SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

Objective

To investigate the possibility that IgG anti–NR2 glutamate receptor antibodies (anti-NR2) derived from patients with systemic lupus erythematosus (SLE) cause an immunologic interaction with endothelial cells (ECs) in the blood–brain barrier, resulting in inflammation of the blood–brain barrier, allowing the entrance of these autoantibodies into the cerebrospinal fluid.

Methods

Purified IgG anti-NR2 antibodies from 14 patients with SLE were tested for their ability to bind to double-stranded DNA (dsDNA) and ECs, to modulate endothelial adhesion molecule expression and cytokine production by ECs, and to activate the NF-κB pathways in the ECs. Purified IgG from 5 normal subjects was used as a negative control.

Results

Purified IgG anti-NR2 antibodies bound to dsDNA in an IgG-dose–dependent manner. This interaction up-regulated the expression of endothelial leukocyte adhesion molecule 1, vascular cell adhesion molecule 1, and intercellular adhesion molecule 1 on the EC surface and increased the production of interleukin-6 (IL-6) and IL-8, but not tumor necrosis factor α or IL-1β, by ECs. Purified IgG anti-NR2 also activated the degradation of cytoplasmic IκB, indicating the activation of NF-κB in the ECs.

Conclusion

EC activation through the NF-κB signaling pathway induced by IgG anti-NR2 antibodies in the central nervous system of SLE patients may lead to inflammation of the blood–brain barrier, initiating the pathogenesis of neuropsychiatric SLE.

Anti–double-stranded DNA (anti-dsDNA) antibodies are specific serologic markers for systemic lupus erythematosus (SLE) and correlate with SLE disease activity (1). Anti-dsDNA antibodies are reported to cause kidney damage by binding either directly to DNA that is present in tissue or by binding to cross-reactive, non-DNA tissue antigens (2).

Several different antigenic cross-reactivities have been identified for anti-dsDNA antibodies (3). The sequence of pentapeptide DWEYS, a molecular mimic of dsDNA, has recently been demonstrated to also be present in the extracellular domain of murine and human N-methyl-D-aspartate (NMDA) receptor subunits NR2a and NR2b (4). SLE patients' anti-dsDNA antibodies that cross-react with NR2 glutamate receptors have been shown to mediate the apoptotic death of neurons in vivo and in vitro (5). In addition, they have been reported to be present in the cerebrospinal fluid (CSF) of SLE patients with progressive cognitive decline and to mediate neuronal death via an apoptotic pathway (5). More recently, we reported a significant relationship between anti–NR2 glutamate receptor antibody titers and the presence of neuropsychiatric SLE (NPSLE), especially of complex presentation, such as when neurologic syndromes of the central nervous system (CNS) and diffuse psychiatric/neuropsychological syndromes occur concurrently in a single patient (6). However, it has not yet been elucidated whether anti-NR2 antibodies are produced outside the CNS and enter the CSF as a result of damage to the blood–brain barrier or whether they are produced locally.

Anti-dsDNA antibodies from patients with SLE have been reported to bind to endothelial cells (ECs) derived from human umbilical veins, although the anti-dsDNA antibody binding site remains unclear (7). Human cerebral ECs were recently reported to express messenger RNA (mRNA) and protein for the NMDA receptor subunits NR2a and NR2b (8). Taken together, we hypothesized that anti-NR2 antibodies might bind to dsDNA and activate ECs, triggering the pathogenesis of NPSLE.

In this report, we provide evidence showing that anti-NR2 antibodies derived from patients with SLE bind to dsDNA and that this interaction activates human umbilical vein ECs by up-regulating adhesion molecule expression on the EC surface and increasing cytokine release from ECs.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

Patients and samples.

In this study, we examined IgG anti-NR2 antibody–positive serum samples from 14 patients who fulfilled the American College of Rheumatology revised criteria for the classification of SLE (9), as well as serum samples from 5 normal control subjects. The study protocol was approved by the ethics committee of Jichi Medical University. Written informed consent for the study was obtained from all of the enrolled patients and controls.

Preparation of the synthetic peptide.

Synthetic DWEYSVWLSN peptide was synthesized by Peptide Institute, using a solid-phase method, as described previously (10). The DWEYS pentapeptide corresponds to human NMDA receptors NR2a and NR2b (amino residues 283–287). The purity of the synthetic peptide was confirmed to be 98.1% by high-performance liquid chromatography, amino acid analysis, and microsequencing. The synthetic DWEYSVWLSN peptide was conjugated to bovine serum albumin (BSA; Sigma) at a 11:1 to 12:1 molar ratio, using glutaraldehyde as described previously (11).

Purification of IgG and IgG anti–NR2 glutamate receptor antibodies.

First, IgG purification from serum samples obtained from 14 SLE patients with IgG anti-NR2 antibodies and from serum samples obtained from 5 normal control subjects was performed using a MabTrap kit (Amersham Pharmacia) according to the manufacturer's instructions. Second, a synthetic DWEYSVWLSN peptide–conjugated Sepharose HP column (HiTrap NHS-activated; Amersham Pharmacia) was used to further purify IgG anti-NR2 antibody from the SLE patients' purified IgG samples, according to the manufacturer's instructions.

Binding of purified IgG anti-NR2 antibodies to a synthetic DWEYSVWLSN peptide.

The wells of 96-well microtiter plates (Falcon Pro-Bind flat-bottomed microtiter plates; Becton Dickinson) were coated overnight at 4°C with 100 μl of DWEYSVWLSN peptide–BSA conjugate at 5 μg/ml of DWEYSVWLSN peptide in phosphate buffered saline (PBS). The wells were then blocked for 2 hours at room temperature with PBS containing 1% BSA (Sigma). Purified IgG anti-NR2 antibodies from 14 patients with SLE and purified IgG from 5 normal control subjects were diluted in PBS containing 3% BSA (1.25–20 μg/ml final IgG concentration). Diluted IgG anti–NR2 glutamate receptor antibodies or normal control IgG (100 μl) was applied to each well, and binding to the synthetic DWEYSVWLSN peptide (optical density [OD] values) was determined as previously described (6).

Binding of purified IgG anti-NR2 antibodies to dsDNA and cardiolipin.

Binding of purified IgG anti-NR2 antibodies and normal control IgG to dsDNA and cardiolipin was performed using Mesacup DNA-II test (double-stranded) and Mesacup cardiolipin test (both from MBL), respectively. Purified IgG anti-NR2 antibodies from 14 patients with SLE and purified IgG from 5 normal control subjects were diluted in PBS (1.25–20 μg/ml final IgG concentration). The binding of IgG anti-NR2 antibodies from SLE patients and of IgG from normal controls to dsDNA antigen or to cardiolipin (OD values) was determined according to the manufacturer's instructions.

Culture of human endothelial cells.

ECs were prepared from a fresh human umbilical vein and cultured as described elsewhere (12). ECs were serially passaged by brief exposure to 0.25% trypsin (Difco) and 0.04% EDTA (Sigma). Only cells from the second passage were used. The cells were positive for the von Willebrand factor.

Binding of purified IgG anti-NR2 antibodies to endothelial cells.

ECs (1 × 105 cells/well) were seeded in 96-well flat-bottomed tissue culture plates (Costar) that had been precoated with 5% gelatin (Sigma). After 2 days in culture, the cells reached confluence as a monolayer, and 200 μl of IgG anti-NR2 antibodies or normal control IgG diluted in medium 199 (Nissui) containing 1% fetal calf serum (FCS; 0.31–20 μg/ml final IgG concentration) (MBL) or medium 199 containing 1% FCS was added to each well, and the wells were incubated for 1 hour at 37°C. Peroxidase-conjugated rabbit antihuman IgG (Dako) diluted 1:1,000 in washing buffer was used as the secondary antibody. The binding of IgG anti-NR2 antibodies and normal control IgG to ECs (OD values) was determined as described previously (13).

ELISAs for adhesion molecule expression and cytokine production by endothelial cells.

ECs (1 × 105 cells/well) were seeded in 96-well flat-bottomed tissue culture plates that had been precoated with 5% gelatin. After 2 days in culture, the cells reached confluence as a monolayer, and 200 μl of IgG anti-NR2 antibodies or normal control IgG diluted in M199 containing 1% FCS (20 μg/ml final IgG concentration), medium 199 containing 1% FCS, or 5 units (17.9 pg)/ml of interleukin-1β (IL-1β) in medium 199 containing 1% FCS was added to each well, and the wells were incubated at 37°C for different periods of time (maximum 24 hours). After incubation, cell-free supernatants were collected and frozen at –30°C until the measurement of cytokine levels. After cell-free supernatants were removed, the EC monolayers were fixed by incubation for 10 minutes at room temperature with 1% paraformaldehyde in PBS. The expression levels of endothelial leukocyte adhesion molecule 1 (ELAM-1), vascular cell adhesion molecule 1 (VCAM-1), and intercellular adhesion molecule 1 (ICAM-1) on the EC surface (OD values) were determined as previously described (13).

Determinations of tumor necrosis factor α (TNFα), IL-1β, IL-6, and IL-8.

Concentrations of TNFα, IL-1β, IL-6, and IL-8 in cell-free supernatants were evaluated by specific immunoassays for human TNFα, IL-1β, IL-6, and IL-8, respectively (Quantikine; R&D Systems). All assays were performed according to the manufacturer's instructions.

Degradation of IκB.

In order to monitor the degradation of IκB, ECs were stimulated for 30 minutes with IL-1β (10 ng/ml), IgG anti-NR2 antibodies from 2 patients with SLE, or IgG from 1 normal control subject (range of IgG concentrations 2.5–20 μg/ml) or were incubated in medium alone, as previously described (13). Western blot analysis was performed by standard methods using an anti-IκBα antibody (SC-371; Santa Cruz Biotechnology), as described previously (14).

Statistical analysis.

Experiments were performed in triplicate. Data are reported as the mean ± SD. Differences between groups were compared by the nonparametric Mann-Whitney U test.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

Binding of IgG anti–NR2 glutamate receptor antibodies to the synthetic DWEYSVWLSN peptide.

IgG anti-NR2 antibodies obtained from 2 patients with SLE bound to the synthetic DWEYSVWLSN peptide in an IgG-dose–dependent manner at IgG concentrations ranging from 1.25 μg/ml to 20 μg/ml (Figure 1). IgG anti-NR2 antibodies from the other 12 patients with SLE showed similar results (data not shown). IgG from the 5 normal control subjects did not bind to a synthetic DWEYSVWLSN peptide (data not shown).

thumbnail image

Figure 1. Binding of IgG anti–NR2 glutamate receptor antibodies from 2 patients with systemic lupus erythematosus to a synthetic DWEYSVWLSN peptide.

Download figure to PowerPoint

Binding of IgG anti-NR2 antibodies to dsDNA and cardiolipin.

IgG anti–NR2 glutamate receptor antibodies from 2 patients with SLE bound to dsDNA in an IgG-dose–dependent manner over a range of IgG concentrations from 1.25 μg/ml to 20 μg/ml (Figure 2). IgG anti-NR2 antibodies from 12 other patients with SLE showed similar results (data not shown). IgG from the 5 normal control subjects did not bind to dsDNA (data not shown).

thumbnail image

Figure 2. Binding of IgG anti–NR2 glutamate receptor antibodies from 2 patients with systemic lupus erythematosus to double-stranded DNA.

Download figure to PowerPoint

IgG anti-NR2 antibodies from all 14 patients with SLE did not bind to cardiolipin, nor did IgG from all 5 normal control subjects (data not shown).

Taken together, the results reported above indicate that purified IgG anti-NR2 antibodies from the 14 SLE patients evaluated in our study react with the common DWEYS sequence, which is a molecular mimic of dsDNA and is also present in the extracellular domain of murine and human NMDA receptor subunits NR2a and NR2b, but not cardiolipin.

Binding of IgG anti-NR2 antibodies to non-fixed ECs.

IgG anti-NR2 antibodies obtained from patients with SLE bound to non-fixed ECs in an IgG-dose–dependent manner at IgG concentrations ranging from 0.31 μg/ml to 20 μg/ml (data not shown). We therefore analyzed the binding of IgG anti-NR2 antibodies from the 14 SLE patients and IgG from the 5 normal control subjects to non-fixed ECs (20 μg/ml of IgG added to non-fixed ECs). The OD value of medium alone (medium 199 containing 1% FCS) was adjusted to 0. The OD values for EC binding (shown as the mean of triplicate measurements) are shown in Figure 3A. The mean ± SD titer of IgG binding to non-fixed ECs derived from 5 normal controls was 0.048 ± 0.041. A titer >3SD above the mean (0.174) was considered a positive result. The mean titer of IgG anti-NR2 antibody binding to non-fixed ECs (0.436 ± 0.149) was significantly higher in the 14 patients with SLE than in the 5 normal controls (P = 0.001), and all 14 SLE patients had positive results (Figure 3A).

thumbnail image

Figure 3. A, Binding of IgG anti–NR2 glutamate receptor antibodies from 14 patients with systemic lupus erythematosus (SLE) and IgG from 5 normal control (NC) subjects to non-fixed endothelial cells (ECs). B–D, Effects of IgG anti–NR2 glutamate receptor antibodies from 14 SLE patients, IgG from 5 normal control subjects, and interleukin-1β (IL-1β) on endothelial leukocyte adhesion molecule 1 (ELAM-1) expression on the EC surface after 4 hours of incubation (B), vascular cell adhesion molecule 1 (VCAM-1) expression on the EC surface after 8 hours of incubation (C), and intercellular adhesion molecule 1 (ICAM-1) expression on the EC surface after 24 hours of incubation (D). Values are the mean of triplicate determinations. Each symbol represents a single subject; horizontal lines show the mean; broken lines show the mean + 3SD in 5 normal controls.

Download figure to PowerPoint

Effects of IgG anti-NR2 antibodies on EC adhesion molecule expression.

IL-1β (5 units/ml) and IgG anti-NR2 antibodies from 3 patients with SLE (20 μg/ml of IgG added to ECs) changed the expression of ELAM-1, VCAM-1, and ICAM-1 on the EC surface in a time-dependent manner. Maximal expression of ELAM-1 on the EC surface induced by IL-1β and IgG anti-NR2 antibodies was reached by 4 hours and decreased thereafter (data not shown). IL-1β and IgG anti-NR2 antibody induction of maximum VCAM-1 expression on the EC surface was reached after 8 hours, decreasing thereafter (data not shown). ICAM-1 expression on the EC surface induced by IL-1β and IgG anti-NR2 antibodies gradually increased in a time-dependent manner (data not shown). We therefore analyzed the expression of ELAM-1, VCAM-1, and ICAM-1 in the presence of IL-1β (5 units/ml), IgG anti-NR2 antibodies from the 14 patients with SLE, and IgG from the 5 normal control subjects (20 μg/ml of IgG added to ECs) after 4, 8, and 24 hours of incubation, respectively.

IL-1β (5 units/ml) enhanced the expression of ELAM-1 (1.478 OD units), VCAM-1 (1.639 OD units), and ICAM-1 (1.314 OD units) (Figures 3B, C, and D). The mean ± SD OD values for ELAM-1, VCAM-1, and ICAM-1 expression in the 5 normal control subjects were 0.163 ± 0.011, 0.190 ± 0.009, and 0.568 ± 0.030; values >3SD above the mean (0.195, 0.216, and 0.657, respectively) were considered positive. The mean OD values for ELAM-1, VCAM-1, and ICAM-1 expression in the 14 SLE patients (0.221 ± 0.021, 0.275 ± 0.065, and 0.746 ± 0.103) were significantly higher than those in the normal controls (P = 0.001 for each comparison), and 14, 11, and 12 of the SLE patients, respectively, had positive results (Figures 3B, C, and D).

Cytokine production by ECs in the presence of IgG anti-NR2 antibodies.

The production of IL-6 and IL-8 by ECs in the presence of IL-1β (5 units/ml), IgG anti-NR2 antibodies from 14 patients with SLE, and IgG from 5 normal control subjects (20 μg/ml of IgG added to ECs) after 24 hours of incubation was analyzed. IL-1β induced largely IL-6 (62.7 pg/105 cells) and IL-8 (300 pg/105 cells) production by ECs (Figures 4A and B). The mean ± SD production of IL-6 and IL-8 by ECs from 5 normal controls was 20.4 ± 0.802 pg/105 cells and 88.1 ± 4.16 pg/105 cells, respectively, and titers >3SD above the mean (22.8 pg/105 cells and 100.6 pg/105 cells) were considered positive (Figures 4A and B). The mean production of IL-6 and IL-8 (31.8 ± 11.1 pg/105 cells and 121.0 ± 36.6 pg/105 cells) was significantly higher in the 14 SLE patients than in the 5 normal control subjects (P = 0.003 and P = 0.012, respectively), and 9 and 10 of the SLE patients, respectively, had positive results (Figures 4A and B). However, IgG anti-NR2 antibodies from 14 patients with SLE (20 μg/ml IgG added to ECs) did not induce the production of TNFα or IL-1β by ECs throughout 24 hours of incubation (data not shown).

thumbnail image

Figure 4. Release of interleukin-6 (IL-6) (A) and IL-8 (B) from endothelial cells (ECs) induced by IgG anti–NR2 glutamate receptor antibodies from 14 patients with systemic lupus erythematosus (SLE), IgG from 5 normal control subjects, and IL-1β after 24 hours of incubation. Values are the mean of triplicate determinations. Each symbol represents a single subject; horizontal lines show the mean; broken lines show the mean + 3SD in 5 normal controls.

Download figure to PowerPoint

IgG anti-NR2 antibody–induced degradation of IκBα.

To investigate whether the EC activation induced by IgG anti-NR2 antibodies is due to stimulation of the NF-κB pathway, we analyzed the degradation of cytoplasmic IκBα that had been pretreated with IgG anti-NR2 antibodies. Coomassie blue staining revealed that an equal amount and quality of proteins were loaded on the gel (data not shown). Stimulation with IL-1β (10 ng/ml; positive control) completely degraded the cytoplasmic IκBα in ECs (Figure 5). Anti-NR2 antibodies obtained from 2 SLE patients substantially decreased in an IgG-dose–dependent manner the levels of cytoplasmic IκBα in ECs (Figure 5). However, IgG from a normal control subject did not affect the protein levels of IκBα. These results strongly suggested that the IgG anti-NR2 antibody–induced enhanced cytokine secretion by ECs and up-regulated expression of adhesion molecules on the EC surface are the result of NF-κB pathway stimulation through the degradation of cytoplasmic IκBα.

thumbnail image

Figure 5. Degradation of cytoplasmic IκBα in endothelial cells induced by IgG anti–NR2 glutamate receptor antibodies obtained from 2 patients with systemic lupus erythematosus (SLE), by medium alone (–), and by interleukin-1β (IL-1β).

Download figure to PowerPoint

Figure 6 is a schematic representation of the mechanism by which IgG anti-NR2 antibodies activate the NF-κB pathway, resulting in the pathogenesis of NPSLE.

thumbnail image

Figure 6. Schematic representation of the mechanism by which IgG anti–NR2 glutamate receptor antibodies activate the NF-κB pathway, resulting in the pathogenesis of neuropsychiatric systemic lupus erythematosus (NPSLE). IgG anti–NR2 glutamate receptor antibodies that are cross-reactive with anti–double-stranded DNA (anti-dsDNA) antibodies might bind to endothelial cells (ECs) through N-methyl-D-aspartate (NMDA) receptor subunits NR2a and NR2b, resulting in activation of NF-κB signaling by ECs, promoting the pathogenesis of NPSLE. Encircled p attached to IκBα indicates phosphorylated IκBα.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

In our study, purified IgG anti–NR2 glutamate receptor antibodies from patients with SLE bound to and activated ECs derived from human umbilical veins by up-regulating adhesion molecule expression and increasing proinflammatory cytokine and chemokine release from ECs. Human cerebral ECs were recently reported to express mRNA and protein for NMDA receptor subunits NR2a and NR2b (8). Thus, it is necessary to elucidate the expression of NMDA receptor subunits NR2a and NR2b on the surface of ECs derived from human umbilical veins.

Anti-DNA antibodies derived from patients with SLE have been reported to bind to human umbilical vein EC membrane proteins of MW 180, 110, 84, 68, 46, 44, and 35–30 kd (7). The molecular weight of NMDA receptor subunits NR2a and NR2b has been reported to be 180 kd, suggesting that the NMDA receptor is indeed one of the cell membrane proteins that react with anti-DNA antibodies (8). The findings of our study as well as published reports suggest the possibility that IgG anti-NR2 antibodies possessing cross-reactivity with anti-DNA antibodies bind to ECs through NMDA receptor subunits NR2a and NR2b, resulting in the activation of NF-κB signaling in ECs, which promote the pathogenesis of NPSLE (Figure 6).

The results of our study suggest the possibility that IgG anti-NR2 antibodies in the serum might bind to and activate ECs in the CNS, resulting in the cerebrovascular endothelial inflammation seen in vivo. Furthermore, this inflammation might promote the diffusion of IgG anti-NR2 antibodies from the serum to the CSF. We recently reported the significant relationship between CSF titers of IgG anti-NR2 antibodies and the presence of NPSLE as well as the significant positive relationship between serum titers of IgG anti-NR2 antibody and CSF titers of IgG anti-NR2 antibody in patients with SLE (6).

Our findings also support the possibility that anti-NR2 antibodies are produced outside the CNS and enter the CSF as a result of the inflammation of cerebral ECs that is caused by anti-NR2 antibodies, although it is unclear whether the function of human umbilical vein ECs is the same as the function of cerebral ECs. The mechanism of EC activation by IgG anti-NR2 antibodies was suggested by the observation that IgG anti-NR2 antibodies bind to NMDA receptor subunits NR2a and NR2b on the EC surface and this binding stimulates the activation of the NF-κB pathway in the ECs (Figure 6). Further studies are required to elucidate whether anti-NR2 antibodies in the CSF bind to neuronal tissue and cause neuronal dysfunction or damage and whether this causes the cognitive disturbance and memory loss commonly observed in SLE patients.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Yoshio had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Yoshio, Okamoto, Hirohata, Minota.

Acquisition of data. Yoshio, Okamoto.

Analysis and interpretation of data. Yoshio, Okamoto.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES