life Technologies), 1 mM sodium pyruvate, 2 mM L-glutamine, 10 mM HEPES, and 100 U/ml penicillin-streptomycin, and grown in humidified 5% CO2. Apoptosis was induced by transferring cells to FBS-free medium, containing 0.5% fatty acid-free BSA (Sigma-Aldrich) and 1 µg/ml staurosporine (Sigma-Aldrich), and incubating for 3 h. Viable cells were maintained in regular medium for the same duration, and used for comparison with apoptotic cells in all experiments.
Induction of apoptosis was confirmed by both fluorescence microscopy and flow cytometry. Apoptosis was detected using the cell-permeant DNA dye Hoechst 33342 (Molecular Probes, Eugene, OR), which stains the nuclei of apoptotic cells more brightly than the nuclei of viable cells, and biotinylated annexin V, which binds to PS on the surface of apoptotic cells. Briefly, cells were washed twice with ice-cold FACS buffer (10 mM HEPES, 140 mM NaCl, 1% fatty acid-free BSA, and 0.02% NaN3 (pH 7.4)), containing 2.5 mM CaCl2 (FACS-CaCl2 buffer). The cell pellet was then resuspended in 50 µl of FACS-CaCl2 buffer, containing biotinylated annexin V (1 µg/ml), and incubated for 15 min on ice. Bound annexin V was detected by incubation with ALEXA-conjugated streptavidin (5 µg/ml; Molecular Probes) for 20 min on ice. After washing, stained cells were incubated with 1 µg/ml of Hoechst 33342 for 10 min at 37°C and visualized on a fluorescence microscope (Diaplan; Leitz, Wetzlar, Germany), using the appropriate filters. Ethidium homodimer-1 (EtdHD; Molecular Probes) was added to the cell suspension at 1 µg/ml immediately before viewing to discriminate between cells with nonintact (postapoptotic or necrotic cells) and intact (early apoptotic or viable cells) cell membranes. Annexin V and EtdHD staining were also assessed by flow cytometry. Both staurosporine-treated (apoptotic) and untreated (viable) cell populations were stained as described above, and analyzed by a FACScan flow cytometer (BD Biosciences, San Jose, CA). Ten thousand gated events were counted for each sample, using the FL1 channel for FITC or ALEXA, and the FL2 channel for EtdHD.
Binding of FITC-conjugated PT (FITC-PT) to Jurkat cells
The binding of FITC-PT to the surface of apoptotic and viable cells was determined by flow cytometry, in the presence or absence of CaCl2. Cells were prepared as described above and washed twice with ice-cold FACS buffer (±2.5 mM CaCl2). Thirty microliters of FITC-PT (100 µg/ml) was added to the cell pellet, and the cells were incubated on ice for 1 h. Then, the cells were washed once with the appropriate FACS buffer, and EtdHD was added at a final concentration of 1 µg/ml immediately before cytometric analysis. FITC-conjugated HSA (FITC-HSA) was used as a negative control in these experiments.
FITC-PT binding to apoptotic cells was also evaluated in the presence of murine mAb. In these experiments, 30 µl of FITC-PT at various concentrations in FACS-CaCl2 was added to the cell pellet, and the cells were incubated for 20 min on ice. Then, 30 µl of purified murine mAb (40 µg/ml) was added to the cell suspension, and incubated for an additional 30 min on ice. The cells were then washed once with FACS-CaCl2 and analyzed as described above.
Binding of Ab to Jurkat cells
Binding of mAb to apoptotic and viable cells was determined by flow cytometry. Cells were prepared as described in Cell culture and induction of apoptosis and washed twice with ice-cold FACS-CaCl2 buffer. Thirty microliters of PT (concentrations ranging from 5 to 40 µg/ml) was added to the cell pellet, and the cells were incubated for 20 min on ice. Then, 30 µl of purified murine mAb (40 µg/ml) was added to the cell suspension, and incubated for an additional 30 min on ice. After washing twice, 100 µl of FITC-conjugated F(ab')2 of goat anti-mouse IgG and IgM H and L chains (IgG + IgM (H + L)) (7.5 µg/ml in FACS-CaCl2 buffer; Jackson ImmunoResearch Laboratories, West Grove, PA) was added to the cell pellet and incubated for 20 min on ice. In experiments using mAb fragments, the following concentrations, which provided the equivalent molarity of Ag binding sites for all samples, were used: IgG (40 µg/ml), IgG' (40 µg/ml), F(ab')2 (28 µg/ml), or Fab' (28 µg/ml). Binding of the fragments was detected by incubation with 100 µl of biotinylated goat-anti-mouse kappa (2 µg/ml in FACS-CaCl2 buffer; Southern Biotechnology Associates, Birmingham, AL) for 20 min on ice. Following two washes with FACS-CaCl2 buffer, cells were incubated with ALEXA-conjugated streptavidin (5 µg/ml in FACS-CaCl2 buffer; Molecular Probes) for 20 min on ice.
In experiments using human Ab, 30 µl of SLE plasma-derived IgG (500 µg/ml) was added in place of the murine mAb, and the secondary Ab was FITC-conjugated F(ab')2 of goat anti-human IgG + IgM (H + L) (Jackson ImmunoResearch Laboratories). Serum from an SLE patient (patient 7) with known high-titer anti-PT Ab and LA activity was used as a positive control in these experiments. All cells were washed once before analysis by flow cytometry, as described in Cell culture and induction of apoptosis.
LA activity
Murine mAb or purified human IgG were tested for LA activity using a dilute activated partial thromboplastin time assay (APTT), as previously described (10). APTT values were considered significantly prolonged if they exceeded the negative control by 
6 s. The presence of LA activity was confirmed by neutralization with hexagonal (II) phase phosphatidylethanolamine (10).
Anti-PT ELISA
Ab reactivity to human PT (anti-PT) was assessed by ELISA. Briefly, human PT was coated at 5 or 20 µg/ml (50 µl/well) in PBS in Immulon-2 plates (Dynex Technologies, Chantilly, VA) and dried for 16 h at 37°C. Both gamma-irradiated (Immulon-2) and nonirradiated (Immulon-1) (Dynex) plates gave equivalent results. The plates were blocked with PBS containing 0.3% (w/v) gelatin for 2 h at 4°C, and then washed three times with TBS. Purified mAb samples were added in duplicate to wells and incubated for 3 h at 25°C. Alkaline phosphatase-conjugated goat anti-mouse IgG (Southern Biotechnology Associates), diluted 1/1000 in PBS containing 0.4% BSA, was added and incubated for 16 h at 4°C. The plates were then developed using p-nitrophenol phosphate substrate at 37°C, and the OD405 was read in an ELISA reader (Model EAR, 400AT; SLT Labinstruments, Grödig, Austria). All samples were also tested in wells coated with 90 µg/ml gelatin, as a control for nonspecific binding.
Anti-PL/PT complex ELISA
Ab reactivity to complexes of PL and PT was determined using a modified anti-PT ELISA. DOPS/DOPC (25%/75% by weight) or DOPC was dried down from chloroform under nitrogen gas, resuspended in TBS at 1 mg/ml, and hydrated at 25°C for 1 h with intermittent vortexing. In certain experiments, DPPC was used in place of DOPC, as it resulted in lower nonspecific binding. DOPS/DOPC and DOPS/DPPC will hereafter be referred to as PSPC (25% PS and 75% phosphatidylcholine (PC) by weight), while DOPC and DPPC will be referred to as PC. PL preparations were diluted to 150 µg/ml in TBS. The PL was mixed with an equal volume of PT (30 µg/ml in TBS containing 10 mM CaCl2; final CaCl2 concentration of mixture was 5 mM), and incubated for 1 h at 25°C. ELISA plates (Immulon-2) were then coated with 50 µl/well of the PL/PT mixture and dried for 16 h at 37°C. The plates were blocked for 2 h at 4°C with TBS containing 5 mM CaCl2 (TBS-CaCl2) and 0.3% gelatin, and then washed three times with TBS-CaCl2. Purified mAb, diluted in TBS-CaCl2 containing 0.3% gelatin, were added and incubated for 3 h at 25°C. Following three washes with TBS-CaCl2, alkaline phosphatase-conjugated goat anti-mouse IgG, diluted in TBS-CaCl2 containing 0.4% BSA, was added and incubated for 16 h at 4°C. The plates were then developed and read as described above.
Reactivity of mAb fragments was also determined by this assay, with the following modifications. Bound fragments, or control undigested IgG or IgG', were detected using biotinylated goat anti-mouse kappa (Southern Biotechnology Associates) diluted 1/5000 in TBS-CaCl2 containing 0.4% BSA, and incubated for 16 h at 4°C. Following three washes with TBS-CaCl2, alkaline phosphatase-conjugated streptavidin (Zymed Laboratories, San Francisco, CA), diluted 1/1000 in TBS-CaCl2 containing 0.4% BSA, was added to the wells and incubated for 30 min at 25°C. The plates were then developed and read as described above.
Reactivity of patient or control sera with PL/PT complexes was evaluated as described in this section for the mAb, with the exception of the use of alkaline phosphatase-conjugated goat anti-human IgG (Fc-specific; 1/1000 dilution; Sigma-Aldrich) for detection of bound Ab. Sera were titrated to determine the dilution that gave 
50% of maximal binding (OD405 reading of 1.2). The following dilutions were used: 1/300 for patient 4, 1/100 for patient 2, 1/3200 for patient 7, and 1/200 for the normal control serum. To evaluate whether the LA mAb affected the binding of each serum to PL/PT-coated plates, purified mAb (0.19–200 µg/ml in TBS-CaCl2 containing 0.3% gelatin) or buffer alone was incubated in the coated and blocked plates for 2 h at 25°C. An irrelevant mAb (anti-2GPI mAb, 12A1-A17.3) was included in these experiments as a control for specificity (24). The plates were washed three times with TBS-CaCl2 and the appropriate dilution of serum was added and incubated for 3 h at 25°C. Bound Ab were detected as described above for the titration of the sera.
Competitive inhibition ELISA
The relative affinities of mAb 29J3-62 and 29I4-24 for soluble PT were determined using a competitive inhibition ELISA (31). Final mAb concentrations giving 50% maximal binding to PT (20 µg/ml PT) in the anti-PT ELISA described in Anti-PT ELISA were used in this competitive inhibition ELISA. Purified 29J3-62 or 29I4-24 (100 µl), diluted in PBS containing 0.1% fatty acid-free BSA (Sigma-Aldrich) to 0.024 and 0.48 µg/ml, respectively, was mixed with an equal volume of PT (at concentrations ranging from 4.1 x 10-7 to 1 x 103 µg/ml (0.0058 pM to 14.3 µM)) diluted in PBS containing 0.1% fatty acid-free BSA, and incubated with gentle shaking for 1 h at 37°C, followed by 16 h at 4°C, to ensure that binding equilibrium was reached. The samples were then incubated on PT-coated (20 µg/ml) ELISA plates for 3 h at 25°C and bound Ab was detected using biotinylated rabbit anti-mouse IgG + IgM (H + L) (Jackson ImmunoResearch Laboratories) diluted 1/1000 in PBS containing 0.4% BSA and incubated for 16 h at 4°C. After washing, alkaline phosphatase-conjugated streptavidin (Zymed), diluted 1/500 in PBS containing 0.4% BSA, was added to the wells and incubated for 30 min at 25°C. The plates were then developed and read as described above.
The reactivity of the mAb with complexes of PL and PT in solution was also assessed by an inhibition assay. Briefly, DOPS/DOPC or DOPC was prepared as described in Anti-PL/PT complex ELISA at 1 mg/ml in TBS, mixed with an equal volume of PT (at concentrations ranging from 4 x 10-6 to 200 µg/ml in TBS containing 0.2% fatty acid-free BSA and 10 mM CaCl2; final concentrations were 0.1% fatty acid-free BSA and 5 mM CaCl2), and incubated for 1 h at 25°C. The mixtures were then added to an equal volume of diluted purified mAb (in TBS containing 0.1% fatty acid-free BSA and 5 mM CaCl2), and preincubated as mentioned above. To prevent interaction between PL in suspension and coated PT, PL was removed from the mixtures by microcentrifugation at 11,600 x g for 30 min at 4°C. The supernatant was then tested for reactivity to PT-coated ELISA plates as described above.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Treatment of Jurkat cells with staurosporine for 3 h was highly efficient at inducing apoptosis, typically yielding up to 97% apoptotic cells, as visualized by fluorescence microscopy. Apoptotic cells displayed condensed and fragmented nuclei, as revealed by their bright staining with Hoechst dye (Fig. 1a, top left). Moreover, all Hoechst-positive cells bound annexin V (Fig. 1a, top right), indicating surface exposure of PS. Importantly, only 12% of Hoechst-positive cells also stained positively with EtdHD (data not shown), indicating that the majority (85%) of cells were in an early stage of apoptosis, preceding loss of membrane integrity. In contrast, 96% of untreated cells were viable, as assessed by dim staining with Hoechst dye (Fig. 1a, bottom left), negative staining for annexin V (Fig. 1a, bottom right), and negative staining with EtdHD (data not shown). Two percent of the untreated cells were apoptotic, as defined by bright Hoechst staining, and 2% were necrotic, as assessed by dim Hoechst staining and positive staining with EtdHD. Induction of apoptosis was also accompanied by a decrease in cell size, as indicated by a reduction in forward scatter by flow cytometry (Fig. 1b, left panels). In all subsequent flow cytometric analyses, EtdHD staining was used to exclude necrotic and postapoptotic cells (apoptotic cells that have lost membrane integrity). Only cells that were negative for EtdHD staining (Fig. 1b, middle panels, region R1) were included in the analysis of the staurosporine-treated (apoptotic) and the untreated (viable) populations. Analysis of annexin V staining of these specific populations by flow cytometry demonstrated that, while apoptotic cells bound annexin V strongly (mean fluorescence intensity (MFI): 124.16), viable cells displayed only low to intermediate staining with annexin V (MFI: 26.15) (Fig. 1b, right panels). Importantly, the apoptotic and viable cell populations were uniform and largely nonoverlapping with respect to annexin V staining, permitting clear discrimination between the two populations.
|
To determine whether PT could bind to the surface of apoptotic cells, apoptotic and viable Jurkat cells were incubated with FITC-PT (100 µg/ml) and analyzed by flow cytometry. As shown in Fig. 2, FITC-PT bound to apoptotic, but not to viable, cells (MFI: 11.62 vs 4.33, respectively). The reactivity of FITC-PT with viable cells was similar to that of the FITC-HSA control (MFI: 4.33 and 4.20, respectively). Binding of PT to apoptotic cells was dependent on the presence of calcium ions (2.5 mM), because no binding was observed in the absence of calcium.
|
The selective binding of PT to apoptotic Jurkat cells suggests that PT-dependent aPL should bind to apoptotic cells in the presence of PT. To investigate whether autoimmune LA Ab react with apoptotic cells, we used IgG isolated from the plasma of six SLE patients with known PT-dependent LA activity (10). Serum from an SLE patient (patient 7) with known high-titer anti-PT Ab and LA activity was included as a positive control in these experiments. Binding to apoptotic and viable cells, in the presence or absence of PT, was assessed by flow cytometry (Fig. 3a). IgG from patients 1, 2, 3, and 4, showed greater binding to apoptotic (MFI range: 23.87–145.21) than to viable (MFI range: 10.66–42.02) cells in the presence of PT. In the absence of PT, binding of these same IgG fractions to apoptotic cells was less than or equal to that to viable cells, and usually similar to that of the normal control IgG. In contrast, IgG from patients 5 and 6 demonstrated slightly greater binding to viable than to apoptotic cells, irrespective of the presence of PT. PT-dependent binding to apoptotic cells correlated with binding to phosphatidylethanolamine-bound PT (Fig. 3), but not with the strength of LA activity or with clinical features of the patients (Ref. 10 and data not shown). Serum Ab from the anti-PT positive control (patient 7) showed strong reactivity with apoptotic cells in the presence, but not the absence, of PT and little or no reactivity with viable cells (with or without PT).
|
Murine LA mAb reactive with PT
The finding that autoimmune LA Ab bound to apoptotic, but not viable, cells in the presence of PT demonstrates that PT-dependent autoimmune aPL can, indeed, bind to apoptotic cells. However, the use of polyclonal IgG isolated from SLE precludes the identification of the particular aPL that are responsible for reactivity with apoptotic cells. To characterize the epitopes responsible for LA Ab binding to apoptotic cells, we used two IgG1 PT-dependent LA murine mAb, 29J3-62 and 29I4-24, derived from mice immunized with DOPS/PT or DOPS/DOPC/PT complexes, respectively. Purified mAb were evaluated for reactivity with CL, DOPS, DOPC, 2GPI, and PT by ELISA. LA activity was evaluated by APTT. Although neither mAb showed anti-CL, anti-DOPS, anti-DOPC, or anti-2GPI activity in ELISA (Table I), both mAb exhibited strong LA activity that was dose dependent and specifically inhibited by hexagonal (II) phase phosphatidylethanolamine (Fig. 4). Furthermore, both 29J3-62 and 29I4-24 reacted with human PT alone (in the absence of PL), although 29I4-24 bound weakly in comparison with 29J3-62. At equivalent Ab concentrations, 29J3-62 bound strongly to PT-coated wells of ELISA plates at high (20 µg/ml) and low (5 µg/ml) coating concentrations, while 29I4-24 showed only weak binding at the high coating concentration and no detectable binding at the low concentration (Fig. 5a).
|
|
|
1.14 nM (0.08 µg/ml)), binding of 29I4-24 to PT was completely inhibited only at very high concentrations of PT (7 µM (500 µg/ml)). The relative affinities (Kd) of the two Ab for soluble PT, defined as the concentration of soluble PT producing 50% inhibition, differed by roughly 9000-fold (0.01 nM for 29J3-62 compared with 90 nM for 29I4-24). Effect of PL on anti-PT reactivity of LA mAb
We next wanted to determine whether the presence of PL would affect the binding of 29J3-62 and 29I4-24 to PT. PT in solution was incubated with either PSPC (25% PS/75% PC w/w) or PC (100% PC) vesicles, in the presence of calcium ions. After coating the PL/PT mixture to an ELISA plate, mAb reactivity was evaluated. 29J3-62, which recognizes soluble and bound PT with high affinity (Fig. 5), bound to all PL/PT- and PT-coated wells equivalently, suggesting that the wells were coated with an equal amount of PT. This allowed for a direct assessment of the contribution of PL to the recognition of PT by the LA mAb. As shown in Fig. 6a, the binding of mAb 29J3-62 to PSPC/PT or PC/PT did not differ significantly from its binding to PT alone. In sharp contrast, the binding of 29I4-24 to PSPC/PT was dramatically increased over that to PT alone or PC/PT. In fact, the binding of 29I4-24 to PSPC/PT was equivalent to or greater than that of 29J3-62 to PT (with or without PL), despite the extremely low reactivity of 29I4-24 to PT alone. Likewise, competitive inhibition of the binding of 29J3-62 to immobilized PT was similar, whether PT was presented with PSPC, PC, or alone (Fig. 6b). In contrast, the binding of 29I4-24 to PT was more strongly competed by PT in the presence of PSPC (10-fold), than by either PT alone or PT in the presence of PC. Although the enhancing effect of PSPC on inhibition of binding of 29I4-24 to PT was several magnitudes less than that observed in the direct binding assay, it is clear from these results that the interaction of 29I4-24 with PT is augmented in the presence of PSPC. There are several potential explanations for the difference in magnitude between the effects of PSPC in the two assays. First, because of Ab bivalency, the degree to which an Ab binds to an Ag attached to a solid phase is, in general, greater than binding to the same Ag in solution or suspension. For this reason, solid phase assays are more sensitive than immunoassays performed in solution. Furthermore, in the direct binding assay, PT was stably anchored to the solid phase, whereas, in the competition assay, it was subject to an equilibrium state (i.e., reversible association to and dissociation from the PSPC vesicles).
|
The demonstration that both LA mAb recognized PL-bound PT suggested that these mAb should be capable of binding to apoptotic cell-bound PT. We evaluated the reactivity of mAb 29J3-62 and 29I4-24 with apoptotic and viable cells by flow cytometry. Both mAb bound strongly to apoptotic cells only in the presence of PT (Fig. 7a). PT-dependent binding of both 29J3-62 and 29I4-24 was 30-fold higher to apoptotic cells than to viable cells (29J3-62, MFI: 1293.06 vs 43.27; 29I4-24, MFI: 952.45 vs 25.26). Titration of PT demonstrated that binding was maximal at 20 µg/ml PT for both mAb (Fig. 7b). Surprisingly, the binding of 29I4-24 to apoptotic cells was nearly equivalent to that of 29J3-62, even at very low concentrations of PT (2.5 µg/ml), despite the great difference in affinities of these two Ab for native PT in the absence of PL. We conclude that the interaction of PT with anionic PL, whether in the form of PSPC or the surface of apoptotic cells, dramatically increases the affinity of 29I4-24 for PT. The affinity of 29J3-62 for PT, in contrast, is minimally affected by the interaction of PT with anionic PL.
|
We next investigated the mechanism(s) underlying the effect of PSPC on recognition of PT by LA mAb. Before discussing the possible mechanisms, it is important to define the terms affinity and avidity. Ab affinity is defined as the strength of binding of a single Ag binding site of the Ab to a monovalent Ag, while Ab avidity is defined as the total binding strength of an Ab having more than one Ag binding site to multivalent Ag (32). Therefore, avidity depends on not only the affinity of each individual Ag binding site, but also the number of binding sites of the Ab engaged by Ag.
At least two processes, not mutually exclusive, may play a role in the binding of IgG LA mAb to PL/PT complexes. First, the interaction of PT with PSPC may result in the expression of a novel epitope on PT. In this case, the strength of interaction of an individual Ag binding site (affinity) of an Ab for that epitope would be increased. Second, PL may provide a surface upon which PT can attach at a sufficiently high density, or in a particular spatial orientation, such that both binding sites of the IgG Ab, each with low affinity for soluble PT, would now interact with PT. In the latter situation, the overall increase in binding of the Ab, due to increased avidity, is dependent on Ab bivalency and occurs when both Ag binding sites simultaneously interact with Ag that is rendered effectively multivalent by attachment to the PL surface.
To determine the relative contribution of these two mechanisms, we tested IgG, IgG', F(ab')2, and Fab' of each mAb for reactivity with PT alone or with PL/PT complexes by ELISA. Based on the model described in the paragraph above, monovalent Fab' of an Ab reactive with a novel epitope on PL/PT should show increased binding to PL/PT, compared with PT. In contrast, if a higher density of PT on the PL surface is the sole factor determining Ab binding, the binding of monovalent Fab' should be equivalent for PL/PT, as compared with PT alone.
For ease of comparison, the data in these studies are expressed as a function of the molarity of Ag binding sites (Fig. 8) and summarized in Table II. First, we evaluated the reactivity of both mAb with PT alone (Fig. 8, right panels). The reactivity of Fab' of 29J3-62 was strong and did not differ significantly from that of intact IgG, IgG', or F(ab')2, indicating that a single Ag binding site of the mAb reacts with PT with a high affinity, and that reduction, alkylation, and fragmentation of the mAb did not affect reactivity. In contrast, binding of the monovalent Fab' of 29I4-24 was significantly lower than that of bivalent intact IgG, IgG', or F(ab')2 and was decreased to background levels. These results clearly demonstrate an important contribution of Ab bivalency for binding of 29I4-24, but not 29J3-62, to PT alone. This is consistent with the low affinity of 29I4-24 IgG for soluble PT (Fig. 5b) and the enhanced reactivity of this mAb for bound PT only in the presence of increased Ag density (Fig. 5a).
|
|
10-9 M for Fab' of both 29I4-24 and 29J3-62. This suggests that the epitope recognized by 29I4-24 is altered by the interaction of PT with PL, and that the affinity of 29I4-24 for this altered epitope is far greater than its affinity for the unaltered epitope. An effect of bivalency on overall binding (avidity) for PSPC/PT was observed for both mAb, in that roughly 2-fold higher concentrations of monovalent Fab' than bivalent F(ab')2 were required to achieve 50% of maximal binding.
Reactivity of the mAb fragments with apoptotic cells by flow cytometry confirmed and extended these results (Fig. 9). As in the ELISA, F(ab')2 of both mAb exhibited similar (in fact, slightly enhanced) binding, compared with their respective IgG and IgG' controls (MFI: 345.99 for 29J3-62; MFI: 239.62 for 29I4-24, at 20 µg/ml PT). More importantly, and as in the ELISA, Fab' of these two mAb had virtually identical reactivities with PT in association with apoptotic cells, despite their extremely different affinities for PT alone. These data with apoptotic cells are consistent with those for PL/PT complexes and suggest that similar PL-dependent epitopes may be expressed on both PSPC- and cell-bound PT. However, in contrast to the results with PL/PT complexes, the reactivity of Fab' of both mAb with PT in association with apoptotic cells was low and dramatically decreased, compared with the reactivity of the F(ab')2. At the highest concentration of PT tested (20 µg/ml), the binding of the Fab' of both 29J3-62 (MFI: 19.15) and 29I4-24 (MFI: 17.61) was similar and 2-fold higher than that of negative control mAb 27D2-83 (MFI: 8.26).
|
Effect of PT-dependent LA mAb on binding of PT to apoptotic cells
Thus far, our data show that PT increases the binding of PT-dependent LA mAb to apoptotic cells. To determine whether PT-dependent LA mAb could similarly augment the binding of PT to apoptotic cells, we compared the binding of FITC-PT to apoptotic cells in the presence and absence of 29J3-62 and 29I4-24 by flow cytometry. Binding of FITC-PT (20 µg/ml) to apoptotic cells was enhanced in the presence of either mAb (29J3-62, MFI: 51.16; 29I4-24, MFI: 38.03) as compared with binding of FITC-PT in the presence of an isotype-matched control Ab (MFI: 7.85) (Fig. 10a). Although the enhancing effect of 29J3-62 was somewhat greater than that of 29I4-24 at all concentrations of FITC-PT, both mAb markedly enhanced binding at higher concentrations (10–20 µg/ml) of FITC-PT (Fig. 10b). This effect was restricted to apoptotic cells, as the mAb did not significantly affect the binding of FITC-PT to viable cells (data not shown).
|
|
To determine whether the murine monoclonal and human LA recognize similar epitopes, a competitive binding assay was performed on plates coated with PL/PT complexes. We examined whether murine LA mAb affected the binding of human SLE serum Ab to PL/PT. As shown in Fig. 12, the SLE sera were affected differently by the murine LA mAb. In one case (patient 4), murine LA mAb 29J3-62 completely inhibited the binding of the human serum IgG Ab to PL/PT. In contrast, for another SLE serum (patient 2), the same murine LA mAb enhanced binding of the serum Ab to PL/PT, compared with buffer (Fig. 12) or an irrelevant mAb control (data not shown). Finally, a third SLE serum (patient 7) was unaffected by murine LA mAb 29J3-62, as was the normal control serum. The second murine LA mAb, 29I4-24, affected only one serum (patient 2) and resulted in virtually identical enhancement of Ab binding to that observed with 29J3-62. The finding that a murine LA mAb was able to completely block the binding of human serum Ab to PL-bound PT suggests recognition of similar or identical epitope(s) by 29J3-62 and some human PT-dependent IgG Ab. In contrast, the ability of the same mAb to enhance the binding of another SLE serum by 50–75% suggests that this serum does not recognize the same epitopes as 29J3-62. Enhancement of SLE Ab reactivity is likely caused by modification of PT upon binding of the mAb to the PL/PT complex, leading to exposure of new or cryptic epitopes. This enhanced binding was not due to reactivity of the anti-human IgG Ab with murine LA mAb. The lack of an effect of either mAb on serum from patient 7 suggests that PT-dependent IgG Ab contained in this serum recognize different epitopes than those recognized by the murine LA mAb. Furthermore, the Ab in serum from patient 7 do not appear to react with neo- or cryptic epitopes exposed by 29J3-62 and 29I4-24, suggesting that they are different from the PT-dependent Ab found in serum from patient 2. Taken together, these findings are consistent with what is currently known about the heterogeneity of human LA autoantibodies. Moreover, they provide confirmation that murine LA mAb and SLE patient LA can recognize similar epitopes.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2GPI and 2GPI-dependent aPL bind specifically to apoptotic, but not viable, cells (24). In this study, we demonstrate that PT behaves similarly to 2GPI, and binds selectively to the surface of apoptotic, but not viable, Jurkat cells. Furthermore, we show that PT supports the binding of SLE-derived polyclonal and murine monoclonal PT-dependent LA Ab to apoptotic cells. To understand the basis for the interaction of PT-dependent LA Ab with apoptotic cells, we characterized the reactivities of two mAb, which had identical LA activity but differed dramatically in their relative affinities for PT. Although one mAb (29J3-62) had a high affinity for PT alone, the other (29I4-24) showed minimal reactivity with PT alone and required PL for elevated binding. Monovalent fragments of 29I4-24 reacted with PT with high affinity in the presence of PSPC, but not PC, suggesting that this mAb recognizes a PL-dependent epitope. In contrast, 29J3-62 recognized PT with high affinity in either the presence of PL (PSPC or PC) or its absence. Despite this heterogeneity, PT-dependent binding of both mAb to apoptotic cells was 30-fold greater than that to viable cells. Strong binding of both mAb to apoptotic cells depended on Ab bivalency, with monovalent fragments exhibiting very low binding, similar in magnitude to that of PT alone. Finally, binding of PT to apoptotic cells was, itself, increased in the presence of bivalent, but not monovalent, forms of either mAb.
The binding of FITC-PT to apoptotic cells, although selective, was only 2- to 3-fold greater than that to viable cells. This low level of binding differs from our previous findings with FITC-labeled 2GPI, which demonstrated 25-fold higher binding of FITC-labeled 2GPI to apoptotic than to viable thymocytes (24), but is similar to the low level of binding of FITC-PT to the surface of damaged endothelial cells, as demonstrated by Zhao et al. (34). Although this difference in binding between FITC-labeled PT and 2GPI may reflect differences in the number of binding sites on the apoptotic cell surface for these two proteins, there may also exist intrinsic differences between these two cell types with respect to their surface membranes as they undergo apoptosis. Thus, while the binding of FITC-labeled 2GPI and PT to apoptotic Jurkat cells was low and similar in magnitude, binding of the same labeled proteins to apoptotic thymocytes differed significantly, with FITC-labeled 2GPI showing a much higher degree of binding (data not shown). Although little is known about the mechanism by which either protein interacts with apoptotic cells, both proteins are known to interact with PS, which is exposed on the apoptotic cell surface. However, a major difference between these two proteins is that the interaction of PT with PS on the surface of apoptotic cells requires calcium, whereas that of 2GPI does not. Finally, differences in the interaction of these two proteins with the apoptotic cell surface may also be attributable to differential effects of FITC-labeling on PT and 2GPI.
PT was first proposed as a possible cofactor for LA Ab by Loeliger in 1959 (5). The precise role of PL in the recognition of PT by LA Ab has been an area of ongoing investigation. In 1991, Bevers et al. (6) demonstrated that some LA Ab recognize a complex of PL-bound human PT, but neither PT nor PL alone. Recent work has indicated that many LA and anti-PT Ab recognize PS- or phosphatidylethanolamine-bound PT preferentially over PT alone in ELISA (9, 10, 35). Furthermore, reactivity with PL-bound PT, but not PT alone, in ELISA correlated significantly with LA activity and clinical manifestations of APS (35). Finally, we have shown that the LA activity of SLE-derived IgG was specifically inhibited by PT in the presence of hexagonal phase phosphatidylethanolamine, but not by PT or phosphatidylethanolamine alone (10). Although these findings with LA Ab from patient samples demonstrate that PL likely plays a crucial role in the antigenicity of PT, the epitopes recognized by LA Ab remain unclear. As PT has been shown to undergo a conformational change upon binding to PS-containing membranes (36), it is possible that some PT-dependent LA Ab are directed against neoepitopes exposed on PT when it interacts with PL. Conversely, the binding of PT to a suitable PL surface may simply result in an increased PT density that is required for LA Ab reactivity.
Therefore, we sought to investigate the precise nature of PL dependence in the recognition of PT by LA Ab. To do this, we used PT-dependent LA mAb that were isolated from mice immunized with PS or PSPC vesicles and PT. Two mAb were selected based on their strong and similar LA activity, and their strikingly different reactivity with PT alone by ELISA. Although 29J3-62 reacted very strongly with PT-coated ELISA plates, 29I4-24 showed only marginal reactivity. Similarly, in a competitive inhibition ELISA, the relative affinity of 29J3-62 for PT in solution was 9000-fold greater than that of 29I4-24. These findings suggested to us that these two mAb might recognize different epitopes on PT. Therefore, we evaluated the effect of PL on the reactivity of these Ab with PT. Although the binding of 29J3-62 to PT was unchanged in the presence of PSPC or PC, the reactivity of 29I4-24 to PT was greatly enhanced by PSPC, but not PC. Competitive inhibition assays provided similar results. Although these data support PL-dependent reactivity of 29I4-24, they do not address the basis for the recognition of PL-bound PT. To determine whether 29I4-24 recognizes an epitope that is truly PL-dependent or whether its reactivity is simply due to an increased density of PT on the PL surface, it was necessary to compare the affinity and avidity of binding of 29I4-24 to PL-bound PT. As defined earlier, Ab affinity denotes the strength of binding of a single Ag binding site of an Ab to a monovalent Ag, while Ab avidity denotes the total binding strength of an Ab having more than one Ag binding site to multivalent Ag (32). Thus, measurements of affinity and avidity are usually done by comparing the reactivities of monovalent and multivalent fragments of the same Ab.
The binding affinity and avidity data for 29I4-24 and 29J3-62 to PT bound to different surfaces are summarized in Table II. We consider first 29I4-24, which reacted weakly to PT alone with low affinity, as measured both by competition with soluble phase PT and by direct binding of monovalent fragments of 29I4-24 to polystyrene-bound PT. Density of PT was important for this recognition because only bivalent mAb showed detectable reactivity to PT alone. In contrast, both monovalent and bivalent fragments of 29I4-24 reacted strongly with PSPC/PT-coated wells. The strong reactivity of monovalent fragments of this mAb demonstrates that this mAb has a high affinity for an epitope on PSPC/PT, and argues against density being the sole determining factor for 29I4-24 binding to PL-bound PT. If 29I4-24 reacted with PSPC-bound PT simply because of an increased density of PT on the PL surface, monovalent fragments of this mAb should react with PSPC/PT with a low affinity similar to that with PT alone. As this was not the case, this provides direct proof for recognition of a PL-dependent epitope and against dependence solely on an increased density of PT. Furthermore, because both monovalent and bivalent 29I4-24 reacted with PSPC/PT, but not with PC/PT or PT alone, recognition of PL-bound PT is specifically dependent on the presence of PS. With respect to 29J3-62, which reacted strongly with PT alone, monovalent and bivalent forms also reacted strongly with PSPC/PT and PC/PT. These data indicate that 29J3-62 has a high affinity and avidity for PT, whether or not it is bound to PL. Taken together, these results confirm the notion that LA Ab are heterogeneous with respect to their reactivity with PT, and underscore the role of PL in this interaction. Based on our data with 29I4-24 and 29J3-62, PT-dependent LA mAb can recognize either PL-dependent or PL-independent epitopes on PT. Dependence on PL, at least for 29I4-24, appears to be due to strong recognition of a new epitope on PL-bound PT, rather than an increase in the density of PT. In contrast, weak recognition of PT alone by this mAb requires a high density of PT.
The finding that both 29I4-24 and 29J3-62 reacted strongly with PL-bound PT suggested that both mAb should recognize PT bound to apoptotic cell membranes. Indeed, both mAb exhibited very strong binding to apoptotic cells that was strictly dependent on the presence of PT. Importantly, despite the 9000-fold difference in relative affinity for PT alone between 29I4-24 and 29J3-62, the binding of these mAb to apoptotic cells differed by only 3- to 4-fold at low concentrations (2–5 µg/ml) of PT and was equivalent at higher concentrations (20 µg/ml) of PT. Similar to the PSPC/PT experiments, we sought to determine the extent to which PL-dependent neoepitopes and Ag density contributed to the binding of our LA mAb to apoptotic cell-bound PT. In contrast to binding to PSPC/PT-coated wells, bivalency was critical for binding of both LA mAb to apoptotic cells. Intact IgG, IgG', or bivalent F(ab')2 of both 29J3-62 and 29I4-24 reacted strongly with apoptotic cell-bound PT, with 29J3-62 showing higher binding than 29I4-24. In contrast, monovalent Fab' of both mAb showed low and similar binding. These data suggest that both mAb have only a low affinity for apoptotic cell-bound PT, but that bivalency results in a high avidity for this same Ag (Table II).
The obvious question that arises from these data is why there is a difference in reactivity between PL-bound and apoptotic cell-bound PT. This could be interpreted to suggest that apoptotic cell-bound PT creates epitopes that differ from those on PSPC-bound PT. Although we cannot formally exclude this possibility, we believe that there is an alternative explanation related to methodological differences between the two assay systems. In the PL-bound system, the PSPC/PT complex was irreversibly bound to a solid phase (polystyrene wells). The PL-bound PT, which cannot dissociate from the surface, therefore provides a stable array of bound PT molecules to which both monovalent and bivalent fragments of 29I4-24 can bind with high affinity and avidity, respectively. In contrast, in the apoptotic cell-bound system, soluble PT was added to apoptotic cells, which allowed PT to associate and dissociate reversibly from the apoptotic cell surface in an equilibrium state. Therefore, a monovalent fragment of mAb that is bound to a molecule of PT attached to the apoptotic cell surface can dissociate from the apoptotic cell in one of two ways. First, it can dissociate from the cell membrane as a complex with the PT molecule to which it is bound. Second, because binding of the mAb to PT is also an equilibrium process, the mAb can dissociate from the PT to which it is bound, also resulting in dissociation of the mAb from the apoptotic cell surface. Although bivalent mAb can also dissociate from the PT to which it is bound, the kinetics of this dissociation are quite different. As both binding sites of the mAb interact with PT, dissociation at only one site would result in the mAb still being tethered to PT. Thus, both binding sites would have to simultaneously release the Ag, for the mAb to be completely dissociated from PT. The existence of two modes (mAb 
PT cell) of dissociation from the apoptotic cell surface, as opposed to only a single mode (mAb PL-bound PT) of dissociation for PL-bound PT attached to a solid phase, most likely explains the lower affinity of monovalent fragments of both mAb for PT bound to apoptotic cells. (Table II).
Several recent reports have shown that LA Ab can enhance the binding of PT to PL (33, 37, 38) and to endothelial cells (34, 37) in vitro. In this way, LA Ab may increase the affinity of PT for the PL or cell surface, so that PT competes with other coagulation factors for the available catalytic surface (38). In fact, enhanced PT binding to endothelial cells has been shown to increase thrombin generation and shorten the clotting time in HUVEC-based coagulation assays (34). To evaluate directly whether our PT-dependent LA mAb were capable of enhancing the binding of PT to apoptotic cells, we incubated FITC-labeled PT with apoptotic cells in the presence or absence of LA-positive mAb. Both LA mAb significantly enhanced the binding of FITC-labeled PT to apoptotic cells, while a control mAb did not. Bivalency was essential for enhanced binding of PT by the mAb. Although intact IgG, IgG', and bivalent F(ab')2 of 29J3-62 and 29I4-24 enhanced binding of FITC-labeled PT to apoptotic cells, monovalent Fab' of these mAb had no enhancing effect whatsoever.
Our findings clearly demonstrate that bivalency is important both for enhanced binding of PT in the presence of LA Ab and for direct binding of LA Ab to apoptotic cells in the presence of PT. Field et al. (33) have similarly shown that bivalent, but not monovalent, fragments of patient-derived LA-positive IgG enhanced binding of PT to PSPC vesicles. They propose a model in which bivalent Ab binds two molecules of PT that are interacting with the PL surface, thus stabilizing the binding of PT to PL. In contrast, binding of monovalent IgG to only one molecule of PT does not have the same stabilizing effect. This type of model can also explain the enhanced binding of PT to apoptotic cells that we have observed in the presence of bivalent, but not monovalent, mAb. It also explains why bivalency is necessary for direct binding of mAb to apoptotic cells. Furthermore, it is important to note that bivalency of 2GPI-dependent aPL has been shown to have an analogous stabilizing effect on the interaction of 2GPI with PL (39, 40, 41). Moreover, while native 2GPI binds to PL membranes of physiologically relevant composition with relatively weak affinity (39, 42), dimeric forms of 2GPI bind strongly (43) and are recognized by aPL with increased avidity (44). In fact, dimeric 2GPI, alone, demonstrates functional LA activity that is enhanced by aPL (43).
The latter findings offer a new understanding of the mechanisms whereby LA Ab promote thrombosis in vivo. As apoptotic cells have been shown to exhibit procoagulant properties in vitro (45), it is possible that they contribute to the thrombotic state associated with APS. Our results imply that interactions between LA Ab and apoptotic cells in the presence of PT may potentiate this procoagulant effect. They also indicate that apoptotic cell-bound PT represents an important target of LA Ab, be they of low or high affinity for PT. Thus, measurement of PT-reactive Ab may not be indicative of their ability to bind to or modulate binding of PT to apoptotic or other cells. Finally, our data provide support for the concept that apoptotic cells play a central role in the induction and/or perpetuation of APS.
In conjunction with our earlier findings on recognition of apoptotic cells by 2GPI-dependent aPL (24), our present results with PT-dependent LA Ab suggest that recognition of apoptotic cells by aPL may be a general paradigm. Thus, both 2GPI-dependent and PT-dependent aPL are reactive with apoptotic cells. Furthermore, 2GPI bound to apoptotic cells is itself immunogenic and can induce the production of aPL (25). We hypothesize that apoptotic cell-bound PT is also capable of inducing aPL. Finally, we propose that other PL-binding proteins, particularly those involved in coagulation, should bind to apoptotic cells or other cells that express anionic PL (e.g., activated platelets). Similar to PT-dependent LA Ab, Ab reactive with these PL-binding proteins could enhance binding of their target proteins to apoptotic cells.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Joyce Rauch, Division of Rheumatology, McGill University Health Center, Montreal General Hospital, 1650 Cedar Avenue, Montreal, Quebec H3G 1A4, Canada. E-mail address: joyce.rauch{at}mcgill.ca
3 Abbreviations used in this paper: aPL, anti-phospholipid Ab; SLE, systemic lupus erythematosus; APS, anti-phospholipid syndrome; PL, phospholipid; 2GPI, 2-glycoprotein I; PT, prothrombin; LA, lupus anticoagulant; PS, phosphatidylserine; DOPC, 1,2-dioleoyl-sn-glycero-3-phosphocholine; DPPC, 1,2-dipalmitoyl-sn-glycero-3-phosphocholine; DOPS, 1,2-dioleoyl-sn-glycero-3-[phospho-L-serine]; CL, bovine heart cardiolipin; HSA, human serum albumin; EtdHD, ethidium homodimer-1; IgG + IgM (H + L), IgG and IgM H and L chains; APTT, dilute activated partial thromboplastin time assay; PC, phosphatidylcholine; PSPC, 25% PS and 75% PC by weight; MFI, mean fluorescence intensity.
Received for publication April 24, 2002. Accepted for publication January 7, 2003.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2-glycoprotein I (apolipoprotein H). Proc. Natl. Acad. Sci. USA 87:4120.2-glycoprotein I (and other proteins). Proc. Natl. Acad. Sci. USA 94:10356.2-glycoprotein I. J. Clin. Invest. 90:1100.
2-glycoprotein I-dependent manner. J. Immunol. 157:2201.[Abstract]
2-glycoprotein I bound to apoptotic thymocytes. J. Autoimmun. 11:413.[Medline]
2-glycoprotein I induces the production of anti-phospholipid antibodies. J. Autoimmun. 15:21.[Medline]
2-glycoprotein I complexes to lipid membranes. Biochemistry 35:13833.[Medline]
2-glycoprotein I (2GPI) monoclonal antibodies with lupus anticoagulant-like activity enhance the 2GPI binding to phospholipids. J. Clin. Invest. 99:2260.[Medline]
2-glycoprotein I binding to phospholipid membranes.. Thromb. Haemostasis 80:610.[Medline]
2-glycoprotein I mimic the in vitro effects of 2-glycoprotein I-anti-2-glycoprotein I antibody complexes. J. Biol. Chem. 276:3060.2-glycoprotein I autoantibodies from patients with the "antiphospholipid" syndrome bind to 2-glycoprotein I with low affinity: dimerization of 2-glycoprotein I induces a significant increase in anti-2-glycoprotein I antibody affinity. J. Immunol. 161:2038.This article has been cited by other articles:
|
|
 |
|
  T. Kawakami, M. Yamazaki, M. Mizoguchi, and Y. Soma Differences in anti-phosphatidylserine-prothrombin complex antibodies and cutaneous vasculitis between regular livedo reticularis and livedo racemosa Rheumatology, May 1, 2009; 48(5): 508 - 512. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Kawakami, M. Yamazaki, M. Mizoguchi, and Y. Soma Antiphosphatidylserine-Prothrombin Complex Antibodies in 3 Patients With Behcet Disease Involving Superficial Vein Thrombophlebitis Arch Dermatol, February 1, 2009; 145(2): 171 - 175. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Kawakami, S. Ooka, M. Mizoguchi, Y. Soma, and M. Yamazaki Remission of Hepatitis B Virus-Related Cryoglobulinemic Vasculitis with Entecavir Ann Intern Med, December 16, 2008; 149(12): 911 - 912. [Full Text] [PDF]
|
|
|
|
|
|
T. Kawakami, M. Yamazaki, K. Kawasaki, M. Mizoguchi, and Y. Soma Therapeutic Effect of Argatroban on Rheumatoid Vasculitis With Antiphosphatidylserine-Prothrombin Complex Antibody Arch Dermatol, August 1, 2008; 144(8): 1075 - 1076. [Full Text] [PDF]
|
|
|
|
 |
|
 T. Horita, K. Ichikawa, H. Kataoka, S. Yasuda, T. Atsumi, and T. Koike Human monoclonal antibodies against the complex of phosphatidylserine and prothrombin from patients with the antiphospholipid antibodies Lupus, July 1, 2007; 16(7): 509 - 516. [Abstract] [PDF]
|
|
|
|
|
|
H. S. Cheong, S. O. Lee, C.-B. Choi, Y.-K. Sung, H. D. Shin, and S.-C. Bae MERTK polymorphisms associated with risk of haematological disorders among Korean SLE patients Rheumatology, February 1, 2007; 46(2): 209 - 214. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Giannakopoulos, F. Passam, S. Rahgozar, and S. A. Krilis Current concepts on the pathogenesis of the antiphospholipid syndrome Blood, January 15, 2007; 109(2): 422 - 430. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Fan, V. A. Patel, A. Longacre, and J. S. Levine Abnormal regulation of the cytoskeletal regulator Rho typifies macrophages of the major murine models of spontaneous autoimmunity J. Leukoc. Biol., January 1, 2006; 79(1): 155 - 165. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. A. Trouw, S. C. Nilsson, I. Goncalves, G. Landberg, and A. M. Blom C4b-binding protein binds to necrotic cells and DNA, limiting DNA release and inhibiting complement activation J. Exp. Med., June 20, 2005; 201(12): 1937 - 1948. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Longacre, J. S. Koh, K. K-H. Hsiao, H. Gilligan, H. Fan, V. A. Patel, and J. S. Levine Macrophages from lupus-prone MRL mice are characterized by abnormalities in Rho activity, cytoskeletal organization, and adhesiveness to extracellular matrix proteins J. Leukoc. Biol., November 1, 2004; 76(5): 971 - 984. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Radic, T. Marion, and M. Monestier Nucleosomes Are Exposed at the Cell Surface in Apoptosis J. Immunol., June 1, 2004; 172(11): 6692 - 6700. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Fan, A. Longacre, F. Meng, V. Patel, K. Hsiao, J. S. Koh, and J. S. Levine Cytokine Dysregulation Induced by Apoptotic Cells Is a Shared Characteristic of Macrophages from Nonobese Diabetic and Systemic Lupus Erythematosus-Prone Mice J. Immunol., April 15, 2004; 172(8): 4834 - 4843. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. W. C. Dieker, J. van der Vlag, and J. H. M. Berden Deranged removal of apoptotic cells: its role in the genesis of lupus Nephrol. Dial. Transplant., February 1, 2004; 19(2): 282 - 285. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) or absence (
) of PT (50 µg/ml); or viable cells in the presence (
) or absence (
) of PT by flow cytometry. SLE patient 7 had known high-titer anti-PT Ab and LA Ab, and was included as a positive control. Purified IgG from pooled normal human plasma was used as a negative control (normal). Reactivity of IgG fractions with phosphatidylethanolamine-bound PT (PL/PT) in ELISA is indicated by a plus (+) sign for positive reactivity or a minus (-) sign for no reactivity, based on previous data (
), and a subclass-matched negative control mAb, 27D2-83 (
), were titrated for LA activity in an APTT. Each value represents the mean of duplicate samples. The inset shows titration of the LA mAb in the presence of HEPES buffer or hexagonal (II) phase phosphatidylethanolamine. The units for the axes in the inset are the same as those in the main figure. The LA activity of both mAb was specifically inhibited by hexagonal (II) phase phosphatidylethanolamine.
) of 29J3-62 (top) and 29I4-24 (bottom) to PSPC/PT (left) or PT alone (right) coated directly onto ELISA plates. IgG and fragments were used at concentrations that provided equivalent molarities of Ag binding sites. Values represent the means of duplicate samples, and the data are representative of two independent experiments.
