An IgG anti-GM1 mAb was provided by Dr Nobuhiro Yuki (Departments of Microbiology and Medicine, National University of Singapore, 2 Science Drive, Singapore 17597), and all samples were stored at −80°C until required. Using an enzyme-linked immunosorbent assay, reactivity was detected with the IgG anti-GM1 mAb, which was in accordance with previous observations by Hotta et al (5).

Drugs were dissolved in the co-culture medium bathing the preparation. ω-conotoxin GVIA and ω-agatoxin IVA (Alomone Labs Ltd., Jerusalem, Israel) were dissolved in a stock solution containing cytochrome c (1 mg/ml; Sigma-Aldrich; St. Louis, MO, USA) to prevent the non-specific binding of the peptide to the chamber walls and tubing.

Pregnant Wistar rats (n=35) were purchased from Japan Laboratory Animals, Inc. (Tokyo, Japan) individually housed under controlled conditions, with a 12-h light-dark cycle and free access to food and water. Experiments were conducted in accordance with the Guidelines for Animal Care of Showa Pharmaceutical University (Tokyo, Japan), in addition to the guidelines for animal use published by the National Institutes of Health (https://grants.nih.gov/grants/olaw/Guide-for-the-Care-and-use-of-laboratory-animals.pdf). This study was conducted with the approval of the Showa Pharmaceutical University Research Ethics Committee (approval no. H18).

A spinal cord-muscle co-culture system was established according to a previously described method by Taguchi et al (10). Briefly, muscle was extracted from prenatal day 17 fetal rats and separated into ~3-mm sections in Tyrode's solution (Sigma-Aldrich) containing 100 µg/ml streptomycin and 100 µg/ml penicillin. The sections were subsequently incubated at 37°C for 20 min in Ca²⁺- and Mg²⁺-free Tyrode's solution containing 1 mg/ml collagenase. For the innervation experiments, explants of the entire transverse slices of fetal spinal cord, including the dorsal root ganglia, were placed onto a collagen-coated 35-mm Petri dish. Individual muscle cells were derived from trituration and placed on the slices of spinal cord for culture. The muscle cells and spinal cords were co-cultured in 67% DMEM (Gibco Life Technologies, Carlsbad, CA, USA), 23% medium 199 (Gibco Life Technologies) and 10% fetal calf serum (Roche Diagnostics, Basel, Switzerland), which was supplemented with 25 ng/ml fibroblast growth factor (Sigma-Aldrich) and 20 µg/ml insulin (Gibco Life Technologies). The spinal cord-muscle co-culture systems were stored in a CO₂ incubator with 5% CO₂ and 95% O₂ at 37°C.

After 1 week of co-culture, the innervated muscle specimens were placed in an experimental chamber on the stage of an inverted microscope (IX-70; Olympus Corporation, Tokyo, Japan). The 1-ml experimental chamber was continuously perfused with medium (67% DMEM and 23% medium 199) at a rate of 1–2 ml/min, with continuous bubbling of 5% CO₂/95% O₂. Glass microelectrodes (GD-1; Narishige Group, Tokyo, Japan) filled with 3 M KCl, with a tip resistance of 20–40 MΩ, were used to record the SMAPs. Each electrode that was connected to a microelectrode amplifier (MEZ-8301; Nihon Koden Corporation, Tokyo, Japan) recorded the electrical activity, which was displayed on an oscilloscope (VC-11; Nihon Koden Corporation). After the stability of muscle action potentials was observed for ~4 min, 10 µl anti-GM1 mAb antibody was applied using a micropipette. Data were transferred to and stored in a computer, using pCLAMP6 software (Molecular Devices, Sunnyvale, CA, USA). All experiments were conducted at 33±1°C.

Hemidiaphragms were used for triple fluorescence labeling to determine the localization of GM1. Tissue samples were incubated in 10% normal goat serum (NGS; Funakoshi Co., Ltd., Tokyo, Japan) in Block Ace blocking agent (Dainippon Sumitomo Pharma Co., Ltd., Tokyo, Japan) for 30 min at room temperature to block non-specific binding, as previously described (6). After blocking, the tissues were incubated for 5 h at 4°C with IgG anti-GM1 antibodies (1:100; provided by Dr Nobuhiro Yuki) in 10% NGS and 10% Block Ace in phosphate-buffered saline (PBS). In order to detect the IgG anti-GM1 mAb, the tissues were incubated for 1 h at 4°C with a fluorescein isothiocyanate (FITC)-conjugated anti-mouse IgG (1:100; Sigma-Aldrich). Subsequently, the hemidiaphragms were incubated with primary anti-Ca_v2.1 (α_1A; 1:1,000; #ACC-001) and anti-Ca_v2.2 (α_1B; 1:1,000; #ACC-002) antibodies (Alomone Labs Ltd., Jerusalem, Israel) for 5 h at 4°C. To detect the primary antibodies, the hemidiaphragms were incubated for 1 h at 4°C with a monoclonal Cy5-conjugated anti-rabbit IgG (1:100; AP182SA6; Chemicon International, Inc., Temecula, CA, USA). Hemidiaphragms were subsequently incubated with rhodamine-α-bungarotoxin (α-BuTx; 1:300; Molecular Probes; Invitrogen Life Technologies, Grand Island, NY, USA) for 5 h at 4°C. Each reaction was terminated by numerous washes with PBS, and the tissue was mounted using Aqua Poly/Mount (PolySciences, Inc., Warrington, PA, USA). All the immunostained sections were observed under an Olympus laser-scanning confocal microscope (Fluoview BW50; Olympus Corporation, Tokyo, Japan) at wavelengths of 488 nm (FITC), 543 nm (rhodamine) or 568 nm (Cy5).

Data are expressed as the mean ± standard error of the mean (SEM). A paired t-test was used to compare the effects of the IgG anti-GM1 mAb, where P<0.05 was considered to indicate a statistically significant difference. Statistical analyses were conducted using Microsoft Excel and statistical add-on software (Microsoft Corporation, Redmond, WA, USA).

After 5–7 days of co-culture, asynchronous contraction of numerous individual muscle fibers was observed at the newly developed NMJs. SMAPs within the innervated muscle cells were recorded with a frequency of 12.7±4.8 sec per 5 sec and an amplitude of 65.4±8.4 mV (n=8, mean ± SEM). IgG anti-GM1 mAb rapidly reduced the number of SMAPs at the NMJs within 1 min (Fig. 1A). Subsequently, SMAPs at the NMJs were blocked completely (Fig. 1B). The inhibitory effect of the IgG anti-GM1 mAb was reversed by washing with the co-culture medium. Thus, IgG anti-GM1 mAb significantly reduced the rate of SMAPs at the NMJs (n=4, 70.8±4.8%; Fig. 1C).

Fig. 2 shows the effect of pretreatment with the N-type calcium channel blocker, ω-conotoxin GVIA (30 nM), and IgG anti-GM1 mAb on the SMAPs. The inhibitory effect of IgG anti-GM1 mAb was completely blocked by pretreatment with ω-conotoxin GVIA in the spinal cord-muscle co-culture systems (Fig. 2A). Fig. 2B shows the time course of the reversal of the IgG anti-GM1 mAb-induced inhibition of SMAPs by pretreatment with ω-conotoxin GVIA. Thus, pretreatment with ω-conotoxin GVIA attenuated the inhibitory effect on SMAPs induced by IgG anti-GM1 mAb (Fig. 2C).

Fig. 3 shows the effect of the P/Q-type calcium channel blocker, ω-agatoxin IVA (10 nM), and IgG anti-GM1 mAb on SMAPs. ω-agatoxin IVA slightly decreased the frequency of SMAPs in the spinal cord-muscle co-culture systems (Fig. 3A). In addition, the inhibitory effect of the IgG anti-GM1 mAb was partially blocked by pretreatment with ω-agatoxin IVA in spinal cord-muscle co-cultures. Fig. 3B shows the time course of the partial reversal of the inhibitory effect on SMAPs produced by IgG anti-GM1 mAb following pretreatment with ω-agatoxin IVA. Therefore, pretreatment with ω-agatoxin IVA slightly attenuated the inhibitory effect on SMAPs induced by IgG anti-GM1 mAb (Fig. 3C).

The localization of IgG anti-GM1 mAb and various calcium channels in the nerve terminal was determined using immunohistochemistry. As shown in Fig. 4A, the rat hemidiaphragm sections were found to stain positively for the anti-Ca_v2.1 (α_1A; P/Q-type calcium channel; 1:1,000; blue) antibody, and the staining overlapped partially with the IgG anti-GM1 mAb staining (green) and α-BuTx staining (red). In addition, anti-Ca_v2.2 (α_1B; N-type calcium channel; 1:1,000; blue) antibody staining overlapped with IgG anti-GM1 mAb staining and α-BuTx staining, similarly to the anti-α_1A antibody staining (Fig. 4B). Thus, IgG anti-GM1 mAb binding was localized at the motor nerve terminal, and the staining corresponded with N-type calcium channels at the motor nerve terminal and partially overlapped with P/Q-type calcium channel staining.