Enigmatic intractable Epilepsy patients have antibodies that bind glutamate receptor peptides, kill neurons, damage the brain, and cause Generalized Tonic Clonic Seizures

Abstract

Epilepsy affects 1-2% of the world population, is enigmatic in 30% of cases, and is often intractable, unresponsive to antiepileptic drugs, and accompanied by cognitive, psychiatric and behavioral problems. Tests for Autoimmune Epilepsy are not performed routinely, and limited to passive diagnosis of known autoimmune antibodies, without essential functional tests to reveal active pathogenic antibodies. We investigated two young Epilepsy patients with different Epilepsy characteristics, repeated intractable seizures, and enigmatic etiology. We suspected Autoimmune Epilepsy. We found that both patients have elevated IgG antibodies, and three types of glutamate receptor antibodies, to: AMPA-GluR3B, NMDA-NR1 and NMDA-NR2 peptides. In contrast, they lack autoantibodies to: LGI1, CASPR2, GABA-RB1, Amphiphysin, CV2, PNMA1, Ri, Yo, Hu, Recoverin, Soxi and Titin. IgG antibodies of both patients bound and killed human neural cells In vitro. Moreover, In vivo video EEG studies in naive rats revealed that patient's IgG antibodies, infused continually into rat brain, bound neural cells in the hippocampus and cortex, caused neural loss in these brain regions, and induced recurrent Generalized Tonic Clonic Seizures. We assume they can do so also in the patient's brain. This is the first model of human Autoimmune Epilepsy in rats. It can serve for discovery of patient's pathogenic antibodies, and drug development. Tests for autoimmune antibodies that bind glutamate receptor peptides, and functional diagnostic tests, are obligatory in all enigmatic intractable Epilepsy patients. Current diagnosis of Autoimmune Epilepsy is insufficient! If pathogenic antibodies are found, intractable patients must receive available, suitable and potentially life-changing immunotherapies for Autoimmune Epilepsy.

Keywords: AMPA-R antibodies; Autoimmune Epilepsy; Epilepsy; General tonic clonic seizures; GluR3B antibodies; Glutamate receptor antibodies; Glutamate receptors; NMDA-R antibodies.

Fig. 1

Electroclinical seizures of the Epilepsy patients IE-3 and IE-15 studied the current research. A Electroclinical seizure of the Epilepsy patient IE-3 showing onset of a generalized seizure that involves both hemispheres. B Electroclinical seizure of Epilepsy patient IE-3 showing continuation of the seizure lasting for ~ 50 s, manifested by loss of consciousness and generalized tonic clonic movement. C Electroclinical seizure of Epilepsy patient IE-15 showing a focal electroclinical seizure that started in the right posterior head region

Fig. 2

Some patients with severe intractable Epilepsy have a high level of IgG antibodies in their blood. The figure shows the concentration (mg/ml) of IgG antibodies in the serum of young patients with severe intractable Epilepsy including IE-3 and IE-15, and of control healthy subjects. The IgG antibodies were purified from serum samples of all the subjects in a professional protein laboratory within the Hebrew University (It should be noted that the IgG levels were tested only once for each Epilepsy patient or healthy subject, in single serum sample. Thus, statistical values could not be calculated

Fig. 3

The intractable Epilepsy patients studied in the present research IE-3 and IE-15, were found to have elevated levels of three types Glutamate receptor antibodies in their serum: GluR3B peptide antibodies (A), NR1 peptide antibodies (B), and NR2A peptide antibodies (C), as compared to ten healthy subjects. Each of the above figures shows the average level of the respective type of Glutamate Receptor antibodies, in serum of 10 healthy subjects (blue bars), IE-3 (red bars), and 5 serum samples of IE-15 (derived from 5 blood withdrawals of IE-3) at different time points, pink). Each serum was tested in the ELISA in duplicate wells, in 3 different serum dilutions (1:10, 1:100 and 1:1000) for its parallel binding to either the respective GluR peptide, or to control PBS + 1% BSA. The figures show the level of the Glutamate receptor antibodies in very low serum dilution of 1:1000. The Y axis shows the value (in OD), calculated for each individual according to the following equation: specific binding (OD) of each individual serum, in each serum dilution, to the GluR peptide = Specific binding (OD) of this serum, in this serum dilution, to either AMPA GluR3B peptide (A) or NMDA-NR1 (B) or NMDA NR2 peptide (C) − (minus) the non-specific binding (OD) of this serum, in this serum dilution, to negative control PBS + 1% BSA (OD). The experimental cutoff seen in the figure shows the following value: average OD + (2 × SD) of the specific binding to the respective GluR peptide, of each serum of either the 7 healthy subjects tested in the same dilution (1:1000) in the same experiment

Fig. 4

The human neural cells culture used in the current study contained both neurons and astrocytes. The human neural cells used in this study in several experiments, to test the binding and killing of the cells by the IgG of the Epilepsy patients, were grown and differentiated from human embryonic stem cells (hESCs), as previously described (Levite et al. ; Itsykson et al. ; Surmacz et al. 2012), and used successfully in various previous studies. The figure shows that this human neural cell culture contains both neurons (A) and astrocytes (B), demonstrated by specific immunostaining for each cell type: β3 tubulin for neurons, and GFAP for astrocytes, and representative confocal microscopy images. The three upper figures in part A of the figure show the β3 tubulin⁺ neurons stained in green, and all the DAPI⁺ cells stained in blue. The three lower figures in part B of the figure, show the GFAP⁺ astrocytes stained in red, and all the DAPI⁺ cells stained in blue

Fig. 5

IgG antibodies of the intractable Epilepsy patient IE-3 bind human neural cells, unlike IgG antibodies of healthy subjects that do not. A Three representative confocal microscopy photos showing that purified IgG antibodies of the Epilepsy patient IE-3 bind human neural cells (pre-fixed before the binding assay). B Three representative confocal microscopy photos showing that purified IgG antibodies of three healthy subjects do not bind human neural cells. In A and B the neural cells bound by the human IgG are shown in pink, the β3 tubulin⁺ neurons are shown in green, and all the DAPI⁺ cells are shown in blue. B, C Similar to A, but of two additional separate binding experiments. The photos show the binding of purified IgG antibodies of the Epilepsy patient IE-3 to human neural cells. In these photos the neural cells bound by the human IgG are shown in red, the β3 tubulin⁺ neurons are shown in green, and all the DAPI⁺ cells are shown in blue

Fig. 6

IgG antibodies of the intractable Epilepsy patient IE-15 bind human neural cells, unlike IgG antibodies of healthy subjects that do not. A Four representative confocal microscopy photos showing that purified IgG antibodies of the Epilepsy patient IE-15 bind human neural cells. B–F Five representative confocal microscopy photos showing that purified IgG antibodies of five different healthy subjects do not bind human neural cells

Fig. 7

IgG antibodies of the intractable Epilepsy patients IE-3 and IE-15 bind and kill living human neural cells, unlike IgG antibodies of healthy subjects that do not. Representative confocal microscopy images show that purified IgG of the Epilepsy patients IE-3 (A) and IE-15 (B) bind (red staining) and kill (Sytox green staining) living (i.e. not pre-fixed, to allow binding and induction of functional effects that may led to cell death) human neural cells, within 1 h. while IgG of healthy subjects HI-5 (C) and HI-7 (D) and IH-9 (E) do not. In the big figures A–E, all the DAPI⁺ cells are shown in blue in the upper left internal small figures; the β3 tubulin⁺ neurons are shown in yellow in the upper middle figures, the Sytox green⁺ dead/necrotic neural cells are shown in green in the upper right figures, and the bound IgG is shown in red in the lower left figures. The lower middle figures show an overlay of all the stained components. The quantitative analysis of the bound and dead neural cells documented in these experiments is shown in Fig. 8

Fig. 8

IgG antibodies of the intractable Epilepsy patients IE-3 and IE-15 bind and kill living human neural cells, unlike IgG antibodies of healthy subjects that do not. A Quantitative analysis of the percentage of human neural cells bound by purified IgG antibodies of the Epilepsy patients IE-3 or IE-15, or of three control healthy subjects. The quantitative analysis was performed based on confocal microscopy photos of the respective immunostainings. Two separate counts were made in each field of each confocal microscopy photographed image: first, all the DAPI⁺ cells were counted (seen in blue in Fig. 7), and secondly, only the cells to which the IgG antibodies bound were counted (seen in red or pink in Fig. 7). According to these numeric values, the percentage of human neural cells bound by the IgG antibodies of the Epilepsy patients or healthy subjects was calculated. B Quantitative analysis of the percentage of human neural cells killed by purified IgG antibodies of the Epilepsy patients IE-3 or IE-15, or of three control healthy subjects. The quantitative analysis was performed based on confocal microscopy photos of the respective immunostainings. Two separate counts were made in each field of each confocal microscopy photographed image: first, all the DAPI⁺ cells were counted (seen in blue in Fig. 7), and secondly, only the dead cells that stained by Sytox Green (seen in green in Fig. 7) were counted. According to these numeric values, the percentage of human neural cells killed by the IgG antibodies of the Epilepsy patients or healthy subjects was calculated

Fig. 9

IgG antibodies of the Epilepsy patients IE-3 and IE-15 induce recurrent seizures in naïve rats, recorded in a novel video EEG animal model. A Schematic illustration of the experimental design. Seizure frequency per week in Sprague–Dawley rats infused with HBSS control (B), or purified IgG antibodies of healthy control subjects HI-5 and HI-19 (C, D, respectively), or purified IgG antibodies of the Epilepsy patients IE-3 or IE-15 (E, F, respectively). G Percentage of rats that developed seizures in the Epilepsy patients and their healthy controls. H Cumulative number of seizures in rats infused with HBSS, IgG antibodies of healthy control subjects HI-5 or HI-19, or the Epilepsy patients IE-3 or IE-15. Data analysis in G was performed using Chi-square and analysis in H was performed by one-way ANOVA followed by Dunnett's comparison test. Differences with *p < 0.05, **p < 0.01, ***p < 0.001 were considered statistically significant

Fig. 10

IgG antibodies of the Epilepsy patients IE-3 and IE-15 induce seizures in naïve rats, and so does Kainic acid used as positive control. A Representative sample traces of spontaneous seizures bursting in rat after status epilepticus induced by kainic acid. B, C Sample traces of the seizures induced in naïve rats by the IgG antibodies of the Epilepsy patients IE-3 and IE-15 respectively. D–F Duration of the seizures induced in naïve rats by the IgG antibodies of the Epilepsy patients IE-3 and IE-15 respectively. Statistical analysis in F was performed by unpaired T-test. Differences with *p < 0.05, **p < 0.01, ***p < 0.001 were considered statistically significant

Fig. 11

IgG antibodies of the Epilepsy patients IE-3 and IE-15 bind and kill neuronal cells in vivo in the hippocampus CA3 region of naïve rats. A Representative images of immunostaining of brain slices of the hippocampus of naïve rats post infusion of purified IgG antibodies either of healthy control subjects (HI) or of the two intractable Epilepsy patients (IE): IE-3 and IE-15. The pictures in the left column, entitled “NeuN” show staining of neuronal cells. The pictures in the second column to the left, entitled “GFAP” show staining of astrocytic cells. The pictures in the third column to the left, entitled “IgG” show staining of the human IgG (patient’s IgG) antibodies. The pictures in the right column, entitled “Merged” show an overlap of the staining. B Percentage of neurons in the CA3 brain region of the rat hippocampus bound by the human IgG of either the Epilepsy patients IE-3 or IE-15, or the healthy control subjects. C Percentage of astrocytes in the CA3 brain region of the rat hippocampus bound by the human IgG of either the Epilepsy patients IE-3 or IE-15, or the healthy control subjects. D Percentage neuronal density in the CA3 region of the rat hippocampus bound by the human IgG of either the Epilepsy patients IE-3 or IE-15, or the healthy control subjects. Scale 50 μm. Data are displayed as the mean ± SEM, *p < 0.05, **p < 0.01, ***p < 0.001 analyzed by one-way ANOVA followed by Dunnett's comparison test

Fig. 12

IgG antibodies of the Epilepsy patients IE-3 and IE-15 bind and kill neuronal cells In vivo in the cortex of naïve rats. A Representative images of immunostaining of brain slices of the cortex of naïve rats post-infusion of purified IgG antibodies either of healthy control subjects (HI) or of two intractable Epilepsy patients IE-3 and IE-15. The pictures in the left column, entitled “NeuN” show staining of neuronal cells. The pictures in the second column to the left, entitled “GFAP” show the staining of astrocytic cells. The pictures in the third column to the left, entitled “IgG” show staining of human IgG (patient’s IgG) antibodies. The pictures in the right column, entitled “Merged” show an overlap of all the staining. B Percentage number of neurons in the cortex of naïve rats bound by the human IgG of either the Epilepsy patients IE-3 or IE-15, or the healthy control subjects. C Percentage number of astrocytes in the cortex of naïve rats bound by the human IgG of either the Epilepsy patients IE-3 or IE-15, or the healthy control subjects. D Percentage neuronal density in the cortex of naïve rats bound by the human IgG of either the Epilepsy patients IE-3 or IE-15, or the healthy control subjects. Scale 50 μm. Data are displayed as the mean ± SEM, *p < 0.05, **p < 0.01, ***p < 0.001 analyzed by one-way ANOVA followed by Dunnett's comparison test

Fig. 13

Graphical summary of the current study on ‘autoimmune Epilepsy’, and of all its findings