The study was approved by the ethics committee of the Medical Faculty of the University of Würzburg and was performed in accordance with the ethical standards of the Declaration of Helsinki of 1964. All patients and controls gave written informed consent to take part in the study.

A total number of 33 patients with MMN attending the Departments of Neurology of the University Hospitals Würzburg and Gießen were prospectively recruited in 2013 and 2014. Diagnosis was based on the EFNS criteria [1]: 25 patients were classified as definite MMN, two as probable MMN and six as possible MMN. Sera were obtained prior to the initiation of treatment in five patients, during a period without treatment in two patients and under immunoglobulin treatment in all other cases. In the latter cases, sera were obtained immediately before application of IVIG, with an interval of at least two weeks from the last application, except for two patients with subcutaneous application of immunoglobulins. Sixty sera of healthy individuals (mean age 54.4 years, 31 males, 29 females) and 10 sera of patients with other autoimmune neurological diseases (myasthenia gravis, multiple sclerosis; mean age 56.9 years, 3 males, 7 females) served as controls. Sera of CIDP patients with anti-contactin-1 and anti-neurofascin-155 auto-antibodies described in previous studies were used as positive controls [6,11]. Routine work-up of the MMN patients included clinical examination by neurologists of the Departments of Neurology of the University Hospitals Würzburg and Gießen and nerve conduction studies according to the EFNS guidelines [1]. Muscle strength was quantified using the Medical Research Council (MRC) scale [12] and weakness was determined as a score of 4 or less. All sera were tested for anti-ganglioside antibodies (GM1 IgM/IgG, GM2 IgM/IgG, GM3 IgM/IgG, GD1a IgM/IgG, GD1b IgM/IgG, GT1b IgM/IgG, GQ1b IgM/IgG; Euroimmun, Lübeck, Germany).

Sciatic nerves were removed from adult female C57BL/6 mice and were pre-fixed in 4% paraformaldehyde, and teased fibers were prepared by gently teasing the single nerve fibers with two fine forceps on a glass slide. Teased fiber preparations were air-dried overnight, fixed with acetone and incubated with blocking solution containing 4% normal goat serum, 4% fetal calf serum, 0.3% Triton-X-100 in PBS. The preparations were incubated with patients’ sera diluted 1:100 overnight at 4°C following incubation with appropriate anti-human Cy3-conjugated secondary antibodies (Dianova, Hamburg, Germany, 1:100) for two hours at room temperature. Double-immunofluorescence staining using patients’ sera and polyclonal rabbit anti-caspr (Abcam, Cambridge, UK, 1:1,000) or polyclonal rabbit anti-pan-neurofascin (Abcam, 1:400) was performed to detect paranodes. Sera of patients with anti-contactin-1-IgG-positive CIDP were used as a positive control. They were mounted with VECTASHIELD with DAPI (Vector Laboratories, Burlingame, CA, USA). Slides were assessed using a fluorescence microscope (Zeiss Ax10, Zeiss, Oberkochen, Germany). The examiner was blinded to the diagnosis and clinical data of the patients.

Maxisorb 96-well-plates (Thermo Fisher Scientific, Waltham, MA, USA) were coated with human contactin-1 (Sino Biological, Beijing, China) diluted at a concentration of 2 μg/ml in 0.1 M carbonate buffer (pH 9.6) and incubated at 4°C overnight. Neurofascin ELISA was performed as previously described using full-length human neurofascin-155 and -186 [6]. All wells were washed with PBS/0.05% Tween-20 and blocked with PBS/3%BSA/0.05% Tween-20 for one hour at 37°C, the wells were then incubated with patients’ samples (1:100) or anti-contactin-1 (mouse monoclonal, Abcam, 1:200) or anti-neurofascin-155 (rabbit anti-neurofascin 155 raised against a peptide derived from the Fibronectin III C domain unique to NF155; 1:200) [13] or anti-neurofascin-186 (rabbit anti-NF186 raised to a peptide derived from the Mucin domain unique to NF186, 1:1000)[13] for 1 h at 37°C. After washing with PBS/0.05% Tween-20, the plate was incubated with appropriate horseradish-conjugated secondary antibodies (anti-human-IgG or IgM for patients material, 1:10,000/1:7,000, DakoCytomation, Glostrup, Denmark; anti-mouse/-rabbit IgG for the control antibody, 1:500). 100 μl of TMB solution (eBioscience, San Diego, CA, USA) were added to each well and the reaction was stopped after fifteen minutes by addition of 50 μl of 1M sulfuric acid. Optical density was measured at 450 nm with Multiscan EX Elisa Reader (Thermo Fisher Scientific, Waltham, MA, USA). Sera of CIDP patients with anti-contactin 1 IgG and anti-neurofascin-155 IgG/IgM were used as controls. Normal controls were run in each assay. The normal value was set at five standard deviations above the mean of all healthy controls, based on the distribution of values of normal controls and of anti-contactin-1-positive/anti-neurofascin positive CIDP patients. IgM and IgG serum levels were determined by human IgM and IgG ELISA kits (AssayPro, St. Charles, MO, USA) according to the instructions of the manufacturer.

HEK293 cells were plated onto poly-D-lysine-coated cover slips in a 24-well-plate at a density of 90,000 cells/well. HEK293 cells were transiently transfected with plasmids of rat contactin-1 [14], human neurofascin-155 and -186 [6] using the calcium-phosphate precipitation method (0.25 μg of DNA/well).[15] Rat contactin-1 is 95% identical to human contactin-1 at the amino acid level, thus rat-contactin-1 is suitable in this assay. With this protocol about 20% of all cells expressed the respective protein on their surface. 48 hours after transfection, cells were fixed with 4% paraformaldehyde and were incubated with patients’ sera (1:500) and anti-contactin-1 (Abcam, 1:200) or anti-neurofascin 155 and 186 (1:100) as a control antibody and appropriate secondary antibodies (Cy3 and AlexaFluor-conjugated anti-human IgG or anti-mouse IgG). Sera of patients with anti-contactin-1-positive CIDP were used as a positive control. Immunostaining was assessed using a fluorescence microscope (Zeiss Ax10, see above).

Thirty-three patients with a diagnosis of MMN were included (mean age 57.1 years, median 57 years, range 33–76 years, 23 men, 10 women), see also Table 1. Disease duration ranged between 0.5 and 27 years, with a mean of 8.8 years and a median of 9 years. Patients were recruited at the University Hospitals of Würzburg (n = 23) and Gießen (n = 10). The majority of patients showed the typical clinical picture of multifocal motor neuropathy with muscle weakness beginning at the hand muscles, spreading to more proximal muscles and to the lower limbs, and muscle atrophy during the course of disease. Onset of muscle weakness at the lower limbs was reported by four patients and remained restricted to the lower limbs in three patients. None of the patients reported any relevant sensory deficits. Conduction blocks were detected in all patients except for five individuals that were classified as possible MMN. All patients had a good response to immunoglobulin treatment.

Anti-ganglioside auto-antibodies were tested in all patients and GM1 IgM auto-antibodies were detected in 14/33 patients. GM2 IgM auto-antibodies were found in 5/33 patients. IgM auto-antibodies against GM2, GD1b, GT1b and GQ1b were found in one patient.

IgG levels were between 17.91 mg/ml and 40.61 mg/ml with a median of 29.76 mg/ml in MMN patients, between 8.84 mg/ml and 24.56 mg/ml in healthy controls (median 15.86 mg/ml) and between 5.47 mg/ml and 23.52 mg/ml (median 11.69 mg/ml) in disease controls. IgM levels ranged from 1.14 mg/ml to 6.3 mg/ml with a median of 3.94 mg/ml in MMN patients, from 1.19 mg/ml and 2.63 mg/ml in healthy controls (median 1.68 mg/ml) and from 1.53 mg/ml to 2.95 mg/ml (median 2.19 mg/ml) in disease controls. Thus IgG and IgM levels were elevated in patients with MMN compared to controls, which is most likely a consequence of the patients’ regular IVIG treatment.

ELISA was performed using sera from all patients and controls. Antibodies against contactin-1, neurofascin-155 and -186 served as positive controls in each assay and showed a clear immunoreaction (Fig 1). All three ELISA protocols had been tested in previous studies, demonstrating efficacy and specificity to detecting the respective auto-antibodies [6]. We did not detect any IgM or IgG auto-antibodies against neurofasin-155, -186 or contactin-1 in any patient with multifocal motor neuropathy. (Fig 2A–2F). Slightly elevated IgG reaction to neurofascin-155 was found by ELISA in one healthy control (Fig 2A).

Cell binding assays were performed with all sera. Antibodies against neurofascin-155, -186 and contactin-1 served as positive controls and showed good binding to transfected cells (Fig 3A, 3D and 3G). However, we did not detect any binding of patients’ or controls’ sera to HEK293 cells transfected with contactin-1, neurofascin-155 or -186 (Fig 3B, 3E and 3H).

We did not detect any specific binding of patients’ sera to the paranodal or nodal region of murine teased fibers resembling the binding pattern of anti-caspr as a paranodal marker (Fig 4A–4C) or anti-neurofascin-155 or -186 (Fig 4D–4F).