Differential binding patterns of anti-sulfatide antibodies to glial membranes

Abstract

Sulfatide is a major glycosphingolipid in myelin and a target for autoantibodies in autoimmune neuropathies. However neuropathy disease models have not been widely established, in part because currently available monoclonal antibodies to sulfatide may not represent the diversity of anti-sulfatide antibody binding patterns found in neuropathy patients. We sought to address this issue by generating and characterising a panel of new anti-sulfatide monoclonal antibodies. These antibodies have sulfatide reactivity distinct from existing antibodies in assays and in binding to peripheral nerve tissues and can be used to provide insights into the pathophysiological roles of anti-sulfatide antibodies in demyelinating neuropathies.

Keywords: Complement; Monoclonal antibody; Myelin; Neuropathy; Sulfatide.

Graphical abstract

Fig. 1

Mice produce anti-sulfatide antibodies in response to immunisations with sulfatide containing liposomes. CST^−/−(n = 6), CST^+/+ (n = 3) and DBA mice (n = 3) were immunised with liposomes comprised of sulfatide, SM, Chol and DCP. Combinatorial glycoarray blots from the terminal bleed sera and plots of the single sulfatide reactivity (measured by fluorescent intensity on array) in sera over time indicate the IgM and IgG antibody responses. (A) IgM antibodies against the liposome components could be detected in sera from all genotypes, particularly against sulfatide and sulfatide-associated complexes. A representative blot from CST^−/− mouse sera is shown. (B) The levels of anti-sulfatide IgM reactivity rose with each consecutive immunisation but there was no significant/measurable difference between the CST^−/− and CST^+/+ sera at any time-point. In contrast, the level of anti-sulfatide antibodies in the DBA mice were significantly higher than the other genotypes on day 21 (two way ANOVA, P < 0.05). (C) IgG antibodies were detected against SM, Chol and associated complexes but negligible reactivity to sulfatide was detected. A representative blot from CST^−/− mouse sera is shown. (D) Anti-sulfatide IgG antibodies only appeared in the DBA sera on Day 28. There was no detectable anti-sulfatide IgG antibodies in the sera of either the CST^−/− or CST^+/+ mice. Array areas spotted with vehicle only are marked with an X. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 2

Monoclonal anti-sulfatide antibodies bind to sulfatide and associated complexes on lipid microarray. Newly generated monoclonal anti-sulfatide antibodies were screened against sulfatide antigen printed on a lipid microarray. The IgM antibodies were screened at 10 μg/ml and the IgG antibodies were screened at 1 μg/ml. Their binding patterns fell into two broad categories. (A) Group A comprises GAME-M5 and GAME-G3 antibodies that bind exclusively to sulfatide and sulfatide complexes that do not contain complex gangliosides at the concentrations used. (B) Group B comprises GAME-M2, GAME-M6 and GAME-M7 antibodies that bind strongly to sulfatide and are not affected by associated complexes. (C) O4 was screened for comparison purposes, and showed binding to sulfatide and associated complexes in a similar pattern as the Group B antibodies. (D) CST^−/− (n = 6) and CST^+/+ (n = 6) mice were passively immunised with 250 μg of GAME-G3 antibody. Anti-sulfatide antibody levels were significantly higher following immunisation on day 1 compared to day −1 (baseline) for both genotypes. Clearance from the circulation does not differ between the two genotypes, implying that clearance is not due to receptor-dependent uptake as has been seen for antibodies against gangliosides. Results represent two independent experiments each with 3 mice per group. * Significance at day 1 versus day −1, two-way ANOVA with Sidak's multiple comparison test, p < 0.05. Array areas spotted with vehicle only are marked with an X.

Fig. 3

Monoclonal anti-sulfatide antibodies bind to myelinating CNS cultures, peripheral nerve sections and ex vivo nerve-muscle preparations. (A) Representative images indicate the relative binding patterns of the Group A, Group B and O4 anti-sulfatide antibodies to myelinating CNS cultures. All antibodies (red) were all able to bind to the oligodendrocytes in CNS cultures to varying degrees; the group B antibodies and O4 displayed the strongest signal. (B) Representative images indicate the relative binding patterns of the anti-sulfatide antibodies (green) to peripheral nerves identified by axonal neurofilament labelling (magenta) in transverse sections. Group A and Group B antibodies bound well to the transverse nerve bundles from CST^+/+ mice, O4 exhibited more diffuse binding. None of the antibodies bound to tissue from CST^−/− mice confirming that they selectively bind to sulfatide. (C) The Group A (GAME-G3) and group B (GAME-M2) anti-sulfatide antibodies (green) bound along the myelin sheath encircling neurofilament labelled axons (magenta) and intensely at the terminal heminode paranodal loops in the mouse distal motor nerve when applied topically to an ex vivo nerve-muscle preparation. In contrast, O4 was unable to bind or bound comparatively weakly to the distal nerve myelin. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 4

– Anti-sulfatide antibodies are capable of binding in vivo and fixing complement in ex vivo nerve-muscle preparations. (A) Binding along the myelin and at the paranodal loops (green, *) was observed in the peripheral nerves (magenta) of CST^+/+ mice after in vivo delivery of 1 mg GAME-G3 i.p. the previous day. No binding was observed in CST^−/− mice demonstrating the specificity of the antibody to sulfatide. (B) 1 mg GAME-G3 was injected i.p. to compare and study access at distal and proximal sites. 24 h later binding could be observed in the distal nerves (whole-mount diaphragm and lumbricals), while binding was absent from proximal nerve trunks (sciatic nerve, spinal root sections). C) GAME-G3 antibody binding (orange) in ex vivo TS muscle was capable of activating the complement cascade in distal motor nerves (magenta), culminating in the deposition of the membrane attack complex (MAC; green). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)