An update on pathobiologic roles of anti-glycan antibodies in Guillain-Barré syndrome

Abstract

Introduction and context

Guillain-Barré syndrome (GBS) is a group of acute, monophasic, and pathophysiologically heterogenous neuropathic disorders. Paralytic forms of the disorder are classified (Table 1) as demyelinating or axonal based on electrophysiology, whereas relatively rare non-paralytic variants include Fisher syndrome (FS), characterized by ophthalomplegia, ataxia, and areflexia [1,2]. With the near eradication of polio, paralytic variants of GBS are the commonest cause of acute flaccid paralysis worldwide.

GBS is considered to be an autoimmune disease triggered frequently by common infections of upper respiratory and gastrointestinal tracts [1-3] in a susceptible host [4]. What constitutes a susceptible host remains enigmatic. Autoantibodies directed against cell surface glycans carried by gangliosides (sialic acid-containing glycosphingolipids enriched in peripheral nerve fibers) have become the main focus of research in GBS. Over the last 15-20 years, several lines of evidence have linked these autoantibodies to the pathogenesis of GBS, particularly to axonal and Fisher variants of the disease. The clinical studies focusing on serological immune markers and GBS phenotype and recovery have identified associations of specific anti-glycan antibodies with different variants of GBS [3] and poor recovery [5,6]. The association studies have implied that specific anti-glycan antibodies not only can induce neuropathy (i.e., injury to intact nerve fibers) but also can adversely affect recovery by inducing more severe neuropathic disease or interfering with the nerve repair process required for recovery (or both). Identification of specific anti-glycan antibodies in GBS patients led to the development of cell culture [7], tissue culture [8-10], and animal [11-14] models that showed the pathogenetic effects of anti-glycan antibodies on intact nerves or nerve cells. Cumulatively, these studies indicate that specific anti-glycan antibodies target relevant antigens in neural cells, especially at motor nerve terminals or nodes of Ranvier (or both) to disrupt the nerve fiber function [15-17]. Human and experimental studies indicate that complement activation is involved in structural injury to the nerve fibers [18-20]. Several issues regarding the anti-glycan antibody-mediated nerve injury remain unresolved [21]. For example, unconditional passive transfer with sera containing anti-glycan antibodies obtained from patients or active immunization animal models of axonal GBS [11,13] has not been reported to induce injury to the intact nerve fibers in experimental animals. This brief review describes some salient recent developments that enhance our understanding of the complex pathobiologic mechanisms involved in anti-glycan antibody-mediated deleterious effects on intact and injured nerve fibers.

Major recent advances

Several recent observations are beginning to unravel the complexity involved in anti-glycan antibody-associated selective nerve fiber injury seen in different variants of GBS. For example, antibodies against GM1 and GD1a or related minor gangliosides are associated with acute motor axonal neuropathy (AMAN) and anti-GQ1b/GT1a with FS [1-3]. The association of specific anti-glycan antibodies with specific GBS variants had raised an important critique; that is, how do specific anti-glycan antibodies induce selective injury to different nerve fibers (e.g., motor versus sensory) or selective topographical involvement of nerves/pathways despite minor or no differences in the biochemical content of gangliosides in different nerves or nerve fibers? The group led by Kusunoki [22], in a series of studies, have presented a novel concept that some GBS sera/anti-glycan antibodies bind to ganglioside complexes (pairs of gangliosides) but not to individual components of ganglioside in solid-phase assays. The authors propose that antibodies against ganglioside complexes recognize new conformational epitope(s) formed by mixing. Whether or not gangliosides and other glycans that constitute cell surface glycocalyx form complexes or unique conformational epitopes in biological/cellular membranes remains to be determined. If this concept is validated, then this has far-reaching implications and could provide an explanation of how different neuronal/nerve fiber populations could be selectively targeted by specific anti-glycan antibodies despite similar biochemical content of individual major gangliosides. That an individual ganglioside can assume a different conformation/orientation in motor and sensory fibers was supported by data published by our group recently [23]. This study focused on anti-GD1a antibodies in the context of AMAN and selective motor fiber injury. We found that some anti-GD1a antibodies selectively bound to motor but not sensory axons and that different anti-GD1a antibodies had different binding patterns to various chemical derivatives of GD1a (fine specificity). On the basis of data derived from biochemical, immunocytochemistry, computer modeling, and enzymatic studies, we proposed a model in which critical GD1a epitopes recognized by selective motor anti-GD1a antibodies are differentially expressed in motor and sensory nerves. Yet another mechanism was suggested by a study showing that motor nerve terminal injury with anti-GM1 antibodies was enhanced with prior enzymatic treatment with sialidase [24]. Based on these findings, it was proposed that a proportion of GM1 ganglioside is cryptic due to masking by other moieties in the cell membrane, including more complex gangliosides. Overall, this set of studies suggests that selective topographical injury of nerves/nerve fibers in GBS is a complex issue and multiple complementary mechanisms are likely to contribute.

Translational progress has also been made in the area of anti-glycan antibody-mediated complement-dependent nerve fiber injury. The Willison group in Glasgow published a series of studies showing that specific anti-glycan antibodies bind to motor nerve terminals or perisynaptic Schwann cells (or both) and activate complement and this not only disrupts neuromuscular transmission but also can induce degeneration of motor nerve terminals (reviewed in [25]). Susuki et al. [17] extended this line of research and examined the role of complement at the nodes of Ranvier in a rabbit model of AMAN. They showed that during the acute phase of limb weakness, deposition of IgG and complement correlates with disruption of voltage-gated sodium channel clusters and paranodal axoglial junctions, the nodal cytoskeleton (Figure 1), and Schwann cell microvilli, all of which stabilize these channels at the nodes. These pathogenic studies led to translational efforts focusing on complement inhibition. Halstead et al., in two separate studies [14,26], show that two inhibitors of C5 (i.e., C5-inhibiting recombinant protein rEV576 and eculizumab [Soliris^®; Alexion Pharmaceuticals, Cheshire, CT, USA], a treatment approved by the US Food and Drug Administration for paroxysmal nocturnal hemoglobinuria) prevented physiological and structural damage to motor nerve fibers in preclinical models. Based on these observations, an argument has been put forth for a clinical trial with eculizumab in GBS.

Open in a separate window

Figure 1.

Disruption of voltage-gated sodium (Na_v) channels and related cytoskeletal molecules at nodes of Ranvier in a rabbit model of acute motor axonal neuropathy (AMAN)

All panels show longitudinal sections of ventral roots, and nerve fibers run horizontally. Sections were immunostained with antibodies to C3 (green), Na_v channels (blue), and βIV spectrin (red). (A) A normal node of Ranvier in a control rabbit, showing absence of complement deposition and normal distribution of the Na_v channel and βIV spectrin. (B) A node of Ranvier in an AMAN rabbit at the acute progressive phase, showing that deposition of complement was associated with disruption of nodal architecture (paranodal axoglial junctions/lengthening) and Na_v channel and βIV spectrin staining. Adapted with permission from Figure 4, Susuki et al. [17]. © Copyright 2007, Society for Neuroscience.

In a multi-investigator collaborative study, Buchwald et al. [27] show that anti-glycan (GM1 and GD1a) antibodies can induce various degrees of synaptic blockade (evoked quantal release) at neuromuscular junctions independently of complement. The same antibodies significantly reduced depolarization-induced calcium influx in neurons. This study was conducted with the following clinical observations in view: (a) some patients with axonal GBS recover too rapidly to explain their recovery on the basis of axon degeneration and regeneration, and (b) some subjects (patients or animals with axonal GBS) with severe clinical weakness were found on autopsy to lack significant pathology that would explain their weakness [28,29]. These experimental studies support the notion that anti-glycan antibody-induced deleterious effects on presynaptic transmitter release is one pathophysiological mechanism of antibody-mediated muscle weakness in AMAN.

Our group examined, for the first time, the effects of anti-glycan antibodies on peripheral nerve repair/axon regeneration in an animal model [30]. We found that passive transfer of anti-GD1a reactive antibodies severely inhibited axon regeneration after peripheral nervous system injury. In mutant mice with altered ganglioside or complement expression, inhibition by antibodies was mediated directly via GD1a and was independent of complement-induced cytolytic injury. The impaired regenerative responses and ultrastructure of injured peripheral axons mimicked dystrophic and stalled growth cones (Figure 2B,C) typically seen after central nervous system injury. We have also seen dystrophic/stalled growth cones in the sural nerve biopsy of a GBS patient who had poor recovery (Figure 2A). These observations support the hypothesis that inhibition of axon regeneration is one mechanism of poor recovery in GBS patients with anti-glycan antibodies. Furthermore, these studies indicate that circulating factors such as autoimmune autoantibodies can inhibit axon regeneration and affect recovery in patients with GBS.

Open in a separate window

Figure 2.

Dystrophic regenerating sprouts in a patient with Guillain-Barré syndrome (GBS) and in an animal model induced with anti-glycan antibodies

(A) Sural nerve biopsy showing dystrophic sprouts (arrows) in a GBS patient with poor recovery. Reduced density of myelinated fibers and myelin debris are also seen. (B) Dystrophic sprouts (arrows) indicating abortive sprouting in a mouse model of sciatic nerve crush and treatment with anti-glycan antibodies. These sprouts are not myelinated and are surrounded by Schwann cell nuclei (*) (C) Electron microscopy image showing dystrophic growth cone with a core of neurofilaments and a surrounding pellet with membranous organelles. Scale bars = 10 μm (A), 20 μm (B), and 2 μm (C). Panels (B) and (C) reproduced with permission from Figure 3, Lehmann et al. [30]. © Copyright 2007, Society for Neuroscience.

Future directions

The notion that anti-glycan antibodies are the primary pathogenetic effectors in GBS continues to be contested. The controversy stems mainly from the fact that experimental modeling in different studies has yielded varied results [31,32]. It is not surprising that experimental modeling is not reproducible given that traditionally only enzyme-linked immunosorbent assay has been used to define the specificity of anti-glycan antibodies for further use in experimental studies. Recent work, some of which is outlined above, indicates that anti-glycan antibody-mediated pathogenecity is quite complex, partly due to subtleties involved in the carbohydrate antigen expression on the neural cell surfaces. Furthermore, anti-glycan antibodies with specificity for a ganglioside are not created equally and these antibodies could be different in terms of fine specificity, tissue-binding patterns, affinity/avidity, Fc-based effector functions, and other as-yet-undefined properties. The development of techniques that allow the examination of different characteristics of glycan antigens and anti-glycan antibodies and increasing use of well-defined monoclonal anti-glycan antibodies is very likely to decrease discrepant experimental findings. The characterization of different properties of anti-glycan antibodies is likely to extend into the clinical arena for delineation of well-defined subgroups of GBS patients (or other immune neuropathies associated with anti-glycan antibodies) for clinical outcome research and enrolment in clinical trials as new therapies emerge.

The elucidation of mechanisms underlying anti-glycan antibody-mediated injury to intact nerve fibers has been a fruitful line of research. A direct result of this is the emergence of complement inhibitors as candidates for clinical trials in GBS and other immune neuropathies [14]. It is anticipated that detailed intracellular signaling mechanisms/pathways mediating anti-glycan antibody-induced injury to intact nerve fibers will be identified. The identification of these pathways may allow the development of rational neuroprotective therapies for use in conjunction with current immunomodulatory treatments to minimize nerve damage during the injury (acute) phase of GBS and other immune neuropathies.

Models of anti-glycan antibody-mediated inhibition of nerve repair have already been established and it is very likely that these models will allow the dissection of molecular mechanisms that prevent successful nerve repair and recovery. Novel therapeutic strategies targeting the signaling pathways that prevent successful regeneration may allow enhanced nerve repair not only in patients with GBS but in those with other neuropathic conditions.

Acknowledgments

Abbreviations

Notes

The electronic version of this article is the complete one and can be found at: http://f1000.com/reports/b/2/21

Notes

References