Analysis of peripheral B cells and autoantibodies against the anti-nicotinic acetylcholine receptor derived from patients with myasthenia gravis using single-cell manipulation tools

Abstract

The majority of patients with myasthenia gravis (MG), an organ-specific autoimmune disease, harbor autoantibodies that attack the nicotinic acetylcholine receptor (nAChR-Abs) at the neuromuscular junction of skeletal muscles, resulting in muscle weakness. Single cell manipulation technologies coupled with genetic engineering are very powerful tools to examine T cell and B cell repertoires and the dynamics of adaptive immunity. These tools have been utilized to develop mAbs in parallel with hybridomas, phage display technologies and B-cell immortalization. By applying a single cell technology and novel high-throughput cell-based binding assays, we identified peripheral B cells that produce pathogenic nAChR-Abs in patients with MG. Although anti-nAChR antibodies produced by individual peripheral B cells generally exhibited low binding affinity for the α-subunit of the nAChR and great sequence diversity, a small fraction of these antibodies bound with high affinity to native-structured nAChRs on cell surfaces. B12L, one such Ab isolated here, competed with a rat Ab (mAb35) for binding to the human nAChR and thus considered to recognize the main immunogenic region (MIR). By evaluating the Ab in in vitro cell-based assays and an in vivo rat passive transfer model, B12L was found to act as a pathogenic Ab in rodents and presumably in humans.These findings suggest that B cells in peripheral blood may impact MG pathogenicity. Our methodology can be applied not only to validate pathogenic Abs as molecular target of MG treatment, but also to discover and analyze Ab production systems in other human diseases.

Fig 1

Scatter plot of flow cytometry analysis and gating strategy for single cell sorting of a) memory B cells with or without fluorescent-labeled antigen and b) plasmablasts.

Fig 2. Binding of positive and negative clones to nAChR-expressing cells determined by flow cytometry based binding assay.

Data shown are from experiments performed in triplicate and are shown as mean values with SD. y-axis: mean fluorescence intensity (MFI) of anti-human IgG-PE. 3B1, 1G3, B12L: positive clone; 3B11: negative clone; mAb35xich1: human chimeric form of rat mAb35 as positive control; isotype ctrl: human isotype-matched control Ab as a negative control. DB40: nAChR expressing cells; TE671: parent cell lines of DB40.

Fig 3. In vitro analysis of human B12L and rat mAb35.

a) Competitive binding of rat mAb35 was monitored by flow cytometer-based binding assay. The indicated concentrations of Abs (mAb35xich1, B12L and Ctrl IgG) were added to DB40 cells prior to spiking with 10 μg/mL of rat mAb35. Data was obtained by monitoring the mean fluorescence intensity (MFI) of anti-rat IgG-PE signal. The percent inhibition was calculated with the following formula: (MFI of rat mAb35 signal without blocking Abs–MFI of rat mAb35 signal with blocking Abs) / (MFI of rat mAb35 signal without blocking Abs–background signal) x 100 (%). All data were obtained from experiments performed in triplicate and are presented as mean values with SD. Ctrl IgG: isotype-matched human IgG for negative ctrl IgG. b) Agonistic and c) antagonistic activities of B12L/mAb35 were monitored by Ca⁺⁺ influx in DB40 cells. Acetylcholine (ACh) was used as a positive control agonist. Atropine was used as an inhibitor of muscarinic type AChR to reduce the background signal for antagonist activity. α-Btx was used as a positive control antagonist. d) Downmodulation of nAChRs induced by B12L, mAb35xich1, and their Fabs was monitored by fluorescence signal of the α-Btx-Alexa Fluor 488 probe. Binding at 100% represents the maximum fluorescence signal of α-Btx-Alexa Fluor 488 bound to nAChRs on the surface of DB40 cells. BG: background signal. Statistical analysis among groups in a) and d) were conducted using Student’s t-test, with p < 0.01 being considered statistically significant.

Fig 4

In vitro competitive assay against a) rat EAMG serum and b) MG plasma. a) The indicated concentrations of Abs (mAb35xich1, B12L and Ctrl IgG) were added to DB40 cells prior to spiking with rat EAMG serum. Data was taken by monitoring the mean fluorescence intensity (MFI) of anti-rat IgG-PE signal. b) The indicated concentrations of Fabs (mAb35xich1-Fab and B12L-Fab) were added to DB40 cells, prior to spiking with plasma from MG donor #1 (diluted 10 times). The percent inhibition was calculated as shown in Fig 3A. All data were obtained in triplicate after monitoring the MFI of anti-human IgG-PE signals through individual experiments, and are shown as mean values with SD. Ctrl IgG: isotype-matched human IgG for negative control IgG. Statistical analysis among groups in (a) and (b) were conducted using Dunnett’s multiple comparison test, with p < 0.01 being considered statistically significant.

Fig 5. Passive transfer experiment for B12L.

a) Clinical scores were recorded at 3 h, 6 h, 9 h, 24 h, 30 h, and 48 h after intravenous administration of B12L or saline (n = 3, each group, mean ± SD). Since the animals administered B12L at a dose of 3 mg/kg were moribund and euthanized 30 h after dosing, their clinical scores were recorded as ‘4’ at 48 h. b) Body weights were measured at 0 h, 24 h and 48 h after administration (n = 3, each group, mean±SD). We could not measure the body weights in the group administered 3 mg/kg of B12L at 48 h, because they were moribund and euthanized at 30 h after administration. c) Representative images of complement deposition in control rats (low magnification [A, B, and C], high magnification [G, H, and I]) and B12L-treated rats (low magnification [D, E, and F], high magnification [J, K, and L]) are shown. C3 (D and J, green) and α-Btx (E and K, red) signals were colocalized in neuromuscular junctions (NMJs) (F and L, merged) at 8 h after administration of B12L at a dose of 1.5 mg/kg. Scale bar = 50 μm. d) The amount of nAChRs in NMJs was calculated by imaging analysis. The graph shows the signal ratio of α-Btx (red) colocalized with SV2A (green) compared to SV2A signal alone in NMJs at 48 h after administration of saline or B12L at a dose of 1.5 mg/kg (n = 4, each group, mean±SD).