Myasthenia Gravis and Related Disorders

© Springer International Publishing AG, part of Springer Nature 2018

Henry J. Kaminski and Linda L. Kusner (eds.)Myasthenia Gravis and Related DisordersCurrent Clinical Neurologyhttps://doi.org/10.1007/978-3-319-73585-6_1

1. Neuromuscular Junction Physiology and Pathophysiology

Jaap J. Plomp¹  

(1)

Department of Neurology, Leiden University Medical Centre, Leiden, The Netherlands

Jaap J. Plomp

Email: [email protected]

Keywords

Acetylcholine receptorAutoimmunityElectrophysiologyEndplateEndplate potentialFatigueMuscle-specific kinaseMyasthenia gravisNeuromuscular junctionNeurotransmitterSafety factorSkeletal muscleSynapse

Introduction

The crucial role of the neuromuscular junction (NMJ) is to reliably transmit action potentials from the motor neuron to the skeletal muscle fiber so that sustained muscle contraction is guaranteed. To be able to understand the NMJ dysfunction in myasthenia gravis (MG) and related disorders, it is important to know how this synapse is structurally organized and the way it achieves successful transmission The goal of this chapter is to provide information about the structure and electrophysiological function of the NMJ and to explain the synaptopathic functional deficits which lead to disturbed muscle contraction in MG and other disorders of the NMJ.

Structural Composition of the Neuromuscular Junction

The Presynaptic Compartment

The NMJ, or endplate, is a tiny structure on skeletal muscle fibers. Depending on the species and muscle type, its length ranges from tens to a few hundreds of microns (Fig. 1.1), while the muscle fiber itself can be as long as several tens of centimeters. At this synaptic connection, the pre- and postsynaptic membranes and their adjacent cytoplasmatic environments are structurally highly specialized and harbor elaborate molecular machineries which govern the transmission of action potentials from motor neuron to muscle fiber. Muscles are innervated by a peripheral nerve which contains the bundled axons of motor and sensory neurons. Within the muscle, a motor axon forms multiple branches, each of which connects to a single skeletal muscle fiber, typically in their central region (see Fig. 1.1a). The most distal portion of the branch becomes devoid of myelin and forms a presynaptic nerve terminal that exactly opposes the specialized postsynaptic area. A small distance of ~50 nm is kept between the pre- and postsynaptic membrane, forming the synaptic cleft (Fig. 1.2). Human NMJs are relatively small (usually <100 μm²) as compared to the most often studied rodent NMJ (which is often >250 μm²) [1]. Furthermore, their geometry differs in that the human nerve terminal forms several interconnected spotlike boutons instead of a much more continuous pretzel shape in rodents (see Fig. 1.1b). NMJs are covered by a few (~3–5) perisynaptic Schwann cells which play (as yet poorly defined) structural and trophic roles in maturation and maintenance of the NMJ and in aiding regeneration of the motor nerve terminal after damage [2, 3].

Fig. 1.1

Morphology of the neuromuscular junction. (a) Branching of distal motor axons in levator auris longus muscle of the mouse, forming nerve terminals on muscle fibers. Transgenic expression of YFP (green) in motor axons, shown with fluorescence microscopy. Muscle fibers visualized with bright-field microscopy. (b) Laser-scanning confocal fluorescence microscopical image of a single, pretzel-shaped mouse diaphragm neuromuscular junction. Staining for acetylcholine receptors with green-fluorescently labeled α-bungarotoxin reveals a fingerprint-like geometric pattern formed by the high density of receptors on the tops of the postsynaptic folds. Courtesy of Lizette van der Pijl. (c) Top: schematic illustration of synaptic transmission of action potentials from motor neuron to skeletal muscle fiber at the neuromuscular junction. Bottom: schematic representation of the pre- and postsynaptic machinery underlying synaptic transmission at the neuromuscular junction. AChE acetylcholinesterase, AChR acetylcholine receptor, ColQ collagen-Q, LRP4 low-density lipoprotein receptor-related protein 4, MuSK muscle-specific kinase

Fig. 1.2

Electron microscopical picture of a detail of a mouse neuromuscular junction. SV synaptic vesicles, SC synaptic cleft, AZ active zone, PF postsynaptic folds. Scale bar = 0.5 μm

Nerve terminals contain the key elements responsible for the controlled release of the neurotransmitter acetylcholine (ACh) (see Fig. 1.1c). ACh is present in quanta, which are more or less fixed amounts of ~10,000 molecules, contained in ~50 nm diameter synaptic vesicles (see Fig. 1.2). Each nerve terminal harbors ~150,000–300,000 vesicles. The phospholipid bilayer membrane of these vesicles contains several transmembrane proteins. These include a transporter protein that loads the vesicle with ACh from the nerve terminal’s cytoplasm where it is biosynthesized. Furthermore present are the SNARE proteins synaptobrevin, which is involved in neuroexocytosis, and synaptotagmin, a Ca²+-sensing protein. These two molecules, in conjunction with the presynaptic membrane SNARE proteins SNAP-25 and syntaxin, form the core of a complex molecular machinery (modulated by many other proteins) which enables Ca²+-dependent fusion of the vesicle and exocytosis of its ACh quantum from the lumen into the synaptic cleft [4]. This process takes place at specialized areas called active zones, where an array of structural proteins forms a scaffold, which anchors the functional proteins needed for regulated neuroexocytosis (see Figs. 1.1 and 1.2) [5]. Active zones exist at the cytoplasmatic side of the presynaptic membrane in a density of ~2.5 per μm², which means that an adult mouse NMJ has ~800 active zones [6]. A central functional protein at active zones is the voltage-gated Ca²+ channel, which at the adult mammalian NMJs is of the CaV2.1 (or P/Q) type [7]. In developing NMJs or NMJs from mice genetically deficient for CaV2.1, other types (CaV1, CaV2.2, and/or CaV2.3) can be present [8]. CaV2.1 channels are positioned at the active zone through binding with the presynaptic protein bassoon and the synaptic cleft molecule laminin-β2 [9]. In this way they are optimally positioned so that the Ca²+ influx resulting from their opening by a motor nerve action potential can evoke quantal ACh release through stimulation of the exocytotic machinery. In the membrane, CaV2.1 channels may reside in lipid rafts, which are cholesterol-rich micro-domains, which contain signaling molecules, as well as the SNARE proteins SNAP-25 and syntaxin [10, 11]. Furthermore, gangliosides are important components of lipid rafts. These are glycosphingolipids enriched in neuronal membranes, especially at synapses [12].

The Synaptic Cleft

The ~50-nm-wide synaptic cleft (see Fig. 1.2) between the pre- and postsynaptic membrane of the NMJ contains several synapse-specific proteins which are anchored in the extracellular basal lamina matrix and are of importance for synaptic structure and function. A key molecule is acetylcholinesterase (AChE), an enzyme responsible for the hydrolytic degradation of released ACh to terminate its signaling action. AChE is embedded in the extracellular matrix through its collagen-Q tail (see Fig. 1.1c), which also connects to perlecan, and possibly via its catalytic domain to laminin [13, 14]. Another important molecule in the synaptic cleft is laminin-β2, which is produced by the muscle fiber and functions as a presynaptic active zone organizer, a process in which its interaction with presynaptic CaV2.1 channels is essential [15]. The synaptic cleft also contains matrix metalloprotease 9 (MMP9), of which the role may be to cleave fragments from the extracellular part of some pre- and postsynaptic membrane proteins which then serve as signaling peptides in synaptic homeostasis [16]. In addition, several types of synapse-specific collagens are present with putative roles in maturation and stability of the NMJ [14].

The Postsynaptic Compartment

The motor nerve terminal is positioned in a primary gutter on the muscle fiber membrane. Ultrastructural studies show the nerve terminal perpendicular to this gutter with the postsynaptic membrane forming 50–100-nm-wide secondary folds extending into the sarcoplasm (see Fig. 1.2). Across species the length of these folds is inversely related to the area of the NMJ, the relatively small human NMJs having deep folds [1]. The tops of the synaptic folds contain the ionotropic ACh receptors (AChRs) of the nicotinic type (see below) in high density of ~10,000 ion channels per μm². When AChRs are labeled with a fluorescent probe and viewed with high-resolution microscopy, the contours of the tops of the folds appear as a fingerprint-like geometric pattern (see Fig. 1.1b). AChR clusters are in dynamic equilibrium; there is intense turnover with equal extents of degradation and insertion of newly synthesized and recycled AChRs so that a constant density is maintained [17]. This balance between AChR insertion and removal is regulated by antagonistic activity of protein kinases A and C [18]. The ultrastructure of the postsynaptic folds and the specific distribution of ion channels facilitate muscle fiber excitation. The confined cytoplasmatic space in the folds increases the electrical resistance, which produces an extra depolarizing effect of the ion current through the AChRs, when being forced to flow through it [19]. At the bottoms of the folds, voltage-gated Na+ channels of the NaV1.4 type are present at higher density than in the extrasynaptic sarcolemma. This relatively high NaV1.4 density lowers the firing threshold and thus facilitates excitability at the postsynaptic membrane (see below) [20].

The postsynaptic folding ultrastructure seems dependent on the large proteins dystrophin and its close homologue utrophin. Dystrophin resides throughout the inside of the sarcolemma where it functions to stabilize muscle fibers by connecting intracellular actin to a complex of intracellular, transmembrane, and extracellular proteins (including syntrophin, dystrobrevin, and dystroglycans), with dystroglycan as a core transmembrane protein. At the NMJ, dystrophin is enriched postsynaptically together with utrophin, where they modulate ultrastructure and function, as demonstrated by altered AChR geometry, reduced ACh sensitivity, and diminished postsynaptic folds in mice deficient for dystrophin alone or both dystrophin and utrophin [21–23].

For the embryonic formation of the postsynaptic membrane specialization, and the stability of the NMJ thereafter, a signaling protein complex exists which consists of the transmembrane receptor tyrosine kinase, muscle-specific kinase (MuSK), the one-pass transmembrane protein, low-density lipoprotein receptor-related protein 4 (LRP4), and an additional number of intracellular postsynaptic proteins (see Fig. 1.1c) [24, 25]. This complex forms a signaling cascade which starts with the binding of neurally released agrin to LRP4. This induces MuSK dimerization, and thereby kinase activation, which in turn stimulates the downstream pathway to finally recruit and cluster AChRs to the tops of the postsynaptic folds. Once expressed and clustered in the membrane, AChRs attract the intracellular protein rapsyn which anchors them to the cytoskeleton [26, 27].

The postsynaptic membrane furthermore contains ErbB receptors which can bind neuregulin-1 released from the motor nerve terminal. Neuregulin-1 (formerly known as AChR-inducing activity, ARIA) was initially thought to be indispensable for expression of AChR subunit genes from subsynaptic nuclei during NMJ formation. However, multiple studies have challenged this view, and it is now believed that neuregulin/ErbB signaling is not essential for AChR gene transcription and formation of AChR clusters in developing NMJs. Rather, neuregulin/ErbB signaling appears to be needed for the maintenance of a normal number of AChRs at the adult NMJ, by inhibiting the removal of AChRs in the recycling process via an effect on α-dystrobrevin1 [28–31].

Normal and Pathological Electrophysiology at the Neuromuscular Junction

The duty of the NMJ is to faithfully transmit each motor neuronal action potential onto the innervated skeletal muscle fiber, where it will spread and stimulate the excitation-contraction system (not discussed here, e.g., for review, see [32]). Although this seems a relatively simple task, the above description of NMJ structure indicates that highly elaborate multi-molecular arrangements are required. Malfunction or deficiency of any of these key factors can influence synaptic function. Although synaptic transmission at the NMJ has a large safety margin (see below), the ultimate consequence of severe defects is block of signal transmission. This will obviously cause muscle weakness because tetanic (i.e., sustained) muscle contraction requires a certain duration of repetitive muscle fiber action potentials at high frequency and that can only be achieved if the NMJ transmission seamlessly follows the trains of high-rate action potential firing of the motor neuron. Defects in key synaptic molecules may arise from genetic mutation, autoimmunity, or intoxication. To understand how these situations lead to block of transmission and consequent muscle weakness, it is important to understand the electrophysiology of the NMJ. Below, the synaptic events at the NMJ are described, as well as their deviations in some prototypical neuromuscular synaptopathies, including MG.

Presynaptic Functional Events

The inside of motor neurons and skeletal muscle fibers is negatively charged as compared to the extracellular space. This resting membrane potential of about −80 mV and the existence of several types of ion channels in the membrane of these cells enable them to generate and transport signals in the form of depolarizations along their membrane. Action potentials are the unitary events (i.e., all-or-none signals of similar magnitude) that are conducted in an active way along the membrane of excitable cells. Their upstroke is caused by influx of Na+ ions through voltage-gated NaV channels, which in the skeletal muscle fiber are of the NaV1.4 type. In the motor axon, action conduction is saltatory, i.e., it hops from node of Ranvier to node of Ranvier where the NaV channel density is relatively high and the lack of a myelin insulation permits ion flux through the membrane. The repolarization phase of action potentials is due to inactivation of NaV channels in conjunction with opening of voltage-gated rectifying K+ channels. The latter channels, or members of the protein complex in which they reside, form the antigenic targets of autoantibodies in neuromyotonia/Isaac’s syndrome, which is characterized by enhanced ACh release at the NMJ and spontaneous and repetitive motor neuronal firing, most likely generated at the motor nerve terminal [33–36]. These presynaptic K+ channels are also the target of dendrotoxin, a toxin component of mamba snake venom [37]. Once a motor axonal action potential reaches the synaptic nerve terminal, it causes opening of the CaV2.1 channels at active zones. This results in Ca²+ influx into the motor nerve terminal, driven by the steep concentration gradient between the extracellular space (~2 mM) and the nerve terminal cytoplasm (~200 nM). The local increase of intraterminal Ca²+ concentration activates the neuroexocytotic machinery at active zones, leading to synchronous exocytosis of several ACh quanta from multiple active zones. The total number of ACh quanta released from the entire nerve terminal per nerve impulse (i.e., the quantal content) varies between NMJs of species and muscle types, as well as with age, but is roughly correlated with NMJ size and thus with the absolute number of active zones [6]. For instance, mouse diaphragm NMJs (being ~250 μm²) have a quantal content of ~50, while human NMJs (being only ~100 μm²) have a lower quantal content of only ~20–30 [23, 38]. From the studies of active zone density and NMJ electrophysiology, it can be deduced that the chance of releasing an ACh quantum from a single active zone in response to an action potential is only ~10% [1]. Most likely this is due to the only brief duration (only ~1 ms) of neural action potentials, which is insufficiently long to activate all CaV2.1 channels, in combination with the cytoplasmatic increase in the concentration of Ca²+ being limited in time and space due to strong cytoplasmatic Ca²+ buffering systems [39].

Several, rare disorders are associated with transmission disturbance at the NMJ due to presynaptic malfunction. The prototypical disorder is Lambert-Eaton myasthenic syndrome (LEMS) , of which most patients have autoantibodies that target CaV2.1 channels [40]. Most often these antibodies have a paraneoplastic origin; many LEMS patients suffer from small-cell lung cancer [41]. It is hypothesized that the autoantibodies cross-link and thereby downregulate CaV2.1 channels, but the exact mechanism is unclear. The result is a reduced presynaptic Ca²+ influx, which causes severe reduction of quantal content. This leads to deminished postsynaptic endplate potentials (EPPs), which do not reach the firing threshold (see below) and thus do not trigger a muscle fiber action potential. Prolonged synaptic activity, however, may cause some degree of quantal content increase due to accumulation of presynaptic Ca²+ at the active zone, which results in some EPP amplitude increase. At NMJs where the initial EPP is just below the firing threshold, this may lead to temporary recovery of neuromuscular transmission. This explains the LEMS characteristic of short-term improvement in muscle force after maximal voluntary activation [40]. The drug 3,4-diaminopyridine provides symptomatic treatment for LEMS. The drug blocks delayed-rectifier voltage-gated K+ channels and thereby broadens the presynaptic action potential, which leads to stimulation of more CaV2.1 channels or opens the available channels for longer duration. The increased Ca²+ influx leads to a higher quantal content and, consequently, an increase in EPP amplitude.

Mutations in the gene CACNA1A, encoding the pore-forming subunit of CaV2.1 channels, are demonstrated in the rare, inherited migraine variant familial hemiplegic migraine (FHM), and also in the clinically overlapping diseases episodic ataxia type-2 (EA-2) and spinocerebellar type-6 (SCA-6) [42]. Besides the central nervous system symptoms of headache, ataxia, and/or epilepsy in these patients, there are indications of subclinical NMJ malfunction in such patients as well as in Cacna1a-mutant mouse strains [8, 43].

The presynaptic motor nerve terminal at the NMJ is the target of Clostridial botulinum neurotoxins, which are the causative agents in the paralytic disorder botulism. The potent toxin binds to gangliosides on the outside of the presynaptic membrane as well as to SV2 and synaptotagmin on the inside of a fusing neuroexocytotic vesicle [44, 45]. The toxin is taken up into the nerve terminal via endocytosis of the synaptic vesicle, and a fragment is released into the cytosol where it cleaves SNARE proteins, thus causing block of ACh release and therewith paralysis. Presynaptic gangliosides may also be the autoimmune target of anti-ganglioside antibodies in the Miller-Fisher syndrome (MFS) variant of the peripheral neuropathy Guillain-Barré syndrome (GBS). Ensuing complement-mediated presynaptic NMJ destruction might contribute to paralytic symptoms of this syndrome [12]. Botulism and MFS/GBS are included in the differential diagnosis of the myasthenic disorders.

Postsynaptic Functional Events

After presynaptic neuroexocytosis, ACh diffuses into the synaptic cleft. A large part of the transmitter is immediately degraded by the AChE anchored in the basal lamina. The ACh molecules that escape destruction can bind to the postsynaptic AChRs. Afterward, when they come off the AChR, they are degraded by AChE as well.

AChRs are pentameric ligand-gated ion channels of the nicotinic type, consisting of two α, one β, one δ, and one ε subunit and having a single-channel conductance of ~60 pS [46, 47]. To induce opening of the pore, an ACh molecule must occupy each of the two binding sites on the AChR. These are localized at an interface of each of the two α subunits with a non-α subunit [48]. AChRs are rather non-specific cation channels; under physiological conditions they allow influx of Na+ ions and to a lesser extent K+ efflux and Ca²+ influx. The predominant Na+ influx through opened AChRs causes a local depolarization at the postsynaptic membrane. The multiple ACh quanta released simultaneously in response to a presynaptic nerve impulse give rise to the EPP, a summed depolarization which stimulates opening of a sufficient number of NaV1.4 channels to form a muscle fiber action potential. This will spread out in two directions over the muscle fiber, away from the NMJ (see Fig. 1.1c), and invade the T-tubuli to switch on the excitation-contraction mechanism that leads to massive Ca²+ release from the sarcoplasmatic reticulum which will evoke contraction of the fiber [49]. Depending on species, age, and muscle type, EPPs are ~15–30 mV in amplitude and have a duration of a few milliseconds with a steep rising phase and an exponential decay phase (Fig. 1.3a). Besides ACh release evoked by nerve activity, a motor nerve terminal spontaneously releases single ACh quanta through exocytosis of single synaptic vesicles. This happens at irregular intervals with an average frequency of ~1–4/s (depending on muscle type) in the adult mouse NMJ and at much lower frequency (~4/min) at the smaller human NMJs. The single ACh quantum released leads to a postsynaptic miniature EPP (MEPP), ranging from ~0.3 to 1.5 mV, depending on species, age, and muscle type and having more or less similar kinetics as the EPP (see Fig. 1.3a). The MEPP is too small to activate many NaV1.4 channels and is thus not capable of inducing a muscle fiber action potential. Presumably, MEPPs form a random spillover from the enormous pool of several hundred thousands of synaptic vesicles present in a motor nerve terminal. While they do not seem to have a functional role, MEPPs are useful in the ex vivo electrophysiological analysis of NMJ function as their mean amplitude forms a measure for the functional AChR density (see below). Furthermore, they can be used to calculate the quantal content at a single NMJ. To this end, the measured EPP amplitude (i.e., the result of simultaneous release of multiple ACh quanta) is divided by the mean measured MEPP amplitude in the same NMJ [50]. First, a correction factor to the EPP amplitude has to be applied, correcting for the nonlinear summation of the depolarizing effect of the multiple ACh quanta [50, 51]. In fruit fly studies a functional role for MEPPs in NMJ maturation has been proposed [52]. The relevance of this finding for the mammalian NMJ remains to be established.

Fig. 1.3

Synaptic electrophysiology of the neuromuscular junction. (a) A nerve stimulation-evoked endplate potential (EPP) and a spontaneous miniature EPP (MEPP) recorded with an intracellular microelectrode from an ex vivo mouse diaphragm muscle fiber at the neuromuscular junction. The triggering of a muscle fiber action potential is prevented during this measurement by the addition of μ-conotoxin-GIIIB to the experimental bath, which selectively blocks the voltage-gated Na+ channels on the muscle fiber membrane. The arrow indicates the moment of nerve stimulation. (b) Schematic illustration of the concept of firing threshold, safety factor, and physiological rundown of EPP amplitude during high-rate synaptic activity at the neuromuscular junction. Asterisks indicate that an EPP will trigger a muscle fiber action potential. In the depicted normal situation, every EPP will trigger a muscle action potential, and sustained tetanic contraction of the muscle fiber will result. (c) Schematic illustration of the situation in myasthenia gravis, where EPPs are reduced in amplitude due to autoimmune-mediated reduction of acetylcholine receptor density at the neuromuscular junction. Due to the reduced safety factor, EPPs will become subthreshold during high-rate synaptic activity, resulting in failures to trigger a muscle fiber action potential. This will lead to muscle contraction fatigue. (c) Example of an intracellular microelectrode recording at the neuromuscular junction of an intercostal muscle biopsy from a myasthenia gravis patient. A reduced safety factor in neuromuscular transmission becomes apparent after some time of 40 Hz nerve stimulation, when EPPs become of peri-threshold amplitude and intermittently fail to trigger muscle fiber action potentials (arrows). Obviously, if such pathophysiological events take place at many neuromuscular junctions in a muscle, there will be fatigable muscle weakness

The firing threshold of a muscle fiber is determined by the density and function of the NaV1.4 channels on the muscle fiber membrane. Because NaV1.4 density is higher at the postsynaptic membrane as compared to the extrasynaptic muscle fiber membrane [53, 54], the firing threshold at the NMJ region is lower than elsewhere. This facilitates the triggering of a muscle action potential by the EPP. A substantial safety factor of neuromuscular transmission exists at the NMJ (see Fig. 1.3b). This means that the actual EPP amplitude is much larger than minimally needed to reach the firing threshold. In rat and mouse NMJs, the minimal EPP amplitude to reach the firing threshold is ~10–12 mV, while the actual EPPs are ~20–35 mV [23, 55]. Thus, the safety factor at these rodent NMJs is ~2–3. In human NMJs the safety factor is thought to be at the lower end of this range (~2) [56]. This renders human neuromuscular transmission somewhat more sensitive to conditions that reduce presynaptic ACh release or postsynaptic ACh sensitivity. The safety factor in neuromuscular transmission is needed to ensure sustained successful transmission, even when upon intense use of the NMJ the ACh release shows a physiological rundown. This is due to factors of the neuroexocytotic system becoming limiting (e.g., Cav2.1 channels which inactivate and/or reduction of the pool of available synaptic vesicles). Tetanic muscle contraction is only possible if the NMJ neatly translates the repetitive motor neuronal action potential firing into a similar pattern of action potentials on the muscle fiber. Motor neurons fire trains of action potentials in the frequency range of 20–100 Hz, depending on muscle fiber type [57, 58]. During such tetanic activity, the EPP amplitude in mouse NMJs diminishes by 20–30% to a more or less constant level after about ten impulses (see Fig. 1.3b). At human NMJs, this EPP amplitude rundown is even somewhat more pronounced, by ~40% [59]. Due to the existence of a safety factor in neuromuscular transmission, the NMJ can cope with such EPP rundown during intense activity; EPPs will remain suprathreshold and each will trigger a muscle fiber action potential, causing sustained tetanic muscle fiber contraction (see Fig. 1.3b).

A hallmark of MG is fatigable muscle weakness, which is due to progressive transmission failure at the NMJ during intense synaptic activity. Most MG patients have IgG1 and IgG3 autoantibodies against postsynaptic AChRs at the NMJ. Their antigenic binding has multiple effects: (1) cross-linking of AChRs, causing increased internalization, (2) direct functional block of the AChR, and (3) complement activation, culminating in focal postsynaptic lysis due to membrane attack complex formation [50, 60, 61]. The primary result of these effects is a reduced postsynaptic ACh sensitivity due to the physical removal and functional block of a proportion of the AChRs. When synaptic signals are electrophysiologically analyzed at NMJ in human MG muscle biopsies or in muscle preparations of MG animal models, this reduced AChR density can be shown to cause a reduction of mean MEPP amplitude [50]. Because MEPPs have no direct physiological role in synaptic transmission the NMJ, this amplitude reduction itself has no acute functional consequences. The extent of reduction, however, forms a good measure for the severity of the autoimmune attack by AChR antibodies. The EPP, caused by the multi-quantal ACh release evoked by a nerve action potential, becomes reduced in amplitude too by the reduced AChR density. This has direct functional consequences for the synaptic function of the NMJ. The safety factor of neuromuscular transmission becomes reduced, and EPPs may be too small to cross the firing threshold of the muscle fiber, especially during intense use of the NMJ when there is EPP amplitude rundown (see Fig. 1.3c). This can lead to a progressive and (intermittent) failure of EPPs to trigger a muscle fiber action potential (see Fig. 1.3c, d) and will result in fatigable muscle weakness. In fact, EPP amplitudes run down more pronouncedly at MG NMJs than at healthy NMJs, presumably due to secondary changes in the presynaptic ACh release mechanism (see below) [59, 62]. This further adds to safety factor reduction in the circumstance of intense synaptic activity.

At severely affected MG NMJs, EPPs may be reduced by such a large extent that the first EPP of a high-frequency train already is subthreshold. Such muscle fibers will not contract at all and will cause an initial/permanent paralysis component of the MG patient, next to the fatigable component caused by the fibers with NMJs that progressively lose transmission during prolonged synaptic activity.

Another factor contributing to the reduced safety factor at MG NMJs is the complement-mediated disruption of the postsynaptic membrane [63]. Besides causing reduction of AChR density, this also leads to concomitant removal of a proportion of the postsynaptic NaV1.4 channels, which are enriched at the bottom of the postsynaptic membrane folds. Because the density of these channels dictates the firing threshold of the muscle fiber at the NMJ, this results in an elevation of the threshold, thereby further reducing the safety factor [64, 65]. Anti-AChR antibodies by themselves do not affect the electrophysiological characteristics of NaV1.4 channels [65]. Furthermore, the complement attack disturbs the geometry of the postsynaptic folds which normally facilitates muscle fiber excitation and thus leads to a less efficient activation of remaining NaV1.4 channels by the ACh-induced ion current through the remaining AChRs, further elevating the firing threshold [65].

The patho-electrophysiology of NMJs in the less frequently occurring MG variants with autoantibodies to postsynaptic antigenic targets other than the AChR, such as MuSK or LRP4, is mostly similar to AChR MG. The precise primary and secondary effects of the non-AChR MG autoantibodies are different from the classical AChR MG autoantibodies [66, 67]. However, the ultimate, secondary effect of these antibodies is dispersal/removal of AChRs, and this leads to reduced EPP amplitude and, consequently, a smaller safety factor in neuromuscular transmission [59, 68–72]. Of note, MuSK MG has two distinctly different features from the classical AChR MG form. First, the MuSK autoantibodies are predominantly of the IgG4, a special IgG subclass, which does not activate the complement cascade. Thus, the observed pathophysiological effects in MuSK MG are most likely complement independent [69]. Second, MuSK MG NMJs seem to lack the phenomenon of compensatory upregulation of ACh release in response to the reduction of postsynaptic AChR density [59, 69, 70, 72, 73]. This homeostatic adaptation of the NMJ in AChR MG is partly compensating for the loss of postsynaptic AChRs and operates at the level of single NMJs via retrograde signaling. Although several post- and presynaptic factors as well as the retrograde signals have been proposed, the mechanism underlying this compensatory response at the MG NMJ is not yet fully understood [74], although it may involve an increase in the size of the pool of releasable ACh vesicles [75]. The homeostatic response might confer some degree of protection against transmission loss in AChR MG NMJs. However, there is concomitant extra rundown of EPPs due to exaggerated quantal content rundown at AChR MG NMJs. This might partially neutralize such a beneficial effect. Because this compensatory increase in ACh release is lacking in MuSK MG, it might be speculated that MuSK, or its intimate binding partner LRP4, are in some way involved in the homeostatic mechanism, e.g., as postsynaptic sensors or retrograde messengers.

In principle, the patho-electrophysiological mechanisms at the NMJ of intoxications with substances that block NMJ AChRs (e.g., d-tubocurarine-like compounds in plants or α-toxins in the venom of snakes or other animals) will be rather similar to the (non-complement-mediated) effects of AChR autoantibodies in MG [37, 50, 62, 76]. In congenital forms of MG with mutations in genes that encode AChR subunits or MuSK or its downstream signaling factors, the patho-electrophysiological mechanisms might be somewhat different and more complex due to the concomitant structural deficits resulting from developmental aberrations of the NMJ [77, 78]. However, transmission block at NMJs will remain the crucial culprit.

In conclusion, the chapter describes the electrophysiological events at the NMJ in relation to its structural background and has briefly discussed the main features of the patho-electrophysiology in some prototypical NMJ synaptopathies. It is hoped that this chapter has increased the reader’s understanding on the physiology of the NMJ and that it will provide a useful background for the reading and interpretation of the following chapters.

Acknowledgments

The author gratefully acknowledges support by the Prinses Beatrix Spierfonds, Stichting Spieren voor Spieren, and L’Association Française contre les myopathies.

References

1.

Slater CR. Structural factors influencing the efficacy of neuromuscular transmission. Ann N Y Acad Sci. 2008;1132:1–12.CrossrefPubMed

2.

Darabid H, Perez-Gonzalez AP, Robitaille R. Neuromuscular synaptogenesis: coordinating partners with multiple functions. Nat Rev Neurosci. 2014;15:703–18.CrossrefPubMed

3.

Kang H, Tian L, Mikesh M, Lichtman JW, Thompson WJ. Terminal Schwann cells participate in neuromuscular synapse remodeling during reinnervation following nerve injury. J Neurosci. 2014;34:6323–33.CrossrefPubMedPubMedCentral

4.

Sudhof TC. Neurotransmitter release: the last millisecond in the life of a synaptic vesicle. Neuron. 2013;80:675–90.CrossrefPubMed

5.

Nishimune H. Active zones of mammalian neuromuscular junctions: formation, density, and aging. Ann N Y Acad Sci. 2012;1274:24–32.CrossrefPubMedPubMedCentral

6.

Chen J, Mizushige T, Nishimune H. Active zone density is conserved during synaptic growth but impaired in aged mice. J Comp Neurol. 2012;520:434–52.CrossrefPubMedPubMedCentral

7.

Uchitel OD, Protti DA, Sanchez V, Cherksey BD, Sugimori M, Llinas R. P-type voltage-dependent calcium channel mediates presynaptic calcium influx and transmitter release in mammalian synapses. Proc Natl Acad Sci U S A. 1992;89:3330–3.CrossrefPubMedPubMedCentral

8.

Kaja S, van de Ven RC, van Dijk JG, Verschuuren JJ, Arahata K, Frants RR, et al. Severely impaired neuromuscular synaptic transmission causes muscle weakness in the Cacna1a-mutant mouse rolling Nagoya. Eur J Neurosci. 2007;25:2009–20.CrossrefPubMed

9.

Chen J, Billings SE, Nishimune H. Calcium channels link the muscle-derived synapse organizer laminin beta2 to Bassoon and CAST/Erc2 to organize presynaptic active zones. J Neurosci. 2011;31:512–25.CrossrefPubMedPubMedCentral

10.

Chamberlain LH, Burgoyne RD, Gould GW. SNARE proteins are highly enriched in lipid rafts in PC12 cells: implications for the spatial control of exocytosis. Proc Natl Acad Sci U S A. 2001;98:5619–24.CrossrefPubMedPubMedCentral

11.

Davies A, Douglas L, Hendrich J, Wratten J, Tran Van MA, Foucault I, et al. The calcium channel alpha2delta-2 subunit partitions with CaV2.1 into lipid rafts in cerebellum: implications for localization and function. J Neurosci. 2006;26:8748–57.CrossrefPubMed

12.

Plomp JJ, Willison HJ. Pathophysiological actions of neuropathy-related anti-ganglioside antibodies at the neuromuscular junction. J Physiol. 2009;587:3979–99.CrossrefPubMedPubMedCentral

13.

Massoulie J, Millard CB. Cholinesterases and the basal lamina at vertebrate neuromuscular junctions. Curr Opin Pharmacol. 2009;9:316–25.CrossrefPubMed

14.

Singhal N, Martin PT. Role of extracellular matrix proteins and their receptors in the development of the vertebrate neuromuscular junction. Dev Neurobiol. 2011;71:982–1005.CrossrefPubMedPubMedCentral

15.

Rogers RS, Nishimune H. The role of laminins in the organization and function of neuromuscular junctions. Matrix Biol. 2017;57–58:86–105.CrossrefPubMed

16.

Wu H, Lu Y, Shen C, Patel N, Gan L, Xiong WC, et al. Distinct roles of muscle and motoneuron LRP4 in neuromuscular junction formation. Neuron. 2012;75:94–107.CrossrefPubMedPubMedCentral

17.

Bruneau E, Sutter D, Hume RI, Akaaboune M. Identification of nicotinic acetylcholine receptor recycling and its role in maintaining receptor density at the neuromuscular junction in vivo. J Neurosci. 2005;25:9949–59.CrossrefPubMed

18.

Martinez P, Pires-Oliveira M, Akaaboune M. PKC and PKA regulate AChR dynamics at the neuromuscular junction of living mice. PLoS One. 2013;8:e81311.Crossref

19.

Vautrin J, Mambrini J. Synaptic current between neuromuscular junction folds. J Theor Biol. 1989;140:479–98.CrossrefPubMed

20.

Ruff RL, Lennon VA. End-plate voltage-gated sodium channels are lost in clinical and experimental myasthenia gravis. Ann Neurol. 1998;43:370–9.CrossrefPubMed

21.

Grady RM, Teng H, Nichol MC, Cunningham JC, Wilkinson RS, Sanes JR. Skeletal and cardiac myopathies in mice lacking utrophin and dystrophin: a model for Duchenne muscular dystrophy. Cell. 1997;90:729–38.CrossrefPubMed

22.

Pilgram GS, Potikanond S, Baines RA, Fradkin LG, Noordermeer JN. The roles of the dystrophin-associated glycoprotein complex at the synapse. Mol Neurobiol. 2010;41:1–21.CrossrefPubMed

23.

van der Pijl EM, van Putten M, Niks EH, Verschuuren JJ, Aartsma-Rus A, Plomp JJ. Characterization of neuromuscular synapse function abnormalities in multiple Duchenne muscular dystrophy mouse models. Eur J Neurosci. 2016;43:1623–35.CrossrefPubMed

24.

Ghazanfari N, Fernandez KJ, Murata Y, Morsch M, Ngo ST, Reddel SW, et al. Muscle specific kinase: organiser of synaptic membrane domains. Int J Biochem Cell Biol. 2011;43:295–8.CrossrefPubMed

25.

Wu H, Xiong WC, Mei L. To build a synapse: signaling pathways in neuromuscular junction assembly. Development. 2010;137:1017–33.CrossrefPubMedPubMedCentral

26.

Chen PJ, Martinez-Pena y Valenzuela I, Aittaleb M, Akaaboune M. AChRs are essential for the targeting of rapsyn to the postsynaptic membrane of NMJs in living mice. J Neurosci. 2016;36:5680–5.CrossrefPubMedPubMedCentral

27.

Gautam M, Noakes PG, Mudd J, Nichol M, Chu GC, Sanes JR, et al. Failure of postsynaptic specialization to develop at neuromuscular junctions of rapsyn-deficient mice. Nature. 1995;377:232–6.CrossrefPubMed

28.

Escher P, Lacazette E, Courtet M, Blindenbacher A, Landmann L, Bezakova G, et al. Synapses form in skeletal muscles lacking neuregulin receptors. Science. 2005;308:1920–3.CrossrefPubMed

29.

Jaworski A, Burden SJ. Neuromuscular synapse formation in mice lacking motor neuron- and skeletal muscle-derived Neuregulin-1. J Neurosci. 2006;26:655–61.CrossrefPubMed

30.

Rimer M. Neuregulins at the neuromuscular synapse: past, present, and future. J Neurosci Res. 2007;85:1827–33.CrossrefPubMed

31.

Schmidt N, Akaaboune M, Gajendran N, Martinez P, Wakefield S, Thurnheer R, et al. Neuregulin/ErbB regulate neuromuscular junction development by phosphorylation of alpha-dystrobrevin. J Cell Biol. 2011;195:1171–84.CrossrefPubMedPubMedCentral

32.

Rebbeck RT, Karunasekara Y, Board PG, Beard NA, Casarotto MG, Dulhunty AF. Skeletal muscle excitation-contraction coupling: who are the dancing partners? Int J Biochem Cell Biol. 2014;48:28–38.CrossrefPubMed

33.

Lang B, Makuch M, Moloney T, Dettmann I, Mindorf S, Probst C, et al. Intracellular and non-neuronal targets of voltage-gated potassium channel complex antibodies. J Neurol Neurosurg Psychiatry. 2017;88:353–61.CrossrefPubMedPubMedCentral

34.

Park SB, Lin CS, Krishnan AV, Simon NG, Bostock H, Vincent A, et al. Axonal dysfunction with voltage gated potassium channel complex antibodies. Exp Neurol. 2014;261:337–42.CrossrefPubMed

35.

Shillito P, Molenaar PC, Vincent A, Leys K, Zheng W, van den Berg RJ, et al. Acquired neuromyotonia: evidence for autoantibodies directed against K+ channels of peripheral nerves. Ann Neurol. 1995;38:714–22.CrossrefPubMed

36.

van SA, Schreurs MW, Wirtz PW, Sillevis Smitt PA, Titulaer MJ. From VGKC to LGI1 and Caspr2 encephalitis: the evolution of a disease entity over time. Autoimmun Rev. 2016;15:970–4.Crossref

37.

Ranawaka UK, Lalloo DG, de Silva HJ. Neurotoxicity in snakebite—the limits of our knowledge. PLoS Negl Trop Dis. 2013;7:e2302.CrossrefPubMedPubMedCentral

38.

Plomp JJ, van Kempen GT, De Baets MB, Graus YM, Kuks JB, Molenaar PC. Acetylcholine release in myasthenia gravis: regulation at single end-plate level. Ann Neurol. 1995;37:627–36.CrossrefPubMed

39.

Samigullin D, Fatikhov N, Khaziev E, Skorinkin A, Nikolsky E, Bukharaeva E. Estimation of presynaptic calcium currents and endogenous calcium buffers at the frog neuromuscular junction with two different calcium fluorescent dyes. Front Synaptic Neurosci. 2014;6:29.PubMed

40.

Titulaer MJ, Lang B, Verschuuren JJ. Lambert-Eaton myasthenic syndrome: from clinical characteristics to therapeutic strategies. Lancet Neurol. 2011;10:1098–107.CrossrefPubMed

41.

Joubert B, Honnorat J. Autoimmune channelopathies in paraneoplastic neurological syndromes. Biochim Biophys Acta. 2015;1848:2665–76.CrossrefPubMed

42.

Plomp JJ, Van den Maagdenberg AM, Molenaar PC, Frants RR, Ferrari MD, Mutant P. Q-type calcium channel electrophysiology and migraine. Curr Opin Investig Drugs. 2001;2:1250–60.PubMed

43.

Maselli RA, Books W, Dunne V. Effect of inherited abnormalities of calcium regulation on human neuromuscular transmission. Ann N Y Acad Sci. 2003;998:18–28.CrossrefPubMed

44.

Bullens RW, O’Hanlon GM, Wagner E, Molenaar PC, Furukawa K, Furukawa K, et al. Complex gangliosides at the neuromuscular junction are membrane receptors for autoantibodies and botulinum neurotoxin but redundant for normal synaptic function. J Neurosci. 2002;22:6876–84.PubMed

45.

Pirazzini M, Rossetto O, Eleopra R, Montecucco C. Botulinum neurotoxins: biology, pharmacology, and toxicology. Pharmacol Rev. 2017;69:200–35.CrossrefPubMedPubMedCentral

46.

Milone M, Monaco ML, Evoli A, Servidei S, Tonali P. Ocular myasthenia: diagnostic value of single fibre EMG in the orbicularis oculi muscle. J Neurol Neurosurg Psychiatry. 1993;56:720–1.CrossrefPubMedPubMedCentral

47.

Newland CF, Beeson D, Vincent A, Newsom-Davis J. Functional and non-functional isoforms of the human muscle acetylcholine receptor. J Physiol. 1995;489:767–78.

48.

Sine SM. End-plate acetylcholine receptor: structure, mechanism, pharmacology, and disease. Physiol Rev. 2012;92:1189–234.CrossrefPubMedPubMedCentral

49.

Bannister RA. Bridging the myoplasmic gap: recent developments in skeletal muscle excitation-contraction coupling. J Muscle Res Cell Motil. 2007;28:275–83.CrossrefPubMed

50.

Plomp JJ, Morsch M, Phillips WD, Verschuuren JJ. Electrophysiological analysis of neuromuscular synaptic function in myasthenia gravis patients and animal models. Exp Neurol. 2015;270:41–54.CrossrefPubMed

51.

McLachlan EM, Martin AR. Non-linear summation of end-plate potentials in the frog and mouse. J Physiol. 1981;311:307–24.CrossrefPubMedPubMedCentral

52.

Choi BJ, Imlach WL, Jiao W, Wolfram V, Wu Y, Grbic M, et al. Miniature neurotransmission regulates Drosophila synaptic structural maturation. Neuron. 2014;82:618–34.CrossrefPubMedPubMedCentral

53.

Flucher BE, Daniels MP. Distribution of Na+ channels and ankyrin in neuromuscular junctions is complementary to that of acetylcholine receptors and the 43 kd protein. Neuron. 1989;3:163–75.CrossrefPubMed

54.

Wood SJ, Slater CR. The contribution of postsynaptic folds to the safety factor for neuromuscular transmission in rat fast- and slow-twitch muscles. J Physiol. 1997;500:165–76.

55.

Wood SJ, Slater CR. Action potential generation in rat slow- and fast-twitch muscles. J Physiol. 1995;486:401–10.

56.

Wood SJ, Slater CR. Safety factor at the neuromuscular junction. Prog Neurobiol. 2001;64:393–429.CrossrefPubMed

57.

Eken T. Spontaneous electromyographic activity in adult rat soleus muscle. J Neurophysiol. 1998;80:365–76.CrossrefPubMed

58.

Hennig R, Lomo T. Firing patterns of motor units in normal rats. Nature. 1985;314:164–6.CrossrefPubMed

59.

Niks EH, Kuks JB, Wokke JH, Veldman H, Bakker E, Verschuuren JJ, et al. Pre- and postsynaptic neuromuscular junction abnormalities in musk myasthenia. Muscle Nerve. 2010;42:283–8.CrossrefPubMed

60.

Gilhus NE, Verschuuren JJ. Myasthenia gravis: subgroup classification and therapeutic strategies. Lancet Neurol. 2015;14:1023–36.CrossrefPubMed

61.

Phillips WD, Vincent A. Pathogenesis of myasthenia gravis: update on disease types, models, and mechanisms. F1000Res. 2016;5:F1000.CrossrefPubMedPubMedCentral

62.

Plomp JJ, van Kempen GT, Molenaar PC. Adaptation of quantal content to decreased postsynaptic sensitivity at single endplates in alpha-bungarotoxin-treated rats. J Physiol. 1992;458:487–99.CrossrefPubMedPubMedCentral

63.

Tuzun E, Christadoss P. Complement associated pathogenic mechanisms in myasthenia gravis. Autoimmun Rev. 2013;12:904–11.CrossrefPubMed

64.

Ruff RL, Lennon VA. How myasthenia gravis alters the safety factor for neuromuscular transmission. J Neuroimmunol. 2008;201–202:13–20.CrossrefPubMedPubMedCentral

65.

Ruff RL. Endplate contributions to the safety factor for neuromuscular transmission. Muscle Nerve. 2011;44:854–61.CrossrefPubMed

66.

Huijbers MG, Zhang W, Klooster R, Niks EH, Friese MB, Straasheijm KR, et al. MuSK IgG4 autoantibodies cause myasthenia gravis by inhibiting binding between MuSK and Lrp4. Proc Natl Acad Sci U S A. 2013;110:20783–8.CrossrefPubMedPubMedCentral

67.

Verschuuren JJ, Huijbers MG, Plomp JJ, Niks EH, Molenaar PC, Martinez-Martinez P, et al. Pathophysiology of myasthenia gravis with antibodies to the acetylcholine receptor, muscle-specific kinase and low-density lipoprotein receptor-related protein 4. Autoimmun Rev. 2013;12:918–23.CrossrefPubMed

68.

Cole RN, Reddel SW, Gervasio OL, Phillips WD. Anti-MuSK patient antibodies disrupt the mouse neuromuscular junction. Ann Neurol. 2008;63:782–9.CrossrefPubMed

69.

Klooster R, Plomp JJ, Huijbers MG, Niks EH, Straasheijm KR, Detmers FJ, et al. Muscle-specific kinase myasthenia gravis IgG4 autoantibodies cause severe neuromuscular junction dysfunction in mice. Brain. 2012;135:1081–101.CrossrefPubMed

70.

Selcen D, Fukuda T, Shen XM, Engel AG. Are MuSK antibodies the primary cause of myasthenic symptoms? Neurology. 2004;62:1945–50.CrossrefPubMed

71.

Shen C, Lu Y, Zhang B, Figueiredo D, Bean J, Jung J, et al. Antibodies against low-density lipoprotein receptor-related protein 4 induce myasthenia gravis. J Clin Invest. 2013;123:5190–202.CrossrefPubMedPubMedCentral

72.

Viegas S, Jacobson L, Waters P, Cossins J, Jacob S, Leite MI, et al. Passive and active immunization models of MuSK-Ab positive myasthenia: electrophysiological evidence for pre and postsynaptic defects. Exp Neurol. 2012;234:506–12.CrossrefPubMed

73.

Morsch M, Reddel SW, Ghazanfari N, Toyka KV, Phillips WD. Pyridostigmine but not 3,4-diaminopyridine exacerbates ACh receptor loss and myasthenia induced in mice by muscle-specific kinase autoantibody. J Physiol. 2013;591:2747–62.CrossrefPubMedPubMedCentral

74.

Plomp JJ. Trans-synaptic homeostasis at the myasthenic neuromuscular junction. Front Biosci (Landmark Ed). 2017;22:1033–51.Crossref

75.

Wang X, Pinter MJ, Rich MM. Reversible recruitment of a homeostatic reserve pool of synaptic vesicles underlies rapid homeostatic plasticity of quantal content. J Neurosci. 2016;36:828–36.CrossrefPubMedPubMedCentral

76.

Barber CM, Isbister GK, Hodgson WC. Alpha neurotoxins. Toxicon. 2013;66:47–58.CrossrefPubMed

77.

Barisic N, Chaouch A, Muller JS, Lochmuller H. Genetic heterogeneity and pathophysiological mechanisms in congenital myasthenic syndromes. Eur J Paediatr Neurol. 2011;15:189–96.CrossrefPubMed

78.

Engel AG, Shen XM, Selcen D, Sine SM. Congenital myasthenic syndromes: pathogenesis, diagnosis, and treatment. Lancet Neurol. 2015;14:420–34.CrossrefPubMedPubMedCentral

© Springer International Publishing AG, part of Springer Nature 2018

Henry J. Kaminski and Linda L. Kusner (eds.)Myasthenia Gravis and Related DisordersCurrent Clinical Neurologyhttps://doi.org/10.1007/978-3-319-73585-6_2

2. Acetylcholine Receptor Structure

Jie Luo¹   and Jon M. Lindstrom²  

(1)

Department of Clinical Sciences and Advanced Medicine, University of Pennsylvania School of Veterinary Medicine, Philadelphia, PA, USA

(2)

Department of Neuroscience, Medical School of the University of Pennsylvania, Philadelphia, PA, USA

Jie Luo

Email: [email protected]

Jon M. Lindstrom (Corresponding author)

Email: [email protected]

Keywords

Nicotinic acetylcholine receptorMyasthenia gravisMyasthenic syndromesAutoimmune responseIon channelGatingMutation

Introduction

Nicotinic acetylcholine receptors (AChRs) are acetylcholine-gated cation channels [1]. They play a critical postsynaptic role in transmission between motor nerves and skeletal muscles and in autonomic ganglia [2, 3]. In the central nervous system, they also act presynaptically and extrasynaptically to modulate transmission by facilitating the release of many transmitters [4, 5]. In the skin [6], lung [7], bronchial and vascular epithelia [8, 9], and several types of immune cells including monocytes [10–12], dendritic cells [13], macrophages [14–17], T-cells [18, 19], and B-cells [20, 21], they also mediate intercellular communication.

Abnormalities of AChRs are responsible for several human diseases . Mutations in AChRs are known to cause congenital myasthenic syndromes [22] and the rare autosomal dominant nocturnal frontal lobe form of epilepsy (ADNFLE) [23–26]. Autoimmune responses to AChRs are known to cause myasthenia gravis (MG) [27], certain dysautonomias [28–30], and some forms of pemphigus [31].

Nicotine acting on AChRs in the brain causes addiction to tobacco [32]. This is by far the largest medical problem in which AChRs play a direct role and the largest preventable cause of disease, accounting for 430,000 premature deaths annually in the United States [33].

Nicotine acting through AChRs has many physiological effects , including beneficial ones such as inducing vascularization, neuroprotection, cognitive enhancement, anxiolysis, and antinociception. Thus, nicotinic agents are lead compounds for the development of drugs to treat many diseases including Alzheimer’s disease, Parkinson’s disease, chronic pain, schizophrenia, and Tourette’s syndrome [34–40] as well as for smoking cessation [41, 42].

There are many known and potential subtypes of AChRs, each defined by the subunits which compose them [1, 3]. All AChRs are formed by five homologous subunits organized around a central cation channel. There are 17 known AChR subunits: α1–10, β1–4, γ, δ, and ε. By contrast with the many subtypes of neuronal AChRs, there are only two subtypes of muscle AChRs. These are a fetal subtype with an (α1)2 β1γδ stoichiometry and an adult subtype with an (α1)2 β1εδ stoichiometry.

AChRs are part of a gene superfamily, which includes the genes for subunits of ionotropic receptors for glycine, γ-amino butyric acid (GABA), and serotonin [1, 43]. The structural homologies of all of these receptors, and the sorts of evolutionary steps which produced this diversity of receptors, have been elegantly illustrated by experiments. One showed that changing only three amino acids in the channel-lining part of an AChR subunit to amino acids found in receptors for GABA or glycine receptors resulted in AChRs with anion-selective channels like those of GABA or glycine receptors [44]. Another experiment showed that a chimera of the extracellular domain of an AChR subunit and the remainder of a serotonin receptor subunit produced an ACh-gated cation channel with the conductance properties of a serotonin receptor [45].

Muscle AChRs are the best characterized members of the AChRs [1]. The presence of a single type of synapse in the skeletal muscle (with the exception of extraocular muscle, Chap. 7) facilitated studies of AChR synthesis, developmental plasticity, and electrophysiological function [46–48]. The presence of large amounts of muscle-like AChR in the electric organs of Torpedo species permitted the purification and characterization of AChRs, partial sequencing of their subunit proteins, cloning of the subunit cDNAs, and low-resolution electron crystallographic determination of their three-dimensional structure [43, 48–50]. Low stringency hybridization , starting with cDNAs for muscle AChR subunits, leads to the cloning of subunits for neuronal AChRs [48]. Immunization with purified electric organ AChRs leads to the discovery of experimental autoimmune myasthenia gravis (EAMG) , the autoimmune nature of MG, and an immunodiagnostic assay for MG [27, 51]. Monoclonal antibodies (mAbs) initially developed as model autoantibodies not only lead to the discovery of the main immunogenic region (MIR) on α1 subunits and the molecular basis of the autoimmune impairment of neuromuscular transmission in MG [27, 52, 53], but also lead to the immunoaffinity purification of neuronal nicotinic AChRs. mAbs have continued to provide useful tools for characterizing AChRs [3].

This chapter reviews the basic structures of muscle and neuronal AChRs. It will describe the antigenic structure of muscle AChRs and consider how this accounts for the pathological mechanisms by which neuromuscular transmission is impaired in MG. This will be briefly contrasted with the antigenic structure of a neuronal AChR involved in autoimmune dysautonomia. This chapter also considers the optimized functional structure of muscle AChRs and how mutations impair AChR function in congenital myasthenic syndromes . The many AChR mutations identified in all of the muscle AChR subunits in myasthenic syndromes are contrasted with the few disease-causing mutations discovered, thus far, in neuronal AChR subunits.

Size and Shape of AChRs

Electron crystallography of two-dimensional helical crystalline arrays of AChRs in fragments of Torpedo electric organ membranes has revealed the basic size and shape of this muscle-type AChR to a resolution of 4 Å, as shown in Fig. 2.1 [50, 55]. Viewed from the side, a Torpedo AChR is roughly cylindrical, about 140 Å long and 80 Å wide. About 65 Å extends on the extracellular surface, 40 Å crosses the lipid bilayer, and 35 Å extends below. Viewed from the top, the extracellular vestibule is a pentagonal tube with walls about 25 Å thick and a central pore about 20 Å in diameter. The channel across the membrane narrows to a close. Other evidence suggests that the open lumen of the channel becomes narrow (perhaps 7 Å across), sufficient only for rapid flow of hydrated cations like Na+ or K+. The five subunits are rodlike, oriented-like barrel staves at a 10° angle around the central channel. Movements of Torpedo AChR subunits associated with activation by ACh have been captured by electron microscopy of crystalline arrays of AChRs in membrane fragments sprayed with ACh and fast frozen [56]. Crystal structures of related receptors have been determined in several conformations, and progress is being rapidly made on relating receptor structure to functional state. For example, glycine receptors have been crystallized in a closed state bound by an antagonist and in an open-channel state bound by glycine [57]. Human neuronal α4β2 AChRs in a desensitized state bound to nicotine have been crystallized with the activation gate in the middle of the cation channel open and a desensitization gate at the cytoplasmic end of the channel closed [58].

Fig. 2.1

Torpedo electric organ AChR structure determined by electron crystallography . (a) The large extracellular domain contrasts with the smaller domain on the cytoplasmic surface. Rapsyn is a 43,000 Da peripheral membrane protein through which muscle AChRs are linked to actin in the cytoskeleton to concentrate them at the tips of folds in the postsynaptic membrane adjacent to active zones in the presynaptic membrane at which ACh is released [54]. (b) Here the polypeptide chains are shown as ribbon structures, highlighting the α1 and δ-subunits, at whose interface one of the two ACh-binding sites in this (α1)2β1γδ AChR is formed. Reprinted from Journal of Molecular Biology, 346(4), Nigel Unwin, Refined Structure of the Nicotinic Acetylcholine Receptor at 4Å Resolution, 967–989, Copyright 2005, with permission from Elsevier

X-ray crystallography of a molluscan glial ACh-binding protein (AChBP) has revealed the structure of the extracellular domain of an AChR-like protein at atomic resolution [59–62] (Fig. 2.2). Snail glia were found to release a water-soluble protein which modulated transmission by binding ACh. The cloned protein showed 24% sequence identity with the extracellular domain of human α7. It provides a good model for the basic structure of the extracellular domains of AChRs and other receptors in their superfamily. This is proven by the demonstration that the AChBP, with slight modification to match their interface, can form a chimera with the transmembrane portion of the 5HT3 receptor to form an ACh-gated cation channel [65]. Figure 2.2 shows that five AChBP subunits assemble as the extracellular domains of AChR subunits would around the vestibule of the channel. A buffer component was initially found to occupy what was expected to be the ACh-binding site, which is formed at the interface between subunits halfway up the side of the assembled protein. All of the contact amino acids for this site corresponded to ones in AChRs, which had been identified by affinity labeling or mutagenesis studies [43, 49, 66]. Subsequently, AChBP was crystallized in combinations with both agonists, such as nicotine [60], and antagonists, such as snake venom toxins [61]. Basically, most recognizable features were found about where they were expected to be from studies of native AChRs, providing confidence that the structure of the AChBP is relevant to that of AChRs.

Fig. 2.2

Mollusk AChBP structure determined by X-ray crystallography . (a) Atomic-resolution structure of the AChBP reveals the basic structure of AChR extracellular domains. It is a 62-Å-high cylinder that is 80 Å in diameter with an 18 Å diameter central hole. The structure is a homopentamer-like α7 AChRs. There are five ACh-binding sites at subunit interfaces, rather than the two ACh-binding sites expected in a muscle AChR heteropentamer at interfaces between α1 and δ-, γ- or ε-subunits. The ACh-binding site is occupied by the buffer component N-2-hydroxyethylpiperazine-N′-2 ethanesulfonic acid (HEPES). The adjacent disulfide-linked cysteine pair corresponding to α1 192–193, which is characteristic of all AChR α-subunits, is on a projection of what can be defined as the + side of the subunit. It intercalates with the − side of the adjacent subunit to form the ACh-binding site. As expected from studies of AChRs [27, 63], the sequence corresponding to the MIR of α1 subunits is located at the extracellular tip and oriented out away from the central axis of the subunit. The Cys-loop linked by a disulfide bond corresponding to that between cysteines 128 and 142 of α1 subunits, the signature loop characteristic of all subunits of the gene superfamily of which AChRs are a member, is located in the AChBP at the base near what would be the transmembrane portion of an AChR or the lipid bilayer. This loop sequence shows little homology with that of AChRs and is more hydrophilic than the sequences characteristic of AChRs. This loop contributes to the water solubility of AChBP but is at the interface between the extracellular and transmembrane domains of AChRs. Reprinted by permission from Macmillan Publishers Ltd: Nature, Brejc K, van Dijk WJ, Klaassen RV, Schuurmans M, van der Oost J, et al., Crystal structure of an ACh-binding protein reveals the ligand-binding domain of nicotinic receptors, 411, copyright 2001. (b) Shows a top view of an AChBP in its resting and active conformations. The C-loop is open in the resting state and closed in the active and desensitized states. Reprinted from Hansen SB, Sulzenbacher G, Huxford T, Marchot P, Taylor P, Bourne Y. Structures of aplysia AChBP complexes with nicotinic agonists and antagonists reveal distinctive binding interfaces and conformations. EMBO J 2005; 24: 3635–3646, with permission from John Wiley and Sons. Large movement of the C loop at the ACh-binding sites is propagated through the AChR to produce movement of the M2 transmembrane domain and ultimately opening of the channel gate located at the middle or bottom of the transmembrane domain [64]

α7 AChR subunits primarily form homomeric AChRs. The extracellular domain cleaved by protein engineering from α7 assembles into water-soluble pentamers with the ligand-binding properties of native α7 AChRs, but it does so inefficiently [67]. Thus, the molluscan AChBP probably contains sequence adaptations for efficient assembly and secretion as a water-soluble protein. Based on this hypothesis, a chimeric α7 ligand-binding domain, which shares 64% sequence identity and 71% similarity with native α7, was generated by combining sequences from α7 AChR with those from AChBP, and crystal structures of the resulting pentamer and its complexes with epibatidine and α-bungarotoxin were determined [68, 69]. This and other features of the structure will be discussed in more detail in subsequent sections.

Structures of AChR Subunits

All AChR subunits share several features . Figure 2.3 diagrammatically shows the transmembrane orientation of the mature polypeptide chain of a generic AChR subunit. To produce the mature polypeptide sequence, a signal sequence of about 20 amino acids is removed from the N-terminus of each subunit during translation as the N-terminal domain crosses the membrane into the lumen of the endoplasmic reticulum.

Fig. 2.3

Aspects of AChR structure. The transmembrane orientation of a generic AChR subunit is depicted in this diagram. The actual structure of the AChR shown in Fig. 2.1b and the large extracellular domain of the AChBP shown in Fig. 2.2 provide more details. The transmembrane domains M1–4 are depicted as largely α-helical. The overall shape of the subunit is depicted as rodlike. Five of these rods assemble in a pentagonal array to form the AChR shown in Fig. 2.1. The subunits are organized around the ion channel so that the amphipathic M2 transmembrane domain from each subunit contributes to the lining of the channel. In muscle AChRs, (e.g., with the subunit arrangement α1γα1δβ), and other heteromeric AChRs (e.g., with the subunit arrangement α4β2α4β2β2), there are only two ACh-binding sites at interfaces between the + side of α-subunits and complementary subunits, but small concerted conformation changes of all subunits are involved in activation and desensitization [55, 65, 70]. Thus, all subunits contribute to the conductance and gating of the channel, even if they are not part of an ACh-binding site. The amino acids lining the ACh-binding site have been identified by affinity labeling and mutagenesis studies [43, 66] and have been found to correspond well to those identified in the crystal structure of the AChBP and α7/AChBP chimera [60, 61, 68, 69]. Notice the predominance of aromatic amino acids in this region. As in ACh esterase [71], the quaternary amine group of ACh is thought to be bound through interactions with π electrons of these aromatic amino acids rather than ionic interactions with acidic amino acids. Note also that the ACh-binding site is formed from amino acids from three different parts of the extracellular domain of the α-subunit interacting with three parts of the complementary subunit and that the interaction is at the interface between the + side of the α-subunit and the − side of the complementary subunit. Thus, the site is ideally positioned to trigger small concerted conformation changes between subunits, thereby permitting low-energy binding events in the extracellular domain to regulate opening, closing, and desensitization of the ion channel gate near the cytoplasmic vestibule of the channel [50, 65, 71]. The amino acids lining the cation channel and accessible either from the extracellular or cytoplasmic surface have been determined largely by the substituted cysteine accessibility method (SCAM) [70]. The channel lining is thought to be formed by M2. The figure depicts the M1–M2 linker at the cytoplasmic end of M1 and M2. In the closed, resting state of the channel, only a short region is occluded and inaccessible to labeling. In the closed desensitized state, a larger region is occluded

The large N-terminal extracellular domain of each subunit, which consists of about 210 amino acids, contains a disulfide-linked loop (the Cys-loop) which is characteristic of all receptors in this superfamily. In α1 subunits, it extends from cysteine 128 to cysteine 142. This sequence is among the most conserved of all AChR subunit sequences. In the AChBP (see Fig. 2.2), this loop is located near what would be the lipid bilayer or extracellular surface of the transmembrane regions [59]. It is hydrophobic in AChRs but hydrophilic in the binding protein. A proline in the loop, which is conserved in all AChR subunits, is missing in the binding protein. Mutating this proline to glycine disrupts assembly of AChR subunits and prevents transport to the surface of assembled AChRs [72]. The extracellular domains of AChR subunits contain one or more glycosylation sites, and in all but the α7–9 subunits, which can form homomeric AChRs, there is an N-glycosylation site at position 141 adjacent to the disulfide bond of the signature loop.

Three closely spaced, highly conserved, largely α-helical transmembrane sequences (M1–M3) corresponding approximately to amino acids 220–310 extend between the large extracellular domain and the largest cytoplasmic domain. The N-terminal third of M1 and the hydrophilic side of M2 form each subunit’s contribution to the lining of the channel [1, 43, 70]. This will be described in slightly more detail in a subsequent section on the channel and gate.

The large cytoplasmic domain between the transmembrane sequences M3 and M4 comprises 110–270 amino acids (α4 having by far the largest). This is the most variable region in sequence between subunits and between species. Consequently, many subunit-specific antibodies bind in this region [73]. The large cytoplasmic domain, in contrast with the