Anti-myelin-associated glycoprotein (MAG) neuropathy is a disabling autoimmune peripheral neuropathy caused by monoclonal IgM autoantibodies that recognize the glycan epitope HNK-1 (human natural killer-1). The glycoepitope consists of the sulphated trisaccharide SO4-3GlcAβ1-3Galβ1-4GlcNAc and is highly expressed on adhesion molecules, such as MAG, present in myelinated nerve fibers. The binding of the anti-MAG IgM autoantibodies to the HNK-1 glycoepitope causes demyelination and axonal damage, so that the patients suffer from severe symptoms including sensorimotor deficits, ataxia, pain and tremors.
Currently, only off-label treatments with limited efficacy are available. We hypothesized that significant clinical improvements could be achieved by selective removal of the pathogenic anti-MAG antibodies with a biodegradable, poly-L-lysine based polymer (PPSGG) presenting multiple copies of HNK 1 glycoepitope mimetic on the peptide backbone.
We assessed the difference in the inhibitory activity of an HNK-1 glycoepitope mimetic and its multivalent presentation on the polymer. The multivalent presentation improved the inhibitory activity towards patients sera, i.e. inhibition of antibody-binding to MAG, by a factor of up to 230,000 per epitope compared to the monomer, resulting in a low nanomolar inhibitory potency (mean IC₅₀ = 3.6 ± 0.4 nM). The pharmacodynamic properties of the glycopolymer were analysed in vitro with patients’ sera using both a MAG-ELISA and an immunofluorescence assay with sciatic nerves from non-human primates, and in vivo using an immunological mouse model. Furthermore, ADME parameters and safety pharmacology of PPSGG were assessed in different species.
PPSGG efficiently inhibited MAG-binding of patients’ antibodies at nanomolar epitope concentrations and antibody-binding to sciatic nerves was strongly inhibited by 62.5 μg/mL and completely inhibited by 125 μg/mL PPSGG. In vivo, single intravenous injections of 10 mg/kg led to significant reductions in anti-MAG IgM titers for up to five days, and weekly treatments resulted in a decreased anti-MAG IgM antibody rebound. A tissue distribution study showed fast uptake and metabolism of both PPSGG and the PPSGG:anti-MAG IgM complex in the mononuclear phagocyte system, which is consistent with the short in vivo plasma half-life of PPSGG in mice of ca. 17 min (t₁/₂ = 16.9 ± 5.5 min). No signs of aggregate formation were detected either in vitro or in vivo. Neither could immunogenicity be induced in the different treated animal species nor could human B cells or dendritic cells be activated ex vivo by the glycopolymer. More importantly, the dose-range finding studies in two animal species showed no safety concerns for PPSGG.
In conclusion, PPSGG efficiently neutralized and removed pathogenic anti-MAG IgM in vitro and in vivo, and had a favorable safety profile. Taken together, the obtained data shows that PPSGG is a very promising drug candidate for clinical trials in anti-MAG neuropathy.