Article Text

Download PDFPDF

Ganglioside complexes containing GQ1b as targets in Miller Fisher and Guillain–Barré syndromes
  1. M Kanzaki1,
  2. K Kaida1,
  3. M Ueda2,
  4. D Morita2,
  5. M Hirakawa2,
  6. K Motoyoshi1,
  7. K Kamakura1,
  8. S Kusunoki2
  1. 1
    Division of Neurology, Internal Medicine 3, National Defense Medical College, Saitama, Japan
  2. 2
    Department of Neurology, Kinki University School of Medicine, Osaka, Japan
  1. Dr Susumu Kusunoki, Department of Neurology, Kinki University School of Medicine, 377-2 Ohno-Higashi, Osaka-Sayama, Osaka 589–8511, Japan; kusunoki-tky{at}umin.ac.jp

Abstract

Background: Serum antibodies to GQ1b are associated with Miller Fisher syndrome (MFS) and Guillain–Barré syndrome (GBS) with ophthalmoplegia. Antibodies to ganglioside complexes (GSCs) have not yet been examined in a large population of patients with MFS or GBS. This study aimed to determine the clinical significance of antibodies to GSCs in MFS and GBS.

Methods: The study investigated serum anti-GSC antibodies and the clinical features in 64 MFS patients, 53 GBS patients with ophthalmoplegia (GBS-OP(+)) and 53 GBS patients without ophthalmoplegia (GBS-OP(−)).

Results: Thirty patients with MFS (47%), 25 with GBS-OP(+) (47%) and none with GBS-OP(−) had antibodies to GSCs containing GQ1b or GT1a. Patients with MFS and GBS-OP(+) were subdivided according to the antibody reactivities; patients with antibodies specific to GQ1b and/or GT1a (without anti-GSCs antibodies) were placed in Group 1, those with antibodies against GSCs with a total of two sialic acids in the terminal residues, such as GQ1b/GM1, were placed in Group 2, and those with antibodies against GSCs with a total of three sialic acids in the terminal residue, such as GQ1b/GD1a, were placed in Group 3. In MFS, sensory disturbances were infrequent in Group 2 compared with the other groups (p<0.0001). Antibodies specific to GQ1b were observed more often in MFS than in GBS-OP(+) (p = 0.0002).

Conclusions: IgG antibodies to GSCs containing GQ1b or GT1a were closely associated with the development of ophthalmoplegia in GBS, as well as MFS. Both GQ1b and clustered epitopes of GSCs containing GQ1b or GT1a may be prime target antigens for MFS and GBS-OP(+).

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Miller Fisher syndrome (MFS) is characterised by ophthalmoplegia, ataxia and areflexia,1 and is thought to be a variant of Guillain–Barré syndrome (GBS). Anti-GQ1b immunoglobulin G (IgG) antibody, an excellent diagnostic marker in MFS, is found in the acute-phase sera of more than 90% of MFS patients.2 Moreover, anti-GQ1b IgG antibody is also associated with ophthalmoplegia in GBS and Bickerstaff’s brain-stem encephalitis.3 4

Glycosphingolipids are known to form microdomains called lipid rafts, together with cholesterol and glycosylphosphatidylinositol (GPI)-anchored proteins.5 Within the microdomains, gangliosides may play an important role in membrane-mediated functions.68 We previously reported antibodies to ganglioside complexes (GSCs) as new target antigens in GBS,9 and that 58% of MFS patients exhibited serum antibodies to GSCs containing GQ1b.10 These studies suggest that the clustered glycoepitopes of GSCs in peripheral nerves may be targets for serum antibodies in acute immune-mediated polyradiculoneuropathy, such as GBS and MFS. We also demonstrated that the specificity of anti-GSC antibodies might be associated with the clinical features of GBS and MFS.

To clarify the clinical significance of antibodies to GSCs containing GQ1b in MFS and GBS, we retrospectively analysed the clinical features of anti-GSC-positive patients with GBS or MFS in a larger population.

METHODS

Patients

We collected data from 64 consecutive patients with MFS and 53 GBS patients presenting with ophthalmoplegia (GBS-OP(+)) who were recruited between January 2003 and June 2005. The serum samples were submitted to us from various teaching and general hospitals for screening of anti-ganglioside antibodies. They were acute-phase sera obtained before specific treatment and within 2 weeks after the onset of the disease. The clinical records were collected with the sera. If the record was insufficient, we sent a questionnaire about clinical findings to the attendant physicians to analyse the clinical features. GBS patients were diagnosed according to the diagnostic criteria of Asbury and Cornblath.11 The diagnosis of MFS was based on acute self-limited ophthalmoplegia, ataxia and areflexia without significant limb weakness, central nervous system (CNS) involvement or other neurological diseases. Patients who developed limb weakness after appearance of ophthalmoplegia, ataxia or areflexia were categorised into GBS-OP(+) (score of 4 or less on the Medical Research Council scale).

We analysed MFS patients who exhibited at least two symptoms in the triad of ophthalmoplegia, ataxia and areflexia. Concerning the ophthalmoplegia, we excluded patients who only exhibited internal ophthalmoplegia. Ataxia was defined as instability in gait and standing, including sensory and cerebellar ataxia. Anti-GSC antibodies were also determined in 53 consecutive GBS patients without ophthalmoplegia (GBS-OP(−)), 20 normal subjects (normal control), and from 88 patients with neurological disorders other than GBS (disease control): multiple sclerosis, 8; myasthenia gravis, 11; amyotrophic lateral sclerosis, 10; spinocerebellar degeneration, 3; Parkinson’s disease, 4; cerebrovascular disease, 8; frontotemporal dementia, 3; brain tumour, 3; myelopathy, 8; chronic inflammatory demyelinating polyradiculoneuropathy, 5; multifocal motor neuropathy, 3; acute cerebellitis, 4; mitochondrial myopathy, encephalopathy, lactic acidosis, and stroke-like episodes, 1; Creutzfeldt–Jakob Disease, 1; and other neuropathy, 16.

The enzyme-linked immunosorbent assay for anti-ganglioside and anti-GSC antibodies

Serum samples were investigated for antibodies to GM1, GM2, GM3, GD1a, GD1b, GD3, GT1a, GT1b, GQ1b and GalNAc-GD1a using the enzyme-linked immunosorbent assay (ELISA). GalNAc-GD1a was isolated in our laboratory from bovine brain,12 and the other gangliosides were purchased from Sigma-Aldrich Co. (St Louis, MO). The patients’ sera (diluted 1:40 with 1% bovine serum albumin (BSA) in phosphate-buffered saline (PBS)) were applied to wells coated with 200 ng of single antigens and uncoated wells as the control. Peroxidase-conjugated anti-human IgG antibodies were purchased from MP Biomedicals (Solon, OH). Anti-human IgG was diluted 1:500 with 1% BSA in PBS, and were applied to each well as the secondary antibodies. We corrected the optical density (OD) values by subtracting those of control wells. The serum was considered to be positive when the corrected OD was more than 0.1. The serum samples were also investigated for anti-GSC antibodies as described elsewhere.9 10 The GSCs used in this study consisted of a mixture of 100 ng of each of two of the seven gangliosides—GM1, GM2, GD1a, GD1b, GT1a, GT1b and GQ1b. The presence of anti-GSC antibodies was established according to previously described criteria.10

Clinical analysis of anti-GSC antibody-positive patients with MFS and GBS with ophthalmoplegia

According to our recent study results,10 we divided MFS patients and patients with GBS-OP(+) into three groups based on the anti-GSC antibodies and anti-GQ1b IgG antibody, as described below in the Results section. The clinical and electrophysiological data of the subjects were obtained from the attending physicians in each hospital. Patients with no information on neurological signs and symptoms were excluded from the analysis. GBS patients’ disabilities were assessed using the Hughes Functional Grading Scale.13

Antecedent Campylobacter jejuni enteritis

IgM and IgG anti-Campylobacter jejuni (C jejuni) antibodies were investigated in serum samples from MFS and GBS-OP(+) patients using an ELISA kit for C jejuni (SERION ELISA classic, C jejuni IgG/IgM; Virion/Serion, Würzburg, Germany). The ELISA was performed on C jejuni antigen-coated plates according to the manufacturer’s instructions, and the results were evaluated. When the patient exhibited antecedent gastrointestinal episodes, such as diarrhoea and abdominal pain, and exhibited positive ELISA results for IgM or IgG antibodies to C jejuni, they were judged to have antecedent gastrointestinal infection associated with C jejuni.

Antibody specificity against GQ1b/GM1 or GQ1b/GD1a complex in MFS

To investigate the specificity of antibodies to GSCs including GQ1b, we assessed reactivity against GQ1b/GM1 or GQ1b/GD1a with ELISA, using representative sera from MFS patients with no anti-GSC antibody, anti-GQ1b/GM1 antibody or anti-GQ1b/GD1a antibody. In addition to GQ1b (200 ng), we added another ganglioside, GM1 or GD1a (0–600 ng)—ie, GQ1b/GM1 or GQ1b/GD1a = 200/0, 200/100, 200/200, 200/400 and 200/600 ng. The ELISA was performed as described above. A monoclonal anti-mouse GQ1b antibody (Seikagaku Corporation, Tokyo, Japan; diluted 1:40 with PBS) was used as a control, and peroxidase-labelled anti-mouse IgA+IgG+IgM antibody (KPL, Inc., Gaithersburg, MD: diluted 1:500 with PBS) was used as the secondary antibody. The corrected OD for each of the values was used to analyse their activities according to the mixture of these antigens.

Statistical analyses

Statistical analyses were performed using SPSS version 12.0J for Windows (SPSS Inc., Chicago). Two-tailed p-values of <0.05 were considered significant. One-way factorial analysis of variance (ANOVA) was used to compare ages. In multiple group comparisons, the Bonferroni test was used as a post-hoc test. The Kruskal–Wallis test was used for nonparametric comparisons of the Hughes functional grading scale. Differences in proportions were examined using contingency tables and the Chi square test or the Fisher’s exact test. When significance was demonstrated, all groups were compared by calculating the odds ratio (OR) with 95% confidence intervals (CIs).

RESULTS

Antibodies to GSCs or single ganglioside antigens

ELISA demonstrated that approximately half of MFS and GBS-OP(+) patients (47%) had one of the IgG antibodies to GSCs containing GQ1b or GT1a (table 1). Patients with antibodies specific to GQ1b (without anti-GSCs antibodies) were categorised into Group 1 (table 2). There were significantly more MFS patients compared with the GBS-OP(+) in Group 1 (table 1). More than one-third of GBS-OP(+) patients did not have either antibodies to GQ1b or GSCs containing GQ1b or GT1a. MFS patients who had at least one of the antibodies to GQ1b/GD1b, GT1a/GM1, GT1a/GD1b or GQ1b/GM1 were categorised into Group 2, because antibodies to GQ1b/GD1b, GT1a/GM1 or GT1a/GD1b were elevated concomitant with anti-GQ1b/GM1 antibody, but not with anti-GQ1b/GD1a antibody,10 and MFS patients who had at least one of the antibodies to GQ1b/GD1a, GQ1b/GT1b, GT1a/GD1a or GT1a/GT1b were categorised into Group 3 (table 2). In short, MFS patients who had antibodies reactive to GSCs with a total of two sialic acids in the terminal residues in GSCs, such as GQ1b/GM1, were classified into Group 2, and MFS patients who had antibodies to GSCs with a total of three sialic acids in the terminal residue, such as GQ1b/GD1a, were classified into Group 3. None of the MFS patients had any different types of anti-GSC antibodies simultaneously. In the same manner, GBS-OP(+) patients could be subdivided into three groups: anti-GQ1b-positive patients without anti-GSC antibodies (Group 1); patients who had antibodies reactive to GSCs with a total of two sialic acids in the terminal residues in GSCs, such as GQ1b/GM1 (Group 2); and who had antibodies reactive to GSCs with a total of three sialic acids in the terminal residue, such as GQ1b/GD1a (Group 3). However, five patients in the GBS Group 2 had IgG antibodies to GQ1b/GD1a, GQ1b/GT1b, GT1a/GD1a or GT1a/GT1b (table 2). These five patients were excluded from clinical analysis between Groups 1, 2 and 3. Three patients in the MFS Group 2, two in the GBS Group 2 and one in the GBS Group 3 had no anti-GQ1b IgG antibodies but had antibodies to GSCs containing GQ1b or GT1a. None of the MFS patients had antibodies to GSCs, consisting of two of the four major gangliosides (GM1, GD1a, GD1b and GT1b), which differed from the GBS patients. No antibodies to GSC were detected in the normal and disease control groups.

Table 1 Anti-GQ1b IgG antibodies and antibodies to ganglioside complexes containing GQ1b or GT1a in MFS, GBS-OP(+) and GBS-OP(−) groups
Table 2 IgG antibodies to GSCs and single ganglioside antigens in MFS, GBS-OP(+) and GBS-OP(−)

As for IgG antibodies to single ganglioside antigens other than GQ1b and GT1a, anti-GT1b IgG antibodies were found in seven MFS patients. In GBS-OP(+) patients, anti-GT1b antibodies were found in 16 patients, anti-GD1a antibodies in 6, anti-GalNAc-GD1a in 5, anti-GD1b in 3, anti-GM1 in 2 and anti-GM2 in 1 patient.

Clinical features in MFS and GBS with ophthalmoplegia

The clinical features of patients with MFS or GBS-OP(+) are shown in tables 3 and 4. Patients who did not have antibodies to GQ1b or GSCs containing GQ1b or GT1a were excluded from clinical analysis. Group 2 MFS patients were characterised by infrequent sensory disturbances (table 3) and tended to exhibit preserved bulbar function. Ataxia was common in the MFS and GBS-OP(+) groups except for GBS Group 3. Antibodies specific to GQ1b were highly associated with disturbances of deep sensation in GBS-OP(+). In GBS-OP(+), the number of patients requiring artificial ventilation was greatest in Group 1; however, there were no significant differences between groups. There were no statistical differences in functional score at the nadir of the disease between each group in GBS-OP(+) (data not shown). MFS and GBS-OP(+) patients who had antibodies to GSCs with a total of two sialic acids in the terminal residues in GSCs (Group 2) tended to suffer from antecedent respiratory infection more frequently than gastrointestinal infection. Antecedent C jejuni enteritis was evident in seven MFS Group 1, one MFS Group 3, three GBS Group 1, two GBS Group 2 and five GBS Group 3 patients. There were no significant differences in the frequency of C jejuni enteritis among these groups. Of 41 MFS patients with available electrophysiological results, 18 patients did not exhibit any abnormalities.

Table 3 Comparison of the clinical features in each MFS group
Table 4 Clinical features of GBS-OP(+) patients

Antibody specificities to GQ1b/GM1 or GQ1b/GD1a in sera from MFS patients

In anti-GSC antibody-negative sera, anti-GQ1b activity decreased with increasing concentrations of GM1 and GD1a (fig 1A), similar to the results with mouse monoclonal anti-GQ1b antibodies (fig 1D). In anti-GQ1b/GM1 antibody-positive sera, antibody activity increased proportionally with GM1 concentrations and decreased with increasing concentrations of GD1a, which differed from the results with anti-GQ1b/GD1a-positive sera (fig 1B, C).

Figure 1 Antibody specificities to GQ1b/GM1 or GQ1b/GD1a in sera from MFS patients. GM1 or GD1a (0–600 ng) was added to GQ1b (200 ng) in a well of an ELISA plate: ie, GQ1b:GM1 or GQ1b:GD1a = 200:0, 200:100, 200:200, 200:400 and 200:600 ng. Anti-GSC-negative serum (Group 1) (A), anti-GQ1b/GM1 antibody-positive serum (Group 2) (B), anti-GQ1b/GD1a antibody-positive serum (Group 3) (C) and monoclonal mouse anti-GQ1b IgG antibody (D) were used. Black squares show the antibody activity to a mixture of GQ1b and GM1. Triangles show the antibody activity to a mixture of GQ1b and GD1a. Each line shows a mean of two measurements.

DISCUSSION

The present study showed that antibodies to GSCs, including GQ1b or GT1a, as well as anti-GQ1b antibodies, are significantly associated with MFS and ophthalmoplegia in GBS. A survey of anti-GSC antibodies made it apparent that MFS-associated antibodies were subdivided into three types based on antibody specificity: ie, antibody specific to GQ1b, antibody reactive to GSCs containing GQ1b or GT1a with a total of two sialic acids in the terminal residues,10 and antibody reactive to GSCs with a total of three sialic acids in the terminal residues. The findings that none of the MFS patients had two or more types of these antibodies supported the validity of the classification. The reason for the diversity of antibody specificity remains to be determined.

IgG antibodies specific to GQ1b were more frequently in MFS than in GBS-OP(+) (Group 1), suggesting a strong association between the anti-GQ1b antibody and the development of MFS (table 1). Our study confirmed a close association between the GQ1b-specific antibody and ataxia or impaired deep sensation in GBS-OP(+), as shown in a previous study.14 Moreover, impairment of deep sensation was infrequent in patients with MFS despite the profound degree of ataxia, as pointed out by Mori et al.15 It is unclear why the loss of deep sensation is less common in MFS than in GBS in Group 1. Impairment of muscle spindle afferents might explain ataxia in MFS without loss of deep sensation.1517 In GBS-OP(+), antiganglioside antibodies other than those to GSCs containing GQ1b or GT1a, might influence neurological dysfunction such as impairment of superficial sensation.

IgG antibodies, which are reactive to GSCs with a total of two sialic acids in the terminal residues, such as GQ1b/GM1, appear to be associated with ophthalmoplegia in GBS as well as in the development of MFS. The present study confirmed that these types of antibodies led to the preservation of sensory function in MFS.

IgG antibodies, which are reactive to GSCs with a total of three sialic acids in the terminal residues, such as GQ1b/GD1a, were infrequent in MFS and we were not able to adequately identify the clinical characteristics in Group 3 patients. In GBS-OP(+) patients, on the other hand, ataxia and loss of deep sensation were attenuated in Group 3 patients. Such clinical differences might result from specific localisation of clustered epitopes consisting of a combination of [NeuAcα2-3Galβ1-3GalNAc] and [NeuAcα2-8NeuAcα2-3Galβ1-3GalNAc]. Thus, each of the three types of the antibodies is likely to be associated with some clinical features. However, clinical prospective studies and experimental studies are needed to confirm such associations.

It is interesting to note whether anti-GSC and anti-GQ1b antibodies bind to identical sites in neuronal membranes. As shown in figure 1, epitopes targeted by each of three types of the MFS-associated antibodies appear to be different.

It is notable that 19 GBS-OP(+) patients (36%) had neither anti-GQ1b IgG antibodies nor antibodies to GSCs containing GQ1b. In these 19 patients, GQ1b antigen may be uninvolved in the immune response leading to the development of ophthalmoplegia. As described recently,18 anti-GalNAc-GD1a antibodies might be associated with ophthalmoplegia in four of the 19 GBS-OP(+) patients with IgG anti-GalNAc-GD1a antibodies. Anti-ganglioside antibodies associated with ophthalmoplegia in GBS may be more diverse than in MFS.

MFS patients had no IgG antibodies to GSCs consisting of two of the four major gangliosides (GM1, GD1a, GD1b and GT1b), and no IgG antibodies to single ganglioside antigens except for GQ1b, GT1a and GT1b. This differs from the situation in GBS patients. The IgG antibodies to GSCs consisting of the four major gangliosides might correlate with development of limb weakness.19

Specific immunoadsorption therapy to remove antibodies could ameliorate the course of GBS and MFS.20 21 The synthetic disialylgalactose immunoaffinity columns with the minimal epitopes of GQ1b and GT1a were useful for eliminating anti-GQ1b antibodies from sera.22 The present and the previous immunoabsorption studies,10 however, suggest that an immunoabsorption column with the epitopes of GSCs, such as GQ1b/GM1 and GQ1b/GD1a, is required in a half of MFS patients to achieve optimal effects.

Acknowledgments

We would like to thank Miss Miwako Suemura for the technical assistance and Drs Hiroshi Ashida and Hitoshi Wakisaka (Division of Biomedical Information Sciences, National Defense Medical College Research Institute) for useful advice on the statistical analyses. This research was supported in part by Grants-in-Aid for Scientific Research (14570581 and 16590854) from the Ministry of Education, Culture, Sports, Science and Technology of Japan, and by a Research Grant for Neuroimmunological Diseases and a Health Sciences Research Grant (Research on Psychiatric and Neurological Diseases and Mental Health) from the Ministry of Health, Labour and Welfare of Japan.

We are grateful to the attending physicians at the following hospitals for the collection of clinical data: Fujita Health University, Hiroshima City Hospital, Hyogo College of Medicine, Jichi Medical University, Kanto Medical Center NTT EC, Kochi Health Sciences Center, Koseiren Takaoka Hospital, Kyushu University, Sekishinkai Sayama Hospital, National Hospital Organization Sendai Medical Center, Teikyo University, Tokyo Metropolitan Geriatric Medical Center, Tokyo Women’s Medical University, University of Tokyo, University of Toyama.

REFERENCES

Footnotes

  • Competing interests: None.