IgM anti-GM2 antibodies in patients with multifocal motor neuropathy target Schwann cells and are associated with early onset

Abstract

Background: Multifocal motor neuropathy (MMN) is a rare, chronic immune-mediated polyneuropathy characterized by asymmetric distal limb weakness. An important feature of MMN is the presence of IgM antibodies against gangliosides, in particular GM1 and less often GM2. Antibodies against GM1 bind to motor neurons (MNs) and cause damage through complement activation. The involvement of Schwann cells (SCs), expressing GM1 and GM2, in the pathogenesis of MMN is unknown.

Methods: Combining the data of our 2007 and 2015 combined cross-sectional and follow-up studies in Dutch patients with MMN, we evaluated the presence of IgM antibodies against GM1 and GM2 in serum from 124 patients with MMN and investigated their binding to SCs and complement-activating properties. We also assessed the relation of IgM binding and complement deposition with clinical characteristics.

Results: Thirteen out of 124 patients (10%) had a positive ELISA titer for IgM anti-GM2. Age at onset of symptoms was significantly lower in MMN patients with anti-GM2 IgM. IgM binding to SCs correlated with IgM anti-GM2 titers. We found no correlation between IgM anti-GM2 titers and MN binding or with IgM anti-GM1 titers. IgM binding to SCs decreased upon pre-incubation of serum with soluble GM2, but not with soluble GM1. IgM anti-GM2 binding to SCs correlated with complement activation, as reflected by increased C3 fixation on SCs and C5a formation in the supernatant.

Conclusion: Circulating IgM anti-GM2 antibodies define a subgroup of patients with MMN that has an earlier onset of disease. These antibodies probably target SCs specifically and activate complement, similarly as IgM anti-GM1 on MNs. Our data indicate that complement activation by IgM antibodies bound to SCs and MNs underlies MMN pathology.

Keywords: Anti-ganglioside antibodies; Complement; IgM anti-GM2; Multifocal motor neuropathy; Schwann cells.

Fig. 1

IgM titers and MMN patient serum-derived IgM binding to SCs and iPSC-MNs. (A) Correlation between IgM anti-GM1 and IgM anti-GM2 titers as determined via ELISA. (B) Sera from 98 MMN patient sera were screened for IgM binding to SCs using flow cytometry. IgM binding is depicted as fold-change (FC) compared to the mean IgM binding of 6 healthy control sera tested in the same assay on the y-axis. IgM binding to iPSC-MNs incubated with MMN sera using microscopy. IgM binding is expressed as FC similarly as in (A). (C) Stratification of MMN patients by IgM anti-ganglioside antibody status determined with ELISA reveals higher IgM binding to SCs in GM2 + versus GM1+/GM2- and -/- patients (Kruskal-Wallis, post-hoc Dunn’s multiple comparisons test). (D) Stratification of MMN patients by IgM anti-ganglioside antibody status indicates higher IgM binding to iPSC-MNs in GM1 + versus GM1- patients. Kruskal-Wallis, post-hoc Dunn’s multiple comparisons test. **** p < 0.0001; *** p < 0.001; FC: fold change; ns: non-significant

Fig. 2

IgM binding to SCs stratified by IgM anti-GM2 and anti-GM1 antibody titer. (A) IgM binding to SCs stratified for IgM anti-GM1 titer. No correlation was found between FC_IgM and IgM anti-GM1 titers (Spearman’s rho r_s=0.1369, p = 0.1790). (B) IgM binding to SCs stratified for IgM anti-GM2 antibody titer group shows a moderate correlation between IgM binding (FC_IgM) and anti-GM2 titer (Spearman’s rho r_s=0.4983, p < 0.0001). FC: fold change

Fig. 3

MMN patient-derived IgM binding to SCs is GM2 specific. Sera from MMN patients with IgM anti-GM2 were selected, incubated with soluble GM1 or GM2 and tested for IgM binding to SCs and in the anti-GM2 ELISA. (A) FC_IgM binding on SCs before (grey bars) and after (orange bars) pre-incubation with GM1. (B) FC_IgM binding on SCs before (grey bars) and after (blue bars) pre-incubation with GM2. (C) Quantification of flow cytometric results depicted in (A and B) as % inhibition. IgM binding is significantly decreased upon pre-incubation with GM2, whereas pre-incubation with GM1 only does not significantly decrease IgM binding. (D) Anti-GM2 specific ELISA showing IgM binding (in d-OD_450nm) without (grey bar) or with pre-incubation with GM1 (orange bars) or GM2 (blue bars) using IgM anti-GM2 positive MMN patient sera. (E) Quantification of ELISA results depicted in (D) as % inhibition. Pre-incubation with GM2 significantly lowers IgM binding. Mean + SD, Kruskal-Wallis, post-hoc Dunn’s multiple comparisons test. **** p < 0.0001; ns: non-significant

Fig. 4

IgM anti-GM2 binding on SCs results in complement activation. IgM binding (FC_IgM, A) and C3 fixation (FC_C3, B) to SCs opsonized with IgM anti-GM2 positive (red bars) or negative (black bars) MMN patient serum and incubated with fresh serum as complement source. Opsonization with anti-GM2 MMN patient serum results in increased IgM binding and increased complement activation. Data are mean + SD of different assays (C) Pooled results of FC_C3 data depicted in (B), Mann-Whitney test. (D) Highly significant strong correlation between IgM anti-GM2 binding to SCs and subsequent complement activation. Red dots IgM anti-GM2 + sera, black dots IgM anti-GM2- sera. Mean + SD, Spearman’s rho r_s=0.9532, p < 0.0001. **** p < 0.0001; FC: fold change

Fig. 5

Complement activation by IgM anti-GM2 bound to SCs in culture. (A) Representative microscopic images of complement activation on SCs opsonized with IgM anti-GM2 positive MMN serum or not, and incubated with fresh serum, heat-inactivated serum (serum HI), or serum pre-incubated with a monoclonal antibody that blocks C5 activation or an isotypic control antibody, as complement source. 20x magnification, scale bar: 50 μm. (B) Quantification of microscopy data using 3 different MMN sera for opsonization.C3 fixation to the cells (quantified as MGV) is expressed as FC (FC_C3) setting the fixation observed with cells not opsonized with MMN serum (striped bar) as 1. Opsonization of the cells with MMN serum results in a significant increase in C3 fixation, which is abrogated when heat inactivated serum is used as complement source, and which is also reduced wen the complement source is pre-incubated with anti-C5 antibody, and not with an isotypic control antibody. (C) Quantification of C5a, depicted as FC_C5a, measured in the supernatant of SC cultures. C5a generation in culture medium is increased upon opsonization of the cells with IgM anti-GM2 positive MMN patient serum and incubation with fresh serum. This C5a generation is inhibited by an anti-C5 antibody added to the complement source but not by a control antibody. Mean + SD, Kruskal-Wallis, post-hoc Dunn’s multiple comparisons test, ** p < 0.01; **** p < 0.0001; FC: fold change; HI: heat inactivated; MMN: multifocal motor neuropathy; ns: non-significant