J7ournal of Neurology, Neurosurgery, and Psychiatry 1994;57: 1303-1307

_7ourl of

NEUROLOGYNEUROSURGERY& PSYCHIATRY

Editorial

Antiglycolipid antibodies in peripheral neuropathy: fact or fiction?

The past decade has seen the emergence of a large fieldof research investigating the concurrence of antiglycolipidantibodies and peripheral neuropathy (see reviews'-4).Initial scepticism about the significance of antigangliosideantibodies is being replaced by a widespread realisationthat these antibodies directly contribute to the pathogen-esis of neuropathy. Two incompletely resolved issueswhich cast doubt on this relation are that raised concen-

trations of antiglycolipid antibodies are very widespreadin normal and disease controls and antiglycolipid anti-bodies may be either present or absent in clinically indis-tinguishable neuropathy syndromes. By contrast, some

syndromes such as Miller Fisher syndrome, are very

tightly linked to a specific antiglycolipid antibody inwhich conclusive passive transfer studies have been per-formed. Refining the clinicoserological associations to aidin diagnosis and to monitor disease progress throughantiglycolipid antibody measurements is becoming more

relevant to patient management. From the research pointof view, considerable efforts are being made to identifynew antibody specificities and to elucidate the precisemechanisms of action of antiglycolipid antibodies. Thepurpose of this editorial is to put a large body of confus-ing and often conflicting data into a balanced clinical per-

spective with particular emphasis on currently recognisedclinical syndromes and guidance about requesting andinterpreting antiglycolipid antibody assays.

Glycolipid terminology, structure, and functionAn early hurdle to cross concerns the complexities of gly-colipid and ganglioside terminology. Glycosphingolipidsare composed of the long chain aliphatic amine sphingo-sine (acylated ceramide) attached to one or more sugars

(hexoses). The ceramide is immersed in the membranelipid bilayer with the carbohydrate structure exposedextracellularly. A single hexose linked to ceramide(monohexosyl-ceramide) is termed a cerebroside, themost abundant of which in humans is galactocerebroside(galactosyl-ceramide). Sulphatide is galactocerebrosidesulphated in the carbon-3 position and is a major compo-

nent of peripheral nerve myelin and an autoantigen inpredominantly sensory neuropathies.5-9

Gangliosides are complex glycosphingolipids that, bydefinition, must contain at least one sialic acid residue.Sialic acid is a generic term for N-acylneuraminic acid,the acyl group generally being acetyl in the human ner-

vous system (as opposed to glycolyl), hence N-acetylneu-raminic acid, commonly abbreviated to NeuNAc or

NANA. The sialic acid(s) are attached to the internal orthe terminal galactose of an oligosaccharide core com-posed of up to four sugars with the following sequence:ceramide-glucose-galactose-N-acetylgalactosamine-galac-tose. Traditionally, gangliosides are named according toSvennerholm'0 II with the following formula: G refers toganglio; M, D, T, and Q refer to the number of sialicacid residues (mono, di, tri, and quad respectively);arabic numerals and lower case letters refer to thesequence of migration as determined by thin layer chro-matography. Although dependent on the solvent systemused, bulky gangliosides with a longer oligosaccharidecore and more sialic acid will migrate more slowly thanthe smaller gangliosides. For example, for three of themonosialogangliosides, GM1 (having a four sugaroligosaccharide core, Cer-Glc-Gal-GalNAc-Gal) migratesmore slowly than GM2 (a three sugar oligosaccharide,Cer-Glc-Gal-GalNAc), which migrates more slowly thanGM3 (a two sugar oligosaccharide, Cer-Glc-Gal).Likewise, GDlb runs ahead of GTlb, which runs aheadof GQlb, as they contain two, three, and four sialic acidsrespectively. The figure shows the structure of GDlb.There are four major gangliosides in the brain (GM1,GDla, GDlb, GTlb) and many minor gangliosides inbrain and peripheral nerve tissues, which are develop-mentally regulated and spatially segregated in a widerange of patterns across different species. Many differentgangliosides can be autoantigens in peripheral neuropa-thy, some of which have not been identified.'

Paragloboside is a neutral glycolipid (Cer-Glu-Gal-GlcNAc-Gal), which, when sialylated on the terminalgalactose (sialosylparagloboside, SGP), is a major periph-eral nerve glycolipid, also known as LMl.*2 Substitutionof the terminal sialic acid for sulphated glucuronic acidforms sulphated glucuronyl paragloboside (SGPG).'3 14Both LMI and SGPG are also important peripheralnerve autoantigens in neuropathy.8 13

All antiglycolipid antibodies associated with neuropa-thy react with epitopes on the carbohydrate region of theglycolipid molecules. Because these carbohydrate struc-tures are often present on several different glycolipids,glycoproteins, and other carbohydrate rich moleculessuch as bacterial lipopolysaccharides, there is consider-able potential for shared reactivity. For example, (a)autoantibodies that react with the sulphated glucuronicacid epitope on SGPG also react with similar epitopes onmany other neural glycoproteins including the myelinassociated glycoprotein (MAG) 13-'5 and the peripheral

1 303

on 26 June 2018 by guest. Protected by copyright.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.57.11.1303 on 1 Novem

ber 1994. Dow

nloaded from

Editorial

Thin layer chromatogram of 1 yg each of asialo-GM1, GM1, GD3,GDlb, GTlb, and GQlb, immunostained with two monoclonal IgMantiganglioside antibodies derivedfrom patients with neuropathy andshowing characteristic patterns of reactivity (see Willison et al 52 formethod). Lane A shows an antibody derivedfrom a patient with achronic ataxic neuropathy and IgM paraproteinaemia reactive withdisialosyl epitopes common to GD3, GDlb, GTlb, and GQlb.52 Lane Bshows an antibody derivedfrom an anti-GM1 positive patient reactivewith the Gal(/31-3)GalNAc epitope common to GM1, GDlb, andasialo-GMI.

nerve compact myelin protein Po06, (b) theGal(,B1-3)GalNAc epitope found on GM1 ganglioside isalso present on asialo-GM1 and GDlb (see figure)17-20and the complete GM1 carbohydrate sequence is presenton Campylobacter jejuni lipopolysaccharide (LPS),2'Campylobacter enteritis being a common prodromal infec-tion in patients with Guillain-Barre syndrome,22-25 a pro-portion of whom have anti-GM1 antibodies.26 27

Glycolipids, including gangliosides, regulate manydiverse physiological processes,28 29 including extensivemodulation of neural cell function.30 They are asymmet-rically located in the plasma membrane, anchored in thebilayer by their ceramide tail, exposing their sialylatedoligosaccharide core extracellularly, a site readily accessi-ble to antibody binding. They are highly enriched inneural membranes, particularly synaptic regions. Theyinteract with growth factor receptors, ion channels, andcell recognition/adhesion molecules and have widespreadeffects on signal transduction systems. At the cellularlevel, they influence such diverse events as cell survival,differentiation, synaptogenesis and synaptic transmission,and neurite outgrowth. In vivo effects include acceleratedrestoration of function in injured tissues and neuro-

protection-for example, against glutamate inducedneuronal injury. Considerable research is currentlyfocused on increasing our understanding of gangliosidefunction.

Antiganglioside antibody assays: methodology andinterpretationAssaying serum samples for antiganglioside and antigly-colipid antibodies is a process with many pitfalls. There isno standardised assay method and the literature on thissubject is crowded with different protocols31 that claimsuperiority. Two multicentre studies with coded sampleshave shown good agreement on clearly positive or nega-tive cases but discrepancies appear with borderline sam-ples, which may nevertheless be of important clinicalrelevance.3233 Most centres use a combination of enzymelinked immunosorbent assay (ELISA) with thin layerchromatography overlay as a confirmatory test. Criticalfactors that influence assay results include (a) the choiceof ELISA plates, taking into account manufacture relatedbatch to batch variations and subsequent storage condi-tions; (b) the temperature at which the assay is per-formed; (c) the duration of serum incubation; (d)washing and blocking buffer composition, particularly thepresence or absence of detergent and the choice of block-ing protein; (e) sample layout on ELISA plates, with par-ticular attention to plate stacking artefacts and edgeeffects. These factors need to be built into stringent qual-ity control practice in all laboratories performing assays,particularly when results influence decisions about diag-nosis and clinical management.

Assay results are usually reported as titres calculatedby end point analysis (the lowest serum dilution that stillgives a raised optical density reading by ELISA). Wheninterpreting an assay result, consideration must be givento the methodology used and the normal range estab-lished for that laboratory. Because antiganglioside anti-bodies form part of the normal antibody/autoantibodyrepertoire,3435 they are commonplace in normal and dis-ease control serum samples.3"0 For example, moderatelyhigh IgM and IgG titres (> 1/103 < 1/104) against asialo-GM1 and sulphatide are typically found in normal serumsamples and are thus not likely to be relevant at this level;by contrast, an anti-GMI IgM titre of 1/103 is likely to berelevant as this is above the upper limit of the normalrange (around 1/500) for this antigen in most laborato-ries.

Antiglycolipid antibody markers for clinicalsyndromesThe table lists the main clinical syndromes with a welldefined antiglycolipid antibody specificity. In some situa-tions, the antiglycolipid antibody assay result will corrob-

Clinical syndromes associated with specific antiglycolipidautoantibodies

Autoantibody specificity andClinical syndrome isotype References

Chronic sensorimnotor Monoclonal IgM antibodiesdemyelinating to SGPG, SGLPG, 1-3, 13-16,neuropathy MAG, and Po 41-46

Chronic axonal sensory IgM antibodies to 5-9neuropathy sulphatide

Chronic large fibre Monoclonal IgM antibodies 47-53sensory neuropathy to gangliosides containingwith ataxia NeuNAc(a2-8)NeuNAc

epitopesMultifocal motor IgM antibodies to GM1 20, 37-40, 58

neuropathy and related Gal(fll-3)GalNAc bearing glycolipids

Miller Fisher syndrome IgG antibodies to GQlb 54-56, 70, 71and GTla

Guillain-Barre syndrome IgG, IgM and IgA antibodies 4, 5, 9, 24, 26,to GMI, GDla, LM1, 37-39, 40,sulphatide, and other 90-100glycolipids

1304

.: ......

on 26 June 2018 by guest. Protected by copyright.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.57.11.1303 on 1 Novem

ber 1994. Dow

nloaded from

Editorial

orate a previously suspected diagnosis, such as multifocalmotor neuropathy, or indeed alert the neurophysiologistto search carefully for conduction block in an apparentcase of lower motor neuron disease in whom multifocalmotor neuropathy was not suspected clinically.Alternatively, a positive result will further refine anexisting diagnosis, such as paraproteinaemic neuropathywith anti-MAG specificity or will help to distinguish animportant differential diagnosis, such as Miller Fishersyndrome, botulism, and brain stem demyelinatingdisease.The first clinical syndrome in which antiglycolipid

antibodies were identified was the IgM paraproteinaemicneuropathy associated with anti-MAG antibodies.4142 Aswell as reacting with MAG and other glycoproteins pre-sent in peripheral nerve, it was established that the anti-MAG paraproteins also reacted with the previouslyunknown acidic glycolipid, SGPG, and its related homo-logue, sulphated glucuronyl lactosaminyl paragloboside(SGLPG).13-15 The clinical syndrome is of a relativelybenign, late onset, chronic sensorimotor demyelinatingneuropathy, often associated with tremor.4344 A patholog-ical feature of note in nerve biopsies is the presence ofwidely spaced myelin.4446 The anti-MAG specificityaccounts for 50% of patients with IgM paraproteinaemicneuropathy; most of the remainder react with otherglycolipids or gangliosides.1 Some anti-MAG parapro-teins cross react with sulphatide67 and in addition, sul-phatide can be a specific and sole antigen in patients withIgM paraproteinaemic, predominantly sensory, neuro-pathy.5-7 It is important to recognise, when evaluatingcases of late onset demyelinating neuropathy, that lowlevel paraproteins may be easily missed by standard elec-tropheretic techniques and that sensitive screeningmethods should be employed.

Another increasingly well defined paraproteinaemicneuropathy syndrome is a chronic large fibre sensoryneuropathy with prominent ataxia, first described in asingle case report almost a decade ago47 and subsequentlyreported by several groups.48-5' In these cases the IgMparaprotein reacts with gangliosides bearingNeuNAc(a2-8)NeuNAc linked disialosyl groups, includ-ing GD3, GDlb, GTlb, and GQlb (figure). There issome variation in the fine specificity of the antibodybetween cases although the clinical pattern is fairly uni-form. Some of these patients also have cold agglutinindisease by virtue of the presence of sialylated glycopro-tein epitopes on human red blood cells. In addition,some cases have been reported with intermittent opthal-moplegia,47 thus producing a clinical syndrome reminis-cent of Miller Fisher syndrome in which antibodies todisialylated gangliosides are also found.5456

Interest in multifocal motor neuropathy with demyeli-nating conduction block57 has been intense since it wasrecognised that patients with this syndrome may haveantibodies to GM1 gangloiside.58 Early reports indicating asignificant association between anti-GM1 antibodies andmotor neuron disease/amyotrophic lateral sclerosis59 60were not confirmed by others3738406'-63 although there doseem to be very rare, unusual cases in which thisoccurs.17-19 We screen all patients with predominantlylower motor neuron syndromes for anti-GMI antibodies.Many patients with multifocal motor neuropathy andanti-GM1 antibodies have a fairly stereotyped clinicalpicture comprising a chronic asymmetric motor syn-drome, usually with distal onset in an upper limb, as pre-viously described.20 64-66 We have not found furthersubdivisions on the basis of antiganglioside serology use-ful, although they have been reported.67 These patientsrespond well, albeit temporarily, to intravenous

immunoglobulin68 which is our routine treatment andcan be regularly repeated. Our own experience withcyclophosphamide has been disappointing (lack of suffi-cient efficacy to warrant the high risks of toxicity)although others report success.69

Antiganglioside antibodies are also found in acuteperipheral neuropathies where they appear transientlyand are predominantly of the IgG class. The most clearlydefined association is between anti-GQlb gangliosideantibodies and Miller Fisher syndrome.54-56 The anti-GQlb antibodies cross react with GTla in the series ofcases reported by Chiba et a170 and in all our cases to date(14/14) and with GDlb, GD3, or both in about half ofour cases. Anti-GQlb reactivity seems to specificallymark the presence of ophthalmoplegia: we have recentlyidentified a case of acute ataxic, areflexic neuropathywithout opthalmoplegia whose serum samples reactedonly with GDlb and GD3, thereby lending support tothis view. A recent report indicates that anti-GQlb anti-bodies are also present in Bickerstaffs brain stemencephalitis, suggesting that there may be a commonaetiological thread to both these conditions.71

In Guillain-Barre syndrome, a wide variety of anti-ganglioside antibodies have been reported in up to 50%of cases in different series (table); although no unifyingpatterns have yet emerged, they may do so in the future.Considerable debate surrounds the significance of anti-GM1 antibodies in Guillain-Barre syndrome. In somestudies, anti-GM1 antibodies mark a particularly severeform of the illness with prominent motor axonal involve-ment and poor recovery,262772 although this association isrefuted by others.7374 This issue may be resolvable by fur-ther subclassification of anti-GM1 antibodies by finespecificity. In this regard, our recent studies on a panel ofcloned human anti-GM1 antibodies75 have shown con-siderable heterogeneity in their ability to bind antigenunder different conditions and in their patterns of tissuereactivity. This may account for the apparent diversity ofclinical expression of anti-GM1 antibody associatedneuropathies and neuronopathies.

The pathogenic relevance of antiglycolipidantibodiesDespite considerable anomalies and complexities, goodevidence is beginning to emerge from a variety of experi-mental models to support a direct role for antigangliosideantibodies in causing neuropathy. Clearly, other factorsincluding cellular immune mechanisms may play a cen-tral part in pathogenesis. The subject is complicated bymany factors common to developing an animal model fordisease including qualitative and quantitative variationsin glycolipid antigen composition in nerves from differentspecies and the lack of well defined human antibodies forpathogenesis studies. The generation of disease associ-ated human monoclonal antibodies should help toresolve this issue.75

Antibodies to SGPG, as found in the "anti-MAG"IgM paraproteinaemic neuropathy have been shown tocause demyelination and other pathological changesincluding widening of myelin lamellae, when injectedlocally into rodent nerve7677 and after systemic passivetransfer to the chicken.78 An intriguing case of anti-MAGIgM paraproteinaemic neuropathy occurring in conjunc-tion with hereditary motor and sensory neuropathy hasbeen described,79 raising the possibility that pre-existingperipheral nerve abnormalities may predispose to thedevelopment of antinerve antibodies, a fact long recog-nised in experimental models of nerve injury.80

Experimental studies on anti-GM1 antibodies have

1305

on 26 June 2018 by guest. Protected by copyright.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.57.11.1303 on 1 Novem

ber 1994. Dow

nloaded from

Editorial

shown a variety of neuropathic effects including demyeli-nation and conduction block81-84 although these studiesare clearly incomplete.8586 Antibodies to GMl/GDlb87and gangliosides bearing disialosyl groups as found inchronic sensory neuropathies,88 are toxic to culturedmotor neurons and dorsal root ganglion neurons respec-tively. It has also recently been shown that Miller Fishersyndrome serum containing anti-GQlb antibody blocksneurotransmitter release in the mouse hemidiaphragmpreparation, which we postulate accounts for the motormanifestations of the syndrome.89 Many groups areactively researching this field, which will undoubtedlyreveal the pathogenic relation between antiglycolipidantibodies and neuropathy in more detail.

ConclusionAntiglycolipid antibodies are here to stay as part of theclinical and experimental investigation of patients withperipheral neuropathy. The extent to which antiglycol-ipid antibody assays are requested, however, will dependon the investigative enthusiasm of individual neurologistsas there are few circumstances in which positive or nega-tive results play a critical part in patient diagnosis andmanagement. The field is still in its infancy and a consid-erable amount of further research is necessary to identifyputative antigens in seronegative cases, or to establishthat such cases are not antibody mediated. This informa-tion will eventually filter into clinical practice and willhopefully direct clinicians towards appropriate treatmentand monitoring of antibody mediated neuropathies.The financial support of the Scottish Motor Neurone

Disease Association and the Guillain-Barre SupportGroup of Great Britain and helpful discussions with DrsNorman Gregson and Donald Marcus are gratefullyacknowledged.

HUGHJ WILLISONUniversity of Glasgow Department of Neurology,

Institute of Neurological Sciences,Southern General Hospital,

Glasgow, G51 4TF, UK

1 Quarles RH, Ilyas AA, Willison HJ. Antibodies to glycolipids in demyeli-nating diseases of the human peripheral nervous system. Chemistry andPhysics of Lipids 1986;42:235-48.

2 Latov N. Antibodies to glycoconjugates in neurological disease. ClinicalAspects ofAutoimmunity 1990;4:18-29.

3 Thomas PK, Willison HJ. Paraproteinaemic neuropathies. In: McLeod J,ed. Inflammatory neuropathies, Balliere's clinical neurology series. London:Balliere Tindall, 1994.

4 Willison HJ, Kennedy PGE. Gangliosides and bacterial toxins in Guillain-Barre syndrome. JNeuroimmunol 1993;46:105-12.

5 van den Berg LH, Lankamp CLAM, de Jager AEJ, et al. Anti-sulphatideantibodies in peripheral neuropathy. J Neurol Neurosurg Psychiatry1994;56:1164-8.

6 Pestronk A, Li F, Griffin J, Feldman EL, et al. Polyneuropathy syndromesassociated with serum antibodies to sulfatide and myelin-associatedglycoprotein. Neurology 1991;41:357-62.

7 Ilyas AA, Cook SD, Dalakas MC, Mithen FA. Anti-MAGIgM parapro-teins from some patients with polyneuropathy associated with IgMparaproteinaemia also react with sulphatide. J Neuroimmunol 1992;37:85-92.

8 Fredman P, Vedeler CA, Nyland H, Aarli JA, Svennerholm L. Antibodiesin sera from patients with inflammatory demyelinating polyradiculo-neuropathy react with ganglioside LM1 and sulphatide of peripheralnerve myelin. JNeurol 199 1;238:75-9.

9 Ilyas AA, Mithen FA, Dalakas MC, et al. Antibodies to sulphated glyco-lipids in Guillain-Barre syndrome. J Neurol Sci 1991;105:108-17.

10 Svennerholm L. Chromatographic separation of gangliosides. J Neurochem1963;1O:613-23.

11 IUPAC-IUB Commission on Biochemical Nomenclature (CBN). Thenomenclature of lipids. EurJ Biochem 1977;79: 11-21.

12 Li Y-T, Mansson J-E, Vanier M-T, Svennerholm L. Structure of themajor glucose-amine containing ganglioside of human tissues. J BiolChem 1973;248:2634-6.

13 Ilyas AA, Quarles RH, MacIntosh TD, Dobersen MJ, Dalakas MC,Brady RO.IgM in a human neuropathy related to paraproteinaemiabinds to a carbohydrate determinant in the myelin-associated glycopro-tein and to a ganglioside.Proc NatlAcad Sci USA 1984;81:1225-9.

14 Chou DKH, Ilyas AA, Evans JE, Costello C, Quarles RH, Jungalwala,FB. Structure of sulfated glucuronyl glycolipids in the nervous systemreacting with HNK-1 antibody and someIgM paraproteins in neuro-pathy. JBiol Chem 1986;261:11717-25.

15 Ariga T, Kohriyama T, Freddo L, et al. Characterization of sulfatedglucuronic acid containing glycolipids reacting with IgM paraproteins inpatients with neuropathy. J Biol Chem 1987;262:848-53.

16 Bollensen E, Steck AJ, Schachner M. Reactivity with the myelin glyco-protein P, in serum from patients with monoclonal IgM gammopathyand polyneuropathy. Neurology 1988;38:1266-70.

17 Freddo L, Yu RK, Latov N, et al. Gangliosides GM1 and GD lb are anti-gens for M-proteins in a patient with motor neuron disease. Neurology1986;36:454-8.

18 Latov N, Hays AP, Donofrio PD, et al. Monoclonal IgM with uniquereactivity to gangliosides GM1 and GDlb and to lacto-N-tetraose intwo patients with motor neuron disease. Neurology 1988;38:763-8.

19 Nardelli E, Steck AJ, Barkas T, et al. Motor neuron disease and mono-clonal IgM with antibody activity against GM1 and GDlb gangliosides.Ann Neurol 1988;23:524-8.

20 Pestronk A. Invited review: motor neuropathies, motor neuron disordersand anti-glycolipid antibodies. Muscle Nerve 1991;14:927-36.

21 Yuki N, Taki T, Inagaki F, et al. A bacterium lipopolysaccharide thatelicits Guillain-Barre syndrome has a GM1 ganglioside-like structure.J Exp Med 1993;178:1771-5.

22 Kaldor J, Speed BR. GBS and Campylobacter jejuni: a serological study.BMJ 1984;288: 1867-70.

23 Winer JB, Hughes RAC, Anderson MJ, Jones DM, Kangro H, WatkinsRFP. A prospective study of acute idiopathic neuropathy. II Antecedentevents. 7 Neurol Neurosurg Psychiatry 1988;51:613-8.

24 Koblar SA, Gregson NA, Hughes RAC, Doherty P, Walsh FS.Camplyobacter neuropathy. Neurology 199 1;41:1327-8.

25 Griffin JW, Ho TW-H. The Guillain-Barre syndrome at 75: TheCampylobacter connection. Ann Neurol 1993;34: 125-7.

26 Yuki N, Yoshino H, Sato S, Miyatake T. Acute axonal polyneuropathyassociated with anti-GM1 antibodies following Campylobacter enteritis.Neurology 1990;40: 1900-2.

27 Gregson NA, Koblar S, Hughes RAC. Antibodies to gangliosides inGuillain-Barre syndrome: specificity and relationship to clinical fea-tures. QJ7Med 1993;86:111-7.

28 Merrill AH, Hannun YA, Bell RM. Introduction: sphingolipids and theirmetabolites in cell regulation. Adv Lipid Res 1993;25:1-21.

29 Hakamori S. Bifunctional role of glycosphingolipids. J7 Biol Chem1990;265:18713-6.

30 Tettemanti G, Riboni L. Gangliosides and modulation of function ofneural cells. Adv Lipid Res 1993;25:235-67.

31 Kornberg AJ, Pestronk A. The clinical and diagnostic role of anti-GM1antibody testing. Muscle Nerve 1994;17: 100-4.

32 Marcus DM, Latov N, Hsi BP, et al. Measurement and significance ofantibodies against GM1 ganglioside. J Neuroimmunol 1989;25:255-9.

33 Zielasek J, Ritter G, Magi S, Hartung HP, Toyka KV, and participatinglaboratories. A comparative trial of anti-glycoconjugate antibody assays:IgM antibodies to GM1. J Neurol 1994;241:475-80.

34 Lee K-W, Inghrami G, Spatz LA, Knowles DM, Latov N. The B-cellsthat express anti-MAG antibodies in neuropathy and non-malignantmonoclonal gammopathy belong to the CD5 subpopulation.J Neuroimmunol 1991;31:83-8.

35 Heidenreich F, Leifeld L, Jovin T. T-cell dependent activity of ganglio-side GM1-specific B cells in Guillain-Barre syndrome and multifocalmotor neuropathy in vitro. J Neuroimmunol 1994;49:97-108.

36 Endo T, Scott DD, Stewart SS, Kundu SK, Marcus DM. Antibodiesto glycosphingolipids in patients with multiple sclerosis and SLE.J7Immunol 1984;132:1793-7.

37 Sadiq SA, Thomas FP, Kilidirias K, et al. The spectrum of neurologicaldisease associated with anti-GM1 antibodies. Neurology 1990;40:1067-72.

38 Adams D, Kuntzer T, Burger D, et al. Predictive value of anti-GM1 anti-bodies in neuromuscular diseases: a study of 180 sera. J Neuroimmunol199 1;32:223-30.

39 Lamb NL, Patten BM. Clinical correlations of anti-GM1 antibodies inamyotrophic lateral sclerosis and neuropathies. Muscle Nerve 1991;14:1021-7.

40 Nobile-Orazio E, Carpo M, Legname G, Meucci N, Sonnino S, ScarlatoG. Anti-GM1 antibodies in motor neuron disease and neuropathy.Neurology 1990;40: 1747-50.

41 Braun PE, Frail DE, Latov N. Myelin-associated glycoprotein is the anti-gen for a monoclonal IgM in polyneuropathy. 7 Neurochem 1982;39:1261-5.

42 Steck AJ, Murray N, Meier C, Page N, Peruisseau G. Demyelinatingneuropathy and monoclonal IgM antibody to myelin-associated glyco-protein. Neurology 1983;33:19-23.

43 Smith IS, Kahn SN, Lacey BW, et al. Chronic demyelinating neuropathyassociated with benignIgM paraproteinaemia. Brain 1983;106:169-95.

44 Yeung KB, Thomas PK, King RMH, et al. The clinical spectrum ofperipheral neuropathies associated with benign monoclonal IgM, IgGand IgA paraproteins. JNeurol 1991;238:383-91.

45 Stefansson KM, Marton LS, AntelJP, et al. Neuropathy accompanyingIgM monoclonal gammopathy. Acta Neuropathol (Berl) 1983;59:255-61.

46 Steck AJ, Murray N. Monoclonal antibodies to myelin-associated glyco-protein reveal antigenic structures and suggest pathogenic mechanisms.Springer Semin Immunopathol 1985;8:29-43.

47 Ilyas AA, Quarles RH, Dalakas MC, Fishman PH, Brady RO.MonoclonalIgM in a patient with paraproteinaemic polyneuropathybinds to gangliosides containing disialosyl groups. Ann Neurol 1985;18:655-9.

48 Arai M, Yoshino H, Kusano Y, Yasaki Y, Ohnishi Y, Miyatake T. Ataxicpolyneuropathy and anti-Pr2 IgMk M proteinaemia.. Neurol 1992;239:147-51.

49 Duane GC, Farrer RG, Dalakas MC, Quarles RH. Sensory neuropathywith monoclonal immunoglobulin M to GDlb ganglioside. Ann Neurol1992;31:683-5.

50 Obi T, KusunokiS, Takatsu M, Mizoguchi K, Nishimura Y.IgM M-protein in a patient with sensory-dominant neuropathy binds preferen-tially to polysialogangliosides. Acta Neyrol Scand 1992;86:215-8.

51 Yuki N, Miyatani N, SatoS, et al. Acute relapsing sensory neuropathyassociated withIgM antibody against B-series gangliosides containing aGalNAcbl-4(Gal3-2aNeuAc8-2aNeuAc)b1 configuration. Neurology1 992;42:686-9.

52 Willison HJ, Paterson G, VeitchJ, Inglis G, Bamett SC. Peripheral neu-

1 306

on 26 June 2018 by guest. Protected by copyright.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.57.11.1303 on 1 Novem

ber 1994. Dow

nloaded from

Editorial

ropathy associated with anti-Pr2 cold agglutinins: biochemical andimmunocytochemical studies. Jf Neurol Neurosurg Psychiatry 1993;56:1178-84.

53 Herron B, Willison HJ, Veitch J, Roelcke D, Illis LS, Boulton FE.Monoclonal IgM cold agglutinins with anti-PrId specificity in a patientwith peripheral neuropathy. Vox Sang 1994;67:58-63.

54 Chiba A, Kusonoki S, Shimizu T, Kanazawa I. Serum IgG antibody toganglioside GQlb is a possible marker of Miller-Fisher syndrome. AnnNeurol 1992;31:677-9.

55 Willison HJ, Veitch J, Paterson G, Kennedy PGE. Miller-Fishersyndrome is associated with serum antibodies to GQlb ganglioside.Neurol Neurosurg Psychiatry 1993;56:204-6.

56 Yuki N, Sato S, Tsuji S, Ohsawa T, Miyatake T. Frequent presence ofanti-GQlb antibody in Fisher's syndrome. Neurology 1993;43:414-7.

57 Lewis RA, Sumner AJ, Brown MJ, Asbury AK. Multifocal demyelinatingneuropathy with persistent conduction block. Neurology 1982;32:958-64.

58 Pestronk A, Comblath DR, Ilyas AA, et al. A treatable multifocal neu-ropathy with antibodies to GM1 ganglioside. Ann Neurol 1988;24:73-8.

59 Pestronk A, Adams RN, Clawson L, et al. Serum antibodies to GM1ganglioside in amyotrophic lateral sclerosis. Neurology 1988;38:1457-61.

60 Pestronk A, Adams RN, Cornblath, et al. Patterns of serum antibodies toGM1 and GDla gangliosides in ALS. Ann Neurol 1989;25:98-102.

61 Shy ME, Evans VA, Dublin FD, et al. Antibodies to GM1 and GDlb inpatients with motor neuron disease without plasma cell dyscrasia. AnnNeurol 1989;25:511-3.

62 Salazas-Grueso EF, Routbort MJ, Martin J, et al. Polyclonal IgM anti-GM1 ganglioside antibody in patients with motor neuron disease andvariants. Ann Neurol 1990;27:558-63.

63 Willison HJ, Chancellor AM, Paterson G, Veitch J, Singh S, Whitelaw J,Kennedy PGE, Warlow CP. Antiglycolipid antibodies, immunoglobu-lins and paraproteins in motor neuron disease: a population based casecontrol study. Neurol Sci 1993;114:209-15.

64 Parry GJ, Clarke S. Multifocal acquired neuropathy masquerading asmotor neuron disease. Muscle Nerve 1988;11:107-16.

65 van den Bergh P, Logigian EL, Kelly JJ Jr. Motor neuropathy with multi-focal conduction blocks. Muscle Nerve 1989;11:26-31.

66 Krarup C, Stewart JD, Sumner AJ, et al. A syndrome of asymetrical limbweakness and motor conduction block. Neurology 1990;40:1 18-27.

67 Pestronk A, Chaudry V, Feldman EL, et al. Lower motor neuronsyndromes defined by patterns of weakness, nerve conduction abnor-malities, and high titres of anti-glycolipid antibodies. Ann Neurol1990;27:316-26.

68 Chaudry V, Corse AM, Comblath DR, et al. Multifocal motor neuropa-thy: response to human immune globulin. Ann Neurol 1993;33:237-42.

69 Feldman EL, Bromberg MB, Albers JW, Pestonk A. Immunosuppressivetreatment in multifocal motor neuropathy. Ann Neurol 1991;30:397-401.

70 Chiba A, Kusonoki S, Obata H, Machinami R, Kanazawa I. Serum anti-GQlb IgG antibody is associated with opthalmoplegia in Miller-Fishersyndrome and Guillain-Barre syndrome. Neurology 1993;43:1911-7.

71 Yuki N, Sato S, Hozumi I, Miyatake T. An immunological abnormalitycommon to Bickerstaffs brain stem encephalitis and Fisher's syndrome.Neurol Sci 1993;118:83-7.

72 Nobile-Orazio E, Carpo M, Meucci N, et al. Guillain-Barre syndromeassociated with high titers of anti-GMl antibodies. Neurol Sci 1992;109:200-6.

73 Vreisendorp FJ, Mishu B, Blaser MJ, Koski CL. Serum antibodies toGM1, GDlb, peripheral nerve myelin and Campylobacter jejuni inpatients with Guillain-Barre syndrome and controls: correlation andprognosis. Ann Neurol 1993;34:130-5.

74 Enders U, Karch H, Toyka KV, et al. The spectrum of immune responsesto Campylobacterjejuni and glycoconjugates in Guillain-Barre syndromeand other neuroimmunological disorders. Ann Neurol 1993;34: 136-44.

75 Willison HJ, Paterson G, Kennedy PGE, Veitch J. Cloning of human anti-GM1 antibodies from motor neuropathy patients. Ann Neurol 1994;35:471-8.

76 Hays AP, Latov N, Takatsu M, Sherman WH. Experimental demyelina-tion of nerve induced by serum of patients with neuropathy and an anti-MAG IgM M-Protein. Neurology 1987;37:242-56.

1307

77 Willison HJ, Trapp BD, Bacher J, Dalakas M, Griffin JW, Quarles RH.Demyelination induced by intraneural injection of human anti-myelin-associated glycoprotein antibodies. Muscle Nerve 1988;1 1: 1169-76.

78 Tatum AH. Experimental paraproteinaemic neuropathy, demyelinationby passive transfer of human IgM anti-myelin-associated glycoproteinantibody. Ann Neurol 1993;33:502-6.

79 Gregory R, Thomas PK, King RHM, Hallam PLJ, Malcolm S, HughesRAC, Harding AE. Coexistence of hereditary motor and sensory neu-ropathy type la and IgM paraproteinaemic neuropathy. Ann Neurol1993;33:649-52.

80 Schwartz M, Seal BA, Eshhar N. Antibodies to gangliosides and myelinautoantigens are produced in mice after sciatic nerve injury.JFNeurochem 1981;38:1192-5.

81 Schwerer B, Lassmann H, Kitz K, Bernheimer H. Ganglioside GM1, amolecufar target for immunological and toxic attacks: similarity of neu-ropathological lesions induced by ganglioside antiserum and choleratoxin. Acta Neuropathol (Berl) 1986;72:55-61.

82 Thomas FP, Trojaborg W, Nagy C, et al. Experimental autoimmuneneuropathy with anti-GM1 antibodies and immunoglobulin deposits atnodes of Ranvier. Acta Neuropathol 199 1;82:378-83.

83 Santoro M, Uncini A, Corbo M, et al. Experimental conduction blockinduced by serum from a patient with anti-GM1 antibodies. AnnNeurol 1992;31:385-90.

84 Arakasi K, Kusonoki S, Kudo N, Kanazawa I. Acute conduction blockin vitro following exposure to anti-ganglioside sera. Muscle Nerve 1993;16:587-93.

85 Parry GJG. Anti-ganglioside antibodies do not necessarily play a role inmultifocal motor neuropathy. Muscle Nerve 1994;17:97-9.

86 Lange DJ, Trojaborg W. Do anti-GMI antibodies induce demyelina-tion? Muscle Nerve 1994;17:105-7.

87 Heiman-Patterson T, Krupa T, Thompson P, Nobile-Orazio E,Tahmoush AJ, Shy ME. Anti-GM1/GD1b M-proteins damage humanspinal cord neurons co-cultured with muscle. J Neurol Sci 1993;120:38-45.

88 Ohsawa T, Miyatake T, Yuki N. Anti-B series ganglioside-recognisingautoantibodies in an acute sensory neuropathy patient cause cell deathof rat dorsal root ganglion neurons. Neurosci Lett 1993;157:167-70.

89 Roberts M, Willison HJ, Vincent A, Newsom-Davis J. Serum factor inthe Miller Fisher variant of Guillain-Barre syndrome blocks neuro-transmitter release. Lancet 1994;343:454-5.

90 Ilyas AA, Willison HJ, Quarles RH, et al. Serum antibodies to ganglio-sides in Guillain-Barre syndrome. Ann Neurol 1988;23:440-7.

91 Ilyas AA, Mithen FA, Chen Z-W, Cook SD. Search for antibodies toneutral glycolipids in sera of patients with Guillain-Barre syndrome._JNeurol Sci 1991;102:67-75.

92 Ilyas AA, Mithen FA, Chen Z-W, Cook SD. Anti-GMl IgA antibodiesin Guillain-Barre syndrome. J Neuroimmunol 1992;36:69-76.

93 Ilyas AA, Mithen FA, Dalakas MC, Chen Z-W, Cook SD. Antibodies toacidic glycolipids in Guillain-Barre syndrome and chronic inflamma-tory demyelinating polyneuropathy. J Neurol Sci 1992;107: 111-21.

94 Gregson NA, Jones D, Thomas PK, Willison HJ. Acute motorneuropathy with antibodies to GM1 ganglioside. Neurology 1991;238:447-51.

95 Svennerholm L, Fredman P. Antibody detection in Guillain-Barre syn-drome. Ann Neurol 1990;27:S36-S40.

96 van den Berg LH, Marrink J, de Jager AEJ, et al. Anti-GMI antibodiesin patients with Guillain-Barre syndrome. JT Neurol Neurosurg Psychiatry1992;55:8-1 1.

97 Walsh FS, Cronin M, Koblar S, et al. Association between glycoconju-gate antibodies and Campylobacter infection in patients with Guillain-Barre syndrome. 7 Neuroimmunol 1991 ;34:43-5 1.

98 Nobile-Orazio E, Carpo M, Meucci N, et al. Guillain-Barre syndromeassociated with high titers of anti-GM1 antibodies. J Neurol Sci 1992;109:200-6.

99 Yuki N, Yoshino H, Sato S, Shinozawa K, Miyatake T. Severe acuteaxonal form of Guillain-Barre syndrome associated with IgG anti-GDla antibodies. Muscle Nerve 1992;15:899-903.

100 Simone IL, Annunziata P, Maimone D, Liguori M, Leante R, Livrea P.Serum and CSF anti-GMI antibodies in patients with Guillain Barresyndrome and chronic inflammatory demyelinating polyneuropathy._JNeurol Sci 1993;114:49-55.

on 26 June 2018 by guest. Protected by copyright.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.57.11.1303 on 1 Novem

ber 1994. Dow

nloaded from