Study design and mice immunization
The study aimed to investigate the effects of active immunization with NMDAR peptides in normal adult mice. C57BL/6 mice (10 weeks old) were immunized with different GluN1 extracellular peptides emulsified in an equal volume of Complete Freund’s Adjuvant supplemented with Mycobacterium tuberculosis H37Ra (4 mg/mL) at a final concentration of 1 mg/mL. Mice were subcutaneously injected with 200 μg of GluN1 peptides or control peptides at the tail base, and received two booster injections with CFA at 4 and 8 weeks after the first immunization. All mice were intraperitoneally injected with 400 ng pertussis toxin at the last immunization. Behavioral tests and histological staining were performed 12 weeks after the first immunization.
Patient sample collection
We collected CSF from patients with high titers of anti-GluN1 ABs (> 1:300) during routine clinical practice. All patients fulfilled the clinical diagnostic criteria for anti-NMDAR encephalitis, revised in 2016 (13). The study protocol was approved by the ethics committee of Nanfang Hospital, Southern Medical University, and written informed consent was obtained from each participant or their guardian.
AB purification
The CSF and serum ABs from patients or immunized mice were purified using protein G Sepharose columns and were then used to treat neurons or brain slices. For the purification process, 2 mL of diluted sample was incubated in a chromatography spin column (Thermo Scientific) of protein G Sepharose beads for 30 min. After three washes with phosphate-buffered saline (PBS), the samples were eluted with elution buffer, dialyzed against PBS, concentrated in stock solutions of 4 mg/mL, and stored at –80 °C until used.
Preparation and staining of GluN1-expressing HEK cells
Human embryonic kidney 293 (HEK293) cells were transiently transfected with NMDAR subunit genes (NR1/NR2A; DsRed2-labeled) as previously described (14). Twenty-four hours later, cells were fixed on coverslips with acetone and incubated overnight at 4 °C with the purified ABs from patient or immunized mice CSF (starting at 1:1) in 0.1% bovine serum albumin (BSA) in PBS. After washing with PBS, the cells were labeled with fluorescein isothiocyanate (FITC)-conjugated anti-human IgG (ab7149, Abcam) or FITC-conjugated anti-mice IgG (ab6785, Abcam) and observed under a fluorescence microscope (BX51, Olympus).
Site-directed mutagenesis
Point mutation were made using the Stratagene QuikChange Mutagenesis kit (210518, Agilent) according to the manufacturer’s instruction. Primers designed for N368Q point mutation: Forward: 5’-gggatgacatgggtaccttggtagatgcccacttgca-3’; Reverse: 5’-tgcaagtgggcatctaccaaggtacccat gtcatccc-3’.
Primary neuronalcultures
Hippocampal neurons were prepared and maintained from embryonic day 18 rat brains as previously described (15). The hippocampi were dissociated with papain at 37 °C for 30 min, and then separated with a fire-polished Pasteur pipette. After centrifugation at 300 g, the cells were resuspended in Neurobasal medium. Cells were counted and plated onto poly-D-lysine-treated 24-well plates. After 6 h, the supernatant was removed and replaced with 500 μL of fresh culture medium. Cells were then cultured for 14 days for subsequent experiments.
Immunocytochemistry
Immunocytochemistry was used to detect binding to brain slices and autoantibodies. On day 14 after immunization, CSF was obtained from mice, and was then purified and incubated with brain slices. The slices were imaged using a confocal microscope (LSM 880, Carl Zeiss, Germany). To stain surface NMDAR, neurons were incubated with ABs derived from the CSF of patients or immunized mice. After incubating for 18 h, the AB-bound surface receptors were labeled with a fluorescent-conjugated secondary AB (1:200, anti-Mouse Alexa Fluor 488, A-11029; 1:200, anti-Human Alexa Fluor 488, A-11013; Invitrogen). Neurons were then fixed, permeabilized, incubated with anti-PSD95 primary AB (1:200, Synaptic Systems, 124-003) to label postsynaptic densities, and visualized after staining with Alexa Fluor 647 (1:200, Invitrogen, A-21244).
Electrophysiological recording
Brain slices preparing and electrophysiological recording were performed as previously described (16). Mouse hippocampal slices (300 μm) were prepared using a vibratome (VT1000S, Leica, Germany). The slices were kept at 30 °C for at least 60 min before experiments in artificial CSF (ACSF; NaCl 124 mM, KCl 2.5 mM, MgSO4 2.0 mM, NaH2PO4 1.25 mM, NaHCO3 26 mM, CaCl2 2 mM, and glucose 10 mM; pH 7.3), bubbled with a mixture of 95% O2 and 5% CO2. Field excitatory postsynaptic potentials were evoked in the CA1 stratum radiatum by stimulating Schaffer collaterals with a two-concentric bipolar stimulating electrode, and were recorded with ACSF-filled glass pipettes. LTP was induced by applying theta burst stimulation. Purified ABs were diluted in ACSF (100 μg/mL) and applied by switching the perfusion from the control ACSF to the AB-containing ACSF. For each recording, the baseline synaptic transmission was monitored for 10 min before AB perfusion, and was washed out with ACSF continuously after theta burst stimulation until the end of the experiment. The acquired data were analyzed with pClamp 10 software (Axon Instruments, USA).
Calcium imaging
Hippocampal neurons were incubated with 20 µg/mL of patient or mice ABs at 37 °C for 18 h. To detect calcium flux, the cells were loaded with Fura-2 (1 µM, Invitrogen, F1221) and incubated at 37 °C for 15–30 min, followed by 15–30 min incubation at room temperature. After washing in Tyrode’s solution, the cells were transferred into wells containing NBQX (10 µM, Tocris, 0373). NMDA (10 µM, Sigma, M3262) was used to stimulate NMDAR-mediated calcium influx. Imaging was performed using an inverted fluorescence microscope (Nikon Eclipse TE2000-U, Japan) with a charge-coupled device camera. Series of images were acquired at 800 ms intervals for 40 s at an excitation wavelength of 470 nm. The fluorescence intensity at each timepoint was measured using ImageJ (NIH).
Behavioral assessments
After 12 weeks of immunization, LE30-immunized mice were tested using a series of behavioral experiments. Mice immunized with a control peptide were used as the control. Behavioral testing was conducted at uniform times, from 09:00 to 12:00, by researchers blinded to the group allocations. Behavioral parameters were recorded using a video tracking system (Smart 3.0, Panlab, Spain). Details of behavioral tests are described in the Supplementary Methods.
Western blot analyses
Minute Plasma Membrane Protein Isolation Kit (SM-005, Invent Biotechnologies) was used to extract membrane and total cell proteins from the primary neurons after incubated with purified human or mice CSF ABs according to the manufacturer's protocol. Proteins in SDS loading buffer were separated on 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred to polyvinylidene fluoride membranes. After blocking with 5% BSA, the membranes were incubated with anti-GluN1 ABs (1:500, Synaptic Systems, 114-011) and a secondary HRP-conjugated AB. The immunocomplexes were detected using enhanced chemiluminescence.
Statistical analyses
Data were presented as mean ± SEM. Independent sample t-tests or Mann–Whitney U tests were used as appropriate for each experiment. Kruskal–Wallis tests were used for nonparametric data. All analyses were performed using SPSS V24.0 (IBM, USA) and the histograms were made using GraphPad Prism 6.0 (GraphPad Software, USA) or ImageJ. P<0.05 was considered statistically significant.