J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.43.7.568 on 1 July 1980. Downloaded from

Journal of , Neurosurgery, and Psychiatry, 1980, 43, 568-576

Experimental autoimmune myasthenia gravis

JON LINDSTROM From the Salk Institute for Biological Studies, San Diego, California, USA

SUMMARY Injection of animals with purified in complete Freund's adjuvant causes development of which crossreact with receptors in muscle. The cross- reacting antibodies impair neuromuscular transmission. Animals with experimental autoimmune myasthenia gravis (EAMG) are excellent models for studying the complex mechanisms by which the autoimmune response to receptor in myasthenia gravis causes muscle . This review first briefly describes the discovery of EAMG. Then, to provide the necessary perspective, receptor structure and function and properties of anti-receptor antibodies are discussed, followed by a brief review of the pathological mechanisms in EAMG.

Introduction labelled AChR to identify antibodies in MG Protected by copyright. patients.6 7 In 1960 John Simpson wrote a paper which argued It had initially been thought that snake toxins that myasthenia gravis (MG) was caused by bound very specifically, but irreversibly to the antibodies to acetylcholine receptors (AChR) acetylcholine binding site of AChR.8 Thus, that impaired neuromuscular transmission by although they would provide a wonderfully specific competitively inhibiting AChR.1 In 1968 my method for localising and quantitating AChR, graduate adviser, Ed Lennox, gave me a copy to they seemed unusable as affinity adsorbents for read. My thesis project was to purify AChR. purifying AChR. In fact, though toxins bind with Many years before, David Nachmanson2 had in- high affinity to AChR, they can be competitively troduced the idea of using fish electric organs as eluted.9 Using this observation, Jim Patrick and I a rich source of AChR, but no suitably specific purified AChR from the electric organs of Elec- method had been devised for biochemically label- trophorus electricus by solubilising the membrane ling AChR or assaying their activity, much less proteins in Triton X-100, absorbing AChR to an purifying them. We were investigating affinity affinity column of toxin-agarose, washing off con- labelling as an approach to identifying AChR and taminants, competitively eluting AChR with http://jnnp.bmj.com/ biochemically,3 but with considerably less success the high affinity antagonist benzoquinonium.10 We than Arthur Karlin was enjoying at that time,4 identified AChR as a single component on sucrose and in ignorance of the really critical observation gradients which bound 1251 toxin and cholinergic by C Y Lee that 1251 labelled snake venom toxins ligands. But this left several unanswered questions. could specifically label AChR.5 The use of fish The most important was, "was this the physiologi- electric organs as a source of AChR and 1251 cally significant AChR or some other binding com- labelled snake venom toxin to identify AChR ponent? " We had no biochemical assay for the

would ultimately permit both the purification and ion channel regulated by acetylcholine binding, on September 28, 2021 by guest. characterisation of AChR and the understanding so we could identify only the binding site, not the of the pathological mechanisms impairing neuro- functional activity of the molecule, as one would muscular transmission in MG. But at the time I with an enzyme. Had we left the ion channel in first read Dr Simpson's article, my interest was in the membrane? We observed two polypeptide using anti-AChR antibodies from MG patients, chains in the purified material. Were these AChR if they existed, to identify AChR. In the natural subunits involved in ligand binding and ion chan- course of things, we ended up much later using nel function or was one or both a contaminant? Were there other subunits we had missed? Address for reprint requests: Dr Jon Lindstrom, The Salk Institute, We decided that the best way to test whether PO Box 85800, San Diego, CA 92138. our purified AChR contained any part of the 568 J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.43.7.568 on 1 July 1980. Downloaded from

Experimental autoimmune myasthenia gravis 569 physiologically significant AChR was to make subunits have unique peptide maps,5' 20 21 and are antibodies against it and see if they blocked the basically immunochemically distinct,22 23 there are function of AChR in intact electric organ cells. some structural similarities. A monoclonal anti- The antibodies did block.1" But more interestingly, body to the 8 subunit also reacts with -y, but with the immunised rabbits became sick and died.12 low affinity.24 Similarly, a monoclonal This immediately brought Dr Simpon's theory to the /3 subunit also reacts with the a subunit, that MG was caused by an autoimmune response but with low affinity.24 The functions of /, y and to AChR1 back to mind. 8 are unknown. Because AChR monomers puri- In the intervening years it has become obvious fied under conditions which prevent denaturation that experimental autoimmune myasthenia gravis of their ion channels retain agonist stimulated (EAMG) and MG are both caused by an auto- cation permeability when reconstituted into model immune response to AChR. The effects of anti- membrane vesicles, it is known that the ion bodies have turned out to be much more complex channel is an integral component of the AChR than simple competitive antagonism of AChR. monomer.25 Thus, some of the AChR subunits The excellent group of co-workers with which I must be involved in the structure and regulation have had the good fortune to be associated has of this channel. done much to provide a fundamental understand- AChR from the electric organ of the fresh ing of both EAMG and MG. These investigators water teleost Electrophorus electricus initially ap- have included Vanda Lennon, Marge Seybold, peared to differ in subunit structure from torpedo Ed Lambert, Andy Engel, Steve Heinemann and AChR.'0 17 Now techniques have been devised many others. Not only have AChR and toxin for preserving four subunits in eel AChR as well.20 been useful probes for studying EAMG and MG, These correspond immunochemically to the four but antibodies to AChR and its subunits have subunits of torpedo AChR.26 27 Protected by copyright. proven very useful probes for answering many of AChR is present at much lower concentration the initial questions about AChR structure and in muscle than in electric organs, and its structure function. is corresponding less certain. Until recently, In the following sections I will briefly outline there had been disagreement over the structure the current state of our knowledge about the even of electric organ AChR. Some reports AChR molecule and its role in EAMG. claimed that it was composed only of a subunits,28 but these erroneous results probably resulted from The AChR molecule selective proteolysis of the higher molecular weight subunits. Antigenic determinants corre- Structure of the AChR molecule has been studied sponding to the four subunits of torpedo AChR most carefully using AChR purified from marine are observed in both bovine27 and human23 muscle. elasmobranch electric organs. AChR synthesis a-like subunits can be affinity labelled with an and destruction has been best studied using acetylcholine binding site directed reagent using chicken and rodent muscle cells in tissue culture. AChR from torpedo, eel, or muscle.27 29 However, AChR function has been best studied using intact in rat muscle,29 but not bovine muscle,27 two sizes http://jnnp.bmj.com/ amphibian and mammalian muscle tissue. It seems of chains were labelled. AChR purified from rat increasingly reasonable to hope that the structure, muscle contains subunits similar in molecular metabolism and function of the AChR is suffici- weight to /3, -y and 8,21 but these have not yet ently universal that species specific variations are been tested to determine whether they correspond relatively small, and information from all these immunochemically to the subunits of torpedo sources can be integrated to provide a reasonable AChR. view of the AChR molecule. AChR metabolism has been most extensively AChR from the electric organ of the marine studied by Fambrough, who has recently reviewed on September 28, 2021 by guest. elasmobranch Torpedo californica contains four this field.30 31 Muscle cells in culture, like dener- strongly associated subunits in the mole ratio vated or foetal muscle tissue, rapidly turn over a2fly8.13-15 The apparent molecular weights of their AChR. AChR at normal, mature synapses these acidic glycoproteinsl5 16 are 38, 50, 57 and are turned over much more slowly, but it is pre- 64X 103 for a, /3, y and 8, respectively.'7 18 There sumed that the same basic mechanisms are in- are two acetylcholine binding sites per AChR volved. The "junctional" AChR present at neuro- monomer, located on the a subunits.'4 17 Torpedo muscular junctions are suspected to differ from AChR normally exist as dimers joined by disulfide "extrajunctional" AChR of denervated muscle by bonds between 8 subunits.19 Although the four a structural modification that alters their ligand J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.43.7.568 on 1 July 1980. Downloaded from

570 Jon Lindstrom affinity, isoelectric point,32 and immunochemical muscle, thereby perhaps affecting acetylcholine properties,27 33 but the basis of this structural dif- metabolism and release (for which there is some ference has not yet been discovered.2' In cultured evidence in MG and EAMG63), nerve ending cells,30 31 AChR are assembled to the point where sprouting (for which there is some evidence64), they can bind 1251 toxin in less than 15 minutes. and perhaps other presynaptic parameters. AChR are first observed in the Golgi apparatus, oriented with their toxin binding sites toward the Immunisation with AChR interior of the membrane vesicles. Completion of AChR assembly requires glycosylation. Two to EAMG has been induced in all species tested by four hours are required before AChR are incor- immunisation with AChR purified from fish porated in the surface membrane, presumably by electric organs. The species tested have included fusion of vesicles of newly synthesised AChR with rabbits,'2 1245-504rats,5' 52.5 2 mice,~~53-57 guinea pigs,5' 58 the surface so that they are oriented right side goats,59 60 monkeys,6' and frogs.62 Detailed fea- out. The half time for destruction of AChR in tures of the characteristic of cell culture is of the order of 20 hours, whereas EAMG vary, depending on the species immunised. at neuromuscular junctions the half time is at least For example, monkeys with EAMG exhibit the 150 hours. The degradation process appears to drooping eyelids characteristic of human MG,6' randomly select AChR. It is an energy-requiring while rodents do not.4558 Most rabbits become process which probably involves endocytosis and nearly moribund 25-30 days after immunisation,'2 certainly involves secondary lysozomes, where while frogs require immunisation prolonged over AChR are degraded to their component amino 4 to 6 months.62 Genetically defined mouse strains acids. Degradation can be easily measured by pre- differ in their response.53 54 Presumably these dif- labeling AChR with 125I toxin and then monitor- ferences in response reflect primarily relatively Protected by copyright. ing release of 125I tyrosine from the cells due to small quantitative differences in degree of cross- 1251 toxin which is degraded along with the AChR reaction of AChR, relative effectiveness of various to which it is attached. About 90 minutes are re- components of the immune response, relative quired for release of 1251 tyrosine after the AChR effectiveness of adaptive responses at the neuro- first disappear from the surface. The rate of muscular synapse, and anatomical differences AChR synthesis or destruction can be influenced rather than fundamental large qualitative differ- by a number of factors, including muscle ences in the component processes triggered by activity,34 hypophysectomy,35 and binding of anti- immunisation with AChR. AChR antibodies.36-44 EAMG has been most thoroughly studied in In conclusion, the AChR molecule in muscle young, female Lewis rats. EAMG has been in- probably is an integral membrane glycoprotein duced in these rats by immunisation with AChR composed of four subunits intimately associated purified from syngenic rats,65 foetal calf muscle, in an a2,8yS complex which includes both the (unpublished) electric eels,5' 52 65 torpedoes,23 27 66 acetylcholine binding site and the cation-specific and the purified subunits of torpedo AChR.2327 channel it regulates. Antibodies can be used to Native torpedo AChR is a very potent immuno- http://jnnp.bmj.com/ demonstrate quite similar basic subunit structures gen. Immunisation with as little as 1 microgram in AChR from different species, despite the fact in complete Freund's adjuvant produces measur- that these antibodies are so specific in their inter- able antibody and loss of muscle AChR.66 In order action with AChR that differences in the amino to achieve relatively high concentrations of anti- acid sequence of these subunits may result in only body cross-reacting with muscle AChR, single 1 % immunological crossreaction with AChR from or multiple doses of 15-30 micrograms are usually a species different from that of the immunogen. used. The degree of crossreaction of electric organ AChR synthesis, localisation, and destruction is AChR with muscle AChR is usually much less on September 28, 2021 by guest. a complex process, potentially subject to regula- than 5%.60 However, high antibody concentrations tion at many points even in the normal neuro- against electric organ AChR are readily achieved, muscular junction. Thus one might expect that and the amount of muscle AChR is very small. the in vivo immune response to AChR which Thus, a rat whose serum can bind 5X10-6 moles occurs in EAMG and MG might have complex of 1251 toxin labelled torpedo AChR per litre may effects on AChR metabolism, localisation, and only cross-react 1% with rat muscle AChR, but function. In addition, of course, one might expect even at 5X 10-8M, anti-muscle antibody still ex- that an immune response to AChR would upset ceeds tenfold the amount of AChR in the rat's the normal trophic interactions between nerve and muscle. J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.43.7.568 on 1 July 1980. Downloaded from

Experimental autoimmune myasthenia gravis 571 Specificities of the antibodies to AChR produced plates.6974 These cells are presumably attracted by immunisation of rats have been studied in by chemotactic fragments released by activated some detail. The most important antigenic deter- complement, and their attack on the postsynaptic minants depend on the native structure of the membrane is triggered by binding to antibody and molecule.23 27 About 50% of the antibodies complement deposited on the membrane. This is directed at the native molecule bind at or near a shown by the observation that the phagocytic in- determinant on the a subunit which is not the vasion which normally occurs after injection of acetylcholine binding site.24 Antisera to native anti-AChR into a normal rat69 is prevented if the AChR cross-react detectably, but not well with rat is first depleted of the C3 component of com- SDS denatured subunits.23 26 27 Interspecies cross- plement. 70 The phagocytes destroy the postsynaptic reaction of antisera to native AChR is greatest membrane71-74 producing functional denervation with denatured a subunits.26 27 Torpedo AChR in many fibres.75 The phagocytes amplify the subunits are less immunogenic and myastheno- effect of the small amount of antibody which is genic than the native molecule by several hundred- actually bound to AChR during acute or passive fold, but immunisation with any of the subunits EAMG.65 69 Large loss of muscle AChR is ob- can induce EAMG.22 Antibodies to the denatured served during the phagocytic invasion.65 69 71 After subunits react well with native AChR,23 27 but two or three days, when the phagocytic invasion recognise antigenic determinants different from is diminishing, the muscle AChR content tran- the conformationally dependent determinants siently increases to more than normal.65 presum- which dominate the immunogenicity of the native ably due to formation of extrajunctional AChR in molecule.24 Because both native AChR and each response to the transient denervation caused in of its four denatured subunits can cause EAMG, many fibres by phagocytic attack. What triggers Protected by copyright. it is clear that there is no single "myasthenogenic" and terminates the phagocytic response in passive antigen critical for inducing EAMG which is and acute EAMG, but prevents phagocytic in- shared by all AChR. vasion during chronic EAMG, is not known. During chronic EAMG antibody and complement Acute, chronic and passive EAMG are present on the postsynaptic membrane in even larger amounts than observed during acute or Three forms of EAMG have been distinguished passive EAMG,76 but phagocytes are not attracted in Lewis rats: acute, chronic and passive.51 67 or activated. It may be that chronic EAMG repre- After a single immunisation with AChR in com- sents a steady state in which much of the bound plete Freund's adjuvant, plus injection of pertussis complement is inactivated by endogenous pro- vaccine at other sites as additional adjuvant, Lewis cesses, but that passive and acute EAMG are non- rats undergo two phases of muscular weakness. equilibrium states caused by the sudden deposition An acute phase lasting 2 to 3 days occurs 8 to 11 of antibody and hence, complement, and the re- days after immunisation.65 If pertussis is not used lease of chemotactic fragments occurs in signifi- as additional adjuvant, an acute phase is not cant amounts before it is terminated by a feed- observed. In any case, after 28 to 30 days chronic back mechanism. There is no obvious equivalent http://jnnp.bmj.com/ muscular weakness begins.65 This may be pro- of acute EAMG in human MG, perhaps because gressive until death, especially at higher AChR the development of anti-AChR antibody is slow doses, or diminish as the response to immunogen and insidious. There is a human parallel of diminishes.68 Antisera from a rat with chronic passive EAMG. Infants born to mothers with MG EAMG can be used to passively transfer EAMG passively acquire a significant fraction of the to a normal rat very efficiently.69-71 The recipient maternal antiAChR concentration over the course rat begins to exhibit weakness within 12 to 24 of gestation, and this is associated with muscular on September 28, 2021 by guest. hours, but if the weakness is not fatal, this weakness, in some cases, that diminishes over a weakness, like that in acute EAMG, passes few weeks after birth as the maternal immuno- within a few days. EAMG can also be transferred globulins are removed.77 It might be expected with lymphocytes, but not very efficiently, and that babies should be less severely affected than the weakness is delayed by many days,67 pre- their mothers, because their AChR metabolism sumably due to the time required for these might be operating at more nearly the high foetal lymphocytes to synthesise sufficient antibody to rate,30 31 permitting them to adapt better to the produce an effect. immune assault. Both acute and passive EAMG are character- Chronic EAMG closely resembles human MG. ised by extensive phagocytic invasion of the end- Morphologically, chronic EAMG is characterised J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.43.7.568 on 1 July 1980. Downloaded from

572 Jon Lindstrom by the presence of a postsynaptic membrane with posite to specialised sites of acetylcholine release27 reduced numbers and sizes of folds.72 74 Phago- in the presynaptic membrane and away from the cytes are not observed. Both antibody and comple- concentrated on the basement ment are bound to the postsynaptic membrane membrane between the folds. Disruption of this and to fragments in the intersynaptic space evi- spatial relationship no doubt further impairs trans- dently shed in the course of focal lysis. 76 The mission, but it is difficult to quantify the effects AChR content of the muscle is reduced to about of focal lysis on neuromuscular transmission. one-third normal, and most of the AChR which Cross-linking of AChR by antibody43 66 causes remain have antibodies bound.65 66 A decrementing their aggregation and internalisation,83 where they electromyogram typical of MG is observed.68 are degraded in lysozomes.39 40 In cell culture the Microelectrophysiological studies show normal internalisation process is energy,36 4144 temperature numbers of quanta released75 but reduced sensi- and cytoplasmic filament dependent,44 and requires tivity of the postsynaptic membrane,79 which about an hour and a half before amino acid resi- accounts for a reduction in the size of miniature dues are released from the cell.66 Lysozomal pro- endplate potentials to about one-third normal. tease inhibitors can prevent destruction of the The muscular weakness characteristic of chronic internalised AChR.39 40 In all these respects, except EAMG can readily be explained by a decrease in for the antibody-induced cross-linking and aggre- amount of active AChR. Simple competitive inhi- gation, the mechanism of antigenic modulation bition of AChR with toxin produces very similar resembles the normal mechanism of AChR de- clinical and electrophysiological signs.80 But this struction.30 31 Antigenic modulation accelerates is not the way antibody to AChR acts. Few" or the rate of AChR destruction by two to threefold none59 of the antibodies are directed at the acetyl- in both cell culture36 4 and organ culture.3740 choline binding site. Antisera can directly inhibit Muscle from rats with EAMG does, in fact, Protected by copyright. AChR function,"1 36 but in general the effects of destroy its AChR at the accelerated rate, using antibody on AChR function are small. Muscle lysozomal enzymes, as expected of antigenic modu- cells in cultures exposed to anti-AChR and then lation.40 Endocytosis may be the rate-limiting step subjected to acetylcholine noise analysis showed of antigenic modulation.66 A threefold increase in approximately a 23% decrease in AChR channel the rate of AChR destruction would be sufficient opening time and a 15% decrease in AChR con- to account for the observation that AChR content ductance when open.36 Because the safety factor in the muscles of rats with EAMG decreases to for neuromuscular transmission is large, this net a minimum of about one-third of normal.66 How- 38% decrease in AChR conductance (if all AChR ever, it is unknown whether in vivo there is an at a normal junction were antibody bound) would adaptive increase in the rate of AChR synthesis not inhibit transmission. This is shown by the from the very low rate characteristic of normal observation that normal rats, depleted of the C3 junctional AChR toward the much higher rate component of complement by treatment with characteristic of extrajunctional AChR. Thus, cobra venom factor and then injected with anti- although it is clear that antigenic modulation is a AChR antibodies, have normal neuromuscular very important mechanism for causing AChR loss http://jnnp.bmj.com/ transmission despite having at least 67% of their and thereby impairing transmission, and quanti- muscle AChR bound with antibody.70 tatively it might be the most important mechan- Complement-mediated destruction of the post- ism, it is not yet possible to quantify its effects synaptic membrane subsequent to anti-AChR with respect to those of antibody dependent, com- antibody binding appears to impair transmission plement-mediated focal lysis. in two ways. Focal lysis of the postsynaptic mem- brane releases membrane fragments containing Concluding remarks AChR, antibody, and complement. 76 Thus, lysis on September 28, 2021 by guest. contributes directly to loss of AChR. The post- No special anti-muscle AChR antibody specificity synaptic membrane must reseal very effectively appears to be required to cause EAMG. As pre- after such lytic attacks, because the resting mem- viously discussed, antibodies against the acetyl- brane potential in chronic EAMG is not signifi- choline binding site are not required. Cross-linking cantly reduced.75 Complement-mediated lysis of of AChR by antibody to trigger antigenic modula- the postsynaptic membrane is probably also re- tion or simple binding of antibody to target fixation sponsible for the destruction of its normal com- of complement should not depend on the part of plex folded structure.7'74 The folded structure the AChR molecule to which the antibody binds. concentrates AChR at the tips of the folds8' op- In fact, immunisation with any of the four de- J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.43.7.568 on 1 July 1980. Downloaded from

Experimental autoimmune myasthenia gravis 573 natured AChR subunits causes EAMG." How- Binding Sites on Membranes. Proc Natl A cad Sci ever, differences in anti-AChR subclass could USA 1970; 67:1688-94. affect the ability to bind complement. And, in 4 Karlin A. Chemical modification of the active theory, some effects of antibody specificity should site of the . J Gen Physiol exist. For example, antibodies to determinants on 1969; 54:245-64. the interior of the cell 5 Lee CY, Tseng LF, Chiu TH. Influence of dener- membrane would not be vation on able to bind in antibodies localization of neurotoxins from clapid vivo, to determinants venoms in rat diaphragms. Nature 1967; 215: in the centre of the molecule might not be able 1177-8. to reach to cross-link AChR, and antibodies to 6 Lindstrom JM, Seybold ME Lennon VA, Whitt- determinants represented twice on the molecule ingham S, Duane DD. Antibody to acetylcholine might not effectively cross-link AChR because receptor in myasthenia gravis: Prevalence, clinical both antibody binding sites could bind within correlates and diagnostic value. Neurology 1976; the monomer. On the other hand, antibodies to 26:1054-9. the acetylcholine binding site or ion channel 7 Lindstrom J. An assay for antibodies to human opening might be very effective at inhibiting acetylcholine receptor in serum from patients AChR with myasthenia gravis. J Clin Immunol & Im- function. Such effects have yet to be munopath 1977; 7:36-43. demonstrated with monoclonal populations of 8 Changeux JP, Meunier JC, Huchet M. Studies anti-AChR antibodies, but if they are, should tell on the Cholinergic Receptor Proteins of Electro- us as much about the AChR molecule as about phorus. Mol Pharmacol 1971; 7:538-53. the pathological mechanisms of EAMG. 9 Patrick J, Heinemann S, Lindstrom J, Schubert D, The discovery of EAMG validated the hypo- Steinbach JH. Appearance of Acetylcholine Re- ceptors Differentiation of a thesis' that the weakness characteristic of MG During Myogenic Protected by copyright. could arise from an autoimmune response to Cell Line. Proc Natl Acad Sci USA 1972; 69: AChR. 2762-6. Further, studies of EAMG provided 10 methods for assay of human AChR,84 Lindstrom J, Patrick J. Purification of the acetyl- anti-AChR choline receptor In: antibody from MG patients,6 7 and for develop- by affinity chromatography. ment of Bennett, ed, Synaptic Transmission and Neuronal other techniques which could be applied Interaction. New York: Raven Press 1974; 191- to human material for the study of MG. It is now 216. evident that MG is in fact caused by an auto- 11 Patrick J, Lindstrom J, Culp B, McMillan J. immune response to AChR. Studies of EAMG Studies on purified eel acetylcholine receptor and have proven very valuable in discovering the anti-acetylcholine receptor antibody. Proc Natl mechanisms by which the autoimmune response Acad Sci USA 1973; 70:3334-8. to AChR impairs transmission in MG. EAMG is 12 Patrick J, Lindstrom J. Autoimmune response to acetylcholine receptor. Science 1973; 180:871-2. probably not an especially useful model for study- 13 ing what causes the Hamilton SL, McLaughlin M, Karlin A. Biochem auto-immune response to Biophys Res Commun 1977; 79:692-9. AChR in MG. And EAMG may be of only 14 Reynolds Karlin A. limited utility in JA, Molecular weight in de- studying therapy of MG because, tergent solution of acetylcholine receptor from http://jnnp.bmj.com/ unlike MG, EAMG does not result from an endo- Torpedo californica, Biochem 1978; 17:2035-8. genous self-sustaining process. But EAMG can be 15 Lindstrom J, Merlie J, Yogeeswaran G. Bio- a very useful model of for immun- chemical properties of acetylcholine receptor sub- ological studies. And EAMG, along with MG, units from Torpedo californica, Biochem 1979; can provide an archetype for recognising and 18:4465-70, studying other auto-immune 16 Vandlen R, Wilson C, Eisenach J, Raftery M. anti-receptor diseases. Studies of Finally, the anti-AChR antibodies produced in the composition of purified Torpedo californica acetylcholine receptor and its subunits. EAMG are proving increasingly valuable as probes Biochem on September 28, 2021 by guest. of AChR structure, 1979; 18:1845-54. function, and metabolism. 17 Karlin A, Weill CL, McNamee MG, Valderrama R. Facets of the structures of acetylcholine re- References ceptors from Electrophorus and Torpedo. Cold Spring Harbor Symposia on Quantitative Biology, 1 Simpson J. Myasthenia gravis: A new hypothesis. 1975; XL:203-10. Scott Med J 1960; 5:419-36. 18 Raftery MA, Vandlen 2 Nachmanson D. Chemical and RL, Reed KL, Lee T. Molecular Basis Characterization of Torpedo californica acetyl- of Nerve Activity, New York: Academic Press, choline receptor: Its subunit composition 1959. ligand and 3 Kiefer H, Lindstrom J, binding properties. Cold Spring Harbor Lennox ES, Singer SJ. Symposia on Quantitative Biology 1975; XL: Photo-Affinity Labeling of Specific Acetylcholine 193-202. J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.43.7.568 on 1 July 1980. Downloaded from

574 Jon Lindstrom 19 Chang HW, Bock E. Molecular forms of the 33 Weinberg C, Hall Z. Antibodies from patients acetylcholine receptor. Effects of calcium ions with myasthenia gravis recognize determinants and a sulfhydryl reagent on the occurrence of unique to extrajunctional acetylcholine receptors. oligomers. Biochem 1977; 16:4513-20. Proc Natl Acad Sci USA 1979; 76:504-8. 20 Froehner SC, Rafto S. Comparison of the subunits 34 Reiness C, Hall Z. Electrical stimulation of de- of Torpedo californica acetylcholine receptor by nervated muscles reduces incorporation of me- peptide mapping. Biochem 1979; 18:301-7. thionine into the ACh Receptor. Nature 1977; 21 Nathanson N, Hall Z. Subunit structure and pep- 268:655-7. tide mapping of junctional and extrajunctional 35 Reiness G, Hogan P, Marshall J, Hall Z, Griffin acetylcholine receptors in rat muscle. Biochem G, Goldberg A. Factors influencing degradation 1979; 18:3392-401. of extrajunctional acetylcholine receptors in 22 Claudio T, Raftery MA. Immunological compari- In: Hall Z, Kelley R, and Fox CF, son of acetylcholine receptors and their subunits eds. Cellular Neurobiology 1977; 207-15. from species of electric ray. Arch Biochem 36 Heinemann S, Bevan S, Kullberg R, Lindstrom J, Biophys 1977; 181:484-9. Rice J. Modulation of the acetylcholine receptor 23 Lindstrom J, Einarson B, Merlie J. Immunization by anti-receptor antibody. Proc Natl Acad Sci of rats with polypeptide chains from torpedo USA 1977; 74:3090-4. acetylcholine receptor causes an autoimmune re- 37 Heinemann S, Merlie J, Lindstrom J. Modulation sponse to receptors in rat muscle. Proc Natl Acad of acetylcholine receptor in rat diaphragm by Sci USA 1978; 75:769-73. antireceptor sera. Nature 1978; 274:65-8. 24 Tzartos S, Lindstrom J. Monoclonal antibodies 38 Reiness G, Weinberg C, Hall Z. Antibody to used to probe acetylcholine receptor structure: acetylcholine receptor increases the degradation localization of the main immunogenic region and of junctional and extrajunctional receptors in

detection of similarities between subunits. Proc adult muscles. Nature 1978; 274:68-70. Protected by copyright. AVatl Acad USA 1979; 77:755-59. 39 Merlie JP, Heinemann SJ, Lindstrom J. Acetyl- 25 Anholt R, Lindstrom J, Montal M. Functional choline receptor degradation in adult rat dia- equivalence of monomeric and dimeric forms of phragms in organ culture and the effect of anti- purified acetylcholine receptor from Torpedo acetylcholine receptor antibodies. J Biol Chem californica in reconstituted lipid vesicles. Eur J 1979; 254:6320-7. Biochem 1979; (in press) 40 Merlie J, Heinemann S, Einarson B, Lindstrom J. 26 Lindstrom J, Cooper J, Tzartos S. Acetylcholine Degradation of acetylcholine receptor in dia- receptors from Torpedo and Electrophorus have phragms of rats with experimental autoimmune similar subunit structures. Biochem 1979; (in myasthenia gravis. J Biol Chem 1979; 254:6328-32. press). 41 Kao I, and Drachman DB. Myasthenic Immuno- 27 Lindstrom J, Walter B, Einarson B. Immuno- globulin Accelerates Acetylcholine Receptor chemical similarities between subunits of acetyl- Degradation. Science 1977; 196:527-9. choline receptors from Torpedo, Electrophorus, 42 Drachman DB, Angus CW, Adams RN, Kao I. and mammalian muscle. Biochem 1979; 18:4470- Effect of myasthenic immunoglobulin on acetyl- 80. choline receptor turnover: Selectivity of degrada-

28 Sobel A, Heidman T, Hofler J, Changeux JP. tion process. Proc Natl Acad Sci USA 1978; 75: http://jnnp.bmj.com/ Distinct protein components from Torpedo mar- 3422-6. morata membranes carry the acetylcholine re- 43 Drachman DB, Angus CW, Adams RN, Michel- ceptor site and the binding site for local anesthe- son JD, Hoffman GJ. Myasthenic antibodies cross tics and histrionicotoxin. Proc Natl Acad Sci USA link acetylcholine receptors to accelerate degra- 1978; 75:510-14. dation. New Eng J Med 1978; 298:1116-22. 29 Froehner S, Karlin A, Hall Z. Affinity alkylation 44 Appel SH, Anwyl R, McAdams MW, Elias S. labels two subunits of reduced acetylcholine re- Accelerated Degradation of Acetylcholine Re- ceptor from mammalian muscle. Proc Natl Acad ceptor from Cultured Rat Myotubes with My- Sci USA 1977; 74:4485-688. asthenia Gravis Sera and Globulins. Proc Natl on September 28, 2021 by guest. 30 Fambrough D, Devreotes P, Cord D, Gardner J, Acad Sci USA 1977; 74:2130-4. Tepperman K. Metabolism of acetylcholine re- 45 Aharonov A, Kalderon N, Silman I, Fuchs S. ceptors in skeletal muscle. Nat! Canc Inst Monogr Preparation and immunochemical characteriza- No 48 1978; 277-94. tion of a water soluble acetylcholine receptor 31 Fambrough D. Control of acetylcholine receptors fraction from the electric organ tissue of the in skeletal muscle. Phy.iol Rev 1979; 59:.165-227. electric eel. Immunochem 1975; 12:765-71. 32 Brockes J, Hall Z. Acetylcholine Receptors in 46 Sugiyama H, Benda P, Meunier J, Changeux J. Normal and Denervated Rat Diaphragm Muscle Immunological characterization of the cholinergic II. Comparison of Junctional and Extrajunctional receptor protein from Electrophorus electricus, AChR. Biochem 1975; 14:2100-6. FEBS Letters 1973; 35:124-8. J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.43.7.568 on 1 July 1980. Downloaded from

Experimental autoimmune myasthenia gravis 575 47 Heilbronn E, Mattsson C, Thornell L, Sjostrom 63 Molenaar PC, Polak RL, Miledi R, Alema S, M, Stallberg E, Hilton-Brown P, Elmquist D. Vincent A, Newsom-Davis J. Acetylcholine in Experimental myasthenia in rabbits: immunologi- Intercostal Muscle from Myasthenia Gravis cal, electrophysiological and morphological aspects. Patients and in Rat Diaphragm after Blockade of NY Acad Sci 1976; 274:337-53. Acetylcholine Receptors. Progress in Brain Re- 48 Penn A, Chang H, Lovelace R, Niemi W, search, Vol 49, 449-58. Miranda A. Antibodies to acetylcholine receptors 64 Bickerstaff ER, Woolf AL. The Intramuscular in rabbits; immunological and electrophysiological Nerve Endings in Myasthenia Gravis. Brain 1960; studies. NY Acad Sci 1976; 274:354-76. 83:10-23. 49 Berti F, Clementi F, Conti-Tranconi B, Folco G. 65 Lindstrom JM, Einarson B, Lennon VA, Seybold A cholinoceptor antiserum, its pharmacological ME. Pathological mechanisms in EAMG I: Im- properties. Br J Pharmac 1976; 57:17-22. munogenicity of syngeneic muscle acetylcholine 50 Green DPL, Miledi R, Vincent A. Neuromuscular receptor and quantitative extraction of receptor transmission after immunization against acetyl- and antibody-receptor complexes from muscles of choline receptors. Proc R Soc Lond 1975; B 189: rats with experimental autoimmune myasthenia 57-68. gravis. J Exp Med 1976; 144:726-38. 51 Lennon VA, Lindstrom J, Seybold ME. Experi- 66 Lindstrom J, Einarson B. Antigenic modulation mental autoimmune myasthenia: A model of and receptor loss in EAMG. Muscle & Nerve myasthenia gravis in rats and guinea pigs. J Exp 1979; 2:173-9. Med 1975; 141:1365-75. 67 Lennon V, Lindstrom J, Seybold M. Experimental Protected by copyright. 52 Lindstrom J, Lennon V, Seybold M, Whittingham autoimmune myasthenia: Cellular and humoral S. Experimental autoimmune myasthenia gravis immune responses. NY Acad Sci 1976; 274:283- and myasthenia gravis. Ann NY Acad Sci 1976; 99. 274:254-74. 68 Lindstrom J. Autoimmune response to acetyl- 53 Fuchs S, Nevo D, Tarrab-Hazdai R. Strain differ- choline receptor in myasthenia gravis and ex- ences in the autoimmune response of mice to perimental autoimmune myasthenia gravis. acetylcholine receptors. Nature 1976; 263:329-30. Proceedings MDA International Scientific Con- 54 Christadoss P, Lennon V, David C. Genetic Con- ference, LP Rowland, ed. Excerpta Medica 1977; trol of Experimental Autoimmune Myasthenia 121-31. Gravis in Mice. J Immunol 1979; 123:2540-3. Lennon R. Acetyl- 69 Lindstrom JM Engel AG, Seybold ME, 55 Fulpius B, Zurn A, Granato D, Leder EH. Pathological mechanisms in receptor myasthenia gravis. NY Acad VA, Lambert choline and EAMG II: Passive transfer of experimental auto- Sci 1976; 274:116-29. anti- Fulpius BW, Moody J. Experimental immune myasthenia gravis in rats with 56 Granato D antibodies. J Exp Med myasthenia in Babl/c mice immunized with rat acetylcholine receptor acetylcholine receptor from rat denervated 1976; 144:739-53. muscle. Proc Natl Acad Sci 1976; 73:2872-6. 70 Lennon VA, Seybold ME, Lindstrom J, Cochrane http://jnnp.bmj.com/ C, Yulevitch R. Role of complement in patho- 57 Lindstrom J. Autoimmune response to acetyl- genesis of experimental autoimmune myasthenia choline receptors in myasthenia gravis and its J Exp Med 1977; 147:973-83. animal model. In: Advances in , gravis. Kunkel HG, Dixon F, eds. New York: Academic 71 Engel A, Sakakibara H, Sahashi K, Lindstrom J, Press, Vol 27, 1979. Lambert E, Lennon V. Passively transferred ex- perimental autoimmune myasthenia gravis. Neur- 58 Tarrab-Hazdai R, Aharonov A, Abramsky 0, ology 1978; 29:179-88. Yaar I, Fuchs S. Passive transfer of experimental 72 Engel AG, Lindstrom JM, Lambert EH, Lennon autoimmune myasthenia by lymph node cells in VA. Ultrastructural localization of the acetyl- on September 28, 2021 by guest. inbred guinea pigs. J Expt Med 1975; 142:785-9. choline receptor in myasthenia gravis and in its 59 Lindstrom J. Immunological studies of acetyl- experimental autoimmune model. Neurology choline receptors. J Supramol Struc 1976; 4:389- 1977; 27:307-15. 403. 73 Engel A, Tsujihata M, Lambert E, Lindstrom J, 60 Lindstrom J Campbell M, Nave B. Specificities Lennon V. Experimental autoimmune myasthenia of antibodies to acetylcholine receptors. Muscle gravis: A sequential and quantitative study of & Nerve 1978; 1:140-5. the ultrastructure and 61 Tarrab-Hazdai R, Aharonov A, Silman I, Fuchs electrophysiologic correlation. J Neuropath Exptl S, Abramsky 0. Experimental auto,immune my- Neurol 1976; 35:569-87. asthenia induced in monkeys by purified acetyl- 74 Engel A, Tsujihata M, Lindstrom J, Lennon V. choline receptor. Nature 1975; 256:128-30. End-plate fine structure in myasthenia gravis and 62 Nastuk W, Niemi W, Alexander J, Chang H, in experimental autoimmune myasthenia gravis. Nastuk M. Myasthenia in frogs immunized NY Acad Sci 1976; 274:60-79. against cholinergic receptor protein. Am J Physioc 75 Lambert EH, Lindstrom JM, Lennon VA. End- 1979; 236:C53-C57. plate potentials in experimental autoimmune my- J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.43.7.568 on 1 July 1980. Downloaded from

576 Jon Lindstrom asthenia gravis in rats. Annals NY Acad Sci gravis. Science 1975; 187:955-7. 1976; 274:300-18. 81 Fertuck HC, Salpeter MM. Localization of acetyl- 76 Sahashi K, Engel AG, Lindstrom J, Lambert EH, choline receptor by 125I-labelled a bungarotoxin Lennon V. Ultrastructural localization of immune binding at mouse motor endplates. Proc Natl complexes (IgG and C3) at the end-pla'e in ex- Acad Sci USA 1974; 71:1376-8. perimental autoimmune myasthenia gravis. J 82 Heuse r J Reese T, Dennis M, Jan Y, Jan L, Neuropath and Exp Neurology 1978; 37:212-23. Evans L. exocytosis captured 77 Keesey J, Lindstrom J, Cokely A. Anti-acetyl- quick freezing and correlated with quantal choline receptor antibody in neonatal myasthenia transmitter release. J Cell Biol 1979; 81:275- gravis. New Eng J Med 1977; 296:55. 300. 78 Seybold M, Lambert E, Lennon V, Lindstrom J. 83 Prives J, Hoffman L, Tarrab-Hazdai R, Fuchs S, Experimental autoimmune myasthenia gravis: Amsterdam A. Ligand induced changes in sta- Clinical, neurophysiologic, and pharmacologic bility and distribution of acetylcholine receptors aspects. Ann NY Acad Sci 1976; 274:275-82. on surface membranes of muscle cells. Life 79 Bevan S, Heinemann S, Lennon VA, Lindstrom Sciences 1979; 24:1713-8. J. Reduced muscle acetylcholine sensitivity in 84 Lindstrom J, Lambert E. Content of acetylcholine rats immunized with acetylcholine receptor. receptor and antibodies bound to receptor in my- Nature 1976; 260:438-9. asthenia gravis, experimental autoimmune my- 80 Satyamurti S, Drachman D, Stone F. Blockade asthenia and in Eaton-Lambert Syndrome. Neur-

of acetylcholine receptors: A model of myasthenia ology 1978; 28:130-8. Protected by copyright. http://jnnp.bmj.com/ on September 28, 2021 by guest.