Ephrin-B2 prevents N-methyl-D-aspartate receptor antibody effects on memory and neuroplasticity

Abstract

Objective: To demonstrate that ephrin-B2 (the ligand of EphB2 receptor) antagonizes the pathogenic effects of patients' N-methyl-D-aspartate receptor (NMDAR) antibodies on memory and synaptic plasticity.

Methods: One hundred twenty-two C57BL/6J mice infused with cerebrospinal fluid (CSF) from patients with anti-NMDAR encephalitis or controls, with or without ephrin-B2, were investigated. CSF was infused through ventricular catheters connected to subcutaneous osmotic pumps over 14 days. Memory, behavioral tasks, locomotor activity, presence of human antibodies specifically bound to hippocampal NMDAR, and antibody effects on the density of cell-surface and synaptic NMDAR and EphB2 were examined at different time points using reported techniques. Short- and long-term synaptic plasticity were determined in acute brain sections; the Schaffer collateral pathway was stimulated and the field excitatory postsynaptic potentials were recorded in the CA1 region of the hippocampus.

Results: Mice infused with patients' CSF, but not control CSF, developed progressive memory deficit and depressive-like behavior along with deposits of NMDAR antibodies in the hippocampus. These findings were associated with a decrease of the density of cell-surface and synaptic NMDAR and EphB2, and marked impairment of long-term synaptic plasticity without altering short-term plasticity. Administration of ephrin-B2 prevented the pathogenic effects of the antibodies in all the investigated paradigms assessing memory, depressive-like behavior, density of cell-surface and synaptic NMDAR and EphB2, and long-term synaptic plasticity.

Interpretation: Administration of ephrin-B2 prevents the pathogenic effects of anti-NMDAR encephalitis antibodies on memory and behavior, levels of cell-surface NMDAR, and synaptic plasticity. These findings reveal a strategy beyond immunotherapy to antagonize patients' antibody effects. Ann Neurol 2016;80:388-400.

FIGURE 1

Soluble ephrin-B2 does not interfere with antibody binding and prevents antibody-induced reduction of surface NMDAR and EphB2 clusters in cultures of neurons. (A) Sections of mouse hippocampus incubated with pooled CSF from patients with anti-NMDAR encephalitis not preabsorbed (left), and pre-absorbed with HEK293 cells expressing (middle) and not expressing (right) the GluN1 subunit of the NMDARs. Preabsorption with GluN1 abrogates patients’ CSF reactivity with brain; scale bar = 200 μm. (B) Cultured rat hippocampal neurons incubated for 1 hour with patients’ or control CSF with or without ephrin-B2 show similar immunolabeling by patients’ antibodies regardless of the presence of ephrin-B2; scale bar = 10 μm. (C) Quantification of intensity of CSF IgG reactivity (10 dendrites per condition); bars show the mean intensity + standard error of the mean in percentage relative to patients’ CSF IgG reactivity. Significance assessed by one-way analysis of variance (ANOVA; p <0.0001) with Bonferroni post-hoc correction; ****p <0.0001. (D) Immunoprecipitation of the neuronal antigen bound to patients’ IgG showed that the target was the NMDAR (band at ~110kDa); the same result was obtained in neurons incubated with patients’ CSF with or without ephrin-B2. (E) Representative dendrite of hippocampal neurons immunostained for surface NMDAR (green) and PSD95 (red) after 24-hour treatment with patients’ CSF antibodies (Pt CSF) or control CSF (Ct CSF) without or with ephrin-B2. Patients’ antibodies reacted with surface NMDAR in nonpermeabilized neurons; synaptic NMDARs were defined by the colocalization of reactivity with PSD95 (yellow). (F) Immunostaining of surface EphB2 after 24-hour treatment with Pt CSF or Ct CSF without or with ephrin-B2. Scale bars = 10 μm. (G) Quantification of the density of surface and synaptic NMDAR: The presence of ephrin-B2 prevented the antibody-mediated reduction of surface and synaptic NMDAR clusters. PSD95 was not affected by ephrin-B2. (H) Quantification of the density of EphB2: The presence of ephrin-B2 prevented the antibody-mediated reduction of EphB2 (n = 10 cells per condition; three independent experiments). Cluster density analysis was performed with a spot detection algorithm from Imaris suite 7.6.4 (Bitplane). (^$) in Y axis represents the number of surface clusters (NMDAR or EphB2) or synaptic clusters (NMDAR) per dendrite (measured three-dimensionally [3D])/length of the dendrite; (^$$) represents the number of intracellular clusters of PSD95 (measured in 3D)/length of dendrite. Significance of treatment effect was assessed by one-way ANOVA (p <0.0001 for NMDAR, synaptic NMDAR, and EphB2) with Bonferroni post-hoc correction; ****p <0.0001. CSF = cerebrospinal fluid; IgG = immunoglobulin G; NMDAR = N-methyl-D-aspartate receptor.

FIGURE 2

Intraventricular infusion of CSF from patients with NMDAR antibodies causes deficits in memory and depressive-like behavior that are prevented by ephrin-B2. (A) Schedule of cognitive testing and animal sacrifice. Memory (novel object recognition [NOR] discrimination index), depressive-like behavior (tail suspension test [TST] and forced swimming test [FST]), and locomotor activity (LOC) were assessed blinded to treatment at the indicated days. The NOR discrimination index was assessed in open field in four different cohorts of mice. Animals were habituated the day before surgery (baseline) to NOR and LOC. Red arrow heads indicate the days of sacrifice for studies of effects of antibodies in brain. (B) NOR index in open-field paradigm in animals treated with patients’ CSF antibodies (Pt CSF; solid circles), Pt CSF + ephrin-B2 (open circles), control CSF (Ct CSF; solid squares), or Ct CSF + ephrin-B2 (open squares). A high index indicates better object recognition memory. (C) Total time of immobility in TST during the infusion period (day 12). (D) Total time of immobility in FST (day 20). Data are presented as mean ± standard error of the mean (median ± interquartile range IQR in C and D). Number of animals: patients’ CSF, n = 12; patients’ CSF + ephrin-B2, n = 13; control CSF, n = 13; and control CSF + ephrin-B2, n = 13. Significance of treatment effect was assessed by repeated-measures two-way analysis of variance (ANOVA; p = 0.0001; B) with post-hoc testing with Bonferroni adjustment (asterisks), or one-way ANOVA (p = 0.0032) and Bonferroni post-hoc correction (C). Patients’ CSF versus control CSF: **p <0.01; ****p <0.0001; patients’ CSF versus patients’ CSF + ephrin-B2: ^$p <0.05; ^$$$$p <0.0001; patients’ CSF versus control CSF + ephrin-B2: ^∘∘p <0.01; ^∘∘∘p <0.001; ^∘∘∘∘p <0.0001. CSF = cerebrospinal fluid; NMDAR = N-methyl-D-aspartate receptor.

FIGURE 3

Animals infused with patients’ CSF have a progressive increase of human anti-NMDAR IgG bound to hippocampus that is not altered by ephrin-B2. (A) Immunostaining of human IgG in sagittal hippocampal sections of mice infused with patients’ CSF antibodies (Pt CSF), Pt CSF + ephrin-B2, control CSF (Ct CSF), and Ct CSF + ephrin-B2, sacrificed at the indicated experimental days. In animals infused with patients’ CSF and patients’ CSF + ephrin-B2, there is a gradual increase of IgG immunostaining until day 18, followed by a decrease of immunostaining. Scale bar: A = 200 μm. (B) Quantification of intensity of human IgG immunostaining in hippocampus of mice infused with patients’ CSF (red bars), patients’ CSF + ephrin-B2 (gray bars), control CSF (blue bars), and control CSF + ephrin-B2 (cyan bars) sacrificed at the indicated time points. For all quantifications, mean intensity of IgG immunostaining in the group with the highest value (animals treated with patients’ CSF and sacrificed at day 18) was defined as 100%. All data are presented as mean ± standard error of the mean. For each time point, 5 animals of each experimental group were examined. Significance of treatment effect was assessed by two-way analysis of variance (ANOVA; time points, treatment, and interaction, all p <0.0001), and post-hoc analyses were performed with Bonferroni correction; ****p <0.0001. (C and D) Demonstration that the human IgG in mouse brain has NMDAR specificity: HEK293 cells expressing the GluN1 subunit of the NMDAR immunolabeled with acid-extracted IgG fractions (top row in C) or pre-extraction fractions (top row in D) from hippocampus of mice infused with patients’ CSF antibodies (Pt CSF), Pt CSF + ephrin-B2, control CSF (Ct CSF), or Ct CSF + ephrin-B2 at day 18. The intense reactivity with GluN1-expressing cells was noted in acid-extracted IgG fractions from Pt CSF and Pt CSF + ephrin-B2 groups (C); none of the pre-extraction fractions from any animal group showed GluN1 reactivity (D). The second row in (C) and (D) shows the reactivity with a monoclonal GluN1 antibody, and the third row the colocalization of immunolabeling. Scale bars = 10 μm. Pt CSF (n = 5), Pt CSF + ephrin-B2 (n = 5), Ct CSF (n = 5), and Ct CSF + ephrin-B2 (n = 5). CSF = cerebrospinal fluid; IgG = immunoglobulin G; NMDAR = N-methyl-D-aspartate receptor.

FIGURE 4

Soluble ephrin-B2 antagonizes the antibody-mediated reduction of NMDAR and EphB2 in mice hippocampus. (A) Hippocampus of mouse immunolabeled for PSD95 and NMDAR. Images were merged (merge) and postprocessed to demonstrate colocalizing clusters. Squares in “merge” indicate the analyzed areas in CA1, CA2, CA3, and dentate gyrus. Each square is a three-dimensional (3D) stack of 50 sections. Scale bar = 200 μm. (B) 3D projection and analysis of the density of total clusters of PSD95 and NMDAR, and synaptic clusters of NMDAR (defined as NMDAR clusters colocalizing with PSD95) in a CA3 region (square in A “merge”) from a representative animal of each experimental group. Merged images (merge: PSD95 [green]/NMDAR [red]) were postprocessed and used to calculate the density of clusters (density = spots/μm³). Scale bar = 2 μm. (C) Density of total clusters of EphB2 and EphB2 colocalizing with NMDAR. Scale bar = 2 μm. (D) Quantification of the density of total (left) and synaptic (right) NMDAR clusters, and (E) total EphB2 and EphB2 colocalizing with NMDAR at day 18 in a pooled analysis of hippocampal areas (CA1, CA2, CA3, and dentate gyrus) in animals treated with patients’ CSF antibodies (Pt CSF; red), Pt CSF + ephrin-B2 (gray), control CSF (Ct CSF; blue), and control Ct CSF + ephrin-B2 (cyan). Mean density of clusters in control CSF treated animals was defined as 100%. Data are presented as scatterplot + mean ± standard error of the mean. For each condition, 5 animals were examined (18 hippocampal areas per animal = 90 hippocampal areas per condition). Significance of treatment effect was assessed by one-way analysis of variance (p <0.0001) and by post-hoc analysis with Bonferroni correction; **p <0.01; ***p <0.001; ****p <0.0001. CSF = cerebrospinal fluid; NMDAR = N-methyl-D-aspartate receptor.

FIGURE 5

Patients’ antibodies cause severe impairment of long-term synaptic plasticity in the hippocampus that is partially prevented by ephrin-B2. (A) The Schaffer collateral pathway (SC, red) was stimulated (Stim) and field potentials were recorded in the CA1 region of the hippocampus (Rec). Long-term potentiation (LTP) was induced by theta-burst stimulation (TBS); DG = dentate gyrus; CA = Cornu Ammonis. (B) Example traces of individual recordings showing average traces of baseline recording before LTP induction (black traces) and after LTP (red traces). Slope and peak amplitude of fEPSPs are increased after TBS in mice infused with control CSF (Ct CSF) and Ct CSF + ephrin-B2, whereas manifestation of LTP is strongly impaired in animals infused with patients’ CSF antibodies (Pt CSF). In mice infused with Pt CSF + ephrin-B2, the increase of slope is improved. Note that initial peak amplitude of fEPSP may vary within individual recordings. (C) Time course of fEPSP recordings demonstrating robust changes in fEPSP slope in the Ct CSF (n = 7 recordings, blue open circles) and Ct CSF + ephrin-B2 group (n = 7, cyan open squares), which is stable throughout the recording period after TBS (arrow). In animals chronically infused with Pt CSF (n = 7, red solid circles), the induction of synaptic LTP is markedly impaired. Recordings from the Pt CSF + ephrin-B2 group (n = 5, gray solid squares) show partially resolved effects on synaptic plasticity after LTP induction. (D) Quantitative analysis of LTP-induced changes in fEPSPs in the plateau interval after TBS depicted as comparison to each individual baseline value (slope increase as median values ± standard error of the mean in the consolidation phase during the last 50 minutes of each recording, starting 15 minutes after TBS). Chronic application of Pt CSF results in marked reduction of LTP (13.3 ± 4.1% slope increase vs 73.6 ± 19.3% and 68.3 ± 14.7% in Ct CSF and Ct CSF + ephrin-B2, respectively). Coadministration of soluble ephrin-B2 improved fEPSP potentiation to levels of 33.7 ± 5.5%. Significance of treatment effect was assessed by two-way analysis of variance (ANOVA; p <0.0001 for treatment group) and by post-hoc analysis with Bonferroni correction; ***p <0.001. (E) Patients’ antibodies do not alter short-term plasticity, as revealed by paired-pulse facilitation. Short-term plasticity in the Schaffer collateral-CA1 synaptic region shows paired pulse facilitation as measured by mean slope values of the first (1st) and second (2nd) stimulus in the group of mice infused with control CSF (Ct CSF, blue), Ct CSF + ephrin-B2 (cyan), patients’ CSF antibodies (Pt CSF, red), or Pt CSF + ephrin-B2 (grey) before (pale color) and after (dark color) LTP induction. Analysis of 2nd versus 1st stimulation reveals a significant increase of fEPSPs in all groups and at both time points before LTP induction (Ct CSF: p <0.0001; Ct CSF + ephrin-B2: p = 0.0002; Pt CSF: p = 0.0012; Pt CSF + ephrin-B2: p = 0.0004) and after LTP (Ct CSF: p <0.0001; Ct CSF + ephrin-B2: p = 0.0009; Pt CSF: p <0.0001; Pt CSF + ephrin-B2: p <0.0001). Comparison of fEPSPs after the first stimulus before and after LTP induction shows a significant increase in the Pt CSF + ephrin-B2 (p = 0.0007) and both control CSF groups (Ct CSF: p = 0.0003; Ct CSF + ephrin-B2: p = 0.0004), but not in the Pt CSF group (p = 0.21). Analysis was performed using repeated-measures two-way ANOVA (p <0.0001 for treatment groups) and post-hoc analysis with Bonferroni correction; **p <0.01; ***p <0.001; ****p <0.0001. (F) Analysis of fEPSP absolute slope values depending on stimulus amplitude. Increasing stimulation leads to higher slope values reaching a plateau phase at stimulation strength >400 μA (maximum slope and peak amplitude of fEPSP). The fEPSP slope is significantly reduced in mice infused with patients’ CSF antibodies (Pt CSF; red circles). In the Pt CSF + ephrin-B2 group (gray circles), the fEPSP slope is partially restored in comparison to mice infused with Pt CSF. Blue and cyan squares indicate animals infused with control CSF (Ct CSF) and Ct CSF + ephrin-B2, respectively. Two-way ANOVA showed p <0.0001 for increasing stimulation and treatment; post-hoc analysis with Bonferroni correction showed difference between the groups of mice infused with Pt CSF compared with Ct CSF with or without ephrin-B2 (p <0.0001); Pt CSF compared with Pt CSF + ephrin-B2 (p = 0.001); and Pt CSF + ephrin-B2 compared with Ct CSF (p = 0.0007). CSF = cerebrospinal fluid.