Native versus deglycosylated IgM in anti-MAG neuropathy: Correlation with clinical status - Study of 10 cases
Graphical abstract
Introduction
The myelin-associated glycoprotein (MAG) is a transmembrane glycoprotein located on periaxonal Schwann cells and oligodendroglial membranes of myelin sheaths. MAG contains a carbohydrate epitope (HNK-1) that is the antigenic target in demyelinating polyneuropathy associated with IgM MG. MAG shares this epitope with other myelin glycoproteins such as P0 glycoprotein, peripheral myelin protein-22 (PMP-22), Contactin 1, Phosphacan and two sulfated glucuronic glycolipids: sulfated glucuronyl paragloboside (SGPG) and sulfated glucuronyl lactosaminyl paragloboside (SLGPG) (Nobile-Orazio, 2004). The pathogenic role of anti-MAG antibodies has been investigated by passive and active immunization of animals (Hays et al., 1987; Willison et al., 1988; Kohriyama et al., 1988) but the immune responses were different according to the molecule showing the same epitope and the animal species, used for immunization. This implies, first, that the conformation of the HNK-1 epitope is influenced by its environment (Hamada et al., 2015) and, second, which some animals such as cats immunized with bovine SGPG, have developed sensory ataxic neuropathy similar to the human form, with high titer anti-SGPG antibodies, demonstrating that bovine SGPG was better immunogenic than human MAG (Ilyas et al., 2008). In a study leaded by Kahn et al. (Kahn et al., 1989), immunization of cats with human MAG did not induce peripheral neuropathy despite the presence of antibody against cat peripheral nerve myelin. According to the authors, cats did not develop antibodies against the HNK-1 oligosaccharide moiety but towards the peptide part of MAG. This study shows that an antibody activity no directed against the HNK1 epitope seems clinically neutral.
MAG belongs to Sialic Acid Binding Immunoglobulin like Lectin type 4a (SIGLEC 4a) family and shows a high affinity for the α(2, 3)-linked sialic acid moieties (Quarles, 2007). Besides, human monoclonal IgM has 5 heavy chain glycosylation sites at Asn 171, 332, 395, 402 and 563 with sialilated oligosaccharides and high-mannose type oligosaccharides. Therefore, given the affinity of MAG for some glyco-sialilated moiety It could be reasonable to hypothesize that IgM may bind to MAG via these glycan moieties located on Fc fragment, as an alternative and additional route of antigen binding other than through the Fab V regions. This MAG-Glycans IgM interaction does not exclude a “real” relationship in Fab V regions, but could lead to an overestimation of the titers of anti-MAG antibodies. Otherwise, as emphasized by Leibiger et al. (1998) the glycosylation of glycoproteins from an organism in a disease state may differ from that under normal conditions. For this reason the glycosylation of human IgM is likely to be different from IgM derived from patients with anti-MAG neuropathy. Furthermore, the authors show the differences and similarities of the glycosylation between normal human IgM and human monoclonal IgM, demonstrating the exclusive presence of α (2, 3)-linked sialic acid within human monoclonal IgM, structure for which, MAG has a strong affinity. Indeed, anti-MAG titers and IgM level at diagnosis are not always associated with disease severity and there is not a good correlation between pre- and post- treatment anti-MAG titers in patients who respond clinically to immunomodulators. Therefore, we hypothesized that deglycosylation of IgM anti-MAG may remove the clinically neutral interactions between MAG and IgM and thereby improve the correlation between serum markers and clinical phenotype in anti-MAG neuropathy because there was evidence that anti-MAG antibodies were pathogenic including induction of a sensory ataxic neuropathy in animal models. This clinical feature was the hallmark in patients with IgM MG reacting with HNK-1 or with gangliosides presenting disialosyl moieties (Willison et al., 2001) .
The aim of this study was a first approach in order to assess the structural features of the immune response, the association, dissociation and affinity constants of IgM vs MAG and the kinetic of this response. Further studies with more important cohorts will be necessary in order to determine whether the anti-MAG IgM level as well as its structure, are only a diagnostic marker or allow a relative monitoring of the disease.
Section snippets
Patients
The ethics committee of our hospital (CPPGHPS: Comité de Protection des Personnes du Groupe Hospitalier Pitié-Salpêtrière) exempted this study from review because we did not conduct human experimentation and the samples received were supplied in anonymous way. Furthermore, these sera came from patients recruited in our institution and in La Timone Hospital, Aix-Marseille University, for two previous studies (Iancu Ferfoglia et al., 2016; Leger et al., 2013) where all patients provided written
α-mannosidase digestions
Human IgM contains five N-glycosylation sites, located in the heavy chain. All N-linked oligosaccharides share the same pentasaccharide core structure beginning by a mannose residue and a wide variety of oligosaccharides can be attached to this core. In order to remove these sugars, jack bean α-mannosidase (Sigma, EC 3.2.1.24) was used at the concentration recommended by the manufacturer. For each extract, 50 μg of IgM were digested in 50 mM sodium citrate buffer, pH 4.6, for 16 h at 37 °C. The
Structural features of the immune response anti-MAG before and after deglycosylation
All IgM extracts were tested in an iso quantitative way with regard to the original serum. Except one (patient 3), all sera showed a decrease of the activity against MAG after deglycosylation, using ELISA and IIF (Fig. 1A,B,C). Staining of IgM with Concanavalin A after enzymatic treatment displayed a strong decrease of intensity for the 8 deglycosylated extracts with regard to the purified native IgM. This decrease of intensity allowed validating the deglycosylation enzymatic reaction. Using
Discussion
Anti-MAG antibodies have been shown to be pathogenic in animal models. Their presence in 12% of patients with IgM MG without neuropathy supported the hypothesis of neutral intermolecular interactions between IgM and MAG, albeit we could not to rule out a subclinical form for these patients as described by Meucci et al. (1999). Conversely, patient 10, from the study group, was followed in our institution for 16 years and has never presented any neurological involvement. Therefore, this profile
Funding sources
This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors.
Declaration of Competing Interest
All authors have no commercial relationship or conflict of interest with the study presented.
Acknowledgments
The authors thank the Biophysical platform at Pasteur Institut, Paris, and Dr. Patrick England for his technical assistance and advices in using the Biacore™ T200 system.
References (35)
Human serum IgM glycosylation: identification of glycoforms that can bind to mannan-binding lectin
J. Biol. Chem.
(2005)Terminal latency index and modified F ratio in distinction of chronic demyelinating neuropathies
Clin. Neurophysiol.
(2001)Predictive factors of efficacy of rituximab in patients with anti-MAG neuropathy
J. Neurol. Sci.
(2017)Binding specificity of anti-HNK-1 IgM M-protein in anti-MAG neuropathy: possible clinical relevance
Neurosci. Res.
(2015)Differential glycosylation of polyclonal IgG, IgG-Fc and IgG-Fab isolated from the sera of patients with ANCA-associated systemic vasculitis
Biochim. Biophys. Acta
(2006)Induction of experimental ataxic sensory neuronopathy in cats by immunization with purified SGPG
J. Neuroimmunol.
(2008)Galactosylation of IgG associated oligosaccharides
Lancet
(1988)Analysis of the feline immune response to human myelin-associated glycoprotein
J. Neurol. Sci.
(1989)Glycosylation analysis of a polyreactive human monoclonal IgG antibody derived from a human-mouse heterohybridoma
Mol. Immunol.
(1995)Glycans in the immune system and the altered glycan theory of autoimmunity: a critical review
J. Autoimmun.
(2015)
A comparative analysis of disease-associated changes in the galactosylation of serum IgG
J. Autoimmun.
Classical and lectin complement pathway activity in polyneuropathy associated with IgM monoclonal gammopathy
J. Neuroimmunol.
A sensitive and reproducible method for the detection of oligoclonal IgM bands
J. Immunol. Methods
Variability in the binding of anti-MAG and anti-SGPG antibodies to target antigens in demyelinating neuropathy and IgM paraproteinemia
J. Neuroimmunol.
Sialylation of N-linked glycans influences the immunomodulatory effects of IgM on T cells
J. Immunol.
Relevance of anti-HNK1 antibodies in the management of anti-MAG neuropathies
J. Neurol.
Neuropathy and IgM M-proteins: prognostic value of antibodies to MAG, SGPG, and sulfatide
Neurology
Cited by (0)
- 1
These authors contributed equally to this work.