Cerebral amyloid angiopathy A microvascular link between parenchymal and vascular dementia.код для вставкиСкачать
EDITORIALS Cerebral Amyloid Angiopathy: A Microvascular Link Between Parenchymal and Vascular Dementia? Why should cerebral amyloid (congophilic) angiopathy (CAA) be of interest to neuroscientists? This peculiar cerebral microvascular lesion is one in which arteriolar media, including its smooth muscle cell (SMC) component, is gradually replaced by fibrillar amyloid (Fig).1 The end result of this process is the presence, within cerebral cortex, of markedly weakened arterial walls, which frequently give rise to (lobar) cerebral hemorrhages.2,3 Less often, severe CAA is associated with ischemic brain lesions and granulomatous angiitis.4,5 That CAA is most common and severe in patients with Alzheimer disease (AD) has been well documented in many studies.6 In this issue of the Annals, Grabowski and colleagues present clinicopathologic details of an Iowa family with autosomal dominant inheritance of a phenotype characterized by progressive aphasic dementia; in one patient who came to necropsy, there was evidence for severe CAA.7 Of interest is that the clinical and neuropathologic syndrome is linked to a mutation (causing a substitution of asparagine for aspartic acid) at codon 694 of the gene that encodes the amyloid precursor protein (APP), from which A␤ is cleaved. The APP codon 694 mutation represents the third APP codon in a row, mutations in which determine a phenotype that includes CAA. A Gln-for-Glu substitution at residue 22 of A␤ (APP codon 693) is linked to hereditary cerebral hemorrhage with amyloidosis, Dutch type (HCHWA-D), a disease in which lobar cerebral hemorrhages and ischemic brain lesions occur in the context of overwhelming meningeal and neocortical CAA.8,9 Although parenchymal amyloid deposits are seen in HCHWA-D brain, neurofibrillary degeneration of neurons is rare and does not appear to correlate with the severity of either microvascular or parenchymal A␤ deposition.10 In nearby Belgium, patients with an APP codon 692 missense mutation manifest massive brain parenchymal and microvascular A␤ deposition, the latter sometimes causing brain hemorrhage, although only a small number of patients with this disease have been studied at autopsy.11 The new “Iowa kindred” will have to be carefully studied using rigorous clinical, neuropsychologic, neuroimaging, and neuropathologic tools, the latter to es- tablish how constant the CAA phenotype is among other family members. In the patient whose brain is described, the morphologic features of CAA were remarkably similar to those noted in HCHWA-D, including radially disposed amyloid fibrils and secondary calcification of (some) affected cerebral microvascular walls.9,12 But can predominantly microvascular (arteriolar and capillary) amyloid deposition lead to dementia and, if so, by what mechanism? The lessons to be learned from patients with HCHWA-D and this newly described family suggest that the answer to the first question is, overwhelmingly, “yes.” A major mechanism causing dementia in Dutch patients appears to be the presence of ischemic brain infarcts and residua of encephalic bleeds.8 –10,12 Retrospective clinicopathologic correlative studies using autopsy material have also shown that cerebrovascular “events” or lesions in individuals with HCHWA-D are a function of the severity of microvascular degeneration linked directly to the presence of CAA.13 Interestly, the single patient from the Iowa kindred examined at necropsy showed widespread abundant neurofibrillary tangles and frequent A␤1– 40-immunoreactive senile plaques, both of which almost certainly contributed to this individual’s dementia syndrome. Small brain infarcts at the cortex–white matter junction are also described, and leukoencephalopathy was noted, ie, the neuropathologic picture showed some features that have been described in patients (lacking CAA) with a relatively “pure” ischemic vascular dementia.14 Patients with a dementia syndrome consistent with AD on occasion have predominantly A␤-immunoreactive CAA as the overwhelming brain lesion present on neuropathologic evaluation. Does this simply represent a microvascular variant of AD resulting from extensive A␤ deposition mainly in capillary and arterial walls (rather than neuropil), or is it a distinctive disease process?15 The existence of such cases should be viewed as an opportunity to explore heterogeneous mechanisms of CNS “amyloidogenesis,” some resulting in parenchymal and others in vascular A␤ deposition.16 Nor is A␤ the only protein that has been implicated in CAA pathogenesis. In CAA associated with AD, gamma trace (cystatin C) is known to be a major constituent of affected arterial walls.17 Comparatively rare forms of familial CAA (fCAA) are associated with arterial medial deposition of unique proteins, the most recently identified one being “ABri,” found in a British kindred of familial dementia linked to a mutation of a gene on chromosome 13.18 Thus, an understanding of CAA may tell us a great deal about novel mechanisms of A␤ deposition and toxicity in the central nervous system, but it might tell us even more about the many types of cerebral microvascular disease characterized by progressive injury and loss of SMCs within the arteriolar me- © 2001 Wiley-Liss, Inc. 691 Brain Research Institute and Neuropsychiatric Research Institute UCLA Medical Center Los Angeles, CA References Fig. (A) A relatively normal arteriole in a nondemented patient. The vessel has been cut at an angle. Perivascular clearing is an artefact of tissue preparation. Arrows indicate arteriolar media with a relatively normal complement of smooth muscle cells (SMCs). (B) An arteriole with severe cerebral amyloid angiopathy, by contrast. Medial SMCs have been replaced by fibrillar amyloid. However, a relatively normal endothelial cell (arrow) is seen adjacent to the lumen. Hematoxylin and eosin. ⫻405 (A), ⫻375 (B). dia (“cerebral angiomyopathies”).19 Finally, patients with severe CAA (including syndromes of fCAA) may offer a unique opportunity to study injurious effects of microvascular amyloid deposition (short of ischemic/ hemorrhagic stroke) on perivascular brain tissue. Harry V. Vinters, MD Departments of Pathology and Laboratory Medicine (Neuropathology) and Neurology 692 Annals of Neurology Vol 49 No 6 June 2001 1. Vinters HV, Secor DL, Read SL, et al. The microvasculature in brain biopsy specimens from patients with Alzheimer’s disease: an immunohistochemical and ultrastructural study. Ultrastruct Pathol 1994;18:333–348. 2. Gilbert GG, Vinters HV. Cerebral amyloid angiopathy: incidence and complications in the aging brain. I. Cerebral hemorrhage. Stroke 1983;14:915–923. 3. Vinters HV. Cerebral amyloid angiopathy—a critical review. Stroke 1987;18:311–324. 4. Cadavid D, Mena H, Koeller K, Frommelt RA. Cerebral beta amyloid angiopathy is a risk factor for cerebral ischemic infarction. A case control study in human brain biopsies. J Neuropathol Exp Neurol 2000;59:768 –773. 5. Anders KH, Wang ZZ, Kornfeld M, et al. Giant cell arteritis in association with cerebral amyloid angiopathy: immunohistochemical and molecular studies. Hum Pathol 1997;28:1237– 1246. 6. Greenberg SM. Cerebral amyloid angiopathy. Prospects for clinical diagnosis and treatment. Neurology 1998;51:690 – 694. 7. Grabowski TJ, Cho HS, Vonsattel JPG, et al. Novel amyloid precursor protein mutation in an Iowa family with dementia and severe cerebral amyloid angiopathy. Ann Neurol 2001;49: 697–705. 8. Bornebroek M, Haan J, Maat-Schieman MLC, Van Duinen SG, Roos RAC. Hereditary cerebral hemorrhage with amyloidosis-Dutch type (HCHWA-D): I—a review of clinical, radiologic and genetic aspects. Brain Pathol 1996;6:111–114. 9. Maat-Schieman MLC, van Duinen SG, Bornebroek M, et al. Hereditary cerebral hemorrhage with amyloidosis-Dutch type (HCHWA-D): II—a review of histopathological aspects. Brain Pathol 1996;6:115–120. 10. Maat-Schieman MLC, Van Duinen SG, Natté R, Roos RAC. Neuropathology of hereditary cerebral hemorrhage with amyloidosis-Dutch type. In: Verbeek MM, de Waal RMW, Vinters HV, eds. Cerebral amyloid angiopathy in Alzheimer’s disease and related disorders. Dordrecht: Kluwer Academic Publishers, 2000:223–236. 11. Roks G, Van Harskamp F, De Koning I, et al. Presentation of amyloidosis in carriers of the codon 692 mutation in the amyloid precursor protein gene (APP692). Brain 2000;123:2130 – 2140. 12. Vinters HV, Natté R, Maat-Schieman MLC, et al. Secondary microvascular degeneration in amyloid angiopathy of patients with hereditary cerebral hemorrhage with amyloidosis, Dutch type (HCHWA-D). Acta Neuropathol 1998;95:235–244. 13. Natté R, Vinters HV, Maat-Schieman MLC, et al. Microvasculopathy is associated with the number of cerebrovascular lesions in hereditary cerebral hemorrhage with amyloidosis, Dutch type. Stroke 1998;29:1588 –1594. 14. Vinters HV, Ellis WG, Zarow C, et al. Neuropathologic substrates of ischemic vascular dementia. J Neuropathol Exp Neurol 2000;59:931–945. 15. Vinters HV. Cerebral amyloid angiopathy and Alzheimer’s disease: two entities or one? J Neurol Sci 1992;112:1–3. 16. Verbeek MM, Eikelenboom P, de Waal RMW. Differences between the pathogenesis of senile plaques and congophilic angiopathy in Alzheimer disease. J Neuropathol Exp Neurol 1997; 56:751–761. 17. Vinters HV, Secor DL, Pardridge WM, Gray F. Immunohisto- chemical study of cerebral amyloid angiopathy. III. Widespread Alzheimer A4 peptide in cerebral microvessel walls colocalizes with gamma trace in patients with leukoencephalopathy. Ann Neurol 1990;28:34 – 42. 18. Frangione B, Vidal R, Rostagno A, Ghiso J. Familial cerebral amyloid angiopathies and dementia. Alzheimer Dis Assoc Disord 2000;14(Suppl 1):S25–S30. 19. Vinters HV, Wang Z. Non-CAA angiopathies and their possible interactions with cerebral amyloid angiopathy. Amyloid: J Protein Folding Disord 8(Suppl 1) (in press, 2001). When Running a Stop Sign May Be a Good Thing Despite the remarkable discoveries regarding the molecular basis of Duchenne and Becker muscular dystrophies (DMD/BMD) and of their animal models during the past decade,1 definitive treatment has not become available.2 Various forms of molecular therapies, such as dystrophin gene replacement or utrophin up-regulation, are still in the preclinical stage.3 Thus, a recent report from Barton-Davis and coworkers4 created excitement by observing a significant number a dystrophin-positive muscle fibers in skeletal muscles of mdx mice after a short course of systemic administration of gentamicin, an aminoglycoside antibiotic. These authors hypothesized that gentamicin interfered with the ability of ribosomes to recognize the pathogenic translational stop codon UAA in the dystrophin mRNA, resulting in a “read through” and the production of stable, full-length dystrophin. In earlier studies, the aminoglycosides’ ability to misread stop codons has been used for negating the deleterious effects of stop mutations of the cystic fibrosis transmembrane regulator in a bronchial cell line vitro5 and in the nasal mucosa of patients in vivo.6 More recently, in a fibroblast cell line in vitro, heterozygous stop mutations of the lysosomal enzyme alpha-L-iduronidase (whose deficiency causes the majority of Hurler’s syndrome cases) were shown to be partially overridden by gentamicin.7 Howard and coworkers,8 using an elegant in vitro system, quantitated the gentamicin-induced read-through for the three different stop codons and also showed a modifying effect of the flanking nucleotides. Barton-Davis et al4 suggested that, despite the relatively low frequency (about 15%) of DMD/BMD patients with a disease-causing nonsense mutation and the notorious toxicity of aminoglycosides, this might be a treatment for selected patients.This challenge was taken up by Wagner et al,9 who, in this issue of the Annals, report their study of 2 DMD and 2 BMD pa- tients who received daily intravenous bolus injections of 7.5 mg/kg gentamicin for 14 days. This produced a peak gentamicin serum concentration similar to that noted in the mdx mice.4 All patients had a nonsense mutation of their dystrophin gene, creating a premature stop codon (2⫻ TGA, TAA, TAG). No oto- or nephrotoxicity was noted. No full-length dystrophin was found in the pre- and posttreatment tibialis anterior muscle biopsies in any of the 4 patients by microscopic immunocytochemistry or Western blot analysis.9 The key issue is what causes the difference between the studies of Barton-Davis et al4 in mdx mice and those of Wagner et al9 in properly selected DMD/ BMD patients despite the close similarity in the peak gentamicin serum levels and the duration of treatment. It appears likely that the difference is related to species and possibly drug composition. There are several relevant parameters that may differ between mice and man. It is believed that the “readthrough” frequency induced by a subtoxic concentration of gentamicin is a relatively rare event.9 In other words, read-through dystrophin mRNA copies may constitute a relatively small percentage of the total number of translated mRNA copies. Furthermore, it is not clear whether pulsed administration of gentamicin producing conspicuous blood level fluctuations is the best way to maximize the read-through efficiency. In addition, the turnover rate of dystrophin is suspected to be slow.10 Thus, it might take more than 2 weeks of treatment to establish a steady state of sarcolemmal dystrophin that is not only detectable but therapeutically useful. In fact, perhaps such a steady state is never attainable in man using subtoxic doses of gentamicin. An important issue may be the brand and composition of gentamicin sulphate used. A given batch of commercially available gentamicin usually contains three major components (gentamicin C1, C1a, and C2) that show only slight differences in their chemical structure but differ significantly in their propensity to induce misreading in bacterial and eukaryotic ribosomes.11,12 Thus, the different results may be attributable to different brands of gentamicin used in the mouse and human studies cited.4,9 Where do we go from here? The first task is to repeat and verify the results of the mdx mouse experiments4 using gentamicin with a defined composition. Importantly, in our laboratories (G.K.) no full-length dystrophin could be induced by the same experimental paradigm in mdx mice using gentamicin from a supplier other than Barton-Davis et al.4 To optimize gentamicin composition for misreading efficiency, Howard and colleagues’8 in vitro paradigm may be used. Further human experiments should be delayed until some of these issues are resolved in animal experiments. However, even if full-length dystrophin production by administration of aminoglycosides becomes possible in © 2001 Wiley-Liss, Inc. 693 DMD patients, it may require high and permanent doses close to the toxic level. A lifelong commitment to such a schedule is a medically and ethically vexing decision even for a disease as pernicious as DMD. George Karpati, MD Montreal Neurological Institute McGill University Montreal, Quebec, Canada Hanns Lochmuller, MD Genzentrum Ludwig-Maximilians-University Munich, Germany References 1. Karpati G, Brown RH Jr. The dawning of a new era in the molecular biology of the muscular dystrophies. Brain Pathol 1996;6:17. 2. Allamand V, Campbell KP. Animal models for muscular dystrophy: valuable tools for the development of therapies. Hum Mol Genet 2000;9:2459 –2467. 3. Hartigan-O’Connor D, Chamberlain JS. Developments in gene therapy for muscular dystrophy. Microsc Res Techn 2000;48: 223–238. 4. Barton-Davis ER, Cordier L, Shoturma DI, et al. Aminoglycoside antibiotics restore dystrophin function to skeletal muscles of mdx mice. J Clin Invest 1999;104:375–381. 5. Howard M, Frizell RA, Bedwell DM. Aminoglycoside antibiotics restore CFTR function by overcoming premature stop codon mutations. Nature Med 1996;2:467– 469. 6. Wilschanski M, Famini C, Blau H, et al. A pilot study of the effect of gentamicin on nasal potential difference measurements in cystic fibrosis patients carrying stop mutations. Am J Respir Crit Care Med 2000;161:860 – 865. 7. Keeling KM, Brooks DA, Hopwood JJ, et al. Gentamicinmediated suppression of Hurler syndrome stop mutations restores a low level of ␣-L-iduronidase activity and reduces lysosomal glycosaminoglycan accumulation. Hum Mol Genet 2001;10:291–299. 8. Howard MT, Shirts BH, Petros LM, et al. Sequence specificity of aminoglycoside-induced stop condon readthrough: potential implications for treatment of Duchenne muscular dystrophy. Ann Neurol 2000;48:164 –169. 9. Wagner KR, Hamed S, Hadley DW, et al. Gentamicin treatment of Duchenne and Becker muscular dystrophy due to nonsense mutations. Ann Neurol 2001;49:706 –711. 10. Ahmad A, Brinson M, Hodges BL, et al. Mdx mice inducibly expressing dystrophin provide insights into the potential of gene therapy for Duchenne muscular dystrophy. Hum Mol Genet 2000;9:2507–2515. 11. Loveless MO, Kohlhepp SJ, Gilbert DN. The influence of aminoglycoside antibiotics on the in vitro function of rat liver ribosomes. J Lab Clin Med 1984;103:294 –303. 12. Yoshizawa S, Fourmy D, Puglisi JD. Structural origins of gentamicin antibiotic action. EMBO J 1998;17:6437– 6448. 694 © 2001 Wiley-Liss, Inc. Variants of the Guillain Barré Syndrome: Progress Toward Fulfilling “Koch’s Postulates” Reports recently published in the Annals of Neurology reflect the extent to which research on uncommon “variants” of the Guillain Barre syndrome has advanced1,2,3,4. Two syndromes are making parallel but complementary advances. The Fisher syndrome (FS) of ataxia, ophthalmoparesis, and areflexia was first described as a variant of GBS by Miller Fisher 45 years ago5. The history of the axonal forms dates to the report by Feasby and colleagues 15 years ago, and the motor form of axonal GBS was described and given the designation acute motor axonal neuropathy (AMAN) only seven years ago6,7. Partly because they are uncommon in the U.S. and Europe, both the Fisher syndrome and AMAN were initially controversial. The AMAN story has been clarified largely by studies from Asia, especially Japan and northern China, where the syndrome is relatively frequent7-13. In spite of their youth, understanding of the pathogenesis of these variants now outstrips the most frequent form of GBS in the U.S., acute inflammatory demyelinating polyneuropathy (AIDP). The unifying hypothesis of pathogenesis being tested in the variant syndromes is molecular mimicry, whereby an antecedent infection generates an immune response to antigenic targets in the organism that are shared by normal tissue components. In this formulation the immune system “sees” and attacks the shared antigenic moieties innocently expressed by an element of the nerve fiber, either axon or Schwann cell. The most frequently recognized antecedent infection in AMAN and FS is Campylobacter jejuni2,11,14-16. Both syndromes are strongly associated with specific anti-ganglioside antibodies. In AMAN these antibodies are against GM1, GD1a, GalNAc-GD1a, and GM1b2,11,12,17-19, while in FS the association is with GQ1b20,21. Lipopolysaccharides of C. jejuni isolates from patients with AMAN and FS carry relevant ganglioside-like moieties16,22-24. Pathological studies in AMAN indicate a non-inflammatory, antibodymediated, complement-dependent mode of injury to motor axons10. In this context the critical missing link– the data needed to satisfy Witebsky’s postulates (the equivalent of Koch’s postulates for an autoimmune pathogenetic sequence) – is induction of clinical and pathological disease by passive transfer of anti- ganglioside antibodies produced by infection with relevant C. jejuni isolates. This goal has been partly reached in a series of studies on FS, using in vitro muscle-nerve preparations as a model for motor nerve terminal transmission1,3,25. Plomp et al. have shown that anti-GQ1b antibody containing sera, IgG fractions, and a human derived monoclonal anti-GQ1b exert a complement-dependent temporary increase in spontaneous neurotransmitter release, followed by block of evoked release resulting in paralysis1. This effect has been reproduced with a set of monoclonal antibodies cross-reactive with GQ1b made by immunizing with the LPS from C. jejuni isolates from GBS and FS patients25. Buchwald and colleagues have produced complementary data, differing in the role of the complement3. Moreover, the observation that the ganglioside GQ1b is enriched in cranial occulomotor nerves provides a satisfying explanation for the association of anti-GQ1b antibodies and ophthalmoplegia seen in FS26. In this issue of Annals, Yuki et al. report a new animal model of axonal injury induced by immunization with a mixture of bovine brain gangliosides or purified GM14. The Yuki group has played a key role in developing the concept of molecular mimicry in GBS, and the present article moves closer the goal of developing a robust model of AMAN. The notable observations include the fact that the antibody responses to ganglioside GM1 were closely associated with axonal and motor injury, that all animals with clinical and pathological disease showed antiganglioside antibody class switching from IgM to IgG, and that, as in AMAN, inflammation was absent in the injured nerves. An important issue regarding this model that needs further study relates to the presence of a significant number of degenerating fibers in the adjuvant immunized control animals. Putting these studies into perspective requires some background information on anti-GM1 antibody associated clinical syndromes, fine specificities of these antibodies, and distribution of GM1 ganglioside in peripheral nerves. Antibodies to GM1 were among the first to be described in patients with GBS and they are usually associated with predominantly motor neuropathies. Anti-GM1 antibodies have attracted controversy because they can be associated with several neuropathic syndromes. IgM anti-GM1 antibodies are present in a significant proportion of patients with multifocal motor neuropathy, a disorder which appears to involve initial demyelination of motor fibers27,28. Anti-GM1 antibodies, particularly of the IgG class, are present in some patients with AMAN, but also some AIDP cases. The specificity of anti-GM1 antibodies also varies among different patients; some antibodies specifically recognize GM1, others recognize the Gal(b1-3)GalNAc moiety shared by GM1, GD1b, and some peripheral nerve glycoproteins29,30. Although anti-GM1 antibodies are associated with predominant motor syndromes, this ganglioside is present in motor and sensory neurons, axons, and associated myelin31-33. With the availability of an animal model with selective motor nerve fiber injury we are closer to understanding several fundamental questions regarding molecular mimicry, anti-ganglioside antibodies, and the pathogenesis of different variants of GBS including AMAN. Some of these issues are also relevant to antibody-mediated neural damage in general: 1) What are the mechanisms underlying generation of IgG antiganglioside antibodies after enteric infection, and why is this response triggered in only a very small proportion of patients after an inciting infection? 2) Why are certain fibers preferentially damaged, for example motor fibers in AMAN or oculomotor fibers in FS? 3) Since gangliosides such as GM1 are expressed by both axon and myelin in myelinated fibers, why are axons or myelin primarily injured in AMAN or AIDP? 4) Are gangliosides the target antigens, or do anti-ganglioside antibodies crossreact with other glycoconjugates? 5) What are the mechanisms of anti-ganglioside antibody mediated nerve fiber injury? Does this nerve fiber injury require complement? 6) What determines the PNS susceptibility? The CNS is also enriched in gangliosides but generally is spared in conditions associated with anti-ganglioside antibodies (a possible exception is the association of anti-GQ1b antibody with Bickerstaff’s encephalitis). 7) Why and how do anti-ganglioside antibodies selectively cross the blood-nerve barrier and access the endoneurial space? Do inflammatory T cells contribute to this selective breakdown of the blood tissue barrier? While these and other issues remain, there is an irony in the fact that 30 years ago it appeared that AIDP was going to be one of the first autoimmune diseases to be fully understood in terms of immunopathogenesis. Yet now, while the pathogenesis of AIDP remains uncertain, we are well along in understanding and modeling the mechanisms leading to two “minor” variants of the Guillain Barre syndrome. Kazim A. Sheikh1, John W. Griffin1-3 Departments of Neurology1, Neuroscience2, and Pathology3 Johns Hopkins Hospital Baltimore, Maryland References 1. Plomp JJ, Molenaar PC, O’Hanlon GM, et al. Miller Fisher anti-GQ1b antibodies: alpha-latrotoxin-like effects on motor end plates [erratum: Ann Neurol 1999 Jun;45(6):823]. Ann Neurol 1999;45:189 –199. 2. Ho TW, Willison HJ, Nachamkin I, et al. Anti-GD1a anti- Editorial: Sheikh and Griffin: GBS Variants 695 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 696 body is associated with axonal but not demyelinating forms of Guillain-Barre syndrome. Ann Neurol 1999;45:168 –173. Buchwald B, Toyka KV, Zielasek J, et al. Neuromuscular blockade by IgG antibodies from patients with Guillain-Barre syndrome: a macro-patch-clamp study. Ann Neurol 1998;44: 913–922. Yuki N, Yamada M, Koga M, et al. Animal model of axonal Guillain-Barre syndrome induced by sensitization with GM1 ganglioside. Ann Neurol 2001;49:712–720. Fisher M. An unusual variant of acute idiopathic polyneuritis (syndrome of ophthalmoplegia ataxia and areflexia). N Engl J Med 1956;255:57– 65. Feasby TE, Gilbert JJ, Brown WF, et al. An acute axonal form of Guillain-Barre polyneuropathy. Brain 1986;109:1115–1126. McKhann GM, Cornblath DR, Griffin JW, et al. Acute motor axonal neuropathy: A frequent cause of acute flaccid paralysis in China. Ann Neurol 1993;33:333–342. McKhann GM, Cornblath DR, Ho TW, Li CY, Bai AY, Wu HS, et al. Clinical and electrophysiological aspects of acute paralytic disease of children and young adults in northern China. Lancet 1991;338:593–597. Griffin JW, Li CY, Ho TW, et al. Guillain-Barre syndrome in northern China: The spectrum of neuropathologic changes in clinically defined cases. Brain 1995;118:577–595. Hafer-Macko C, Hsieh S-T, Li CY, Ho TW, Sheikh K, Cornblath DR, et al. Acute motor axonal neuropathy: an antibodymediated attack on axolemma. Ann Neurol 1996; 40:635– 644. Yuki N, Yoshino H, Sato S, Miyatake T. Acute axonal polyneuropathy associated with anti-GM1 antibodies following Campylobacter enteritis. Neurology 1990;40:1900 –1902. Kuwabara S, Yuki N, Koga M, et al. IgG anti-GM1 antibody is associated with reversible conduction failure and axonal degeneration in Guillain-Barre syndrome. Ann Neurol 1998;44:202– 208. Hao Q, Saida T, Yoshino H, et al. Anti-GalNAc-GD1a antibody-associated Guillain-Barre syndrome with a predominantly distal weakness without cranial nerve impairment and sensory disturbance. Ann Neurol 1999;45:758 –768. Rees JH, Soudain SE, Gregson NA, Hughes RA. C. jejuni infection and Guillain-Barre syndrome. N Engl J Med 1995;333: 1374 –1379. Oomes PG, Jacobs BC, Hazenberg MPH, et al. Anti-GM1 IgG antibodies and C. jejuni bacteria in Guillain-Barre syndrome: evidence of molecular mimicry. Ann Neurol 1995;38:170 –175. Yuki N, Taki T, Takahashi M, et al. Molecular mimicry between GQ1b ganglioside and lipopolysaccharides of C. jejuni isolated from patients with Fisher’s syndrome. Ann Neurol 1994;36:791–793. Kusunoki S, Chiba A, Kon K, et al. N-acetylgalactosaminyl GD1a is a target molecule for serum antibody in Guillain-Barre syndrome. Ann Neurol 1994;35:570 –576. Ang CW, Yuki N, Jacobs BC, et al. Rapidly progressive, predominantly motor Guillain-Barre syndrome with anti-GalNAcGD1a antibodies. Neurology 1999;53:2122–2127. Annals of Neurology Vol 49 No 6 June 2001 19. Yuki N, Ho TW, Tagawa Y, et al. Autoantibodies to GM1b and GalNAc-GD1a: relationship to C. jejuni infection and acute motor axonal neuropathy in China. J Neurol Sci 1999; 164:134 –138. 20. 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;31:677– 679. 21. Willison HJ, Veitch J, Patterson G, Kennedy PGE. Miller Fisher syndrome is associated with serum antibodies to GQ1b ganglioside. J Neurol Neurosurg Psychiatry 1993;56:204 –206. 22. Yuki N, Handa S, Taki T, et al. Cross-reactive antigen between nervous tissue and a bacterium elicits Guillain-Barre syndrome: Molecular mimicry between gangliocide GM1 and lipopolysaccharide from Penner’s serotype 19 of C. jejuni. Biomed Res 1992;13:451– 453. 23. Sheikh KA, Nachamkin I, Ho TW, et al. C. jejuni lipopolysaccharides in Guillain-Barre syndrome: molecular mimicry and host susceptibility. Neurology 1998;51:371–378. 24. Nachamkin I, Ung H, Moran AP, et al. Ganglioside GM1 mimicry in Campylobacter strains from sporadic infections in the United States [erratum: J Infect Dis 1999 Jun;179(6): 1593]. J Infect Dis 1999;179:1183–1189. 25. Goodyear CS, O’Hanlon GM, Plomp JJ, et al. Monoclonal antibodies raised against Guillain-Barre syndrome-associated C. jejuni lipopolysaccharides react with neuronal gangliosides and paralyze muscle-nerve preparations [erratum: J Clin Invest 1999 Dec;104(12):1771]. J Clin Invest 1999;104:697–708. 26. Chiba A, Kusunoki S, Obata H, et al. Ganglioside composition of the human cranial nerves, with special reference to pathophysiology of Miller Fisher syndrome. Brain Res 1997;745:32– 36. 27. Baba H, Daune GC, Ilyas AA, et al. Anti-GM1 ganglioside antibodies with differing fine specificities in patients with multifocal motor neuropathy. J Neuroimmunol 1989;25:143–150. 28. Pestronk A, Choski R. Multifocal motor neuropathy. Serum IgM anti-GM1 ganglioside antibodies in most patients detected using covalent linkage of GM1 to ELISA plates. Neurology 1997;49:1289 –1292. 29. Gregson NA, Koblar S, Hughes RAC. Antibodies to gangliosides in Guillain-Barre syndrome: Specificity and relationship to clinical features. Quart J Med 1993;86:111–117. 30. Apostolski S, Sadiq SA, Hays A, et al. Identification of Gal(beta1–3)GalNAc bearing glycoproteins at the nodes of Ranvier in peripheral nerve. J Neurosci Res 1994;38:134 –141. 31. O’Hanlon GM, Paterson GJ, Wilson G, et al. Anti-GM1 ganglioside antibodies cloned from autoimmune neuropathy patients show diverse binding patterns in the rodent nervous system. J Neuropath Exp Neurol 1996;55:184 –195. 32. Sheikh KA, Deerinck TJ, Ellisman MH, Griffin JW. The distribution of ganglioside-like moieties in peripheral nerves. Brain 1999;122:449 – 460. 33. Corbo M, Quattrini A, Latov N, Hays AP. Localization of GM1 and Gal(beta1–3)GalNAc antigenic determinants in peripheral nerve. Neurology 1993;43:809 – 814.