Anti-GD1a Antibody Is Associated with Axonal But Not Demyelinating Forms of Guillain-Barré Syndrome

Transcription

1 Anti-GD1a Antibody Is Associated with Axonal But Not Demyelinating Forms of Guillain-Barré Syndrome T. W. Ho, MD,* H. J. Willison, FRCP, I. Nachamkin, DrPH, C. Y. Li, MD, J. Veitch, FIMLS, H. Ung, BS, G. R. Wang, R. C. Liu, BS, D. R. Cornblath, MD,* A. K. Asbury, MD, J. W. Griffin, MD,* and G. M. McKhann, MD* Immunopathological studies suggest that the target of immune attack is different in the subtypes of Guillain-Barré syndrome (GBS). In acute motor axonal neuropathy (AMAN), the attack appears directed against the axolemma and nodes of Ranvier. In acute inflammatory demyelinating polyneuropathy (AIDP), the attack appears directed against a component of the Schwann cell. However, the nature of the antigenic targets is still not clear. We prospectively studied 138 Chinese GBS patients and found that IgG anti-gd1a antibodies were closely associated with AMAN but not AIDP. With a cutoff titer of greater than 1:100, 60% of AMAN versus 4% of AIDP patients had IgG anti-gd1a antibodies; with a cutoff titer of greater than 1:1,000, 24% of AMAN patients and none of the AIDP patients had IgG anti-gd1a antibodies. In contrast, low levels of IgG anti-gm1 antibodies (>1:100) were detected in both the AMAN and the AIDP forms (57% vs 35%, NS). High titers of IgG anti-gm1 (>1:1,000) were more common in the AMAN form (24% vs 8%, NS). Serological evidence of recent Campylobacter infection was detected in 81% of AMAN and 50% of AIDP patients, and anti-ganglioside antibodies were common in both Campylobacter-infected and noninfected patients. Our results suggest that IgG anti-gd1a antibodies may be involved in the pathogenesis of AMAN. Ho TW, Willison HJ, Nachamkin I, Li CY, Veitch J, Ung H, Wang GR, Liu RC, Cornblath DR, Asbury AK, Griffin JW, McKhann GM. Anti-GD1a antibody is associated with axonal but not demyelinating forms of Guillain-Barré syndrome. Ann Neurol 1999;45: Guillain-Barré syndrome (GBS) is a heterogeneous disorder with different clinical, electrophysiological, and pathological subtypes. In North America, Western Europe, and Australia, the major subtype is acute inflammatory demyelinating polyneuropathy (AIDP). AIDP is characterized electrophysiologically by demyelination of both motor and sensory nerves and pathologically by various degrees of lymphocytic infiltration and demyelination. 1 4 In northern China, Japan, and Mexico, acute motor axonal neuropathy (AMAN) is frequently encountered. 5 9 AMAN is characterized electrophysiologically by low motor-evoked amplitudes without features of demyelination and pathologically by macrophage-mediated attack on motor axons especially at the nodes of Ranvier, with variable amounts of Wallerian-like degeneration, but little inflammation or demyelination. 5 7,10 Several studies have shown that infection with Campylobacter jejuni has been associated with AIDP and AMAN. 5,11 14 Because anti-ganglioside antibodies are found in many GBS patients 5,11 13,15,16 and the lipopolysaccharide of some C. jejuni strains isolated from GBS patients contains ganglioside-like epitopes, molecular mimicry between epitopes on the surface of C. jejuni and the neural targets has been proposed as a possible mechanism for C. jejuni associated GBS. 20 The difference in the pattern of nerve damage and differential localization of antibody and complement in AIDP and AMAN suggest different targets of immune attack. 21,22 The targets of immune attack have been proposed to be gangliosides and/or structurally related glycoconjugates that are localized either at the abaxonal surface of the Schwann cell in the case of AIDP or at the motor nodes of Ranvier and adjacent axolemma in the case of AMAN. 21,22 Many anti-glycoconjugate antibodies have been From the *Department of Neurology, Johns Hopkins University School of Medicine, and Johns Hopkins University Zanvyl Krieger Mind Brain Institute, Baltimore, MD; University Department of Neurology, Southern General Hospital, Glasgow, UK; Departments of Pathology and Laboratory Medicine, and Neurology, University of Pennsylvania School of Medicine, Philadelphia, PA; and Department of Neurology, Second Teaching Hospital, Hebei Medical School, Shijiazhuang, People s Republic of China. Received May 11, 1998, and in revised form Jul 1 and Aug 8. Accepted for publication Aug 11, Address correspondence to Dr Ho, Department of Neurology, Pathology 509, 600 North Wolfe Street, Baltimore, MD Copyright 1999 by the American Neurological Association

2 described in GBS patients. The strongest association is between the Fisher syndrome and IgG anti-gq1b However, the association of anti-glycoconjugate antibodies with AIDP and AMAN is variable. 11,12,15,16,26 28 We examined the anti-glycoconjugate profile in a group of well-characterized, prospectively studied GBS patients with a balanced proportion of AMAN and AIDP subtypes. We then compared our findings with those of GBS patients from the United States. Patients and Methods Patients The study population included all patients with clinically defined GBS admitted to the Second Teaching Hospital of Hebei Medical University (Shijiazhuang, People s Republic of China) between January 1, 1992, and December 31, To be included, patients had to have their first neurological symptoms within 30 days before admission. One hundred seventy-eight patients with clinically defined GBS were admitted; of these, 138 patients who fulfilled the inclusion criteria were enrolled in the study. After informed consent, each patient underwent serial clinical examinations, serological and electrodiagnostic studies, and stool cultures as previously described. 5 All studies were approved by the appropriate institutional review boards. Forty-three GBS patients, from 1996 to 1997, admitted to Johns Hopkins Hospital and Hospital of the University of Pennsylvania who fulfilled the same inclusion criteria, were also tested for anti-ganglioside and anti-c. jejuni antibodies. Control Subjects During the study, control sera were collected from the following groups: (1) 39 age-matched and sex-matched controls from two Chinese villages during the summer, the peak season for GBS in China; (2) 10 hospitalized Chinese patients with other neurological diseases; (3) 18 patients from the United States with culture-positive uncomplicated Campylobacter infection; and (4) 13 family members of patients. Electrophysiological Studies Motor and sensory conduction studies were performed with a Dantec Cantata (Skoulunde, Denmark) machine by standard methods. The diagnostic definitions for AIDP and AMAN were based on our previous criteria, except that nerves with very low compound motor-evoked amplitude ( 10% of lower limit of normal) were classified as equivocal. 5 This change was made because of the recognition that the latency and conduction velocity may be abnormal and in the demyelinating range as a result of loss of fast conducting fibers alone. 29 Anti-Glycolipid Serology Enzyme-linked immunosorbent assays were performed on the initial acute sera from GBS patients, and control groups according to Willison and colleagues. 30 GM1, asialo-gm1 (GA1), GD1a, and GD1b (Sigma, St Louis, MO) were tested and a positive serological response analyzed by using two cutoff values for titers greater than 1:100 and greater than 1:1,000. Anti C. jejuni Antibodies Antibodies against C. jejuni were measured as previously described. 5 Sera from GBS patients and controls obtained in 1992 were reported previously. 5 Positivity was defined as an optical density ratio (ODR) of either 1 or more, or 2 or more, in two or more immunoglobulin classes. Previous studies of patients with culture-confirmed infections and of healthy controls showed that using an ODR definition of 1 or more was 100% sensitive and 90% specific, and 2 or more was 65% sensitive but 97% specific. 31 Statistical Analysis For statistical analysis, only patients with AIDP or AMAN were included; patients with inexcitable nerves or equivocal physiological studies were not included. Comparative analyses were done in a univariate fashion by using the Mantel Haenszel 2 and two-tailed Fisher exact tests, when indicated (Stata 5.0, College Station, TX). Because of multiple Table 1. Comparison of Anti-GM1 and Anti-GD1a Antibodies in AMAN and AIDP Pattern of GBS AMAN (n 68) AIDP (n 26) Odds Ratio (95% Confidence) p Titers 1:100 IgG anti-gd1a 60% 4% a ( ) IgG anti-gm1 57% 35% (1 6.5) Titers 1:1,000 IgG anti-gd1a 24% 0% 7.7 b a (1 61) b IgG anti-gm1 24% 8% (0.8 17) a Significance after Bonferroni corrections. b Estimated odds ratio and 95% confidence interval, assuming the 0% cell is 1. AMAN acute motor axonal neuropathy; AIDP acute inflammatory demyelinating polyneuropathy; GBS Guillain-Barré syndrome. Ho et al: Anti-GD1a Antibody in GBS 169

3 Table 2. Relationship of Anti-Ganglioside Antibodies to C. jejuni Serology in Patients with GBS C.J. Positive (n 99) C.J. Negative (n 33) Odds Ratio (95% Confidence) p Titers 1:100 Any anti-ganglioside 82% 61% ( ) IgG anti-gd1a 49% 33% 2 NS ( ) IgG anti-gm1 56% 33% ( ) Titers 1:1,000 Any anti-ganglioside 54% 24% a ( ) IgG anti-gd1a 18% 9% 2.2 NS (0.6 8) IgG anti-gm1 28% 15% 2.2 NS ( ) a Significance after Bonferroni correction; NS not significant. GBS Guillain-Barré syndrome; C.J. Campylobacter jejuni. comparisons, the significant levels were adjusted to p for Table 1 (eight comparisons) and p for Table 2 (10 comparisons), using the Bonferroni correction. Results Of the 138 Chinese patients enrolled in the study, 72 were classified as AMAN (52%), 26 as AIDP (19%), 33 as equivocal (24%), and 7 had inexcitable motor nerves (5%). The median ages for AMAN, AIDP, equivocal, and inexcitable were 13, 18, 22, and 7 years, respectively. Anti-Glycolipid Serology IgG anti-gd1a, GM1, GD1b, and GA1 antibody titers for the different groups are shown in the Figure. IGG ANTI-GD1A. This antibody was the most specific for the AMAN group, with good sensitivity (see Table 1). With a low cutoff, 60% of AMAN patients compared with 4% of AIDP patients were positive for IgG anti-gd1a antibodies (p ; odds ratio, 38; 95% confidence interval, 5 297). With the higher cutoff, the sensitivity for AMAN decreased to 24%, but specificity was 100% (p 0.005). Only 1 of the family controls showed a low titer (1:390) of IgG anti- GD1a antibodies. None of the other controls had detectable anti-gd1a antibodies. IGG ANTI-GM1. Like anti-gd1a, anti-gm1 antibodies were statistically more common in the GBS patients than in the control groups. Only 3 of the village controls (p ) and 1 of the family controls (p 0.002) had low-titer anti-gm1 antibodies compared with GBS patients. When analyzed by GBS subtypes, IgG anti-gm1 antibodies were more common in the AMAN group than in the AIDP group. With the low cutoff, 57% of the AMAN patients versus 35% of AIDP patients were positive for IgG anti-gm1 antibodies (not significant [NS]). With a higher cutoff, 24% versus 8% were positive for IgG anti-gm1 antibodies (NS). Thirty-eight percent of AMAN and 4% of AIDP patients were positive for both IgG anti- GD1a and anti-gm1 antibodies with the lower cutoff. IGG ANTI-GD1B. The presence of IgG anti-gd1b antibodies was not significantly different between GBS patients and controls groups with either cutoff (see Fig, C). When analyzed within the subgroups of GBS patients, and with a high cutoff, however, IgG anti- GD1b antibodies were more common in the AMAN group than in the AIDP group (19% vs 0%, respectively; NS). OTHER ANTI-GLYCOLIPID ANTIBODIES. IgG anti-ga1 antibodies were frequently detected in both GBS patients and in control groups and did not differ significantly (see Fig, D). The levels of other classes of antiglycolipid antibodies (IgA and IgM) did not reach statistical significance (results not shown). Comparison with US GBS Patients Forty-three US GBS patients were tested for the presence of anti-glycolipid antibodies; 36 exhibited demyelinating electrophysiology and 2 had axonal electrophysiology. Five more met clinical criteria for Fisher syndrome. IgG anti-gd1a antibodies at greater than 1:1,000 were detected only in 1 of the 2 axonal patients and in none of the other 36 AIDP patients (NS). In addition, 2 AIDP and 1 Fisher patient had low titers ( 1:400) of IgG anti-gd1a antibodies. IgG GM1 170 Annals of Neurology Vol 45 No 2 February 1999

4 Fig. IgG anti-ganglioside antibodies in different groups. AMAN acute motor axonal neuropathy; AIDP acute inflammatory demyelinating polyneuropathy; Equiv. equivocal; Inexcit. inexcitable; OND other neurological disease; C.J. C. jejuni enteritis; 5 patient samples; US axonal Guillain-Barré syndrome (US GBS). antibodies were present in 3 patients, of which 1 was an AMAN case. Relationship with Anti C. jejuni Serology With positive defined as an ODR 1 or more in two or more immunoglobulin classes, 74% of Chinese patients with GBS (81% of patients with AMAN and 50% of patients with AIDP) and 16% of Chinese village control subjects were seropositive. With the more stringent definition of positive (ODR 2.0 in two or more immunoglobulin classes), 45% of patients with GBS (47% of patients with AMAN and 27% of patients with AIDP) and 4% of Chinese village controls were seropositive. Anti-glycolipid antibodies were common in both C. jejuni positive and C. jejuni negative groups (see Table 2). However, C. jejuni positive patients were more likely to have anti-ganglioside antibodies (at low-titer cutoff, p 0.02, not significant after Bonferroni correction; p 0.005, at high-titer cutoff). In a similar manner, IgG anti-gm1 antibodies were slightly more frequent in the C. jejuni positive group than in the C. jejuni negative group (NS). No significant difference was observed for IgG anti-gd1a antibodies. Discussion In this prospective study, high-titer IgG anti-gd1a antibodies were strongly correlated with Chinese AMAN but not with AIDP patients or control subjects. In the US patients studied, high titers of IgG anti-gd1a antibodies were found in 1 of the 2 axonal cases but in none of the AIDP cases. Although case reports have suggested that IgG anti-gd1a can occur in GBS, most studies have not found this association. 13,14,16,37 We suggest the association between AMAN and IgG anti-gd1a was not recognized, as most studies involved GBS populations consisting predominantly, if not exclusively, of AIDP patients. Forty percent of AMAN patients were negative for anti-gd1a antibodies. We did not find any statistically significant differences between the anti-gd1a positive or anti-gd1a negative AMAN cases in terms of the time when the first blood samples were taken, age, season, or prior C. jejuni infection. It is possible that other antigens or other pathogenetic mechanisms may be involved in these patients. A more sensitive assay, however, to test for anti-gd1a antibodies may be needed. IgG anti-gm1 antibodies were found more frequently in AMAN than in AIDP patients. The differ- Ho et al: Anti-GD1a Antibody in GBS 171

5 ence is greatest when the high-titer cutoff is used as previously reported by Kornberg and colleagues. 27 Similar to results from our previous study, however, IgG anti-gm1 antibodies are neither as specific nor as sensitive as IgG anti-gd1a antibodies. 5 Enders and colleagues 12 and Vriesendorp and associates 11 found no correlation between anti-gm1 antibodies and GBS subtypes as determined electrophysiologically. Rees and co-workers 16 detected IgG anti-gm1 antibodies in 25% of their AIDP cases versus 57% in AMAN/acute motor sensory axonal neuropathy (AMSAN) (p 0.07). These studies suggest that anti-gm1 antibodies are only modestly correlated with different subtypes of GBS. The epitope(s) involved in AIDP remains elusive. Anti-GM1, GD1b, and GA1 antibodies can occur in both axonal and demyelinating patients. Although there may be many different forms of AIDP, each with its own antibodies, the presence of many of these antiglycoconjugate antibodies in axonal and in demyelinating cases of GBS and in controls suggests questions about their specificity and their biological significance. Antecedent C. jejuni infection is common in all subtypes of GBS including both AIDP and AMAN. 5,38,39 In the 10 strains of C. jejuni we have isolated from GBS patients, two are from patients with AIDP, seven are from patients with AMAN, and one is from a patient with Fisher syndrome. 40 Rees and colleagues 14 reported that in GBS patients from England, 76% of C. jejuni positive patients had AIDP and 24% had AMAN/AMSAN. AMAN/AMSAN patients were frequently C. jejuni positive (6 of 7 patients), but because of the greater frequency overall of AIDP patients, most of their C. jejuni positive patients had AIDP. In a similar manner, studies in Europe and the United States, by Enders and associates, 12 by Vriesendorp and collaborators, 11 and by Jacobs and colleagues, 41 have shown that GBS associated with C. jejuni is predominantly associated with AIDP. We did not find any relationship between antiganglioside antibody and C. jejuni infection. In a single strain of C. jejuni, the terminal sugar structure in the lipopolysaccharide contains many ganglioside-like epitopes, mimicking among others GM1, GM1b, and GD1a. 17,19 Therefore, it is not surprising that many GBS patients with prior C. jejuni infection have raised antibodies against these epitopes. Anti-ganglioside antibodies are common in both C. jejuni infected and C. jejuni noninfected patients, suggesting that other inciting factors can also elicit anti-ganglioside antibody responses. In conclusion, IgG anti-gd1a antibodies are strongly associated with the AMAN subtype of GBS. Our study suggests that antibody against a GD1a-like epitope is important in the pathogenesis of axonal GBS. Our previous studies have shown that the axolemma at motor nodes of Ranvier is the most likely target for antibody-mediated immune attack. 21 The present results suggest that a GD1a-like epitope may be present on the node of Ranvier and subject to immune attack in axonal GBS. Further studies are needed to investigate the localization of GD1a in peripheral nerve and to determine the pathogenetic significance of IgG anti-gd1a antibodies in animal models of GBS. This study was supported in part by National Institutes of Health RO1 NS34846, RO1 NS31528, and KO8 NS , the GBS Support Group (UK), and the Wellcome Trust. Dr Willison is a Wellcome Trust Senior Fellow. References 1. Haymaker W, Kernohan JW. The Landry-Guillain-Barré syndrome: a clinicopathologic report of fifty fatal cases and a critique of the literature. Medicine 1949;28: Asbury AK, Arnason BG, Adams RD. The inflammatory lesion in idiopathic polyneuritis: its role in pathogenesis. Medicine 1969;48: Prineas JW. Acute idiopathic polyneuritis. An electron microscope study. Lab Invest 1972;26: Prineas JW. Pathology of the Guillain-Barré syndrome. Ann Neurol 1981;9(Suppl):S6 S19 5. Ho TW, Mishu B, Li CY, et al. Guillain-Barré syndrome in northern China: relationship to Campylobacter jejuni infection and anti-glycolipid antibodies. Brain 1995;118: McKhann GM, Cornblath DR, Griffin JW, et al. Acute motor axonal neuropathy: a frequent cause of acute flaccid paralysis in China. Ann Neurol 1993;33: McKhann GM, Cornblath DR, Ho T, et al. Clinical and electrophysiological aspects of acute paralytic disease of children and young adults in northern China. Lancet 1991;338: Sobue G, Li M, Terao S, et al. Axonal pathology in Japanese Guillain-Barré syndrome: a study of 15 autopsied cases. Neurology 1997;48: Ramos-Alvarez M, Bessudo L, Sabin A. Paralytic syndromes associated with noninflammatory cytoplasmic or nuclear neuronopathy: acute paralytic disease in Mexican children, neuropathologically distinguishable from Landry-Guillain-Barré syndrome. JAMA 1969;207: Griffin JW, Li CY, Ho TW, et al. Guillain-Barré syndrome in northern China: the spectrum of neuropathological changes in clinically defined cases. Brain 1995;118: Vriesendorp FJ, Mishu B, Blaser MJ, Koski CL. Serum antibodies to GM1, GD1b, peripheral nerve myelin, and Campylobacter jejuni in patients with Guillain-Barré syndrome and controls: correlation and prognosis. Ann Neurol 1993;34: Enders U, Karch H, Toyka KV, et al. The spectrum of immune responses to Campylobacter jejuni and glycoconjugates in Guillain-Barré syndrome and in other neuroimmunological disorders. Ann Neurol 1993;34: Walsh FS, Cronin M, Koblar S, et al. Association between glycoconjugate antibodies and Campylobacter infection in patients with Guillain-Barré syndrome. J Neuroimmunol 1991;34: Rees JH, Soudain SE, Gregson NA, Hughes RA. Campylobacter jejuni infection and Guillain-Barré syndrome. N Engl J Med 1995;333: Gregson NA, Koblar S, Hughes RA. Antibodies to gangliosides 172 Annals of Neurology Vol 45 No 2 February 1999

6 in Guillain-Barré syndrome: specificity and relationship to clinical features. Q J Med 1993;86: Rees JH, Gregson NA, Hughes RA. Anti-ganglioside GM1 antibodies in Guillain-Barré syndrome and their relationship to Campylobacter jejuni infection. Ann Neurol 1995;38: Aspinall GO, Fujimoto S, McDonald AG, et al. Lipopolysaccharides from Campylobacter jejuni associated with Guillain- Barré syndrome patients mimic human gangliosides in structure. Infect Immun 1994;62: Aspinall GO, McDonald AG, Pang H, et al. Lipopolysaccharides of Campylobacter jejuni serotype O:19: structures of core oligosaccharide regions from the serostrain and two bacterial isolates from patients with the Guillain-Barré syndrome. Biochemistry 1994;33: Yuki N, Taki T, Takahashi M, et al. Penner s serotype 4 of Campylobacter jejuni has a lipopolysaccharide that bears a GM1 ganglioside epitope as well as one that bears a GD1 a epitope. Infect Immun 1994;62: Yuki N, Taki T, Inagaki F, et al. A bacterium lipopolysaccharide that elicits Guillain-Barré syndrome has a GM1 ganglioside-like structure. J Exp Med 1993;178: Hafer-Macko C, Hsieh ST, Li CY, et al. Acute motor axonal neuropathy: an antibody-mediated attack on axolemma. Ann Neurol 1996;40: Hafer-Macko CE, Sheikh KA, Li CY, et al. Immune attack on the Schwann cell surface in acute inflammatory demyelinating polyneuropathy. Ann Neurol 1996;39: Yuki N, Sato S, Tsuji S, et al. Frequent presence of anti-gq1b antibody in Fisher s syndrome. Neurology 1993;43: Willison HJ, Veitch J, Paterson G, Kennedy PG. Miller Fisher syndrome is associated with serum antibodies to GQ1b ganglioside. J Neurol Neurosurg Psychiatry 1993;56: Chiba A, Kusunoki S, Shimizu T, Kanazawa I. Serum IgG antibody to ganglioside GQ1b is a possible marker of Miller- Fisher syndrome. Ann Neurol 1992;31: Yuki N, Yoshino H, Sato S, Miyatake T. Acute axonal polyneuropathy associated with anti-gm1 antibodies following Campylobacter enteritis. Neurology 1990;40: Kornberg AJ, Pestronk A, Bieser K, et al. The clinical correlates of high-titer IgG anti-gm1 antibodies. Ann Neurol 1994;35: Visser LH, Van der Meche FG, Van Doorn PA, et al. Guillain- Barré syndrome without sensory loss (acute motor neuropathy): a subgroup with specific clinical, electrodiagnostic and laboratory features. Dutch Guillain-Barré study group. Brain 1995; 118: Cornblath DR, Kuncl RW, Mellits ED, et al. Nerve conduction studies in amyotrophic lateral sclerosis. Muscle Nerve 1992;15: Willison HJ, Chancellor AM, Paterson G, et al. Antiglycolipid antibodies, immunoglobulins and paraproteins in motor neuron disease: a population based case-control study. J Neurol Sci 1993;114: Mishu B, Ilyas AA, Koski CL, et al. Serologic evidence of previous Campylobacter jejuni infection in patients with the Guillain-Barré syndrome. Ann Intern Med 1993;118: Lugaresi A, Ragno M, Torrieri F, et al. Acute motor axonal neuropathy with high titer IgG and IgA anti-gd1a antibodies following Campylobacter enteritis. J Neurol Sci 1997;147: Yuki N, Yoshino H, Sato S, et al. Severe acute axonal form of Guillain-Barré syndrome associated with IgG anti-gd1a antibodies. Muscle Nerve 1992;15: Yuki N, Yamada M, Sato S, et al. Association of IgG anti- GD1a antibody with severe Guillain-Barré syndrome. Muscle Nerve 1993;16: Bollensen E, Schipper HI, Steck AJ. Motor neuropathy with activity of monoclonal IgM antibody to GD1a ganglioside. J Neurol 1989;236: Carpo M, Nobile-Orazio E, Meucci N, et al. Anti-GD1a ganglioside antibodies in peripheral motor syndromes. Ann Neurol 1996;39: Svennerholm L, Fredman P. Antibody detection in Guillain- Barré syndrome. Ann Neurol 1990;27(Suppl):S36 S Kuroki S, Saida T, Nukina M, et al. Campylobacter jejuni strains from patients with Guillain-Barré syndrome belong mostly to Penner serogroup 19 and contain -N-acetylglucosamine residues. Ann Neurol 1993;33: Yuki N, Takahashi M, Tagawa Y, et al. Association of Campylobacter jejuni serotype with antiganglioside antibody in Guillain-Barré syndrome and Fisher s syndrome. Ann Neurol 1997;42: Sheikh KA, Nachamkin I, Ho TW, et al. Campylobacter jejuni lipopolysaccharides in Guillain-Barré syndrome: molecular mimicry and host susceptibility. Neurology 1998;51: Jacobs BC, van Doorn PA, Schmitz PI, et al. Campylobacter jejuni infections and anti-gm1 antibodies in Guillain-Barré syndrome. Ann Neurol 1996;40: Ho et al: Anti-GD1a Antibody in GBS 173