IgA NMDA receptor antibodies are markers of synaptic immunity in slow cognitive impairment

Abstract

Objective: To report that antibodies to synaptic proteins may occur in association with slow, progressive cognitive decline.

Methods: A total of 24 patients with progressive cognitive dysfunction of unclear etiology were examined for onconeuronal and synaptic receptor antibodies. The effect of serum was examined in cultures of dissociated mouse hippocampal neurons.

Results: Seven patients had immunoglobulin A (IgA), but no immunoglobulin G (IgG), antibodies against NMDA receptor (NMDAR). Anti-NMDAR IgA positive patients' serum, but not serum from control individuals, caused dramatic decrease of the levels of NMDAR and other synaptic proteins in neurons, along with prominent changes in NMDAR-mediated currents. These effects correlated with the titer of IgA NMDAR antibodies and were reversed after removing patients' serum from the culture media. When available, comprehensive clinical assessment and brain metabolic imaging showed neurologic improvement after immunotherapy.

Conclusions: A subset of patients with slowly progressive cognitive impairment has an underlying synaptic autoimmunity that decreases the density of NMDAR and other synaptic proteins, and alters synaptic currents. This autoimmunity can be demonstrated examining patients' serum and CSF for NMDAR IgA antibodies, identifying possible candidates for immunotherapy.

Figure 1. Intense downregulation of NMDA receptors (NMDAR) by immunoglobulin A (IgA) NMDAR antibodies

(A) Immunostaining of HEK cells transfected with the NR1 subunit of NMDAR and probed with patient serum and anti-IgA secondary antibodies. (B) No staining was observed using an anti–immunoglobulin G (IgG) secondary antibody. (C) Nontransfected HEK293 cells served as negative control. Strong immunopositive staining of primate cerebellum (D) and rat hippocampus (E) with IgA-positive serum (n = 3), but not with control serum (F). (G) Primary mouse hippocampal neurons were incubated for 3 days with patient serum (1:100 dilution). Following removal of incubation medium cells were cultured for further 4 days with growth serum alone (rescue). A membrane fraction was obtained from harvested cells and processed for Western blotting. Staining against NR1 subunits revealed a strong downregulation of NMDAR following incubation with index patient serum. Removal of patient serum restored initial NMDAR levels. Incubation with control serum (CTL) or media (no additions) had no effect on NMDAR expression. Actin was used for loading control. (H) Example trace illustrating the NMDAR current response following a brief UV flash-triggered glutamate uncaging (arrow) onto the cell soma. Cells had been incubated for 3 days in control serum. (I) Note smaller synaptic NMDAR current as a response to the UV flash stimulus after 3-day exposure to serum from the index patient. (J) Grand averages of NMDAR-mediated synaptic responses triggered to UV flash onset including all cells investigated (black and red for control and index patient serum, respectively). (K, left) Cumulative probability plots showing the peak amplitude distributions of NMDAR-mediated currents following UV pulse-triggered glutamate uncaging. Note the systematic reduction of responses in index patient serum treated cells (red distribution) compared to control (p = 0.0008, Kolmogorov-Smirnov test). (K, right) Comparison of group averages demonstrates a ∼90% reduction of NMDAR currents upon glutamate uncaging (111.3 ± 19.4 pA vs 11.7 ± 8.1 pA for control and patient sera; p = 0.0007, rank sum test; n = 10 and n = 7 cells, respectively).

Figure 2. Immunoglobulin A (IgA) antibody titers, immunomodulatory treatment, and titer-dependent pathogenic effect of antibodies

(A) Plasma exchange (PE) was most efficient for short-term reduction of serum IgA antibodies (note log scale of antibody titer). (B) Marked atrophy of the temporal lobes in brain MRI at clinical presentation of the index patient (left). Following immunotherapy, no significant increase in brain atrophy was observed during 6 months (right; identical timescale as in A) as judged by MRI-based volumetry of gray matter (GM) and CSF (C, left). (C, right) Compared to healthy age-matched controls, the reduction of GM volume was most pronounced in the hippocampus (detail) and in frontal and parietal cortex areas (orange). (D) The pathogenic effect of the patient serum on synaptic protein expression was largely reversible after PE: cultivated hippocampal neurons were incubated for 3 days with control serum (left), index patient serum before (middle; 1:100 dilution) and after PE (right). Fixed cells were stained for synapsin (arrows indicate synapsin dots). Incubation with serum before PE resulted in a markedly reduced synapsin expression compared with control serum (similar to figure 3B; p = 0.000008), while serum after PE exhibited only a very slight reduction in the synapsin immunoreactivity, thus rescuing the effect of the pre-PE serum (p = 0.000016). Scale bar: 50 μm. (E) Temporal and frontal brain hypometabolism (FDG-PET) in the index patient and partial improvement following immunotherapy. (a) Transversal slices through the scaled and stereotactically normalized FDG-PET images of the index patient at baseline (before PE) showing frontal and temporal areas of hypometabolism. (b) Comparison of the index patient with stereotactically normalized FDG-PET images of the control group revealed significant hypometabolism (blue) in the frontotemporal brain areas. (c) Transversal images of the index patient immediately after PE (3 weeks after baseline PET in [a]) using the same color table as in (a). (d) Subtraction analysis of changes in scaled FDG uptake from before PE/after PE ([a] − [c]). The cold color table indicates decrease of scaled FDG uptake, whereas the hot color table indicates increase, e.g., arrows pointing to increased metabolism after PE in the temporal region. The subtraction maps are given in units of the SD of change in brain regions with stable FDG uptake and are thresholded at 2.0 SD. The subtraction maps are overlaid to the baseline image. (e) Subtraction analysis of the baseline PET and follow-up imaging 4 months later with increase of metabolism in posterior brain areas.

Figure 3. Effect of immunoglobulin A (IgA)–positive serum is not confined to NMDA receptors (NMDAR)

(A) The same hippocampal cultures used in figure 3A were stained for synaptophysin, a general synaptic vesicle protein. The effects observed were qualitatively the same as observed for NMDAR, namely downregulation after 3 days of incubation with index patient serum and restoration of immunoreactivity following a 4-day rescue with growth medium. (B) Downregulation of synaptic proteins by anti-NMDAR antibodies of the IgA class. Cultured hippocampal neurons were incubated for 3 days with no additions (no add.), control (CTL), or index patient serum (1:100). Fixed cells were stained for the general synaptic proteins synapsin and synaptophysin (shown in green) as well as for the vesicular GABA transporter (VGAT) and the vesicular glutamate transporter 1 (VGLUT1), markers of inhibitory or excitatory synaptic inputs (shown in red). Incubation with patient serum resulted in a reduced expression of all 4 proteins. Incubation with control serum did not affect expression of the proteins. Insets show the merged details of the boxed areas depicted in the immunofluorescence and phase contrast images. Scale bar: 50 μm. (C) Example current traces to illustrate differences in spontaneously occurring excitatory postsynaptic currents (sEPSCs) from cells exposed to control serum (left panel) and index patient serum (right panel). Both recordings were performed at −67 mV to isolate excitatory, presumably AMPAR-mediated, synaptic inputs. (D) Group data analysis: cumulative probability plots of the cell-wise averages for sEPSCs. Note the systematic reduction of values from cells incubated with serum from the index patient (red distribution) compared to control (4.0 ± 1.5 Hz vs 1.0 ± 0.4 Hz, p = 0.02, Kolmogorov-Smirnov test). (E) Cumulative distributions of sEPSC amplitudes. The systematic negative shift of values corresponds to highly significantly different distributions (57.4 ± 6.2 pA vs 40.1 ± 7.9 pA, p = 2.3E–50, Kolmogorov-Smirnov test; n = 18 and n = 11 cells for control and patient serum-treated cells).

Figure 4. Presynaptic binding of immunoglobulin A (IgA) antibodies and pathogenic effect of purified IgA

(A) Double staining of hippocampal neurons with antisynapsin antibodies (red) and human serum followed by antihuman IgA (green). (B) Enlarged insets from A demonstrate colabeling of human IgA antibodies with the presynaptic marker synapsin (arrowheads). (C) In contrast, incubation with control serum (CTL) or media (no add.) did not result in positive staining. (D) Downregulation of synaptic proteins by purified IgA antibodies (1:20) after 3 days of incubation. No effect was observed after incubation with elution buffer (buffer) or control serum (CTL). (D, right) Quantification of synapsin downregulation (p < 0.001) and cell counts. No differences in the number of cells were observed.