NR2-reactive antibody decreases cell viability through augmentation of Ca2+ influx in systemic lupus erythematosus

Authors


Abstract

Objective

Anti–N-methyl-D-aspartate (anti-NMDA) receptor subunit NR2–reactive antibody may play a crucial role in neuronal manifestations of systemic lupus erythematosus (SLE). However, how NR2-reactive antibody acts as a critical modulator of the NMDA receptor is unknown. This study was undertaken to investigate the biologic function of NR2-reactive antibody in patients with SLE.

Methods

The study included 14 patients with SLE, 9 of whom had NR2-reactive antibody. We analyzed the effects of NR2-reactive antibody on cell viability and intracellular Ca2+ level. We also investigated the efficacy of zinc as a modulator of the intracellular Ca2+ level in the presence of NR2-reactive antibody.

Results

There was a significant inverse correlation between the NR2-reactive antibody titer and cell viability (R2 = 0.67, P < 0.0001; n = 23), and there was a significant association between the NR2-reactive antibody titer and the intracellular Ca2+ level in NR1/NR2a-transfected HEK 293 cells (R2 = 0.69, P < 0.0001). Intracellular Ca2+ levels were significantly higher in cells incubated with IgG derived from NR2-reactive antibody–positive SLE patients than in those incubated with IgG derived from NR2-reactive antibody–negative SLE patients (P = 0.0002). The addition of zinc decreased the intracellular Ca2+ level in a dose-dependent manner. NR2-reactive antibody–positive SLE IgG weakened the efficacy of zinc as a negative modulator of the intracellular Ca2+ level.

Conclusion

Our findings indicate that NR2-reactive antibody decreases cell viability by Ca2+ influx in SLE through inhibition of the binding capacity of zinc.

N-methyl-D-aspartate (NMDA) receptors are ligand-gated ion channels that play crucial roles in synaptic transmission and central nervous system (CNS) plasticity. The receptors are heterodimers of NMDA receptor subunits NR1, which bind glycine, and NR2 (NR2a, NR2b, NR2c, or NR2d), which bind glutamate (1). NMDA receptor dysfunction is implicated in multiple brain disorders, including stroke, chronic neurodegeneration, epilepsy, and schizophrenia (2–5). In contrast, NMDA receptor–reactive antibody is observed in various autoimmune disorders, and anti-NR1/NR2 antibody–associated encephalitis has recently been described by several researchers (6–8).

Systemic lupus erythematosus (SLE) is a multisystem inflammatory disorder characterized by the presence of autoantibodies directed against double-stranded DNA (dsDNA). Some anti-dsDNA antibodies cross-react with NR2 and damage neuronal cells via an apoptotic pathway (9). Not all anti-dsDNA antibodies cross-react with NR2 to the same degree. We previously found no association between the anti-dsDNA antibody titer and the NR2-reactive antibody titer in 107 patients with SLE (10). The frequency of serum NR2-reactive antibody positivity is ∼30–40% in patients with SLE (10–14).

Associations between serum NR2-reactive antibody positivity and neuropsychiatric SLE (NPSLE) have been demonstrated in some previous studies (10, 12, 14). In contrast, other studies showed no significant association between serum NR2-reactive antibody positivity and cognitive dysfunction (11, 13, 15, 16). DeGiorgio and colleagues have shown that NR2-reactive antibody breaching the blood–brain barrier can cause neuronal damage via an apoptotic pathway (9, 17). The existence of NR2-reactive antibody in cerebrospinal fluid is an important factor in neuronal damage in SLE. However, how NR2-reactive antibody breaches the CNS through the blood–brain barrier and causes neuronal damage via an apoptotic pathway is not known. Recently, Faust and colleagues demonstrated that NR2-reactive antibody acts as a positive modulator of NMDA receptor–mediated synaptic responses and toxicity and preferentially binds to the open NMDA receptor pore (18). These results suggest that NR2-reactive antibody binding to NMDA receptor prolongs the open state, which would increase Ca2+ influx into cells.

Each NR2 subunit consists of a large extracellular amino-terminal domain, a bilobed agonist-binding domain, a transmembrane domain, and an intracellular C-terminal domain. The amino-terminal domain is composed of the first ∼350 amino acids of the protein (19). Additionally, the amino-terminal domain interacts with various extracellular allosteric modulators, such as zinc in the case of NR2a/NR2b, and plays an important role in fine-tuning the functional properties of the NMDA receptor (19). In enzyme-linked immunosorbent assays (ELISAs), NR2-reactive antibody in patients with SLE can react against the peptide DWEYSVWLSN (9, 16, 17). DWEYS has been reported as the common sequence of residues 283–287 of NR2a/NR2b, which are included in the amino-terminal domain (9, 19). However, the actual sequence of residues 283–287 is DWDYS in NR2a and DEWDY in NR2b, according to the NCBI (http://www.ncbi.nlm.nih.gov/). Asp283 in NR2a and Glu284 in NR2b are considered zinc-binding sites. Zinc binding to the NR2a/NR2b amino-terminal domain modulates NMDA receptor–mediated synaptic responses (20). Therefore, the sequence near residue 283 must be an important region of NR2. We speculated that NR2-reactive antibody could react with the zinc-binding site of NR2 to promote intracellular Ca2+ signaling and damage neuronal cells.

To determine the precise functions of NR2-reactive antibody in patients with SLE, we analyzed the effects of NR2-reactive antibody on cell viability and Ca2+ influx. We also investigated the efficacy of zinc as a modulator of intracellular Ca2+ levels in the presence or absence of NR2-reactive antibody.

MATERIALS AND METHODS

Materials.

Patients were diagnosed as having SLE based on the American College of Rheumatology classification criteria (21). Sera were obtained from 9 NR2-reactive antibody–positive patients, 5 NR2-reactive antibody–negative patients, and 9 healthy controls. IgG was extracted and purified from sera by standard column-based methods. Each IgG sample obtained was used in every experiment. HEK 293 cells were purchased from RIKEN Cell Bank. The plasmid constructs pcDNA3.1-NR1, pcDNA3.1-NR2a, and pcDNA3.1 were generous gifts from Dr. Jon W. Johnson (University of Pittsburgh, Pittsburgh, PA). Dulbecco's modified Eagle's medium (DMEM) was obtained from Gibco BRL. Fulo-3 acetoxymethyl ester was obtained from Molecular Probes. NMDA was purchased from Sigma. Other chemicals were of the highest purity commercially available.

Measurement of NR2-reactive antibody.

We used the peptide ISVSYDDWDYSLE to measure NR2-reactive antibody by ELISA. This peptide is the sequence of NR2a residues 277–289. Serum samples (100 μl) diluted 1:1,000 in 10% Block-Ace (Dainippon) were added to each well, as previously described (10).

Expression of NMDA receptors in HEK cells.

HEK 293 cells were plated at a density of 2.5 × 104 cells in a 4-well dish. HEK 293 cells were grown in DMEM supplemented with 5% fetal bovine serum for 24 hours before transfection (22). We used rat NMDA receptor subunits cloned into expression vectors as previously described (23). Cells were transfected at a 1:3 ratio with NR1 and NR2a subunit expression vector by the calcium phosphate coprecipitation method, followed by further culture for an additional 24 hours (24).

Determination of cell viability.

Cell viability was measured by MTT reduction colorimetric assays with minor modifications (23, 25). As described above, HEK 293 cells were incubated for 24 hours after transfection. Cells were then cultured for another 24 hours in DMEM containing zinc at 1 μM, 10 μM, or 100 μM and purified IgG obtained from sera at 0.1 mg/ml. NR1/NR2a-transfected cells were then washed once with phosphate buffered saline (PBS) and incubated with 0.5 mg/ml MTT in PBS for 2 hours. NR1/NR2a-transfected cells were then solubilized by the addition of a lysis solution containing 99.5% isopropanol and 0.04M HCl. The amount of MTT formazan product was determined by measuring the absorbance at 550 nm on a microplate reader. Relative values were calculated as percentages above the value obtained in cells with empty vector (control group). Additionally, the background value obtained under cell-free conditions was subtracted from the total value.

Measurement of intracellular Ca2+ level.

Intracellular Ca2+ was measured as previously described (24, 26). After transfection, incubated cells were washed twice with recording medium containing 129 mM NaCl, 4 mM KCl, 1 mM MgCl2, 2 mM CaCl2, 4.2 mM glucose, and 10 mM HEPES (pH 7.4), followed by incubation at 37°C for 1 hour in recording medium, which contained 60 nM Pluronic F-127, and 3 μM Fluo-3. Cells were washed with recording medium once, followed by settling for 1 hour in recording medium containing 0.1 mg/ml purified IgG obtained from sera. Confocal laser scanning microscopy was used to observe the intracellular free Ca2+ level. The medium was changed once more, followed by exposure to NMDA at 0.1 μM, 1.0 μM, 10 μM, 100 μM, or 1,000 μM. NMDA was prepared in recording medium immediately before each use. The calcium ionophore A23187 was then added at 10 μM to obtain the maximum fluorescence for quantitative normalization. Fluorescence was recorded using an excitation wavelength of 488 nm. The data obtained were normalized to the basal level of fluorescence intensity in cells exposed to 10 μM A23187.

Evaluation of the effect of zinc on intracellular free Ca2+ levels.

As described above, after transfection, HEK 293 cells were washed with recording medium, followed by incubation at 37°C for 1 hour in recording medium containing 60 nM Pluronic F-127 and 3 μM Fluo-3. Incubated cells were washed with recording medium once, followed by settling for 1 hour in recording medium containing 0.1 mg/ml purified IgG obtained from sera. The medium was changed once more. NMDA (10 μM) was added, followed by exposure to zinc at 0.01 μM, 0.1 μM, 1.0 μM, 10 μM, or 100 μM. Intracellular free Ca2+ was measured as described above.

Statistical analysis.

The t-test was used to compare mean values, and the Mann-Whitney U test was used to compare median values. Regression analysis was performed when appropriate. Correlations were measured with Pearson's correlation coefficient. The data were analyzed with JMP software (SAS Institute). P values less than 0.05 were considered significant.

RESULTS

Association between anti-dsDNA antibody and NR2-reactive antibody in SLE patients.

Serum NR2-reactive antibody was measured by ELISA, using the DWDYS peptide, which comprises residues 283–287 of NR2a (10). NR2-reactive antibody positivity was defined as an optical density (OD) of >0.62. This cutoff value was based on a mean ± SD OD of 0.62 ± 4 determined in 74 non-SLE serum samples, including samples from 21 patients with rheumatoid arthritis, 19 patients with systemic sclerosis, 22 patients with polymyositis/dermatomyositis, and 12 healthy controls. Anti-dsDNA antibody was measured by radioimmunoassay. (The normal value is <6 IU/ml.) Neuropsychiatric symptoms were observed in 5 (56%) of 9 patients with NR2-reactive antibody–positive SLE.

Five NR2-reactive antibody–negative patients were also enrolled in the study in order to evaluate the association between anti-dsDNA antibody and NR2-reactive antibody. There was no significant difference in anti-dsDNA antibody titer between the 9 patients with NR2-reactive antibody and the 5 patients without NR2-reactive antibody (P = 0.20). There was no statistically significant correlation between NR2-reactive antibody titer and anti-dsDNA antibody titer (R2 = 0.24, P = 0.09; n = 14).

Association between NR2-reactive antibody titer and NMDA receptor–induced cell viability.

To estimate the effect of NR2-reactive antibody in patients with SLE on cell viability, NR1/NR2a-transfected HEK 293 cells were incubated with healthy control IgG (0.1 mg/ml; n = 9), NR2-reactive antibody–negative SLE IgG (0.1 mg/ml; n = 5), or NR2-reactive antibody–positive SLE IgG (0.1 mg/ml; n = 9). Cell viability was calculated as the percentage above the value obtained in control cells that were transfected with empty vector alone. The concentration of NR2-reactive antibody is expressed as the OD value. Regression analysis was performed with NR2-reactive antibody as the independent variable and cell viability as the dependent variable. There was a statistically significant association between NR2-reactive antibody titer and cell viability (R2 = 0.67, P < 0.0001; n = 23) (Figure 1A). There was no significant association between anti-dsDNA antibody titer and cell viability in SLE (R2 = 0.05, P = 0.46; n = 14) (Figure 1B). These results indicate that NR2-reactive antibody inhibits cell viability in a dose-dependent manner.

Figure 1.

A, Significant association between N-methyl-D-aspartate receptor subunit NR2–reactive antibody titers and cell viability. B, Lack of a significant association between anti–double-stranded DNA (anti-dsDNA) antibody titers and cell viability. Values on the x-axis in B are the base 10 logarithm. Circles represent individual samples. OD = optical density.

Effects of NR2-reactive antibody on NMDA receptor–related Ca2+ influx into cells.

Because NMDA receptor is involved in the regulation of intracellular Ca2+ levels, we investigated the effect of NR2-reactive antibody on intracellular Ca2+ levels. NR1/NR2a-transfected HEK 293 cells were incubated with healthy control IgG (0.1 mg/ml; n = 9), NR2-reactive antibody–negative SLE IgG (0.1 mg/ml; n = 5), or NR2-reactive antibody–positive SLE IgG (0.1 mg/ml; n = 9). The HEK 293 cells in conditioned media were stimulated with 10 μM NMDA, and the intracellular Ca2+ levels were evaluated by Fluo-3 and confocal laser scanning microscopy. Data were normalized to the fluorescence intensity in cells exposed to 10 μM A23187. The value of NR2-reactive antibody is expressed as the OD value. Regression analysis was performed with NR2-reactive antibody as the independent variable and intracellular Ca2+ level as the dependent variable. We also investigated the association between anti-dsDNA antibody titer and intracellular Ca2+ level. Anti-dsDNA antibody was not detected in the 9 healthy controls.

There was a significant association between the NR2-reactive antibody titer and the intracellular Ca2+ level (R2 = 0.69, P < 0.0001; n = 23) (Figure 2A). In contrast, there was no significant association between the anti-dsDNA antibody titer and the intracellular Ca2+ level in SLE (R2 = 0.12, P = 0.22; n = 14) (Figure 2B). These results indicate that NR2-reactive antibody specifically increases the intracellular Ca2+ level in a dose-dependent manner.

Figure 2.

A, Significant association between NR2-reactive antibody titers and intracellular Ca2+ levels. B, Lack of a significant association between anti-dsDNA antibody titers and intracellular Ca2+ levels. Values on the x-axis in B are the base 10 logarithm. Circles represent individual samples. See Figure 1 for definitions.

Comparison of the intracellular Ca2+ level in NR1/NR2a-transfected HEK 293 cells incubated in the presence or absence of NR2-reactive antibody–positive IgG.

Figure 3 shows the intracellular Ca2+ level in NR1/NR2a-transfected HEK 293 cells incubated with healthy control IgG (0.1 mg/ml; n = 9), NR2-reactive antibody–negative SLE IgG (0.1 mg/ml; n = 5), or NR2-reactive antibody–positive SLE IgG (0.1 mg/ml; n = 9). NR1/NR2a-transfected HEK 293 cells treated with the various IgG fractions were exposed to the indicated concentrations of NMDA.

Figure 3.

Comparison of intracellular Ca2+ levels in HEK 293 cells transfected with N-methyl-D-aspartate (NMDA) receptor subunits NR1/NR2a and incubated with and without NR2-reactive antibody. A, Intracellular Ca2+ levels as determined by confocal laser scanning microscopy in the absence of stimulation and in the presence of 10 μM or 100 μM NMDA or 10 μM A23187. Representative results are shown. Bar = 100 μm. B, Comparison of intracellular Ca2+ levels in each subset. As the NMDA concentration increased, the level of intracellular free Ca2+ increased in NR1/NR2a-transfected HEK 293 cells treated with each IgG. The intracellular Ca2+ level was significantly higher in NR1/NR2a-transfected HEK 293 cells treated with IgG from NR2-reactive antibody–positive systemic lupus erythematosus (SLE) patients than in those treated with healthy control IgG or IgG from NR2-reactive antibody–negative SLE patients in the presence of 10 μM NMDA. Bars show the mean ± SEM. ∗ = P < 0.0001 versus healthy control IgG; ∗∗ = P = 0.0002 versus NR2-reactive antibody–negative SLE IgG. Color figure can be viewed in the online issue, which is available at http://onlinelibrary.wiley.com/journal/10.1002/(ISSN)1529-0131.

Figure 3A shows the intracellular Ca2+ level as determined by confocal laser scanning microscopy in unstimulated cells and after the addition of 10 μM or 100 μM NMDA or 10 μM A23187 to each subset. Figure 3B shows that as the concentration of NMDA increased, the level of intracellular free Ca2+ increased in NR1/NR2a-transfected HEK 293 cells treated with each IgG. The intracellular Ca2+ level was significantly higher in NR1/NR2a-transfected HEK 293 cells treated with NR2-reactive antibody–positive SLE IgG than in those treated with healthy control IgG or NR2-reactive antibody–negative SLE IgG in the presence of 10 μM NMDA (P < 0.0001 and P = 0.0002, respectively) (Figures 3A and B). The intracellular Ca2+ level was higher in NR1/NR2a-transfected HEK 293 cells treated with NR2-reactive antibody–positive SLE IgG than in those treated with the other 2 IgG samples in the presence of 100 μM or 1,000 μM NMDA, although the differences were not statistically significant.

Effects of zinc on cell viability.

Amino acid sequencing of NR2 showed that the receptor had a zinc-binding site, and that the binding contributed to the regulation of Ca2+ influx through NMDA receptor signaling. Moreover, NR2-reactive antibody detected in patients with SLE recognizes the amino acids that include the zinc-binding site. These findings prompted us to investigate the effects of zinc on the viability of NR1/NR2a-transfected HEK 293 cells exposed to NR2-reactive antibody. NR1/NR2a-transfected cells were cultured in DMEM containing zinc at 1 μM, 10 μM, or 100 μM. Cell viability was calculated as the percentage above the value obtained in control cells (transfected with empty vector). As shown in Figure 4, as the concentration of zinc increased, greater cell viability was recovered in NR1/NR2a-transfected HEK 293 cells. However, cell viability was significantly lower in NR1/NR2a-transfected HEK 293 cells treated with NR2-reactive antibody–positive SLE IgG than in those treated with healthy control IgG or NR2-reactive antibody–negative SLE IgG in the presence of 10 μM zinc (P < 0.0001 for each comparison). These findings indicate that NR2-reactive antibody interacts with the zinc-binding site, resulting in a decrease in the protection of cell viability provided by zinc to NR1/NR2a-transfected HEK 293 cells.

Figure 4.

Effect of zinc on N-methyl-D-aspartate receptor subunit NR1/NR2a–transfected HEK 293 cell viability in the presence of IgG from each subset. As the concentration of zinc increased, the cell viability in NR1/NR2a-transfected HEK 293 cells treated with IgG from each subset increased. Cell viability was significantly lower in NR1/NR2a-transfected HEK 293 cells treated with IgG from NR2-reactive antibody–positive systemic lupus erythematosus (SLE) patients than in those treated with healthy control IgG or IgG from NR2-reactive antibody–negative SLE patients in the presence of 10 μM zinc. Bars show the mean ± SEM. ∗ = P < 0.0001 versus healthy control IgG and NR2-reactive antibody–negative SLE IgG.

Efficacy of zinc as a modulator of NMDA receptor–stimulated intracellular Ca2+ levels.

Figure 5 shows the efficacy of zinc as a modulator of the intracellular Ca2+ level related to NMDA activity in the presence or absence of NR2-reactive antibody. NR1/NR2a-transfected HEK 293 cells treated with each IgG were exposed to 0.01 μM, 0.1 μM, 1 μM, 10 μM, or 100 μM zinc. The intracellular Ca2+ level was evaluated as described above and is expressed as the mean ± SEM.

Figure 5.

Efficacy of zinc as a modulator of NMDA receptor–induced intracellular Ca2+ level. A, Intracellular Ca2+ levels as determined by confocal laser scanning microscopy in the absence of stimulation and in the presence of both 10 μM NMDA and 10 μM zinc or in the presence of 10 μM A23187 alone. Representative results are shown. Bar = 100 μm. B, Efficacy of zinc as a modulator of intracellular Ca2+ level in the presence or absence of NR2-reactive antibody. NR1/NR2a-transfected HEK 293 cells treated with IgG from each subset were exposed to 10 μM NMDA, followed by exposure to zinc at 0.01 μM, 0.1 μM, 1.0 μM, 10 μM, or 100 μM. As the concentration of zinc increased, the level of intracellular free Ca2+ decreased in NR1/NR2a-transfected HEK 293 cells treated with IgG from each subset. Intracellular Ca2+ was higher in NR1/NR2a-transfected HEK 293 cells treated with NR2-reactive antibody–positive SLE IgG than in those treated with healthy control IgG or NR2-reactive antibody–negative SLE IgG in the presence of 10 μM zinc and in the presence of 100 μM zinc. Values are the mean. ∗ = P = 0.013 versus healthy control IgG and P = 0.0157 versus NR2-reactive antibody–negative SLE IgG; ∗∗ = P = 0.019 versus healthy control IgG and P = 0.047 versus NR2-reactive antibody–negative SLE IgG. See Figure 3 for definitions.

Figure 5A shows the intracellular Ca2+ levels determined by confocal laser scanning microscopy when each subset of cells was exposed to media alone, to both 10 μM NMDA and 10 μM zinc, or to 10 μM A23187 alone. Figure 5B indicates that as the concentration of zinc increased, the levels of intracellular Ca2+ decreased in NR1/NR2a-transfected HEK 293 cells treated with IgG from each subset. After exposure to 10 μM zinc, the intracellular Ca2+ levels were significantly higher in NR1/NR2a-transfected HEK 293 cells treated with NR2-reactive antibody–positive SLE IgG than in those treated with healthy control IgG or NR2-reactive antibody–negative SLE IgG (P = 0.013 and P = 0.0157, respectively) (Figure 5B). This finding indicates that NR2-reactive antibody–positive SLE IgG weakens the efficacy of zinc as a negative modulator of intracellular Ca2+ levels. When 100 μM zinc was used, the concentrations of intracellular Ca2+ were significantly higher in NR1/NR2a-transfected HEK 293 cells treated with NR2-reactive antibody–positive SLE IgG than in those treated with healthy control IgG or NR2-reactive antibody–negative SLE IgG (P = 0.019 and P = 0.047, respectively).

DISCUSSION

We have demonstrated that NR2-reactive antibody derived from SLE patients inhibits cell viability through Ca2+ influx. DeGiorgio and colleagues demonstrated that NR2-reactive antibody mediates neuronal death via an apoptotic pathway in vitro and in vivo (9). Kowal et al reported that NR2-reactive antibody causes cognitive impairment when it accesses the CNS through a breach in the blood–brain barrier (17). Faust and colleagues have demonstrated that NR2-reactive antibody stimulates NMDA receptor–mediated synaptic responses and excitotoxicity through enhanced mitochondrial permeability. In the present study, we did not determine whether NR2-reactive antibody also reacts with NR2b. However, Faust and colleagues have shown that human SLE autoantibodies bind both NR2a and NR2b (18).

NR2a-containing NMDA receptor and NR2b-containing NMDA receptor play different roles both in vitro and in vivo (27). In mature cortical cultures, activation of NR2a-containing NMDA receptor promotes neuronal survival and exerts a neuroprotective action against neuronal damage. In contrast, activation of NR2b-containing NMDA receptor results in excitotoxicity, increasing neuronal apoptosis (27). These findings may indicate that human SLE autoantibodies cause neuronal damage through inhibition of NR2a-containing NMDA receptor signaling or activation of NR2b-containing NMDA receptor signaling. On the other hand, overstimulation of NR2 can cause excitotoxic neuronal death through excessive entry of Ca2+ into cells (9, 28–30).

Although the appropriate intracellular Ca2+ level is important for cell survival, an excessive Ca2+ load can trigger different cell death programs, such as activation of protease, caspase, and other catabolic processes (31). Intracellular Ca2+ is stored in the endoplasmic reticulum or mitochondria. The intracellular Ca2+ level in unstimulated cells is maintained at <100 nM by both uptake into the endoplasmic reticulum and extrusion into the extracellular space by the plasma membrane Ca2+ ATPase (32). Excessive Ca2+ influx into cells triggered by several agents (e.g., Ca2+ ionophores) promotes opening of the mitochondrial permeability transition pore, resulting in the release of cytochrome c and other proapoptotic proteins from mitochondria and the induction of apoptosis (32). In the present study, NR2-reactive antibody titer was correlated with intracellular Ca2+ level. Our results are consistent with those reported by Faust and colleagues (18).

The ligands of NR1 and NR2 are glycine and glutamate, respectively. NMDA receptor is composed of glycine-binding NR1 and glutamate-binding NR2 subunits. The binding of both glycine and glutamate activates intracellular Ca2+ signaling (1, 19). In addition, the gating of NR2 is controlled by the region formed by the NR2 amino-terminal domain and the linker connecting the NR2 amino-terminal domain to the NR2 agonist-binding domain (33).

Zinc is an allosteric inhibitor of NMDA receptor. Allosteric inhibitors likely are critical in the regulation of NMDA receptor activity (33–35). In the present study, it was clear that higher concentrations of zinc inhibited intracellular Ca2+ influx in NR1/NR2a-transfected HEK 293 cells through the addition of NMDA. We also showed that the efficacy of zinc as a negative modulator of intracellular Ca2+ influx was significantly weaker in HEK 293 cells treated with NR2-reactive antibody–positive IgG than in those treated with NR2-reactive antibody–negative IgG. In addition, the viability of HEK 293 cells significantly decreased in NR2-reactive antibody–positive IgG compared to NR2-reactive antibody–negative IgG. These findings indicate that NR2-reactive antibody decreases the efficacy of zinc in regulating NMDA receptor activity. The NR2 amino-terminal domain sequence around Asp283 is important because zinc binding to Asp283 modulates Ca2+ signaling in cells expressing NR2a/NR2b. Our observations suggest that NR2-reactive antibody blocks the zinc-binding site, promotes intercellular Ca2+ influx, and induces apoptosis. Unexpectedly, anti-NR antibody did not directly stimulate the NMDA receptor (NR1 or NR2) to increase Ca2+ influx, suggesting that NR2-reactive antibody in SLE patients is neither agonistic nor antagonistic.

Some anti-dsDNA antibodies cross-react with NR2 and damage neuronal cells via an apoptotic pathway (9). Not all anti-dsDNA antibodies cross-react with NR2 to the same degree. The frequency of serum NR2-reactive antibody positivity is ∼30–40% in patients with SLE (10–14). In the present study, we found no statistically significant correlation between NR2-reactive antibody titers and anti-dsDNA antibody titers. Kowal and colleagues showed that NR2-reactive antibody that reaches the CNS through the blood–brain barrier may potently damage neuronal cells (17). Two main mechanisms for blood–brain barrier damage in SLE have been described: microthrombi in cerebral vessels and immune-mediated attack of the endothelium (36). Anti-NR1/NR2 antibody has been detected in paraneoplastic encephalitis associated with ovarian teratoma and is believed to be produced to cross-react with teratoma as an antigen (6). Both the induction of NR2-reactive antibody in peripheral organs and the impairment of the blood–brain barrier could lead to neuropsychiatric syndrome in SLE patients. In NR2-reactive antibody–associated NPSLE, treatments should be considered that not only eliminate NR2-reactive antibody, but also protect the integrity of the blood–brain barrier and increase the zinc concentration in the CNS.

In conclusion, NR2-reactive antibody has a unique function that binds to the zinc-binding site of NR2 and that inhibits the biologic effects of zinc. NR2-reactive antibody decreases cell viability by Ca2+ influx in SLE.

AUTHOR CONTRIBUTIONS

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. Kawaguchi 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. Gono, Kawaguchi, Yoneda, Yamanaka.

Acquisition of data. Gono, Takarada, Fukumori, Kaneko, Hanaoka.

Analysis and interpretation of data. Gono, Takarada, Fukumori, Kawaguchi, Katsumata.

Ancillary