Disorders of Neuromuscular Junction Ion Channels

PHYSIOLOGY IN MEDICINE In collaboration with The American Physiological Society, Thomas E. Andreoli, MD, Editor

Disorders of Neuromuscular Junction Ion Channels Kanokwan Boonyapisit, MD, Henry J. Kaminski, MD, Robert L. Ruff, MD, PhD Ion channel defects produce a clinically diverse set of disorders that range from cystic fibrosis and some forms of migraine to renal tubular defects and episodic ataxias. This review discusses diseases related to impaired function of the skeletal muscle acetylcholine receptor and calcium channels of the motor nerve terminal. Myasthenia gravis is an autoimmune disease caused by antibodies directed toward the skeletal muscle acetylcholine receptor that compromise neuromuscular transmission. Con-

genital myasthenias are genetic disorders, a subset of which are caused by mutations of the acetylcholine receptor. LambertEaton myasthenic syndrome is an immune disorder characterized by impaired synaptic vesicle release likely related to a defect of calcium influx. The disorders will illustrate new insights into synaptic transmission and ion channel structure that are relevant for all ion channel disorders. Am J Med. 1999;106:97–113. q1999 by Excerpta Medica, Inc.

T

ANATOMY AND PHYSIOLOGY OF THE NEUROMUSCULAR JUNCTION

his article discusses genetic and acquired disorders of ion channels at the skeletal muscle neuromuscular junction. The most common and best understood of these disorders is myasthenia gravis (MG), which is also the most completely characterized autoimmune disease. Antibodies directed toward the skeletal muscle acetylcholine receptor (AChR) cause MG and lead to a reduction in the endplate potential and subsequent failure of action potential generation. The congenital myasthenias (CMs) are a heterogeneous group of genetic disorders that compromise neuromuscular transmission (1–3). A subset of the CMs are caused by mutations of the AChR and will be described in detail in this review. Lambert-Eaton myasthenic syndrome (LEMS) is an immune disorder, usually paraneoplastic in etiology, that affects the nerve terminal’s mechanism of calcium-dependent synaptic vesicle release (1,3). A review of the architecture and function of the neuromuscular junction (NMJ) with a particular focus on the structure–function relations of the AChR is presented followed by a discussion of the three disorders.

Am J Med. 1999: From the Departments of Neurology (KB, HJK, RLR) and Neurosciences (HJK, RLR), Case Western Reserve University School of Medicine, Department of Veterans Affairs Medical Center in Cleveland, University Hospitals of Cleveland, Cleveland, Ohio. Supported by the Office of Research and Development, Medical Research Service of the Department of Veterans Affairs (RLR, HJK) and NIH Grant EY-00332 (HJK). Requests for reprints should be addressed to Henry J. Kaminski, MD, Department of Neurology, University Hospitals of Cleveland, 11100 Euclid Avenue, Cleveland, Ohio 44106. q1999 by Excerpta Medica, Inc. All rights reserved.

Architecture of the Neuromuscular Junction The communication point between the nerve and muscle is designed to translate an electrical impulse to a chemical signal to an electrical stimulus that initiates muscle contraction, a feat achieved in approximately 200 ms. An action potential is propagated down the axon of the myelinated motor nerve by saltatory conduction. The neuronal sodium channel is the main ionic channel responsible for action potential conduction (4). The distal portion of a motor nerve fiber branches to provide a single nerve terminal to each of the many muscle fibers that are innervated by that nerve fiber. The terminal branches of the nerve fiber, which are each up to 100 mm long, are unmyelinated (5). The terminal branches of the motor nerve fibers contain delayed rectifier and inward rectifier potassium channels as well as sodium channels (4). Therefore, the amplitude and duration of the action potential in the terminal nerve fibers is controlled by potassium channels as well as by sodium channels. Acetylcholine (ACh) is stored in vesicles within the nerve terminal (5). The AChcontaining vesicles are aligned near release sites in the nerve terminal, and with depolarization of the nerve, plasma membrane vesicles fuse to release their contents (5). The release sites lie in direct opposition to the tops of the secondary synaptic folds of the postsynaptic muscle membrane (5–7). The NMJ is a complex structure consisting of the motor nerve terminal, postsynaptic muscle surface, specialized basal lamina, and associated Schwann cell (Figure 1). 0002-9343/99/$–see front matter 97 PII S0002-9343(98)00374-X

Neuromuscular Junction Ion Channels/Boonyapisit et al

Figure 1. Diagram of neuromuscular junction. See text for details. NO 5 nitric oxide; NOS 5 nitric oxide synthase.

The nerve terminal branches lie in depressions of the postsynaptic membrane called primary synaptic clefts. The space between the nerve terminal and the postsynaptic membrane is about 50 nm (5). The postsynaptic muscle surface area is increased by invaginations of the plasma membrane into secondary synaptic clefts or folds. The AChRs are concentrated at the tops of the secondary synaptic folds and firmly anchored to the dystrophinrelated protein complex through rapsyn (8,9). Rapsyn is important in clustering AChRs at the endplate during synaptogenesis, and rapsyn-deficient transgenic mice do not cluster AChRs, utrophin, and other dystrophin-related complex (DRC) proteins (9 –12). The concentration of AChRs at the endplate is about 15,000 to 20,000 receptors/mm2 (13). Outside the endplate region the concentration of AChRs is 1,000-fold less (14). The high concentration of AChRs and other synapse-specific proteins is partially due to localized gene transcription by the subsynaptic nuclei. AChRs continually turn over, with old receptors internalized and degraded. Removed receptors are replaced with new receptors. Early in development the half-life of AChRs is 13 to 24 hours. At a mature endplate the half-life of AChRs is 8 to 11 days. Skeletal muscle sodium channels are concentrated at the depths of the synaptic folds, and as will be discussed, the clustering of these ion channels also has important functional consequences (Figure 1) (9,15,16). Acetylcholinesterase (ChE) is primarily synthesized by muscle and located in the basal lamina of the secondary synaptic folds. The concentration of ChE is approximately 3,000 molecules/mm2 of postsynaptic membrane, which is about five- to eightfold lower than the concentration of AChRs (13). ChE terminates the action of ACh and prevents AChRs from being repeatedly activated (17). The DRC is composed of several proteins and spans the muscle membrane. On the intracellular surface the DRC is bound to utrophin, which forms a scaffold with other DRCs. The interaction of all these proteins links the basal 98

January 1999

THE AMERICAN JOURNAL OF MEDICINEt Volume 106

lamina to the muscle cytoskeleton (18 –20). Nitric oxide synthase is concentrated at the NMJ, where it is bound to syntrophin, a member of the DRC (21,22). Nitric oxide synthase produces the free radical gas nitric oxide, which participates in cell signaling of many cellular processes. The presence of nitric oxide synthase at the NMJ suggests that nitric oxide could diffuse from its site of synthesis to affect target proteins in the nerve and muscle. The mechanism by which the complex of synapse-specific proteins is concentrated at the NMJ is not known. Agrin, a 200 kD glycoprotein, is found in the basal lamina of the cleft and is primarily synthesized by motor neurons. Alpha-dystroglycan on the extracellular surface binds to the DRC and appears to act as the agrin receptor to induce AChR clustering. Agrin induces clustering of AChRs, ChE, rapsyn, and other proteins on the postsynaptic membrane (23). Agrin binding to a dystroglycan leads to an intracellular signaling cascade that acts through a muscle-specific, tyrosine kinase receptor (24). This signaling cascade presumably mediates the clustering of synapse-specific proteins. Transgenic mice without agrin have very poor clustering of AChR, but other postsynaptic proteins do show localization (25). Therefore, other clustering signals may exist.

Mechanism of Synaptic Vesicle Fusion The first step in generation of the muscle endplate potential is fusion of synaptic vesicles with nerve terminal membrane leading to the release of ACh. Calcium influx through voltage-gated calcium channels into the nerve terminal initiates synaptic vesicle fusion. There are several calcium channel subtypes distinguished by their sensitivity to toxins or blocking agents (26 –28). P-type channels are blocked by v-agatoxin IVA (from American funnel web spider) and are associated with vesicle fusion in mammalian nerve terminal. The subunit structure has not been characterized precisely, but the channel appears to consist of distinct a subunits, b, d, and g subunits.


Figure 2. Diagrammatic representation of synaptic vesicle release. After formation and packaging, vesicles must be brought close to the nerve terminal membrane (docking) and made competent for membrane fusion (priming). After nerve terminal depolarization occurs, voltage-gated channels allow an influx of calcium that serves as the signal for release of ACh. Fused synaptic vesicle membrane undergoes endocytosis, and the membrane is recycled.

The calcium channels responsible for the vesicle fusion are located near the release sites, termed “active zones.” At the active zone, calcium channels are arranged in parallel double rows. Each row contains approximately five channels with 20-nm spacing between rows and about 60 nm between double rows. When an action potential reaches the nerve terminal, calcium channels are activated and allow calcium entry. Calcium concentration reaches 0.1 to 1 mM in regions where vesicle fusion occurs and triggers fusion. There is a greater than second order relationship between calcium current into the nerve terminal and the size of the endplate current elicited in the muscle membrane (6,7,29). In the vertebrate NMJ, a normal nerve terminal action potential does not fully activate the nerve terminal calcium channels, because the duration of the action potential is #1 ms and the nerve terminal calcium channels are activated with a time constant of $1.3 ms (30). Increasing the duration of the nerve terminal action potential by blocking delayed rectifier potassium channels with tetraethylammonium or 3,4-diaminopyridine (DAP) increases calcium entry and subsequent ACh release (30,31).

diffusion (35). Inactivation of ChE slows the decay of the ACh-induced endplate current.

Acetylcholine Receptor Structure–Function Relations The AChR (Figure 3) is composed of four subunits and in mammals exists in two isoforms (36,37). The adult AChR is composed of two a subunits and one copy of each of the b, d, and e subunits (37). The fetal AChR has a g subunit in place of the « subunit. Two isoforms of the a subunit exist, but the significance of the most recently described isoforms is not known. There is homology in the amino acid sequences among the subunits of the AChR (37). Each of the subunits appears to have its N-terminal and C-terminal regions in the extracellular space. In addition, each subunit contain four a helices, called M1 to M4, that probably span the membrane (38,39). The extracellular portions of the subunits consisting of the N- and C-terminal regions and the region between M2 and M3 form a large extracellular vestibule that surrounds the channel extracellular orifice. The regions between M1 and M2 and

ACh Action Each synaptic vesicle fusion releases about 10,000 ACh molecules into the synaptic cleft (32). ATP is also released by synaptic vesicle fusion, and the released ATP may modulate transmitter sensitivity (33). The depolarization response of the postsynaptic membrane to the release of the contents of a single vesicle of ACh is called the miniature endplate potential (MEPP), and the net postsynaptic depolarization produced by the release of all the vesicles triggered by a nerve action potential is the endplate potential. The quantal content corresponds to the number of transmitter vesicles released by a nerve terminal action potential. An action potential propagating into the nerve terminal stimulates the fusion of 50 to 300 synaptic vesicles (ie, the normal quantal content is 50 to 300) (5). The diffusion of ACh across the synaptic cleft is very rapid due to the small distance to be traversed and the relatively high diffusion constant for ACh (34). ACh is removed from the synaptic cleft by hydrolysis due to ChE and by

Figure 3. Pentameric subunit structure of the AChR. January 1999

THE AMERICAN JOURNAL OF MEDICINEt

Volume 106 99


between M3 and M4 form a smaller vestibule around the intracellular orifice of the ion channel (38,40,41). Studies in which cDNA associated with specific subunits was injected into Xenopus oocytes indicated that few functional channels were produced if the a, b, or d subunit was omitted (37). Although the d subunit can substitute for the g or « subunits, AChRs are more efficiently made from a combination of a, b, d, and g or « subunits (42,43). Muscle fibers express AChRs before there is any contact between motor axons and the muscle fibers (44). AChRs collect into clusters prior to neuronal contact. Contact between a neurite growth cone and a muscle fiber enhances AChR clustering in the vicinity of the contact. Before and after the initial contacts of nerve fibers on a muscle fiber, both the fetal- and adult-type AChRs are expressed with the fetal-type receptor predominating. Maturation of the synapse leads to suppression of the g subunit and increased «-subunit expression (37). After denervation the g subunit is again expressed, and fetaltype receptors reappear on muscle fibers at and away from the endplate. Adult and fetal AChRs can be distinguished by their single-channel conductance and channel open times. Adult-type channels have shorter mean open times and a single-channel conductance that is about 50% larger than is found with fetal-type channels (44). The differences between adult and fetal AChR channels in mammals are due to replacement of the g with the « subunit (37). Phosphorylation and other forms of posttranslational modification can alter properties of the AChR (44–47). In particular, subunit phosphorylation appears to regulate agonist-induced desensitization (45,46). Electrophysiologic studies suggest that mature tonic-fiber synapses in reptiles and amphibians have both adult and fetal AChR (42,48,49). In mammals, only a subset of extraocular muscle fibers express both AChR receptor isoforms (50–52).

Ion Passage through the AChR The AChR forms a cation-selective channel that is relatively nonselective among cations compared with other voltage-gated cation-selective channels, such sodium, calcium, and potassium channels (38,53–55). The selectivity of ionic channels is based partly on the size of the pore. The AChR ion pore at its narrowest point is approximately 6.5 Å, termed the selectivity filter, which is considerably larger than that of the voltage-gated ion channels, which are highly ion selective (53). Ions do not pass through the channel freely in a bulk flow fashion. Instead ions bind to specific sites within the pore and traverse the channel by moving single file from one site to the next. To traverse the channel pore, an ion must be stripped of accompanying water molecules, which requires energy (54). In order for an ion to traverse the AChR channel, it must pass through the selectivity filter, and the free energy of the ion bound to a binding site within the channel 100

January 1999


Volume 106

must be less than or similar to the free energy of the hydrated ion. The structure of the binding sites within the pore is critical to ion passage (55). If the interaction between a binding site and ion is too weak, the ion will not pass through the channel. If the interaction is too strong, then the ion will remain fixed at the binding site and effectively block the channel. Several noncompetitive AChR inhibitors appear to block current flow by binding to and remaining at sites within the pore (56). The charge of the amino acid side chains that line the pore and that form the inner and outer vestibule contribute to the ion selectivity of the channel (54,55,57,58). The binding of ACh to the AChR likely leads to a change in pore structure that allows ions to traverse the channel (40,41). ACh binding may also alter the properties of ion binding sites within the pore (37).

ACh Binding Site

Each a subunit contains one binding site for ACh. The binding sites exist at the interface between 1) the a subunit and the d subunit and 2) the a subunit and either the g subunit or « subunit. AChR channel opening usually requires that two ACh molecules bind to the AChR (56). There is positive cooperativity for ACh binding (59). The difference in the binding affinity for the first and second ACh molecule could result from a conformation change induced by binding of the first ACh molecule or a difference in the subunits surrounding each a subunit (59 – 61). The high affinity agonist binding site is probably the a–g binding site, and the low affinity site is probably the a–d binding site. Binding of agonist to the binding sites formed by a and g subunits or a and d subunits could move the subunits to change the conformation of the entire AChR (37). Thus, the sequential binding of agonist molecules to the AChR triggers the change in channel conformation that results in channel opening. In a2b«dtype AChRs the two agonist binding sites are probably formed between one a subunit and the « subunit (highaffinity site) and the other a subunit and the d subunit (low-affinity site). The a-subunit portion of the ACh binding site is located close to a reducible disulfide bond that is formed between cysteines at positions 192 and 193. Reduction of the disulfide bond greatly alters ACh binding affinity (56). The binding site for a bungarotoxin is also located on the N-terminal region of the a subunit, close to, but distinct from, the ACh binding site (62). Amino acid substitutions in the N-terminal region can alter the binding affinity of ACh, other nicotinic agonists, or competitive inhibitors such as a bungarotoxin or d tubocurarine (63).

Ionic Pore Several experiments suggest that the M2 segments of the subunits form the ion channel. Noncompetitive AChR


inhibitors, which enter and block the AChR ionic channel, bind to specific amino acids in the M2 segments of specific subunits such as the serine in the M2 segment of the d subunit at position 262 (d 262) (64 – 66). The effects of the amino acid substitutions in M2 on the binding affinity of noncompetitive AChR inhibitors for the channel support the hypothesis that the M2 segments contribute to the lining of the ion channel (37). Amino acids in the M1 segments near the extracellular side of the membrane also line the channel. The exposure of amino acids in M1 to the channel lumen varies depending on whether the channel is open, closed, or desensitized. Therefore, the conformation of the ion pore changes based on the gating state of the channel. The data are consistent with a model in which portions of the M1 and M2 segments form a gate that occludes the channel orifice in the closed state (67– 69).

Site-Directed Mutagenesis and Ion Passage Experiments using site-directed mutagenesis demonstrated that the net amount of negative charges in the intra- and extracellular vestibules surrounding the AChR ion pore strongly affect the size of the single-channel currents. Negative charges in the vestibules attract cations, thereby increasing the concentration of current carrying ions at the pore orifices (70). Imoto et al (71) made a series of site-directed mutations of subunits in the inner and outer vestibules surrounding M2. They found that changing the charge in the inner vestibule altered the conductance for outward-flowing currents and changing the charge of the outer vestibule altered the channel conductance for inward-flowing currents. Increasing the net negative charge in a vestibule increased the single-channel current. The adult form of the bovine AChR with the « subunit has a larger single-channel conductance than the bovine channel with the g subunit. The inner vestibule of the g subunit has two net positive charges compared with one net positive charge for the « subunit. The outer vestibule of the bovine g subunit has two net positive charges compared with no net charge for the « subunit (72). The difference in single-channel conductance between the adult and fetal isoforms of the rat AChR are due to a combination of the amino acid composition of M2 and the amount of positive charge in the segments flanking M2 (73).

Effects of Site-Directed Mutagenesis on Open Time The amino acid sequence in M2 may affect the stability of the open state and hence the open time of the channel. AChR channels composed with the « subunit (adult type) have a shorter open time compared with channels composed with the g subunit (fetal type). There are two amino acid differences in the M2 segments between the g subunit and « subunit (37). The first is that an alanine in

the g subunit is replaced by a serine in the « subunit. The second difference is that a valine in the g subunit is replaced by an isoleucine. Alanine, valine, and isoleucine are nonpolar amino acids; serine is a polar amino acid. Hence, the switch of alanine for serine changes the polarity of the M2 segment, whereas the exchange of valine for isoleucine does not change the polarity of the M2 segment. Replacing a nonpolar amino acid with a polar amino acid in the M2 segment in a critical region near the cytoplasmic orifice of the channel may destabilize the open state resulting in shorter open times. When a nonpolar phenylalanine in the b-subunit M2 region is replaced by a polar serine, the open time of the channel is reduced (66). When a polar serine in the a subunit or d subunit was replaced by a nonpolar alanine, the channel open time increased (66). Though circumstantial, the evidence suggests that the polarity of amino acids near the cytoplasmic end of M2 may regulate the channel open time (74,75). Since the M1 segments contribute to the lining of the ion channel, it is not surprising that mutations in M1 can change the stability of the open state, hence the channel open time (67– 69). The intracellular segment between M3 and M4 and two amino acids in M4 produce the longer open times of AChRs composed with g subunits. The intracellular M3–M4 linker segment is shorter in the « subunit compared with the g subunit. By using chimeric subunits composed of combinations of the g and « subunits, Bouzat et al (76) showed most of the difference in the open times of AChRs formed with g or « subunits resulted from a 30-amino-acid segment in the intracellular linker between M3 and M4 and two amino acids, leucine at position 440 and methionine at position 442, in M4 of the g subunit.

Main Immunogenic Region The main immunogenic region (MIR) of the AChR is designated as such because most antibodies in clinical and experimental autoimmune myasthenia gravis bind to this region (77,78). The MIR is on the extracellular Nterminal segment of the a subunit of the AChR at positions 61 to 76 (a 61 to a 76). Each MIR is close to, but distinct from, the ACh binding site (79). The MIRs are on the upper outer portion of the extracellular domain of the AChR, which affords easy access for antibodies (80).

Safety Factor for Neuromuscular Transmission A useful concept to understand the basis of neuromuscular transmission disorders is the safety factor for neuromuscular transmission. The safety factor is defined as the ratio of the endplate potential to the difference between the membrane potential and the threshold potential for initiating an action potential (81). As long as the threshold potential is achieved, the action potential will initiate calcium release from the sarcoplasmic reticulum and normal contraction occurs. Quantal release, AChR conJanuary 1999


Volume 106 101


duction properties, AChR density, and ChE activity contribute to the endplate potential and therefore the safety factor transmission. The density of sodium channels affects the safety factor through the action potential threshold. The following disorders illustrate the significance of each of these factors.

MYASTHENIA GRAVIS Clinical Aspects The hallmark of MG is weakness that fatigues. The majority of patients present with ocular muscle manifestations, roughly equally distributed between ptosis and diplopia (82,83). Within 6 months of presentation, slightly more than 50% of patients develop generalized disease, and 75% of patients will have bulbar or extremity weakness within the first year. After 3 years of onset, only 6% of ocular myasthenics develop generalized disease. In 15% of myasthenics clinical involvement is restricted to ocular muscles (83). Despite the frequent occurrence of ocular manifestations, the presentation of MG may be extremely varied (83). Patients may present with dysarthria, dysphagia, respiratory muscle weakness, limb weakness, or a mixture of these signs and symptoms. Rarely, isolated weakness of certain muscles may occur. Similar to all autoimmune disorders, women are affected more than men in younger age groups, and the incidence of MG increases in both sexes with age (1). Approximately 15% of myasthenics harbor a thymoma. Diagnostic testing involves the edrophonium test, electromyography, and serum testing for AChR antibodies. Intravenous injection of edrophonium is most useful when improvement in ptosis or the strength of an EOM is demonstrated (84). Difficulties arise in evaluation of the test results when attempting to evaluate improvement in limb strength or bulbar function (85). False-positive edrophonium tests may occur with motor neuron disease, LEMS, intracranial mass lesions, and rarely other processes (85). False-negative tests are relatively common, and repeated tests are of value. Clinically, anti-AChR antibodies are detected by a radioimmunoassay, which uses human denervated leg muscle as an antigenic source. Antibodies are present in up to 80% of patients, but only 50% of ocular myasthenics possess anti-AChR antibodies (86). Immunoassays that use cell lines engineered to produce high concentrations of adult AChR detect AChR antibodies at higher frequencies than the standard assay (87). Interestingly, levels of AChR antibodies correlate poorly with clinical status. Antibodies to other skeletal muscle proteins are detected among myasthenics and most frequently in those with associated thymomas. The pathogenic significance of these antibodies is not clear, but surprisingly, antibodies toward the ryanodine receptor correlate better 102

January 1999


Volume 106

with clinical status than AChR antibodies (88). Patients with seronegative MG may represent a special subset of acquired autoimmune MG. Sera from seronegative myasthenics infused into mice may impair neuromuscular transmission and reduce AChR density (89). The EMG is often helpful in distinguishing MG from myopathic and neuropathic conditions. A decremental response of the compound muscle action potential (CMAP) with 5-Hz stimulation can be identified in at least 74% of myasthenics (86). Ocular myasthenics, despite lack of generalized manifestations, may demonstrate a decremental response but in a lower percentage than generalized myasthenics (1). Single-fiber EMG is the most sensitive test for detecting neuromuscular transmission abnormalities. However, single-fiber examinations should be performed only by experienced electromyographers and may be falsely positive in other neurologic conditions (86). Care of the individual myasthenic depends on the severity of weakness, age, gender, and associated medical conditions (1,78). Most patients institute some lifestyle modification to limit physical activity. Initial therapy with ChE inhibitors, usually pyridostigmine bromide, is standard. ChE inhibitors affect muscarinic synapses and therefore may lead to cholinergic side effects. Gastrointestinal complaints of nausea, vomiting, diarrhea, and cramps are most common but may be controlled with administration of atropine and glycopyrrolate. The development of ChE inhibitor–induced weakness is rarely encountered. If cholinergic weakness is seriously considered, ChE therapy should be temporarily discontinued; improvement occurs rapidly after discontinuation. Patients with generalized weakness and significantly compromised function that is not improved by ChE inhibitors warrant immune therapy (83,90,91). Prednisone is effective therapy and improves strength in the majority of patients but must be maintained for years with its concomitant side effects (78). Azathioprine is an effective treatment, alone or in combination with prednisone, but weakness may take several months to improve (1,78). Cyclosporine may be used in a similar fashion to azathioprine but usually takes effect within months. Cyclophosphamide improves strength in some patients resistant to other therapies but has worse side effects than the other agents. Thymectomy is the treatment for patients with thymoma, although such patients often have a poor response. Individuals under 55 years of age in good health likely benefit from removal of the thymus. Rates of stable remission after thymectomy vary widely, ranging from 15% (90) to 64% (92), and some reports state that significant improvement occurs in over 90% of thymectomized patients (92). Plasmapharesis and intravenous immunoglobulin (IVIg) are usually used for rapid improvement of severe


weakness, prior to thymectomy, and rarely as chronic therapies (1,78). Plasmapheresis is more consistently effective. However, its use is often restricted to specialized centers. Exchange columns specific for immunoglobulins have been used but are not clearly superior to standard columns. IVIg has been used in myasthenics with severe disease and poor responses to other treatments. IVIg therapy appears to improve strength rapidly, within 5 days of initiation, and to lower anti-AChR antibodies, but the response is short-lived and not uniformly observed.

Pathogenesis MG is one of the few diseases that fulfills strict criteria for an immune-mediated disorder. First, the presumed antigen can be administered to an animal and induce a disease similar to the human form (93). Second, antibodies may be passively transferred to induce the disease (94,95). Finally, immunoglobulin can be identified at the presumed site of the disease. IgG is found bound to the NMJ in the myasthenic, and AChR isolated from myasthenic tissue has IgG bound to it (96). AChR antibodies appear to produce the neuromuscular transmission defect in MG by 1) binding to the AChR and affecting its function, 2) accelerating the degradation rate of AChR and thereby lowering the concentration of AChR, and 3) causing complement-mediated lysis of the muscle endplate (97,98). The first mechanism appears to be relatively unimportant, but dramatic exceptions do exist that may be clinically important (99,100). Antibodies that block binding of bungarotoxin correlate with the clinical severity of MG. Monoclonal antibodies that bind to the ACh binding site produce EAMG in animals within hours of injection (101). Their effect appears to be a direct blockade of ACh-induced opening of the ion channel. Monoclonal antibodies have been produced that modify AChR electrophysiologic characteristics (99). In patients such antibodies appear to be only a small percentage of total AChR antibodies, and their effect on the endplate potential is small. Antibody can cross-link AChRs and increase their degradation rate. The ability of AChR antibody to increase degradation appears to correlate with clinical manifestations to some degree (99). The reduction of AChR density at the endplate is brought about in part by an increase in AChR degradation. Antibody can cross-link AChR and increase its degradation rate, and the ability of AChR antibody to increase degradation appears to correlate with clinical manifestations to some extent better than antibody titer (99). Degradation of AChR on cultured rat or human skeletal muscle is increased by addition of myasthenic sera to the media. AChR degradation rates are increased in EAMG animals in organ culture systems treated with myasthenic sera and in whole animals treated with myasthenic serum (102). The half-life of the fetal AChR at the

NMJ is shorter than that of the adult AChR. This may be an additional reason for the susceptibility of EOM to MG. By far the most important effect of AChR antibody is complement-mediated destruction of the NMJ. Engel and Fumagalli (98) demonstrated deposition of complement components and activation of the lytic phase at the NMJ. The C9 component of complement is associated with degraded membranous material, and the abundance of C9 correlates with destruction of junctional folds. Loss and simplification of junctional folds leads to loss of AChR-rich membrane. Alteration in the architecture of the junctional folds decreases the amount of membrane surface available for AChR insertion and may affect AChR turnover. All these factors would lead to a decrease in the safety factor for neuromuscular transmission. Sodium channels are lost from the endplate in MG due to complement-mediated endplate damage (103). The loss of endplate sodium channels reduces safety factor for neuromuscular transmission in MG. Voltage-gated sodium channels are concentrated at the endplate in the depths of the secondary synaptic folds (Figure 1) (104). The high concentration of endplate sodium channels reduces the action potential threshold at the endplate, which increases the safety factor for neuromuscular transmission by more than 50% (103,105). In muscle fibers obtained from patients with MG or rats with passively transferred MG, the density of sodium channels at the endplate was reduced (Figure 4). The action potential threshold at the endplate was larger than in normal muscle fibers. The increased action potential threshold reduced the safety factor for myasthenic fibers by about 33% (103,106). The endplate does appear to respond to the loss of AChR. Choline acetyltransferase activity and ACh release are increased at the myasthenic endplate (107). These alterations could lead to an increase in the endplate potential. The transcription of AChR subunit genes is increased in response to experimental MG in animals (108), but the increased synthesis does not appear to lead to sufficient synthesis of AChR to balance the loss produced by the disease. It is interesting to note that experimental MG does not induce transcription of the g-subunit gene (108). Therefore, MG does not appear to produce a functional denervation of muscle.

MG Immune Reaction The nature and the extent of the immune reaction in MG are beyond the scope of this review but are important to appreciate in how channel differences and the immune reaction may produce the unique clinical presentation of each patient. Anti-AChR antibodies are polyclonal IgG subclasses that bind to many different sites on the AChR (99). Investigators using monoclonal antibodies toward the AChR have defined immunogenic regions on the a subunit of the AChR. The majority of myasthenic antiJanuary 1999


Volume 106 103


Figure 4. Diagrammatic representation of the distribution of Na1 channels (circles) and AChRs (triangles) on muscle fiber membrane. Top: The density of Na1 channels is higher at the endplate compared with extrajunctional membrane. Middle: AChRs are concentrated on the tops of the secondary synaptic folds close to the nerve terminal, and Na1 channels are concentrated in the depths of the synaptic folds. Bottom: Loss of endplate membrane in MG results in simplification of the synaptic folds. Endplate membrane containing AChRs and Na1 channels is lost.

bodies and monoclonal antibodies raised against native AChR bind to the MIR. Vincent (100) identified a greater frequency of antibody binding to a particular region in patients with thymoma compared with other myasthenics. Antibodies to another ion channel, the skeletal muscle ryanodine receptor, correlate with the clinical severity of MG in patients with thymomas (88). The pathologic significance of this is not clear, especially since the normal thymus and thymomas express the ryanodine receptor (109). Variations in binding characteristics of antibodies may reflect differences in the ontogeny of the disorder in subgroups of patients and partially explain differences in clinical characteristics and response to treatment. The autoantibody synthesis in MG is under the influence of autoreactive T cells (1,78,110). Helper T cells are found in patients with MG, which can increase the production of anti-AChR antibodies, and suppressor T cells are present, which can decrease antibody production. T cells bind AChR epitopes only after processing by proteolytic cleavage and association with a major histocompatibility molecule by antigen-presenting cells.

Ion Channel Differences among Neuromuscular Junctions MG demonstrates a striking variation of weakness among muscles. The reasons for this differential involvement may partially arise from antibody specificities of individual patients (as described above), but differences in the 104

January 1999


Volume 106

ion channel properties of NMJs among muscle groups are increasingly becoming appreciated. The adult extraocular muscles persistently express the fetal AChR, which could provide a unique target for immune-mediated damage of these muscles. More subtle differences in AChR function exist between fast and slow muscle fibers, as evidenced by variations in ACh sensitivity. Sodium channel density differs among muscles and likely contributes to variations in the safety factor of junctions (111). Several investigations have identified fetal and adult AChR expression in mature EOM (Figure 6) (50,52, 87,112). The majority of studies are consistent with EOM being the only muscle that expresses the fetal AChR at adult NMJs (82,113). The presence of fetal-type channels at certain EOM synapses may facilitate tonic force generation, because the channels have a longer mean open time, which would tend to produce a more uniform spread of depolarization along the fiber. Evidence exists that the immune process of MG may specifically target EOM synapses. Some studies demonstrate that ocular myasthenics’ sera reacts more strongly when human EOM is used as a source of antigen (114,115). Some seronegative ocular myasthenics are positive when tested against EOM-derived antigen (82). Serum from myasthenic patients contains antibodies that react selectively with fetal AChR (100,102). Whether fetal antibodies correlate with EOM weakness among myasthenics has not been clearly established (87). However, T cells from myasthenics react against fetal AChR epitopes, and the reaction correlates with ocular manifestations (116,117). The linkage of fetal AChR expression and antibody specificity appears particularly important for neonatal MG. Neonatal MG refers to the development of weakness, usually transient, among infants born to myasthenic mothers. Neonatal MG does not correlate with the severity of clinical manifestations of the mother or absolute levels of AChR antibodies, but the development of neonatal MG correlates with higher ratios of fetal to adult AChR antibodies (118). Presence of fetal AChR antibodies also is associated with arthrogryposis multiplex congenita, and administration of fetal AChR antibodies to pregnant mice reproduces the disorder in the offspring (119).

CONGENITAL MYASTHENIAS PRODUCED BY ACHR MUTATIONS Congenital myasthenias are clinically similar to immunemediated MG but are caused by genetic defects of the presynaptic or postsynaptic apparatus. The underlying abnormalities in these disorders have been variably characterized, and mutations of a number of synaptic proteins could be hypothesized to produce a myasthenic condition (120). The following section emphasizes de-


fects of the AChR that provide particular insight into structure–function relationships. For more detailed discussion of other congenital myasthenias, see Vincent et al (120), Engel (3) or Kaminski and Ruff (1). Patients with congenital myasthenia have fatiguing weakness, and electromyography (EMG) often reveals a decremental response to repetitive stimulation. However, antibodies toward the AChR are not present. Manifestations of the illness may be traced to birth, and symptoms do not respond to immunosuppressive therapies. Congenital myasthenias should be differentiated from neonatal MG, which is caused by maternal transfer of anti-AChR antibodies across the placenta (121).

Slow Channel Syndrome Patients with slow channel syndrome have intermittent weakness on the backdrop of a slowly progressive myopathy. They can present from infancy to adulthood and most often demonstrate an autosomal dominant mode of inheritance (122,123). Involvement is usually greater in the arms and trunk compared with the legs. Weakness does not improve and may worsen with anti-ChE medications. Repetitive stimulation produces a decremental response, and a single electrical stimulation leads to repetitive CMAPs. MEPP duration is prolonged with reduced amplitudes. The EMG and in vivo studies resemble those of ChE deficiency. However, ChE stains reveal the presence of functional ChE. Therefore, the prolongation of the MEPP most likely results from an increase of the open time of the AChR. An increased open time would prolong the depolarization of the membrane beyond the refractory period for action potential generation and produce the observed repetitive CMAPs. In patients with slow channel syndrome, single nucleotide mutations have been found in the a, b, and « subunits of the AChR (Figure 5). The majority of mutations lie within the pore-forming regions of the channel and appear to stabilize the open state of the channel. Croxen et al (124) described an a-subunit mutation that prolongs channel open time by a glycine to serine substitution at amino acid 153 that reduces the dissociation of ACh from the binding pocket and therefore increases the number of channel reopenings. Another point mutation at position 156 leads to a substitution of valine for a methionine that appears to stabilize the open state. These findings are consistent with mutational analysis that demonstrates this region of the AChR around positions 153 to 156 in the initial extracellular loop being involved in both ACh binding and open channel stabilization. A mutation that lies between the M2 and M3 domains in the extracellular loop also prolongs channel open time, but the molecular mechanism by which this occurs is not clear. Gomez et al (125) described a mutation in the M2 region of the b subunit that produced a methionine for leucine substitution. The substitution occurred in the re-

Figure 5. Axial cross section of AChR. The M2 domains of a and « subunits are shown lining the ion pore. The solid ring indicates the location of leucine side chains that are thought to form the channel gate. Known mutations that produce slowchannel syndrome are shown at their presumed locations in the quaternary structure of the AChR. The subunit affected, the native amino acid’s single letter code and position in the amino acid sequence, and the resulting amino acid substitution are indicated.

gion of the AChR pore that Unwin (74,75) proposed forms a ring of leucine side chains among the subunit M2 regions and is critical in stabilization of the open state. Ohno et al (126) described a mutation of the «-subunit M2 region that substituted a threonine with a proline residue. This substitution should disrupt the a-helical structure of M2. The substitution resulted in very long channel openings. Another «-subunit phenylalanine for leucine substitution in the M2 region (although outside the leucine ring) produced a prolonged channel open time and spontaneous openings in the absence of ACh.

High-Conductance Fast Channel Syndrome A nine-year-old girl had delayed motor development and fatiguing weakness since infancy, and a sibling had similar symptoms (127). Her symptoms responded partially to anti-ChE medication, and an EMG revealed a decremental response but only after exercise. Routine histochemical studies of the muscle were normal, but electron micrographs revealed a significant reduction in area of the primary and secondary junctional folds. The density of AChR was normal, and ChE was present at the endplate. MEPP amplitude was large, and the rate of decay was fast. Channel conductance was increased, and open time was reduced. The most likely explanation for the abnormalities in the patient is a mutation of the AChR. The increase in conductance may be explained by an increase in the net negative charge of the surface of the January 1999


Volume 106 105


external surface of the AChR channel. The shorter channel open time could result if a mutation destabilized the open state conformation of the AChR.

Short Channel Open Time and AChR Deficiency Engel et al (128) described a patient with an AChR deficiency and decreased open time. At birth the girl required mechanical ventilation, was hypotonic, and diffusely weak. Fatiguing weakness improved with anti-ChE medications, but motor development was delayed. EMG was consistent with a myasthenic disorder. Morphologic studies demonstrated an enlargement of the endplate region, and some junctional folds were simplified with a reduction of AChR density. MEPP amplitude was low but increased with neostigmine treatment. The AChR open time was decreased, and the conductance was normal. The reduced MEPP amplitude was probably caused by a reduced number of AChRs. The shorter channel open time and MEPP duration suggest a mutation that destabilized the open state. The abnormality of the AChR in this patient also led to a decrease in AChR density, which could have been due to an increase in degradation or decreased synthesis.

AChR Deficiency Caused by «-Subunit Mutations Engel et al (129) described two patients with electrophysiologic abnormalities consistent with a mutation of the « subunit. Patients had generalized weakness that dated to early childhood, although one patient was diagnosed with a myasthenic disorder in their fourth decade. Both responded somewhat to cholinesterase inhibitors and had a decremental response to repetitive stimulation. The endplate morphology of one patient was normal, but the other had many degenerating junctional folds and mitochondria, and necrotic nuclei. Both patients had reductions of AChR density and MEPP amplitude. One patient had MEPP decay times and noise analysis data indicating the presence of two channel types, one with a normal channel open time and the other with a markedly prolonged channel open time. Analysis of channel characteristics of the other patient indicated a single channel with a markedly prolonged open time. In both patients the channel conductance was decreased. Structural analysis using antibodies directed toward epitopes on the AChR suggested alteration of the cytoplasmic loop of the «-subunit between the M3 and M4 membrane-spanning domains. The distinction between a mutation lying in this region or altering tertiary structure of this area could not be determined using the experimental methods. From mutagenesis studies, it is not clear how an abnormality of this region would produce the observed channel characteristics. These patients also had a significant reduction in AChR density. Mutations of amino acid residues 106 and 115 of the rat and mouse « subunit led an alteration of AChR assembly and a decrease in expression of AChR on 106

January 1999


Volume 106

the cell surface (130). Perhaps, the mutation in « subunit of these patients also affected AChR assembly. Ohno et al (131) described three patients with nonsense and mis-sense mutations in each of their «-subunit genes. The mutations led to a reduction of endplate AChR expression. The non-sense mutations prevented expression of full-length transcripts, and mis-sense mutations inhibited incorporation of «-subunit– containing channels into the membrane. In two of the patients, persistent expression of the fetal AChR and multiple endplates were identified that appeared to compensate for the reduction of adult AChRs.

Abnormality of ACh and AChR Interaction Since birth, a 21-year-old had severe weakness with fatigue that responded poorly to anti-ChE (132). EMG revealed a decremental response to 2-Hz stimulation. MEPP amplitudes were decreased, but no morphologic abnormalities of the endplate, reduction in AChR density, or synaptic vesicle abnormality was identified to explain the MEPP amplitude reduction. Single-channel conductance was normal, but channel closure was complex with two time constants. A reduced affinity of the receptor for ACh could explain the decreased MEPP in the setting of normal AChR conductance and a normal number of AChRs, and the improvement with cholinesterase inhibitors. Mutational analysis of the Torpedo and mouse AChR demonstrates the replacement of cystine residues 192 or 193 of the a subunit by a serine or tyrosine 190 by phenylalanine would lead to a reduction of ACh affinity by 10- to 50-fold (132,133). The ACh binding site appears to be close to peptide loops that include tryptophan 86 and 149 and tyrosine 93 and 151. This syndrome could be explained by mutations in the loops that form the ACh binding site but do not inhibit a-bungarotoxin binding.

AChR with Altered d-Tubocurarine Binding Morgan-Hughes et al (134) described an adult with 14 months of fluctuating weakness. An EMG showed a slight decrement in CMAP with slow stimulation. Light microscopic analysis revealed tubular aggregates in the muscle. Some endplates revealed reduced synaptic vesicles and short, irregular junctional folds. The AChR density was reduced. The AChR affinity for binding of d tubocurarine was increased compared to normal and myasthenic controls. Presumably, a change in AChR structure led to altered sensitivity to d tubocurarine and a reduction of AChR density and secondary alterations of the endplate. Mutagenesis studies have demonstrated that binding of antagonists may be modified without altering ACh binding (133).

Endplate Myopathy Many congenital myasthenias demonstrate a degeneration of the postsynaptic region and simplification of junc-


tional folds often with a concomitant reduction of AChRs. The reduction of postsynaptic area likely leads to a reduction of sodium channels. However, this possibility has not been investigated. The cause of the membrane damage has been hypothesized to result from the prolonged depolarization of the endplate that would be produced by an increase in the channel open time of the AChR or ChE deficiency (3). The AChR is permeable to calcium, and up to 6% of the inward current may be passed by calcium ions. Mechanisms that prolong the time or frequency of channel openings would increase calcium influx. Excess calcium could lead to activation of calcium-sensitive proteases and inhibition of mitochondrial respiration. Nitric oxide synthase that is concentrated at the NMJ could also lead to free radical damage of the endplate (22). Prolonged depolarization of the endplate region may also lead to inactivation of sodium channels that are concentrated in the depths of the junctional folds.

LAMBERT-EATON MYASTHENIC SYNDROME Weakness and fatigeability of the trunk and proximal limb muscles characterize LEMS. Oropharyngeal muscles may be affected but not as prominently as in MG. Ocular muscle involvement occurs in more than 50% of patients (135). Autonomic nervous system involvement is common with decreased salivation, impaired lacrimation, reduced sweating, orthostatic hypotension, and impotence. Impaired pupillary constriction may be evident. Respiratory muscle involvement with respiratory failure is uncommon but may occur (136,137). The characteristic feature of the weakness in LEMS is that strength is most severely reduced in rested muscles, and strength increases over a period of seconds if the patient is able to initiate voluntary contraction. Paradoxical lid elevation after sustained upgraze may be evident and differentiate the condition from MG (138). Tendon reflexes may be diminished or absent in rested muscles and then increased after a brief maximal voluntary contraction (135). The diagnosis of LEMS is made based on clinical and EMG data with serologic testing adding support (139). A high suspicion of an underlying malignancy, particularly small-cell lung cancer, should be maintained in patients with LEMS 1) above the age of 40, 2) who have significant risk factors for lung cancers, 3) when symptoms have been presented for less than 2 years (140), and 4) in patients who complain of a severe deep aching pain in association with the proximal muscle weakness and fatigeability and autonomic impairment. LEMS occurs most frequently in mid to late life and is more frequent in men in most series (135,141,142). Most LEMS is associated with an underlying malignancy, but

spontaneous, autoimmune LEMS occurs and is more often observed in younger women (135,143). Rare patients may have both LEMS and MG (143). Small-cell lung cancers are responsible for 80% of paraneoplastic LEMS (135,142,144), but numerous other cancers have been associated (140). The incidence of LEMS in association with small-cell lung cancer has been estimated at 6%, but the true incidence is unknown since LEMS often goes unappreciated.

Pathophysiology EMG shows low amplitude motor unit potentials that increase in amplitude with continued activity. The amplitude of the CMAP evoked by a single nerve stimulus to rested muscle is extremely small. After a brief period of maximal voluntary activity, the CMAP amplitude increases. Stimulation at 10 Hz or higher frequencies usually produces a progressive increase of the CMAP amplitude. Similar stimulation in normal subjects has little effect on the CMAP amplitude (1). In contrast, stimulation at about 5 to 10 Hz in patients with MG produces a decremental response. Singlefiber EMG shows increase jitter and frequent blocking. The abnormalities improve after stimulation or voluntary contraction. In contrast in MG, blocking increases with activity or stimulation (145). The neuromuscular transmission defect in LEMS is that two few synaptic vesicles are released in response to nerve stimulation so that the quantal content of the endplate potential is abnormally low (1). The relationship between the number of vesicles released in response to nerve stimulation and extracellular calcium is disturbed in LEMS, indicating calcium entry into the nerve terminals is compromised (146). MEPP amplitudes are normal, indicating that the amount of ACh in the synaptic vesicles and the postsynaptic sensitivity are normal in LEMS (141,147). The incremental response to repetitive stimulation seen in LEMS can be understood in the following context (3). With repetitive stimulation, two competing forces act on the nerve terminal. Stimulation depletes the pool of readily releaseable synaptic vesicles. This depletion effect reduces transmitter release by reducing the number of vesicles that are released in response to a nerve terminal action potential. The opposite is that with repeated stimulation, calcium can accumulate within the nerve terminal, thereby increasing the probability that a vesicle will fuse with the nerve terminal membrane. In a normal nerve terminal, the effect of depletion of readily releaseable synaptic vesicles predominates, so with repeated stimulation the number of vesicles released or quantal content will decrease. In LEMS very few vesicles are released, so depletion of vesicles is not a prominent effect. With rapid repetitive stimulation, the calcium influx is enhanced and calcium concentration in the nerve terminal can rise high enough to stimulate synaptic vesicle fusion. A sufficient number of synaptic vesicles releases January 1999


Volume 106 107


their contents to result in a supra-threshold endplate potential. Since calcium diffuses from the nerve terminal in 100 to 200 msec, stimulation at a frequency lower than 5 Hz will allow enough time for calcium to diffuse from the nerve terminal so significant calcium accumulation does not occur. Therefore, slow repetitive stimulation results in further decrement of the CMAP amplitude in LEMS (55,57). The autoimmune etiology of LEMS is suggested by 1) the response of the nonneoplastic form to glucocorticoids (148), 2) nonneoplastic LEMS is associated with other autoimmune disorders (135,149,150), 3) plasmapheresis and immunosuppressive therapy improve both neoplastic and nonneoplastic LEMS (151), and 4) IgG from LEMS patients injected into rodents reproduces the electrophysiologic features of LEMS (149). A striking structural feature of LEMS is that the number of active zones in the density of active zone particles are markedly reduced (152). Injection of IgG from LEMS patients into mice also results in depletion of active zone particles (153). The active zone particles are thought to be associated with the calcium channels responsible for transmitter release or possibly the calcium channels themselves (154). The LEMS IgG binds to active zones at the nerve terminal (97,155,156). The antigenic modulation that results in depletion of the active zone particles requires that the LEMS IgG be able to crosslink active zone particles (157). As was true for AChRs and MG, crosslinking of channels by antibodies results in clustering of the active zone particles (153) followed by a depletion of active zone particles, which probably results from accelerated internalization of the calcium channels that is insufficiently compensated by increased channel production in the nerve cell (97,155,156). Injection of LEMS IgG into mice leads to impaired transmitter release. LEMS IgG also impairs calcium channel function of cultured cells from small-cell lung cancers (158) and other tissues (159,160). LEMS antibodies differ in their abilities to block or bind subtypes for calcium channels. The LEMS IgG binds directly to several classes of calcium channels, including N-type (161–164), L-type (161,165,166), and P-type channels (161). Thus far, no effects on T-type channels have been demonstrated. The P-type channels are those reconfined to NMJ, and therefore, the antibodies directed against P-type calcium channels may be most important in compromising transmitter release (161,167,168). P-type channels are most likely those expressed on small-cell lung carcinomas (161,169). Lennon et al (170) found that the P-type VGCC antagonists were the most potent inhibitors of calcium influx in cultured small-cell lung cancer lines, and anti-P-type VGCC antibodies were present in all patients and in 91% with LEMS without cancer. At least a fraction of the antibodies recognize the b subunit of the calcium channel, which exists only on the cytoplasmic surface (171). Anti108

January 1999


Volume 106

bodies against other synaptic proteins are found in LEMS patients, but their pathogenic significance is not certain (172–174). Mechanisms that initiate the autoimmune reaction in LEMS are still not well understood. In paraneoplastic LEMS associated with small-cell lung cancer, the most plausible explanation is induction of antibodies to calcium channels on the cancer that crossreact with channels at the NMJ. In LEMS associated with cancer, treatment of the underlying tumor may improve symptoms (175). Cholinesterase inhibitor drugs may improve strength (176). Guanidine inhibits uptake of calcium by subcellular organelles, such as mitochondria, which raise the intracellular calcium concentration and increase the release of ACh from the nerve terminal (177). Guanidine increases the quantal content of the endplate potential (177) and can improve symptoms in LEMS (135,176). However, guanidine has a poor side-effect profile (178). DAP prolongs the duration of the presynaptic action potential by blocking delayed rectifier potassium channels (179). The prolonged action potential increases calcium entry into the nerve terminal, which increases quantal release (32). DAP in doses of up to 100 mg per day can relieve the motor and autonomic symptoms of LEMS (180). The concurrent administration of pyridostigmine produces greater symptomatic improvement and permits reduction of DAP dose. The side effects of DAP consist of transient perioral and digital parasthesia. Seizures have been reported in patients who are taking the dose of 100 mg/day (180).

SUMMARY This article discussed disorders of the NMJ. It demonstrated that basic understanding of receptor physiology advances knowledge of diseases. Careful analysis of the impact of disease-associated mutations in acetylcholine receptor subunits produced novel insights on the structure–function relationships of the acetylcholine receptor. The autoimmune disorders myasthenia gravis and Lambert-Eaton myasthenic syndrome illustrate different ways that the complex interactions among receptors, ion channels, and complex intracellular processes can be disrupted in antibody-mediated diseases. Diseases of the NMJ are of general interest, because an understanding of their pathophysiology may provide insight into disorders at other neural synapses, such as epileptic or psychiatric diseases.

REFERENCES 1. Kaminski HJ, Ruff RL. The myasthenic syndromes. In: Schultz SG, Andreoli TE, Brown A, et al, eds. Molecular Biology of Membrane Transport Disorders. New York: Plenum Publishing, 1996;565– 594. 2. Kaminski HJ, Ruff RL. Congenital disorders of neuromuscular transmission. Hosp Prac. 1992;39:73– 86.

Neuromuscular Junction Ion Channels/Boonyapisit et al 3. Engel A. Myasthenic syndromes. In: Engel A, Franzini-Armstrong C, eds. Myology. New York: McGraw-Hill, 1994;1798 –1835. 4. Black JA, Kocsis JD, Waxman SG. Ion channel organization of the myelinated fiber. Trends Neurosci. 1990;13:48 –54. 5. Salpeter MM. The Vertebrate Neuromuscular Junction. New York: Alan R. Liss, 1987. 6. Augustine GJ, Adler EM, Charlton MP. The calcium signal for transmitter secretion from presynaptic nerve terminals. Ann NY Acad Sci. 1991;635:365–381. 7. Smith SJ, Augustine GJ. Calcium ions, active zones and synaptic transmitter release. Trends Neurosci. 1988;10:458 – 464. 8. Gautam M, Noakes PG, Mudd J, et al. Failure of postsynaptic specialization to develop at neuromuscular junctions of rapsyndeficient mice. Nature. 1995;377:232–236. 9. Flucher BE, Daniels MP. Distribution of Na1 channels and ankyrin in neuromuscular junctions is complementary to that of acetylcholine receptors and the 43 kD protein. Neuron. 1989;3: 163–175. 10. Sanes JR, Johnson YR, Kotzbauer PT, et al. Selective expression of an acetylcholine receptor-lacZ transgene in synaptic nuclei of adult muscle fibers. Development. 1991;113:1181–1191. 11. Martinou JC, Falls DI, Fischback GD, Merlie JP. Acetylcholine receptor-inducing activity stimulates expression of the epsilonsubunit gene of the muscle acetylcholine receptor. Proc Natl Acad Sci USA. 1991;88:7669 –7673. 12. Merlie JP, Sanes JR. Concentration of acetylcholine receptor mRNA in synaptic regions of adult muscle fibers. Nature. 1985; 317:66 – 68. 13. Land BR, Salpeter EE, Salpeter MM. Kinectic parameters for acetylcholine interaction in intact neuromuscular junction. Proc Natl Acad Sci USA. 1981;78:7200 –7204. 14. Kuffler SW, Yoshikami D. The distribution of acetylcholine sensitivity at the post-synaptic membrane of vertebrate skeletal twitch muscles: iontophoretic mapping in the micron range. J Physiol (Lond). 1975;244:703–730. 15. Haimovich B, Schotland DL, Fieles WE, Barchi RL. Localization of sodium channel subtypes in rat skeletal muscle using channelspecific monoclonal antibodies. J Neurosci. 1987;7:2957–2966. 16. Ruff RL, Whittlesey D. Na1 current densities and voltage dependence in human intercostal muscle fibres. J Physiol (Lond). 1992; 458:85–97. 17. Colquhoun D, Sakmann B. Fast events in single-channel currents activated by acetylcholine and its analogues at the frog muscle end-plate. J Physiol (Lond). 1985;369:501–557. 18. Gee SH, Montanaro F, Lindenbaum MH, Carbonetto S. Dystroglycan-a, a dystrophin-associated glycoprotein, is a functional agrin receptor. Cell. 1994;77:675– 686. 19. Campanelli JT, Roberds SL, Cambell KP, Scheller RH. A role for dystrophin-associated glycoproteins and utrophin in agrin-induced AChR clustering. Cell. 1994;77:663– 674. 20. Sealock R, Froehner SC. Dystrophin-associated proteins and synapse formation: is a-dystroglycan the agrin receptor? Cell. 1994; 77:617– 619. 21. Brenman JE, Chao DS, Gee SH, et al. Interaction of nitric oxide synthase with the postsynaptic density protein PSD-95 and a1syntrophin mediated by PDZ domains. Cell. 1996;84:757–767. 22. Kusner LL, Kaminski HJ. Nitric oxide synthase is concentrated at the skeletal muscle endplate. Brain Res. 1996;730:238 –242. 23. Hall ZW, Sanes JR. Synaptic structure and development: the neuromuscular junction. Cell. 1993;72:99 –121. 24. Kleiman RJ, Reichardt LF. Testing the agrin hypothesis. Cell. 1996; 85:461– 464. 25. Gautam M, Noakes PG, Moscoso L, et al. Defective neuromuscular synaptogenesis in agrin-deficient mutant mice. Cell. 1996;85: 525–535.

26. Cherksey BD, Sugimori M, Llinas RR. Properties of calcium channels isolated with spider toxin, FTX. Ann NY Acad Sci. 1991;635: 80 – 89. 27. Nowycky MC, Fox AP, Tsien RW. Three types of neuronal calcium channel with calcium agonist sensitivity. Nature. 1985;316: 440 – 443. 28. Nowycky MC. Two high-threshold Ca21 channels contribute Ca21 for depolarization-secretion coupling in the mammalian neuropophysis. Ann NY Acad Sci. 1991;635:45–57. 29. Llinas RR. Depolarization release coupling: an overview. Ann NY Acad Sci. 1991;635:3–17. 30. Stanley EF. Presynaptic calcium channels and the transmitter release mechanism. Ann NY Acad Sci. 1993;681:368 –372. 31. Lindgren CA, Moore JW. Calcium current in motor nerve endings of the lizard. Ann NY Acad Sci. 1991;635:58 – 69. 32. Miledi R, Molenaar PC, Polak RL. Electrophysiological and chemical determination of acetylcholine release at the frog neuromuscular junction. J Physiol (Lond). 1983;334:245–254. 33. Etcheberrigaray R, Fielder JL, Pollard HB, Rojas E. Endoplasmic reticulum as a source of Ca21 in neurotransmitter secretion. Ann NY Acad Sci. 1991;635:90 –99. 34. Land BR, Harris WV, Salpeter EE, Salpeter MM. Diffusion and binding constants for acetylcholine derived from the falling phase of miniature endplate currents. Proc Natl Acad Sci USA. 1984;81: 1594 –1598. 35. McMahan UJ, Sanes JR, Marshall LM. Cholinesterase is associated with the basal lamina at the neuromuscular junction. Nature. 1978;271:172–174. 36. Hall ZW. The nerve terminal. In: Hall ZW, ed. An Introduction to Molecular Neurobiology. Sunderland, MA: Sinauer Associates, 1992:148 –180. 37. Kaminski HJ, Ruff RL. Insights into possible skeletal muscle nicotinic acetylcholine receptor (AChR) changes in some congenital myasthenias from physiological studies, point mutations, subunit substitutions of the AChR. Ann NY Acad Sci. 1993;681:435– 450. 38. Guy HR, Hucho F. The ion channel of nicotinic acetylcholine receptor. Trends Neurosci. 1987;10:318 –322. 39. Chavez RA, Hall ZW. Expression of fusion proteins of the nicotinic acetylcholine receptor from mammalian muscle identifies the membrane-spanning regions in the a and d subunits. J Cell Biol. 1992;116:385–393. 40. Toyoshima C, Unwin N. Ion channel of acetylcholine receptor reconstructed from images of postsynaptic membranes. Nature. 1988;336:247–250. 41. Unwin N, Toyoshima C, Kubalek E. Arrangement of the acetylcholine receptor subunits in the resting and desensitized states, determined by cryoelectron microscopy of crystallized torpedo postsynaptic membranes. J Cell Biol. 1988;107:1123–1138. 42. Henderson LP, Brehm P. The single-channel basis for the slow kinetics of synaptic currents in vertebrate slow muscle fibers. Neuron. 1989;2:1399 –1405. 43. Brehm P, Henderson L. Regulation of acetylcholine receptor channel function during development of skeletal muscle. Dev Biol. 1988;129:1–11. 44. Brehm P. Resolving the structural basis for developmental changes in muscle ACh receptor function: it takes nerve. Trends Neurosci. 1989;12:174 –177. 45. Qu Z, Moritz E, Huganir RL. Regulation of tyrosine phosphorylation of the nicotinic acetylcholine receptor at the rat neuromuscular junction. Neuron. 1990;2:367–378. 46. Safran A, Provenzano C, Sagi-Eisenberg R, Fuchs S. Phosphorylation of membrane-bound acetylcholine receptor by protein kinase C: characterization and subunit specificity. Biochem. 1990;29: 6730 – 6734. 47. Rozental R. In vitro denervation of frog skeletal muscle: expresJanuary 1999


Volume 106 109


48.

49. 50.

51.

52.

53.

54.

55. 56.

57.

58.

59.

60. 61.

62.

63.

64.

65.

66.

67.

68. 110

sion of several conductance classes of nicotinic receptors. Neurosci Lett. 1991;133:65– 67. Dionne VE. Two types of nicotinic acetylcholine receptor channels at slow fibre end-plates of the garter snake. J Physiol. 1989;409: 313–331. Ruff RL, Spiegel P. Ca sensitivity and AChR currents of twitch and tonic snake muscle fibers. Am J Physiol. 1990;259:C911–C919. Kaminski HJ, Kusner LL, Block CH. Expression of acetylcholine receptor isoforms at extraocular muscle endplates. Invest Ophthalmol Vis Sci. 1996;37:345–351. Kaminski HJ, Kusner LL, Nash KV, Ruff RL. The g-subunit of the acetylcholine receptor is not expressed in the levator palpebrae superioris. Neurology. 1995;45:516 –518. Horton RM, Manfredi AA, Conti-Tronconi BM. The “embryonic” gamma subunit of the nicotinic acetylcholine receptor is expressed in adult extraocular muscle. Neurology. 1993;43:983–986. Dwyer TM, Adams DJ, Hille B. The permeability of the endplate channel to organic cations in frog muscle. J Gen Physiol. 1980;75: 469 – 492. Dani JA, Eisenman G. Monovalent and divalent cation permeation in acetylcholine receptors channel. J Gen Physiol. 1987;89: 959 –983. Ruff RL. Ionic channels I. The biophysical basis for ion passage and channel gating. Muscle Nerve. 1986;9:675– 699. Karlin A, Kao PN, DiPaola M. Molecular pharmacology of the nicotinic acetylcholine receptor. Trends Pharmacol. 1986;4:304 – 308. Ruff RL. Ionic channels II. Voltage- and agonist-gated and agonist-modified channel properties and structure. Muscle Nerve. 1986;9:767–786. Dani JA. Site-directed mutagenesis and single-channel currents define the ionic channel of the nicotinic acetylcholine receptor. Trends Neurosci. 1989;12:125–130. Sigworth FJ, Sine SM. Data transformations for improved display and fitting of single-channel dwell time histograms. Biophys J. 1987;52:1047–1054. Sine SM, Steinbach JH. Activation of a nicotinic acetylcholine receptor. Biophys J. 1984;45:175–185. Auerbach AB, Sachs F. Single channel currents from acetylcholine receptors in embryonic chick muscle. Kinetic and conductance properties of gaps within bursts. Biophys J. 1984;45:187–198. McLaughlin J, Hawrot E, Yellen G. Covalent modification of engineered cysteines in the nicotinic acetylcholine receptor agonistbinding domain inhibits receptor activation. Biochem J. 1995;310: 765–769. Neumann D, Barchan D, Horowitz M, Kochva E, Fuchs S. Snake acetylcholine receptor: cloning of the domain containing the four extracellular cysteines of the a-subunit. Proc Natl Acad Sci USA. 1989;86:7255–7259. Giraudat J, Dennis M, Heidmann T, Chang J-Y, Changeux J-P. Structure of the high-affinity binding site for noncompetitive blockers of the acetylcholine receptor: serine-262 of the d subunit is labeled [3H]chlorpromazine. Proc Natl Acad Sci USA. 1986;83: 2719 –2723. Hucho F, Oberthu¨r W, Lottspeich F. The ion channel of the nicotinic acetylcholine receptor is formed by the homologous helices M II of the receptor subunits. FEBS Lett. 1986;205:137–142. Leonard RJ, Labarca CG, Charnet P, Davidson N, Lester HA. Evidence that the M2 membrane-spanning region lines the ion channel pore of the nicotinic receptor. Science. 1988;242:1578 –1581. Akabas MH, Stauffer DA, Xu M, Karlin A. Acetylcholine receptor channel structure probed in cysteine-substitution mutants. Science. 1992;258:307–310. Akabas MH, Karlin A. Identification of acetylcholine receptor January 1999


Volume 106

69.

70.

71.

72.

73.

74. 75. 76.

77.

78. 79. 80.

81.

82. 83.

84. 85. 86.

87.

88.

89.

90.

channel-lining residues in the M1 segment of the a-subunit. Biochem. 1995;34:12496 –12500. Karlin A, Akabas MH. Toward a structural basis for the function of nicotinic acetylcholine receptors and their cousins. Neuron. 1995; 15:1231–1244. Imoto K, Methfessel C, Sakmann B, et al. Location of a d subunit region determining ion transport through the acetylcholine receptor channel. Nature. 1986;324:670 – 674. Imoto K, Busch C, Sakmann b, et al. Rings of negatively charged amino acids determine the acetylcholine receptor channel conductance. Nature. 1988;305:645– 648. Murray N, Zheng Y-C, Mandel G, et al. A single site on the « subunit is responsible for the change in ACh receptor channel conductance during skeletal muscle development. Neuron. 1995; 14:865– 870. Herlitze S, Villarroel A, Witzemann V, Koenen M, Sakmann B. Structural determinants of channel conductance in fetal and adult rat muscle acetylcholine receptors. J Physiol (Lond). 1996;492: 775–787. Unwin N. Acetylcholine receptor channel imaged in the open state. Nature. 1995;373:37– 43. Unwin N. Nicotinic acetylcholine receptor at 9 Å resolution. J Mol Biol. 1993;229:1101–1124. Bouzat C, Bren N, Sine SM. Structural basis of the different gating kinetics of fetal and adult acetylcholine receptors. Neuron. 1994; 13:1395–1402. Lindstrom JM, Seybold MD, Lennon VA, Whittingham S, Duane DD. Antibody to acetylcholine receptor in myasthenia gravis: prevalence, clinical correlates, and diagnostic value. Neurology. 1976;26:1054 –1059. Drachman DB. Myasthenia gravis. NEJM. 1994;330:1797–1810. Karlin A. Explorations of the nicotinic acetylcholine receptor. Harvey Lect Ser. 1991;85:71–107. Beroukhim R, Unwin N. Three-dimensional location of the main immunogenic region of the acetylcholine receptor. Neuron. 1995; 15:323–331. Banker BQ, Kelly SS, Robbins N. Neuromuscular transmission and correlative morphology in young and old mice. J Physiol (Lond). 1983;339:355–375. Kaminski H. Acetylcholine receptor epitopes in ocular myasthenia. Ann NY Acad Sci. 1998;841:309 –319. Grob D, Arsura EL, Brunner NG, Namba T. The course of myasthenia gravis and therapies affecting outcome. Ann NY Acad Sci. 1987;505:472– 499. Daroff RB. The office tensilon test for ocular myasthenia gravis. Arch Neurol. 1986;43:843– 844. Oh SJ, Cho HK. Edrophonium responsiveness not necessarily diagnostic of myasthenia gravis. Muscle Nerve. 1990;13:187–191. Oh SJ, Kim DE, Kuruoglu R, Bradley RJ, Dwyer D. Diagnostic sensitivity of the laboratory tests in myasthenia gravis. Muscle Nerve. 1992;15:720 –724. MacLennan C, Beeson D, Buijs A-M, Vincent A, Newsom-Davis J. Acetylcholine receptor expression in human extraocular muscles and their susceptibility to myasthenia gravis. Ann Neurol. 1997;41: 423– 431. Mygland A, Aarli J, Matre R, Gilhus N. Ryanodine receptor antibodies related to severity of thymoma associated myasthenia gravis. J Neurol Neurosurg Psychiatr. 1994;57:843– 846. Mossman S, Vincent A, Newsom-Davis J. Myasthenia gravis without acetylcholine receptor antibody: a distinct disease entity. Lancet. 1986;i:116 –119. Beghi E, Antozzi C, Batocchi AP, et al. Prognosis of myasthenia gravis: a multicenter follow-up study. J Neurol Sci. 1991;106:213– 220.

Neuromuscular Junction Ion Channels/Boonyapisit et al 91. Rowland LP. Controversies about the treatment of myasthenia gravis. J Neurol Neurosurg Psychiatr. 1980;43:644 – 659. 92. Olanow CW, Wechsler AS, Sirotkin-Roses M, Stajich J, Roses AD. Thymectomy as primary therapy in myasthenia gravis. Ann NY Acad Sci. 1987;505:595– 606. 93. Patrick J, Lindstrom J. Autoimmune response to acetylcholine receptor. Science. 1973;180:871– 872. 94. Lindstrom JM, Einarson BL, Lennon VA, Seybold ME. Pathological mechanisms in experimental autoimmune myasthenia gravis. I. Immunogenicity of syngeneic muscle acetylcholine receptor and quantitative extraction of receptor and anti-receptor complexes from muscle of rats with experimental autoimmune myasthenia gravis. J Exp Med. 1976;144:726 –738. 95. Toyka KV, Drachman DB, Griffin DE, Pestronk D. Myasthenia gravis study of humoral immune mechanisms by transfer to mice. NEJM. 1977;296:125–131. 96. Engel AG, Lambert EH, Howard FM. Immune complexes (IgG and C3) at the motor end-plate in myasthenia gravis. Ultrastructure and light microscopic localization and electrophysiological correlations. Mayo Clin Proc. 1977;52:267–280. 97. Engel AG. The molecular biology of end-plate diseases. In: Salpeter MM, ed. The Vertebrate Neuromuscular Junction, vol 23. New York: Alan R. Liss, 1987:361– 424. 98. Engel AG, Fumagalli G. Mechanisms of acetylcholine receptor loss from the neuromuscular junction. Ciba Found Symp. 1982;90: 197–224. 99. Scho¨nbeck S, Chrestel S, Hollfeld R. Myasthenia gravis: prototype of the antireceptor autoimmune diseases. Int Rev Neurobiol. 1990; 32:175–200. 100. Vincent A. Disorders affecting the acetylcholine receptor: myasthenia gravis and congenital myasthenia. J Recept Res. 1987;7:599 – 616. 101. Mihovilovic M, Donnelly-Roberts D, Richman D, Martinez-Carrion M. Pathogenesis of hyperacute experimental autoimmune myasthenia gravis. J Immunol. 1994;152:5997– 6002. 102. Tzartos S, Seybold M, Lindstrom J. Specificities of antibodies to acetylcholine receptors in sera from myasthenia gravis patients measured by monoclonal antibodies. Proc Natl Acad Sci. 1982; 1982:188 –192. 103. Ruff RL, Lennon VA. Endplate voltage-gated sodium channels are lost in clinical and experimental myasthenia gravis. Ann Neurol. 1998;43:370 –379. 104. Kaminski H, Suarez J, Ruff R. Neuromuscular junction physiology in myasthenia gravis: isoforms of the acetylcholine receptor in extraocular muscle and the contribution of sodium channels to the safety factor. Neurology. 1997;48 (suppl):8 –17. 105. Ruff RL. Electrophysiology of postsynaptic activation. Ann NY Acad Sci. 1998;841:57–70. 106. Richardson RF, Ruff RL. Anti-AchR antibodies in passively transferred autoimmune myasthenia gravis reduce endplate sodium currents by decreasing the density of endplate sodium channels, not by blocking sodium channels. Neurology. 1998;50:A433. 107. Molenaar PC. Synaptic adaptation in diseases of the neuromuscular junction. Prog Brain Res. 1990;84:145–149. 108. Asher O, Neumann D, Witzemann V, Fuchs S. Acetylcholine receptor gene expression in experimental autoimmune myasthenia gravis. FEBS Lett. 1990;261:231–235. 109. Kusner L, Mygland A, Kaminski H. Ryanodine receptor gene expression in thymus and thymomas. Muscle Nerve. 1998;21:1299 – 1303. 110. Smiley JD, Moore SE Jr. Molecular mechanisms of autoimmunity. Am J Med Sci. 1988;295:478 – 496. 111. Ruff RL, Whittlesey D. Na1 currents near and away from endplates on human fast and slow twitch muscle fibers. Muscle Nerve. 1993;16:922–929.

112. Missias AC, Chu GC, Klocke BJ, Sanes JR, Merlie JP. Regulation of the acetylcholine receptor gamma subunit gene in developing skeletal muscle: analysis with subunit-specific antibodies, transgenic mice, and cultured cells. Dev Biol. 1996;179:223. 113. Kaminski H, Ruff R. Ocular muscle involvement by myasthenia gravis. Ann Neurol. 1997;41:419 – 420. 114. Compston DAS, Vincent A, Newsom-Davis J, Batchelor JR. Clinical, pathological, HLA antigen, and immunological evidence for disease heterogeneity in myasthenia gravis. Brain J. 1980;103:579 – 601. 115. Vincent A, Newsom-Davis J. Acetylcholine receptor antibody characteristics in myasthenia gravis. I. Patients with generalized myasthenia or disease restricted to ocular muscles. Clin Exp Immunol. 1982;49:257–265. 116. Protti MP, Manfredi AA, Howard JF, Conti-Tronconi BM. T cells in myastheni gravis specific for embryonic acetylcholine receptor. Neurology. 1991;41:1809 –1814. 117. Yuen M, Macklin K, Conti-Fine B. MHC class II presentation of human acetylcholine receptor in myasthenia gravis: binding of synthetic gamma subunit. J Autoimmunol. 1996;9:67–77. 118. Gardnerova M, Eymard B, Morel E, et al. The fetal/adult acetylcholine receptor antibody ratio in mothers with myasthenia gravis as a marker for transfer of the disease to the newborn. Neurology. 1997;48:50 –54. 119. Vincent A, Newland C, Brueton L, et al. Arthrogryposis multiplex congenita with maternal autoantibodies specific for the fetal antigen. Lancet. 1995;346:24 –25. 120. Vincent A, Newland C, Croxen R, Beeson D. Genes at the junction— candidates for congenital myasthenic syndrome. Trends Neurosci. 1997;20:15–22. 121. Morel E, Eymard B, Vernet-der Garabedian B, Pannier C, Dulac O, Bach JF. Neonatal myasthenia gravis: a new clinical and immunological appraisal on 30 cases. Neurology. 1988;38:138 –142. 122. Oosterhuis HJGH, Newsom-Davis J, Wokke JHJ, et al. The slow channel syndrome. Two new cases. Brain. 1987;110:1061–1079. 123. Engel AG, Lambert EH, Mulder DM, et al. A newly recognized congenital myasthenic syndrome attributed to a prolonged open time of the acetylcholine-induced ion channel. Ann Neurol. 1982; 11:553–569. 124. Croxen R, Newland C, Beeson D, et al. Mutations in different functional domains of the human muscle acetylcholine receptor a subunit in patients with the slow channel syndrome congenital myasthenic syndrome. Hum Mol Genet. 1997;6:767–774. 125. Gomez C, Maselli R, Gammack J, et al. A b-subunit mutation in the acetylcholine receptor genes reveal heterogeneity in the slow channel congenital myasthenic syndrome. Ann Neurol. 1996;39: 1217–1227. 126. Ohno K, Hutchinson D, Milone M, et al. Congenital myasthenic syndrome caused by prolonged acetylcholine receptor channel openings due to a mutation in the M2 domain of the epsilon subunit. Proc Natl Acad Sci. 1995;92:758 –762. 127. Engel AG, Uchitel OD, Walls TJ, Nagel A, Harper CM, Bodensteiner J. Newly recognized congenital myasthenic syndrome associated with high conductance and fast closure of the acetylcholine receptor channel. Ann Neurol. 1993;34:38 – 47. 128. Engel AG, Nagel A, Walls TJ, Harper CM, Waisburg HA. Congenital myasthenic syndromes: I. deficiency and short open-time of the acetylcholine receptor. Muscle Nerve. 1993;16:1284 –1292. 129. Engel AG, Hutchinson DO, Nakano S, et al. Myasthenic syndromes attributed to mutations affecting the epsilon subunit of the acetylcholine receptor. Ann NY Acad Sci. 1993;681:496 –508. 130. Gu Y, Camacho P, Gardner P, Hall ZW. Identification of two amino acid residues in the subunit that promote mammalian muscle acetylcholine receptor assembly in COS cells. Neuron. 1991;6:879 – 887. January 1999


Volume 106 111

Neuromuscular Junction Ion Channels/Boonyapisit et al 131. Ohno K, Quiram PA, Milone M, et al. Congenital myasthenic syndromes due to hereroallelic nonsense/missense mutations in the acetylcholine receptor « subunit gene: identification and functional characterization of six new mutations. Hum Mol Genet. 1997;6:754 –766. 132. Uchitel O, Engel AG, Wals TJ, Nagel A, Atassi MZ, Bril V. Congenital myasthenic syndromes: II. syndrome attributed to abnormal interaction of acetylcholine with its receptor. Muscle Nerve. 1993;16:1293–1301. 133. Changeux J-P, Devillers-Thiery A, Galzi J-L, Bertrand D. New mutants to explore nicotinic receptor functions. Trends Pharmocol Sci. 1992;13:299 –301. 134. Morgan-Hughes JA, Lecky BRF, Landon DN, Murray NMF. Alterations in the number and affinity of junctional acetylcholine receptors in a myopathy with tubular aggregates. A newly recognized receptor defect. Brain. 1981;104:279 –295. 135. O’Neill JH, Murray NMF, Newsom-Davis J. The Lambert-Eaton myasthenic syndrome. A review of 50 cases. Brain. 1988;111:577– 596. 136. Beydoun SR. Delayed diagnosis of Lambert-Eaton myasthenic syndrome in a patient presenting with recurrent refractory respiratory failure. Muscle Nerve. 1994;17:689 – 690. 137. Barr CW, Claussen G, Thomas D, Fesenmeier JT, Pearlman RL, Oh SJ. Primary respiratory failure as the presenting symptom in Lambert-Eaton myasthenic syndrome. Muscle Nerve. 1993;16: 712–715. 138. Breen L, Gutman L, Brick J, Riggs J. Paradoxical lid elevation with sustained upgaze: a sign of Lambert-Eaton syndrome. Muscle Nerve. 1991;14:863– 866. 139. Engel AG. Myasthenic syndromes. In: Engel AG, Banker BQ, eds. Myology. New York: McGraw-Hill, 1986:1955–1990. 140. Sanders DB. Lambert-Eaton myasthenic syndrome: clinical diagnosis, immune-mediated mechanisms, and update on therapies. Ann Neurol. 1995;37(suppl 1):S63–73. 141. Elmqvist D, Lambert EH. Detailed analysis of neuromuscular transmission in a patient with the myasthenic syndrome sometimes associated with bronchogenic carcinoma. Mayo Proc Clin. 1968;43:689 –713. 142. Lambert EH, Lennon VA. Neuromuscular transmission in nude mice bearing oat-cell tumors from Lambert-Eaton myasthenic syndrome. Muscle Nerve. 1982;5:S39 –S45. 143. Newsom-Davis J, Leys K, Vincent A, Ferguson I, Modi G, Mills K. Immunological evidence for the co-existence of the LambertEaton myasthenic syndrome and myasthenia gravis in two patients. J Neurol Neurosurg Psychiatr. 1991;54:452– 453. 144. Oguro-Okamoto M, Griesman GE, Wieben ED, Slaymaker SJ, Snutch TP, Lennon VA. Molecular diversity of neuronal-type calcium channels identified in small cell lung carcinoma. Mayo Clin Proc. 1992;67:1150 –1159. 145. Trontelj JV, Stalberg E. Single motor endplates in myasthenia gravis and LEMS at different firing rates. Muscle Nerve. 1990;14:226 – 232. 146. Cull-Candy SG, Meledi R, Trautmann A, Uchitel OD. On the release of transmitter at normal, myasthenia gravis and myasthenic syndrome affected human end-plates. J Physiol. 1980;299:621– 638. 147. Lambert EH, Elmqvist D. Quantal components of end-plate potentials in the myasthenic syndrome. Ann NY Acad Sci. 1971;183: 183–199. 148. Streib EW, Rothner AD. Eaton-Lambert myasthenic syndrome: long-term treatment of three patients with prednisone. Ann Neurol. 1981;10:448 – 453. 149. Lang B, Newsom-Davis J, Wray DW, Vincent A. Autoimmune aetiology for myasthenic (Lambert-Eaton) syndrome. Lancet. 1981;ii:224 –226. 112

January 1999


Volume 106

150. Lennon VA, Lambert EH, Whittingham S, Fairbainks V. Autoimmunity in the Lambert-Eaton myasthenic syndrome. Muscle Nerve. 1982;5:S21–S25. 151. Newsom-Davis J, Murray NMF. Plasma exchange and immunosuppressive drug treatment in the Lambert-Eaton myasthenic syndrome. Neurology. 1984;34:480 – 485. 152. Fukunaga H, Engel AG, Osame M, Lambert EH. Paucity and disorganization of presynaptic membrane active zones in the Lambert-Eaton myasthenic syndrome. Muscle Nerve. 1982;5:686 – 697. 153. Fukunaga H, Engel AG, Lang B, Newsom-Davis B, Vincent A. Passive transfer of Lambert-Eaton myasthenic syndrome IgG from man to mouse depletes the presynaptic membrane active zones. Proc Natl Acad Sci USA. 1983;80:7636 –7640. 154. Engel AG. Review of evidence for loss of motor nerve terminal calcium channels in Lambert-Eaton myasthenic syndrome. Ann NY Acad Sci. 1991;635:246 –258. 155. Fukuoka T, Engel AG, Lang B, Newsom-Davis J, Prior C, Wray DW. Lambert-Eaton myasthenic syndrome: I. early morphologic effects of IgG on the presynaptic membrane active zones. Ann Neurol. 1987;22:193–199. 156. Fukuoka T, Engel AG, Lang B, Newsom-Davis J, Vincent A. Lambert-Eaton myasthenic syndrome: II. immunoelectron microscopy localization of IgG at the mouse motor end-plate. Ann Neurol. 1987;22:200 –211. 157. Nagel A, Engel AG, Lang B, Newsom-Davis J, Fukuoka T. Lambert-Eaton syndrome IgG depletes presynaptic membrane active zone particles by antigenic modulation. Ann Neurol. 1988;24:552– 558. 158. Lang B, Vincent A, Murray NMF, Newsom-Davis J. LambertEaton myasthenic syndrome: immunoglobulin G inhibition of Ca21 flux in tumor in cells correlates with disease severity. Ann Neurol. 1989;25:265–271. 159. Kim YI, Neher E. IgG from patients with Lambert-Eaton syndrome blocks voltage-dependent calcium channels. Science. 1988; 239:405– 408. 160. Peers C, Lang B, Newsom-Davis J, Wray DW. Selective action of Lambert-Eaton myasthenic syndrome antibodies on calcium channels in a rodent neuroblastoma x glioma hybrid cell line. J Physiol (Lond). 1987;421:293–308. 161. Lang B, Johnston I, Leys K, et al. Autoantibody specificities in Lambert-Eaton myasthenic syndrome. Ann NY Acad Sci. 1993; 681:382–391. 162. Lennon VA, Lambert EH. Antibodies bind solubilized calcium channel-omega-conotoxin complexes from small cell lung carcinoma: a diagnostic aid for Lambert-Eaton myasthenic syndrome. Mayo Clin Proc. 1989;64:1498 –1504. 163. Leys K, Lang B, Johnston I, Newsom-Davis J. Calcium channel autoantibodies in the Lambert-Eaton myasthenic syndrome. Ann Neurol. 1991;29:307– 414. 164. Sher E, Gotti C, Canal N, et al. Specificity of calcium channel autoantibodies in Lambert-Eaton myasthenic syndrome. Lancet. 1989;ii:640 – 643. 165. Blandino JKW, Kim YI. Lambert-Eaton syndrome IgG inhibits dihydropyridine-sensitive, slowly inactivating calcium channels in bovine adrenal chromaffin cells. Ann NY Acad Sci. 1993;681: 394 –397. 166. Viglione MP, Blandino JKW, Kim YI. Effects of Lambert-Eaton syndrome serum and IgG on calcium and sodium currents in small-cell lung cancer cells. Ann NY Acad Sci. 1993;681:418 – 421. 167. Protti DA, Sanchez VA, Cherksey BD, Sugimori M, Llinas RR, Uchitel OD. Mammalian neuromuscular transmission blocked by funnel web toxin. Ann NY Acad Sci. 1993;681:405– 407. 168. Motomura M, Johnston I, Lang B, Vincent A, Newsom-Davis J. An improved diagnostic assay for Lambert-Eaton myasthenic syndrome. J Neurol Neurosurg Psychiatr. 1995;58:85– 87.

Neuromuscular Junction Ion Channels/Boonyapisit et al 169. Johnston I, Lang B, Leys K, Newsom-Davis J. Heterogeneity of calcium channels autoantibodies detected using a small cell lung cancer line from a Lambert-Eaton myasthenic syndrome patient. Neurology. 1994;44:34 –38. 170. Lennon VA, Kryzer TJ, Griesmann GE, et al. Calcium-channel antibodies in the Lambert-Eaton syndrome and other paraneoplastic syndromes. NEJM. 1995;332:1467–1474. 171. Rosenfeld MR, Wong E, Dalmau J, et al. Sera from patients with Lambert-Eaton myasthenic syndrome recognize the b-subunit of Ca21 channel complexes. Ann NY Acad Sci. 1993;681:408 – 411. 172. Leveque C, Hoshino T, David P, et al. The synaptic vesicle protein synaptotagmin associates with calcium channels and is a putative Lambert-Eaton myasthenic syndrome antigen. Proc Natl Acad Sci. USA 1992;89:3625–3629. 173. Hajela RK, Atchison WD. The proteins synaptotagmin and syntaxin are not general targets of Lambert-Eaton myasthenic syndrome autoantibody. J Neurochem. 1995;64:1245–1251. 174. Takamori M, Hamada T, Komai K, Takahashi M, Yoshida A. Synaptotagmin can cause an immune-mediated model of LambertEaton myasthenic syndrome in rats. Ann Neurol. 1994;35:74 – 80.

175. Chalk CH, Murray NMF, Newsom-Davis J, O’Neill JH, Spiro SG. Response of the Lambert-Eaton myasthenic syndrome to treatment of associated small-cell lung carcinoma. Neurology. 1990;40: 1552–1556. 176. Lambert EH, Rooke ED, Eaton LM, Hodgson CH. Myasthenic syndrome occasionally associated with bronchial neoplasm: neurophysiologic studies. In: Viets HR, ed. Myathenia Gravis. Springfield: Charles C. Thomas, 1961:362– 410. 177. Otsuka M, Endo M. The effect of guanidine on neuromuscular transmission. J Pharmacol Exp Ther. 1960;128:273–282. 178. Cherington M. Guanidine and germaine in Lambert-Eaton syndrome. Neurology. 1976;26:944 –946. 179. Saint DA. The effects of 4-aminopyridine and tetraethylammonium on the kinetics of transmitter release at the mammalian neuromuscular synapse. Can J Physiol Pharmacol. 1989;67:1045– 1050. 180. Sanders DB, Howard JF Jr, Massey JM. 3,4-Diaminopyridine in Lambert-Eaton myasthenic syndrome and myasthenia gravis. Ann NY Acad Sci. 1993;681:588 –590.

January 1999


Volume 106 113