myastenia gravis- past, present and future

8/8/2019 Myastenia Gravis- Past, Present and Future

1/12


2/12

science in medicine

2844 TheJournalofClinicalInvestigation http://www.jci.org Volume 116 Number 11 November 2006

Figure 1Structure of the NMJ. As it enters the muscle and approaches its target fibers, each motor neuron axon divides into branches that innervate

many individual muscle fibers. Each branch loses its myelin sheath and further subdivides into many presynaptic boutons, which contain ACh-

loaded synaptic vesicles and face the surface of the muscle fiber at the NMJ. The synaptic bouton and the muscle surface are separated by the

synaptic cleft, which contains AChE and proteins and proteoglycans involved in stabilizing the NMJ structure. The NMJ postsynaptic membrane

has characteristic deep folds, and the AChR is densely packed at the fold top. When the nerve action potential reaches the synaptic bouton, ACh

is released into the synaptic cleft, where it diffuses to reach and bind the AChR. ACh binding triggers the AChR ion channel opening, permitting

influx of Na+ into the muscle fiber. The resulting EPP activates voltage-gated Na+ channels at the bottom of the folds, leading to further Na+ influx

and spreading of the action potential along the muscle fiber. Other proteins, including Rapsyn, MuSK, and agrin, which are involved in AChR

clustering, are also present on the muscle membrane in close proximity to the AChR. MASC, myotube-associated specificity component; RATL,

rapsyn-associated transmembrane linker. Figure modified with permission from Lippincott Williams and Wilkins (126).


3/12

science in medicine

TheJournalofClinicalInvestigation http://www.jci.org Volume 116 Number 11 November 2006 2845

stabilizing the NMJ structure. The NMJ postsynaptic membranehas characteristic deep folds: the AChR is densely packed (about12,000 molecules per m2) in highly ordered hexagonal lattices ofmolecules at the tops of the folds.

Muscle AChR molecules are transmembrane proteins formed

by 5 subunits: 2 identical subunits, which contribute importantstructural elements to the ACh-binding sites, and 3 different buthomologous subunits, termed , (or ; see below), and . Muscleexpresses 2 developmentally regulated AChR isoforms. Embryonicmuscle expresses AChRs formed by, , , and subunits. Afterinnervation, expression of the subunit gene is substituted by thatof the homologous subunit gene to yield the adult AChR iso-form, a complex of, , , and subunits. Some adult muscles,notably the EOM, still express embryonic AChR (11).

When the nerve action potential reaches the synaptic bouton, thedepolarization opens voltage-gated Ca2+ channels on the presynap-tic membrane. This Ca2+ influx triggers fusion of synaptic vesicleswith the presynaptic membrane and ACh release. Quantal contentof anerve impulse refers to the number of ACh vesicles (quanta) releasedby that impulse. The ACh diffuses into the synaptic cleft (where itcan be hydrolyzed by AChE) and reaches and binds to AChR, therebytriggering the opening of its cation channels and influx of Na+ intothe muscle fiber. The resulting endplate potential (EPP) activates

voltage-gated Na+ channels, leading to further influx of Na+ andspreading of the action potential along the muscle fiber.

The postsynaptic transmembrane protein, muscle-specific tyro-sine kinase (MuSK) (Figure 1), is the main autoantigen in someMG patients (12). MuSK expression in both developing and maturemuscle is similar to that of AChR. In mature muscle, MuSK is pres-ent prominently only at the NMJ, where it is part of the receptorfor agrin. Agrin is a protein synthesized by motor neurons andsecreted into the synaptic basal lamina. The signaling mediated by

agrin/MuSK interaction triggers and maintains rapsyn-dependentclustering of AChR and other postsynaptic proteins (13). Rapsyn,a peripheral membrane protein exposed on the cytoplasmic sur-face of the postsynaptic membrane, is necessary for clustering of

AChR, with which it coclusters. Rapsyn and AChR are present inequimolar concentrations at the NMJ, and they may be physicallyassociated. Rapsyn causes clustering of NMJ proteins other thanthe AChR, including MuSK. Mice lacking agrin or MuSK fail toform NMJs and die at birth of profound muscle weakness, andtheir AChR and other synaptic proteins are uniformly expressedalong the muscle fibers (14).

NMJ properties that influence susceptibility to muscle weakness in MG. TheEPP generated in normal NMJs is larger than the threshold needed

to generate an action potential. This difference may vary in differentmuscles, as discussed below. Neuromuscular transmissionsafety fac-toris defined as the ratio between the actual EPP and the thresholdpotential required to generate the muscle action potential. Its reduc-tion is the electrophysiological defect that causes MG symptoms.

The quantal content of an impulse, the conduction propertiesand density of postsynaptic AChR, and the activity of AChE in thesynaptic cleft all contribute to the EPP (15). Also, the postsynapticfolds (Figure 1) form a high-resistance pathway that focuses end-plate current flow on voltage-gated Na+ channels in the depths ofthe folds, thereby enhancing the safety factor. A reduction in thenumber or activity of the AChR molecules at the NMJ decreases theEPP, which may still be adequate at rest; however, when the quantal

release of ACh is reduced after repetitive activity, the EPP may fallbelow the threshold needed to trigger the action potential.

NMJ properties vary among muscles and may influence musclesusceptibility to MG (15). This is well illustrated by the NMJ of theEOMs, which are especially susceptible to developing myasthenicweakness. The NMJs of EOM differ from those of skeletal musclein several ways. They have less prominent synaptic folds, and there-

fore fewer postsynaptic AChRs and Na+

channels, and a reducedsafety factor (16). They are subject to very high neuronal firingfrequency, making them prone to fatigue. Also, they express lessintrinsic complement regulators, making them more susceptibleto complement-mediated injury (17). In skeletal muscles, fast-twitch fibers have NMJs with greater quantal contents, a greaterdegree of postsynaptic folding (18), and higher postsynaptic sensi-tivity to ACh than slow-twitch NMJs (19), and they have increasedNa+ current in the NMJ region (20). These properties may makefast-twitch skeletal muscle f ibers less susceptible to myasthenicfailure than slow-twitch fibers.

Effector mechanisms of anti-AChR Abs . Anti-AChR Abs affect neuro-muscular transmission by at least 3 mechanisms: (a) binding andactivation of complement at the NMJ; (b) accelerated degradationof AChR molecules crosslinked by Ab (a process known as anti-genic modulation); and (c) functional AChR block (Figure 2).

The NMJs of MG patients and EAMG animals contain activa-tion fragments of complement component 3 (C3), the terminaland lytic complement component 9 (C9), and the membraneattack complex (MAC) (21). Different lines of indirect evidencesuggest that complement activation at the NMJ might be the pri-mary cause of AChR loss and failure of neuromuscular transmis-sion (Figure 2A): (a) complement depletion protects animals fromEAMG (22); (b) administration of Abs that block complementcomponent 6 (anti-C6) (23) or a complement inhibitor (solubleCR1) (24) protects rodents from EAMG; (c) mice with a reducedcomplement function because of a genetic deficit of complement

components are resistant or less susceptible to EAMG inductionthan mice with normal complement (25); (d) IL-12deficient mice,which synthesize Th1-driven, complement-fixing Abs poorly (26),develop minimal EAMG symptoms after AChR immunizationin spite of robust anti-AChR Ab synthesis; moreover, their NMJscontain Abs but not complement, suggesting that anti-AChR Absthat do not activate complement do not effectively compromiseneuromuscular transmission.

Cells are protected from activation of autologous complementon their surfaces by the so-called intrinsic complement regulators.These include the decay-accelerating factor (DAF or CD55), themembrane cofactor protein (MCP or CD46), and the membraneinhibitor of reactive lysis (MIRL or CD59) (2729). Consistent

with an important role of complement in EAMG, passive transferof EAMG with anti-AChR Abs causes more severe muscle weaknessin DAF-deficient mice than in wild-type mice (30).

Antigenic modulation is the ability of an Ab to cross-link 2 anti-gen molecules, thereby triggering a cellular signal that causes accel-erated endocytosis and degradation of the cross-linked molecules(Figure 2B). IgG from MG patients causes antigenic modulation ofmuscle AChR in vivo and in vitro (31). If accelerated degradationis not compensated by increased AChR synthesis (32), it will leadto a reduction of the available AChR molecules at the NMJ andmyasthenic symptoms. This property can be used as a diagnostictest for MG (32). However, not all anti-AChR Abs cause antigenicmodulation because, even though all IgG Abs have 2 antigen-bind-

ing sites, the epitope location on the AChR surface may restrict theability of Abs to cross-link a second AChR molecule (33).


4/12

science in medicine


Functional AChR block due to Ab binding to the ACh-binding site(Figure 2C) is an uncommon pathogenic mechanism in MG, yet itmay be clinically important. This is because the presence of Ab to theACh-binding site of the AChR causes acute, severe muscle weaknessin rodents without either inflammation or necrosis of the NMJ (34).Many MG patients have low levels of anti-AChR Abs that recognizethe ACh-binding site (35); these might block the AChR in spite oftheir low concentration and contribute to acute myasthenic crises.

Role of CD4+

T cells in MG. Pathogenic anti-AChR Abs are high-affinity IgGs, whose synthesis requires that activated CD4+ T cells

interact with B cells, resulting in low-affinity anti-AChR Abs. Thistriggers somatic mutations of the Ig genes, leading to synthesis ofhigh-affinity Abs. B cells secreting low-affinity anti-AChR Abs arecommon; for example, about 10% of monoclonal IgGs in multiplemyeloma patients bind muscle AChRs (36). Myelomas are rarelyassociated with MG, perhaps because of the low aff inity of theiranti-AChR Abs (36).

MG patients have AChR-specific CD4+ T cells with T helper

function in the blood and thymus (37), and their symptomsimprove after thymectomy (38) or treatment with anti-CD4 Abs

Figure 2Effector mechanisms of anti-AChR Abs. (A)Ab binding to the AChR activates the complement cascade, resulting in the formation of membrane

attack complex (MAC) and localized destruction of the postsynaptic NMJ membrane. This ultimately leads to a simplified, altered morphology of the

postsynaptic membrane of the NMJ of MG patients, which lacks the normal deep folds and has a relatively flat surface. (B)Abs cross-link AChR mol-

ecules on the NMJ postsynaptic membrane, causing endocytosis of the cross-linked AChR molecules and their degradation (antigenic modulation).This ultimately leads to a reduced number of AChR molecules on the postsynaptic membrane. (C)Ab binding the ACh-binding sites of the AChR

causes functional block of the AChR by interfering with binding of ACh released at the NMJ. This results in failure of neuromuscular transmission.


5/12

science in medicine


(39). Moreover, in MG patients with AIDS, a reduction in CD4+T cells correlates with myasthenic symptom improvement (40).Studies in experimental systems directly demonstrated thatCD4+ T cells are necessary for the development of MG symp-toms. SCID mice engrafted with blood lymphocytes from MG

patients produce antihuman AChR Abs and develop MG symp-toms only if the grafted cells include CD4+ T cells (41). Also,mice genetically deficient in functional CD4+ T cells do notdevelop EAMG (42). Healthy subjects may have AChR-specificCD4+ T cells, which do not cause a clinically significant autoim-mune response, probably because of mechanisms of immuno-logical tolerance, which fail in autoimmunity.

Blood CD4+ T cells from both gMG and oMG patients respondto the AChR in vitro. Those from gMG patients respond to all the

AChR subunits (43), and their epitope repertoire expands as thedisease progresses (44). A few AChR sequences were recognizedby most gMG patients (e.g., see ref. 45). CD4+ T cell lines specif-ic for these universal AChR epitopes, when grafted into SCIDmice, support anti-AChR Ab production by B cells, which result inMG symptoms (41). The responses to AChR and AChR epitopesof CD4+ T cells from oMG patients were weaker and less stableover time than in those of gMG patients (44). Also, CD4 + T cellsfrom individual oMG patients rarely recognize all AChR subunits,even when the disease has lasted for many years (44). It is not clearwhether CD4+ T cells from oMG patients recognize the embryonic or the adult subunit or both (44).

The pathogenic role of anti-AChR CD4+ T cells in MG and EAMGexplains the important role of MHC class II molecules, which pres-ent the antigen epitopes to the specific CD4+ T cells. In mice, sus-ceptibility to EAMG correlates with the class II molecule allelesthat they express (46). Moreover, a mutation of the gene encodingthe subunit of the I-Ab molecule converts the highly susceptible

C57BL/6 strain of mice into the EAMG-resistant BM12 strain ofmice (47). MG patients, like patients with other autoimmune dis-eases, express some MHC (HLA) alleles with higher frequency thanexpected in the general population. HLA gene products found fre-quently in MG patients include the B8 and A1 class I molecule, theDR3/DW3 class II molecule, and certain DQ allele products. Somestudies have used mice that express individual DR or DQ allelestransgenically, to determine whether some DR or DQ moleculesinfluence the development of EAMG. Those studies confirmedthat expression of the DQ8 and DR3 molecules correlated withEAMG susceptibility and expression of the DQ6 molecule corre-lated with resistance (48, 49).

Role of CD4+ T cell subtypes and cytokines in MG and EAMG. Differenti-

ated CD4+ T cells are classified into subtypes based on the cytokinesthey secrete. Among them, Th1 and Th2 cells have different and attimes opposing functions (50) (Figure 3). Th1 cells secrete proin-flammatory cytokines, such as IL-2, IFN-, and TNF-, which areimportant in cell-mediated immune responses. Th2 cells secreteantiinflammatory cytokines, such as IL-4, IL-6, and IL-10, which arealso important inducers of humoral immune responses. Moreover,IL-4 stimulates differentiation of Th3 cells, which secrete TGF-and are involved in immunosuppressive mechanisms (51). Both Th1and Th2 cytokines may induce Ab synthesis. However, they supportthe synthesis of different Ig types. In mice (and likely in humans)Th1 cells induce IgG subclasses that bind and activate complementefficiently whereas Th2 cells induce Ig isotypes and IgG subclasses

that fix complement poorly or not at all. In rats, both Th1 and Th2cells induce complement-fixing IgG subclasses (52).

MG patients have abundant anti-AChR Th1 cells in the bloodthat recognize many AChR epitopes (e.g., see ref. 53) and inducesynthesis of pathogenic anti-AChR Abs when grafted together withB cells and macrophages from the same patient into SCID mice (41).MG patients also have anti-AChR Th2 and Th3 cells in the blood

(54). Mice congenitally lacking or overexpressing a cytokine havebeen used to study the role of cytokines in EAMG. Those studiessuggested that Th1 cells and their cytokines are needed for EAMGdevelopment, probably because they induce expression of comple-ment-binding pathogenic anti-AChR Abs (e.g., see refs. 26, 55).Theresistance to EAMG induction conferred by genetic deficiency inTNF- or TNF receptor proteins (56) is overcome by treatmentwith IL-12, confirming that sensitization and differentiation ofTh1 cells is important for EAMG development (57). Other studiesdemonstrated the important role of Th1 cells in EAMG by chang-ing the concentration of Th1 cytokines in rodents with normalgenes for cytokines and their receptors. For example, treatment ofrats with antiTNF-Abs suppresses EAMG development (58), andtreatment of mice with a soluble recombinant form of the humanTNF receptor, able to outcompete mouse TNF- for binding to themouse receptor, significantly improves symptoms of establishedEAMG (59). Moreover, estrogen enhances EAMG development inmice by promoting augmented IL-12 production by AChR-specificTh1 cells, suggesting that estrogens mediate sex differences in auto-immunity because of a Th1-mediated mechanism (60) (Figure 3).Proinflammatory Th1 cytokines induce expression of MHC class IImolecules in muscle, thereby facilitating presentation of muscle

AChR epitopes and further expansion of activated anti-AChR CD4 +T cells (61). Increased IFN- production may explain the increasedexpression of IFN-induced chemokines and monokines and theirreceptors in muscle, thymus, and lymph nodes in MG patients andrats with EAMG. A decrease in chemokine expression correlates

with decreased severity of symptoms (62). Anti-AChR Th2 cells have complex and contrasting roles in

EAMG. They can be protective (e.g., refs. 63, 64), but the Th2cytokines IL-5, IL-6, and IL-10 also foster EAMG development (e.g.,refs. 6567). The resistance to EAMG of mice genetically deficientin IL-6 is associated with a reduced formation of germinal centersin the spleen and a reduced synthesis of anti-AChR IgG Abs whilethe anti-AChR IgM response is normal, suggesting a defect in Tcell help and in the switching from IgM to IgG isotypes (67).

Other CD4+ T cell subtypes may have a role in MG. CD4+ T cellsthat express the CD25 marker and the transcription factor Foxp3are known as Tregs and areimportant in maintaining self toler-ance (Figure 3). Tregs in MG patients may be functionally impaired

(68). In addition, the number of circulating Tregs has been shownto increase after thymectomy, and the increase correlated withsymptom improvement (69).

Role of NK and NKT cells in MG and EAMG. CD1-drestricted NKTcells may be involved in maintaining self tolerance. In EAMG andMG, NKT cells and Tregs may cooperate in regulating the anti-

AChR response. In AChR-immunized mice, activation of NKT cellsby a synthetic glycolipid agonist inhibits EAMG development; thesetherapeutic effects are likely mediated by the increase in numberand modulatory function of Tregs induced by the glycolipid (70).

NK cells can also influence the development of EAMG and pos-sibly MG. In mice, NK cells are necessary for EAMG development(71). The permissive role of NK cells in EAMG is due to their

secreting IFN-, thereby permitting and enhancing the sensitiza-tion of Th1 cells. IL-18 is an important growth and differentiation


6/12

science in medicine


factor for both NK cells and Th1 cells, especially in cooperationwith IL-12 (55). Thus, IL-18 may be especially important in MGand EAMG pathogenesis. This is supported by the finding that

IL-18deficient mice are resistant to EAMG and pharmacologicblock of IL-18 suppresses EAMG (55, 71). The finding that MGpatients have increased serum levels of IL-18, which are higherin gMG than in oMG patients and tend to decrease with clinicalimprovement, supports a role for IL-18 in human MG (72).

Other autoantigens in MG. Up to 20% of MG patients do not haveanti-AChR Abs and are consequently known as seronegativepatients (73). Many seronegative patients (31%41% in most stud-ies) (74) develop Abs against MuSK. Lower frequencies of patientswith anti-MuSK Abs among seronegative MG patients may occurin particular ethnic groups or geographic locations (e.g., Chinese,Norwegians) (74, 75); this might reflect environmental or geneticsusceptibility factors. MG patients with anti-MuSK Abs never have

anti-AChR Abs, with the notable exception of a group of Japanesepatients investigated in one study (74, 76).

The agrin/MuSK signaling pathway likely maintains the struc-tural and functional integrity of the postsynaptic NMJ apparatusalso in adult muscle. Anti-MuSK Abs and IgG from MG patients

who have anti-MuSK Abs disrupt AChR aggregation in myotubesfrom MuSK-immunized animals and block agrin-induced AChRclustering (12, 77). Also, immunization of animals with MuSKfragments induces myasthenic symptoms (77). This suggests thatanti-MuSK Abs affect the agrin-dependent maintenance of AChRclusters at the NMJ, ultimately leading to reduced AChR numbers.Complement-mediated damage might also be responsible for lossof AChRs and NMJ damage. However, some MG patients withanti-MuSK Abs do not experience AChR loss at the NMJ (78), per-haps because anti-MuSK Abs are mainly IgG4, which do not bindcomplement efficiently (79). Moreover, some studies have foundthat anti-MuSK Abs do not cause substantial AChR loss, comple-ment deposition, or morphologic damage at the NMJ (78). Anti-

MuSK Abs may have other pathogenic mechanisms. In this regard,a recent study that investigated the effects of anti-MuSK Abcon-

Figure 3Cytokine network and cells involved in the pathogenesis and immunoregulation of MG. Th1 cytokines stimulate production of IgG subclasses

that bind and activate complement effectively, whereas Th2 cytokines stimulate the production of Ig classes and IgG subclasses that do not. The

Th2 cytokine IL-4 is also a differentiation factor for Th3 cells, immunosuppressive cells that secrete TGF-. The Th1 cytokine IFN- stimulates

expression of MHC class II molecules on the muscle cell membrane, thus facilitating presentation of muscle AChR. The IL-18 secreted by APCs

favors the differentiation of Th1 cells both directly and indirectly through the action of NK cells. CD1-drestricted NKT cells can activate Tregs,

thereby inhibiting autoimmune processes. See text for further details.


7/12

science in medicine


taining sera from MG patients on muscle cell cultures found thatsome sera inhibit cell proliferation by causing cell cycle arrest andcause downregulation of the expression of AChR subunits, rapsyn,

and other muscle proteins (80).Some seronegative patients who do not have either anti-AChR or

anti-MuSK Abs might have a plasma factor that activates a secondmessenger pathway in the muscle, resulting in phosphorylationand inactivation of the AChR (81).MG patients may also synthe-size Abs against nonmuscle-specific proteins, such as myofibril-lar proteins (82). Some of those Abs, especially anti-myosin Absand antifast troponin Abs, may cross-react with the AChR (83).MG patients with thymoma have Abs against titin (84) and theryanodine receptor (85).

Diagnosis of MG

Textbook descriptions suggest that clinical diagnosis of MG should

be straightforward. However, this is not always the case. Delayedor missed diagnoses occur frequently. This is because MG is rela-tively rare and therefore unfamiliar to practitioners, and its fluc-tuating muscular weakness can be puzzling. Once suspected, thediagnosis of MG relies on serological tests that detect anti-AChR oranti-MuSK Abs and sometimes Abs against other muscle proteins(actin, striational protein) and electrodiagnostic tests that detectcharacteristic defects in neuromuscular transmission (Table 1).

The discovery of anti-AChR Abs yielded useful diagnostic tests.The demonstration of serum anti-AChR Abs proves the diagno-sis of MG. However, their absence does not exclude it becauseanti-AChR Abs are detectable only in 80%90% of gMG patientsand 30%50% of oMG patients (86). The most commonly used

anti-AChR Ab assay measures the serum levels of an autoanti-body that precipitates muscle AChR extracted from appropriate

human cell lines or amputated tissue. The AChR is detected bythe binding of a radiolabeled, irreversible cholinergic antagonist,-bungarotoxin (86). Thus, this assay cannot reveal anti-AChR

Abs against the cholinergic site, which can be assayed by testingthe ability of the patients IgG to competitively inhibit the bind-ing of cholinergic ligands (86). Yet another serologic test mea-sures the ability of the patients sera or IgG to induce antigenicmodulation of the AChR in cell cultures (86). In patients withMG symptoms who do not have detectable anti-AChR Abs (73),serologic diagnosis of MG can be attempted by determining thepresence of anti-MuSK Abs (12). About 5% of MG patients haveneither anti-AChR nor anti-MuSK Abs.

Standard electrodiagnostic tests that utilize repetitive stimula-tion of peripheral nerves are commonly used to detect neuromus-cular transmission defect in MG. They are relatively sensitive andreliable (Table 1). The most sensitive electrodiagnostic test for MG

is single-fiber electromyography, which reveals deficits of neuro-muscular transmission in 95%99% of MG patients and excludesthe diagnosis of MG when it yields normal results (87). It selectivelyrecords action potentials from a small number (usually 2 or 3) ofmuscle fibers innervated by a single motor unit. The amount of

ACh released at the NMJ at different times has a small variability,resulting in comparable variations in the rise of EPP and the musclefiber pair interpotential intervals. This variability is highly sensitiveto neuromuscular transmission abnormalities and is increased inMG patients. Neuromuscular blockingis the failure of transmis-sion of one of the potentials, when one of the muscle fibers fails totransmit an action potential because the EPP does not reach thenecessary threshold. The American Association of Neuromuscular

& Electrodiagnostic Medicine has developed guidelines for electro-diagnostic testing for evaluation of MG (88).

Table 1

Diagnostic tests for MG

Test Sensitivity Comments

Clinicaltests

Edrophonium Detectable in 80%90% of MG patients Edrophonium testing is safe. However, because of known adverse

effects, alternative nonpharmacological tests have been developed.

Sleep Criteria poorly defined

Ice-pack test Criteria poorly defined

AssaysofserumAbs

Anti-AChR Detectable in approximately 80%90% of May also be observed in patients with Lambert-Eaton syndrome and

gMG patients and 30%50% of oMG patients motor neuron disease, thymoma patients without MG, and relatives

of MG patients.

Anti-MuSK Detectable in approximately 30%40% of

anti-AChR Abnegative gMG patients and

rarely in oMG patients

Anti-striational protein Detectable in 80% of thymomatous MG More common in older patients. First autoantibody detected in MG.

patients and 30% of nonthymomatous However, poor specificity makes this test nondiagnostic.

MG patients

Electrodiagnostictests

Repetitive stimulation Positive in approximately 90% of gMG

of peripheral nerves patients and 30%60% of oMG patients

Single-fiber Positive in 95%99% of MG patients Despite its sensitivity, single-fiber electromyography is not the test of

electromyography choice because it is dependent on operator skills and patient

cooperation. Also, results are abnormal in neuropathies and motor

neuron and muscle diseases.


8/12

science in medicine


Therapeutic MG managementThe current management of MG includes the use of anticholines-terase drugs for temporary improvement of neuromuscular trans-mission, removal of anti-AChR Abs by plasma exchange or specificimmunoadsorption procedures, use of nonspecific immunosup-pressants or immunomodulators to curb the anti-AChR response,and thymectomy (Table 2). No current treatment targets the autoim-mune defect of MG selectively. However, the advances in understand-ing the pathogenesis of autoimmunity and MG suggest that newapproaches that will curb or even eliminate the anti-AChR responsespecifically will be forthcoming (see Table 2 and The future).

Anticholinesterase drugs improve myasthenic symptoms innearly all patients, but they fully relieve the symptoms in only a

few. Thus, most patients require additional immunosuppressivetreatment (see below). A novel approach to long-term therapeuticinhibition of AChE activity in MG patients is based on the obser-

vation that in the NMJ of both MG patients and EAMG animals,there is enhanced transcription and altered splicing of AChEpre-mRNA, with accumulation of a normally rare readthrough

AChE-R variant (89). The commonly occurring synaptic AChE-Svariant forms membrane multimers. In contrast, AChE-R existsas soluble monomers that lack the carboxyterminal cysteineneeded for membrane attachment. Thus, AChE-R permeates thesynaptic space and degrades ACh before it reaches the postsyn-aptic membrane, thereby compromising AChR activation. Theseobservations prompted the design and use of EN101, an antisense

oligonucleotide that suppresses the expression of AChE-R. EN101normalizes neuromuscular transmission in EAMG by modulat-

ing the synthesis of AChE variants, thereby affecting the rate ofACh hydrolysis and the efficacy of AChR activation (90). EN101 isundergoing human trials.

Despite the lack of large controlled trials, corticosteroids are theimmunosuppressive agent most frequently used for the treatmentof MG and the most consistently effective (91). They are adminis-tered at high doses for several months and at low doses for years.

Anti-AChR Ab levels decrease in the first months of therapy. Mostpatients obtain a clinical benefit, which may be related to reduc-tion of lymphocyte differentiation and proliferation, redistribu-tion of lymphocytes into tissues that are not sites of immunore-activity, changes in cytokine expression (primarily of TNF, IL-1,and IL-2), inhibition of macrophage function and of antigen pro-

cessing and presentation, or a possible increase in muscle AChRsynthesis. The shortcoming of corticosteroid treatment is the fre-quent steroid-related complications. This has motivated the useof other immunosuppressants, either as steroid-sparing agentsor as a substitute for corticosteroids; a proportion of MG patientscan be treated successfully without corticosteroids.

Azathioprine, a purine analog, reduces nucleic acid synthesis,thereby interfering with T and B cell proliferation. It has beenutilized as a single immunosuppressant agent in MG since the1970s, and large, retrospective reports support its efficacy (92). Arandomized, double-blind trial demonstrated its efficacy also as asteroid-sparing agent. Its major disadvantage is the delayed clini-cal response, which may take up to 15 months (93).

Cyclophosphamide administered intravenously and orallyis an effective treatment for MG (94); more than half of the

Table 2

Current and potential therapies for MG management

Approach Therapeutics Mechanismofaction

Current

Modulation of neuromuscular Cholinesterase inhibitors Prolong ACh activity

transmission EN101 Antisense nucleotide that suppresses the synthesis of AChE-R, the soluble

and most effective form of synaptic AChE

Immunomodulation Thymectomy Probably multiple effects

Plasma exchange Removal of Ab

IVIg Multiple effects

Immunoadsorption Removal of anti-AChR Ab

General immunosuppression Prednisone Multiple effects

Azathioprine Purine analog. Inhibits T and B cell proliferation

Cyclosporine, Tacrolimus Blocks T cell activation and growth

Mycophenolate mofetil Inhibits guanosine nucleotide synthesis and selectively inhibits activated T cells

Etanercept Recombinant soluble TNF receptor that competitively inhibits TNF- binding

Potential

Complement inhibition Anti-C6 Abs Abs that block complement component 6 and protect rodents from EAMGCR1 Soluble recombinant receptor that competitively inhibits complement

Induction of tolerance to AChR Multiple approaches Administration of AChR or portion of sequence by tolerance-inducing routes

(e.g., oral, nasal)

T cell vaccination

Use of modified APCs loaded with AChR epitopes

Depletion of AChR-specific B Multiple approaches For example, AChR toxin to eliminated B cells, Fas ligand conjugates targeted

or T cells to T cells

Interruption of MHC class II, Multiple approaches For example, APL (peptide epitope analogs that cannot activate CD4+ T cells)

epitope peptide, T cell receptor,

and CD4+ complex


9/12

science in medicine


patients became asymptomatic after 1 year of treatment. Itsdelayed effect and undesirable side effects (hair loss, and lessfrequently, nausea, vomiting, anorexia, and skin discoloration)limits its use to the management of patients who do not respondto other immunosuppressive treatments.

Cyclosporine is a cyclic undecapeptide that blocks the synthesisof cytokines (and especially IL-2), IL-2 receptors, and other pro-teins critical in the function of CD4+ T cells. Its efficacy in MG wasfirst suggested by a small, randomized, placebo-controlled study,which has not been followed by similar studies on larger groups ofpatients (95). However, larger retrospective studies have supportedits use as a steroid-sparing agent (96).

Tacrolimus, a macrolide antibiotic, is similar to cyclosporine inits biological activity. No large controlled studies on the eff icacyof tacrolimus in MG are available. However, a large retrospec-tive study of treatment-resistant patients supports its use as asteroid-sparing agent (97).

In the US, mycophenolate mofetil is increasingly used both as asteroid-sparing agent and as a stand-alone immunosuppressant.Unfortunately, the investigations that support its use studiedsmall patient groups or were of a retrospective nature; thereforethe value of their conclusions is limited (98). Those studies sug-gest that mycophenolate has limited toxicity, consistent with itsselective inhibitory activity of guanosine nucleotide synthesis inactivated T and B cells, which should limit its effects on other celltypes (99). Randomized, controlled trials to evaluate mycopheno-late mofetil usefulness in MG treatment are under way.

For some MG patients, symptoms do not improve with the use ofcorticosteroids or one of the immunosuppressive agents describedabove, alone or in combination, and they may develop intolerableside effects prompting consideration of other therapeutic options.MG patients resistant to therapy have been successfully treated

with cyclophosphamide in combination with bone marrow trans-plant (100) or with rituximab, a monoclonal Ab against the B cellsurface marker CD20 (101). Etanercept, a soluble, recombinantTNF receptor that competitively blocks the action of TNF-, hasbeen shown to have a steroid-sparing effects in studies on smallgroups of patients (102). Plasma exchange and i.v. Ig (IVIg) (dis-cussed below) can be used as chronic therapy although their ben-efit has not been documented in rigorously designed studies.

Plasma exchange (plasmapheresis) and IVIg are used for acutemanagement of severe muscular weakness. Plasmapheresis involvesreplacing 11.5 times the plasma volume with saline, albumin, orplasma protein fraction, leading to a reduction of serum AChR

Ab levels (103). For IVIg therapy, Ig isolated from pooled human

plasma by ethanol cryoprecipitation is administered for 5 days(0.4 g/kg/day). Fewer infusions at higher doses are also used. Themechanism of action of IVIg is complex and likely includes inhibi-tion of cytokines, competition with autoantibodies, inhibition ofcomplement deposition, interference with binding of Fc receptoron macrophages and Ig receptor on B cells, and interference withantigen recognition by sensitized T cells (104). More specific tech-niques to remove pathogenic anti-AChR Abs utilizing immunoad-sorption have been developed recently, which offer a more targetedapproach to MG treatment (105).

The observations that lead to therapeutic ablation of the thymusin MG have been mentioned above. However, the clinical efficacyof thymectomy has been questioned because the evidence support-

ing its use is not solid (106). Retrospective studies yielded widelydifferent conclusions, but they included patients whose thymi were

removed by different surgical approaches, and they used differ-ent definitions of remission and limited statistical analysis (107).Moreover, corticosteroids and other immunosuppressive agents arecommonly used by MG patients undergoing thymectomy, makingit difficult to ascertain whether thymectomy provides additional

benefits. Usually, thymectomy is performed on patients early in thecourse of their disease and restricted to patients younger than 60. Areview by the American Academy of Neurology concluded that, whilethymectomy should be considered a treatment option, its benefits innonthymomatous MG have not been firmly established (106) andthat a prospective, controlled, randomized study with standardizedmedical therapy for all patients is needed. Also, the NIH is sponsor-ing a clinical trial to determine whether the extended transsternalthymectomy reduces corticosteroid requirements for patients with

AChR Abpositive nonthymomatous gMG (108).Thymectomy may not be a viable therapeutic approach for anti-

MuSK Abpositive patients because their thymi lack the germinalcenters and the infiltrates of lymphocytes that characterize thymifrom patients who have anti-AChR Abs. This suggests that a differ-ent pathologic mechanism occurs in anti-MuSK Abpositive andanti-AChR Abpositive MG.

Lack of well-designed investigations with appropriate statisticalpower undermines the rationale for use of the current immuno-therapies for MG. Moreover, there are no rigorous trials that com-pare the efficacy of the various immunosuppressive approaches.Different factors contribute to this situation, including the vari-able clinical presentations and time course of MG and its low inci-dence; a lack of agreed-upon clinical scales to measure the outcomeof the treatment; and a suboptimal cooperation among the majorcenters that care for MG patients during clinical trials (109).

The future

Complement inhibition is an attractive future therapeutic approachfor MG because it is effective in rodent EAMG (e.g., ref. 24). More-over, anti-C5 inhibitors show short-term safety and are effective in a

variety of human disorders, including myocardial infarction (110),coronary artery bypass graft surgery (111), and lung transplanta-tion (112). Thus, therapeutic approaches based on inhibition ofcomplement activation will likely be tried for MG in the future.

However, the ultimate goal for MG treatment is to eradicate therogue anti-AChR autoimmune response specifically and reestab-lish tolerance to the AChR without affecting the other functions ofthe immune system or causing other adverse effects. Such targetedimmunosuppressive approaches (Table 2) are still far from clinicaluse. However, their success in EAMG suggests that approaches for

specific modulation of the autoimmune anti-AChR response maybecome part of MG patient care in the next decade. We will sum-marize here the different approaches that have proven successful forthe prevention and treatment of EAMG induced by immunizationwith AChR. We will also analyze the possible technical and biologicallimitations to their application for the treatment of human MG.

Approaches that have proven successful in rodent EAMG includethe following: (a) administration of AChR or parts of its sequencein a manner known to induce tolerance; (b) depletion of AChR-specific B cells or T cells; and (c) interference with formation of thecomplex between MHC class II molecules, epitope peptide, T cellreceptor, and CD4 molecule.

Antigen presentation under special circumstances may lead to

antigen-specific tolerance in adult animals rather than activatedCD4+ T cells. Earlier studies showed that in rats, presentation of


10/12

science in medicine


AChR epitopes by unsuitable APCs (fixed B cells that had beenincubated with AChR under conditions favoring AChR uptake andprocessing) caused unresponsiveness of the AChR-specific CD4+T cells to further stimulation with AChR (113). More recently,several studies have demonstrated that DCs, especially after treat-

ment with TGF-, IFN-, or IL-10, when injected into rats withdeveloping or ongoing EAMG, suppressed or ameliorated themyasthenic symptoms (114116). The effect was correlated witha reduced production of anti-AChR Abs without a reduced prolif-erative response of T cells to the AChR. Approaches based on theuse of tolerance-inducing APCs, which should present all AChRepitopes and therefore influence all AChR-specific T cells, mightbe useful for the treatment of MG. Should pulsing of the APCswith human AChR be needed, biosynthetic human AChR subunitscould be used as antigens.

Mucosal or subcutaneous administration of AChR or syntheticor biosynthetic AChR peptides to rodents approaches known toinduce antigen-specific tolerance in adult animals prevented ordelayed EAMG development (e.g., see refs. 117119). Dependingon the dose of the antigen administered, anergy/deletion of anti-gen-specific T cells (at high doses) and/or expansion of cells pro-ducing immunosuppressive cytokines (TGF-, IL-4, IL-10) (at lowdoses) are major mechanisms in mucosal tolerance induction. Theuse of mucosal tolerization procedures in human MG, however,is problematic because those procedures can be a double-edgedsword (120); they reduce AChR-specific CD4+ T cell responses butmay also stimulate AChR-specific B cells to produce Abs, therebyworsening the disease. Also, a large amount of human AChRswould be required, which may be difficult to obtain.

Conjugates of a toxin with AChR or synthetic AChR sequences,when administered to animals with EAMG, eliminated B cellsproducing anti-AChR Abs (121). This is probably because the

AChR moiety of the conjugate docks onto the membrane-boundAbs of AChR-specific B cells, which can then be killed by the toxicdomain. This approach has 2 caveats. First, the toxin may damageother cells. Second, anti-AChR CD4+ T cells can recruit new B cellsto synthesize more anti-AChR Abs.

AChR-specific CD4+ T cells can be specifically eliminated invitro by APCs genetically engineered to express relevant portionsof the AChR, Fas ligand (to eliminate the activated AChR-specificT cells with which they interact), and a portion of Fas-associateddeath domain, which prevents self-destruction by the Fas ligand(122). It is not known yet whether this strategy can be safely usedto modulate EAMG in vivo.

Activation of CD4+ T cells requires interaction and stable bind-

ing of several proteins on the surfaces of the CD4 + T cell and ofthe APC. In experimental systems, interfering with formation ofthis complex usually reduced the activity of autoimmune CD4+ Tcells. This may be obtained by administering or inducing Abs thatrecognize the binding site for the antigen of the T cell receptor

(known as T cell vaccination) (123). T cell vaccination is alreadyused in clinical trials for the treatment of multiple sclerosis, rheu-matoid arthritis, and psoriasis (124). It is effective in EAMG, and itis a promising future strategy for the treatment of MG (124). Themechanisms of action of T cell vaccination are complex, and they

likely include the induction of modulatory CD4+

and CD8+

T cells(124). Another approach used synthetic peptide analogs of an epi-tope recognized by autoimmune CD4+ T cells that bind the MHCclass II molecules but cannot stimulate the specific CD4+ cells.These are known as altered peptide ligands (APLs). APLs com-pete with peptide epitopes derived from the autoantigen, therebyturning off the autoimmune response. APLs might also stimulatemodulatory antiinflammatory CD4+ T cells or anergize the patho-genic CD4+ T cells (125). The rich epitope repertoire of anti-AChRCD4+ T cells in MG patients reduces the therapeutic potential ofapproaches that interfere with activation of specific CD4+ T cells;targeting only a few epitopes may not significantly reduce the anti-

AChR response. Moreover, these treatments are likely to produceonly transient improvement that ceases when administration ofthe antiT cell Ab is discontinued.

MG and EAMG have offered unique opportunities to investi-gate the molecular mechanisms of an Ab-mediated autoimmunedisease. Many factors have contributed to making MG the bestunderstood human autoimmune disease. These include the sim-plicity of the pathogenic mechanism in MG, where NMJ failureexplains all symptoms; the deeper understanding of the structureand the function of the NMJ and its molecular components, mostnotably, the AChR; and the increasing understanding of the mech-anisms that modulate immune responses and maintain tolerance.Hopefully increasing knowledge of the immunobiology of MGwill form a foundation for designing new and specific therapeuticapproaches aimed at curbing the rogue autoimmune response and

reestablishing immunological tolerance without interfering withthe other immune functions.

If this expectation is fulfilled, MG, which has been a benchmarkto understanding autoimmunity in humans, will become a refer-ence point for the design of specific immunosuppressive treat-ments of other autoimmune Ab-mediated diseases.

Acknowledgments

The authors are supported by grants from the NIH (NS23919 toB.M. Conti-Fine; EY013238 and R24EY014837 to H.J. Kaminski)and from the Muscular Dystrophy Association of America (to M.Milani and B.M. Conti-Fine).

Address correspondence to: Bianca M. Conti-Fine, Departmentof Biochemistry, Molecular Biology, and Biophysics, University ofMinnesota, 6-155 Jackson Hall, 321 Church St. SE, Minneapolis,Minnesota 55455, USA. Phone: (612) 624-6796; Fax: (612) 625-2163; E-mail: [email protected].

1. Robertson, D.N. 2000. Enumerating neurology.Brain.123:663664.

2. Marsteller, H.B. 1988. The first American case ofmyasthenia gravis.Arch. Neurol.45:185187.

3. Pascuzzi, R.M. 1994. The history of myastheniagravis.Neurol. Clin.12:231242.

4. Keesey, J.C. 2002. Crisis in myasthenia gravis: anhistorical perspective.Muscle Nerve.26:13.

5. Walker, M.B. 1935. Case showing the effect ofprostigmin on myasthenia gravis.Proc. R. Soc. Med.

28:759761.6. Nastuk, W.L., Strauss, A.J., and Osserman, K.E.

1959. Search for a neuromuscular blocking agentin the blood of patients with myasthenia gravis.

Am. J. Med.26:394409.7. Simpson, J.A. 1960. Myasthenia gravis, a new

hypothesis. Scott. Med. J.5:419436.8. Patrick, J. , and Lind strom, J. 19 73. Autoim-

mune response to acetylcholine receptor. Science.180:871872.

9. Rowland, L.P. 1980. Controversies about the treat-ment of myasthenia gravis. J. Neurol. Neurosurg.

Psychiatr.43:644659.10. Lehmann, H.C., Hartung, H.P., Hetzel, G.R., Stuve,

O., and Kieseier, B.C. 2006. Plasma exchange inneuroimmunological disorders: part 2. Treat-ment of neuromuscular disorders. Arch. Neurol.63:10661071.

11. Horton, R.M., Manfredi, A.A., and Conti-Tron-coni, B. 1993. The embryonic gamma subunit ofthe nicotinic acetylcholine receptor is expressed inadult extraocular muscle.Neurology.43:983986.

12. Hoch, W., et al. 2001. Auto-antibodies to the recep-tor tyrosine kinase MuSK in patients with myas-

thenia gravis without acetylcholine receptor anti-bodies.Nat. Med.7:365368.


11/12

science in medicine


13. Ruegg, M.A., and Bixby, J.L. 1998. Agrin orches-trates synaptic differentiation at the vertebrateneuromuscular junction. Trends Neurosci.21:2227.

14. Glass, D.J., et al. 1996. The receptor tyrosine kinaseMuSK is required for neuromuscular junction for-mation and is a functional receptor for agrin. ColdSpring Harb. Symp. Quant. Biol.61:435444.

15. Hughes, B.W., Kusner, L.L., and Kaminski, H.J.2006. Molecular architecture of the neuromuscu-lar junction.Muscle Nerve.33:445461.

16. Khanna, S., and Porter, J.D. 2002. Conservationof synapse-signaling pathways at the extraocularmuscle neuromuscular junction. Ann. N. Y. Acad.Sci.956:394396.

17. Kaminski, H.J., Li, Z., Richmonds, C., Lin, F., andMedof, M.E. 2004. Complement regulators inextraocular muscle and experimental autoimmunemyasthenia gravis.Exp. Neurol.189:333342.

18. Wood, S.J., and Slater, C.R. 1997. The contributionof postsynaptic folds to the safety factor for neuro-muscular transmission in rat fast- and slow-twitchmuscles.J. Physiol.500:165176.

19. Sterz, R., Pagala, M., and Peper, K. 1983. Postjunc-tional characteristics of the endplates in mammali-

an fast and slow muscles.Pflugers Arch.398:4854.20. Ruff, R.L. 1996. Single-channel basis of slow inacti-vation of Na+ channels in rat skeletal muscle.Am. J.Physiol.271:C971C981.

21. Sahashi, K., Engel, A.G., Lambert, E.H., and How-ard, F.M., Jr. 1980. Ultrastructural localization ofthe terminal and lytic ninth complement compo-nent (C9) at the motor end-plate in myastheniagravis.J. Neuropathol. Exp. Neurol.39:160172.

22. Lennon, V.A., Seybold, M.E., Lindstrom, J.M.,Cochrane, C., and Ulevitch, R. 1978. Role of comple-ment in the pathogenesis of experimental autoim-mune myasthenia gravis.J. Exp. Med.147:973983.

23. Biesecker, G., and Gomez, C.M. 1989. Inhibition ofacute passive transfer experimental autoimmunemyasthenia gravis with Fab antibody to comple-ment C6.J. Immunol.142:26542659.

24. Piddlesden, S.J., Jiang, S., Levin, J.L., Vincent, A.,

and Morgan, B.P. 1996. Soluble complementreceptor 1 (sCR1) protects against experimentalautoimmune myasthenia gravis. J. Neuroimmunol.71:173177.

25. Christadoss, P. 1988. C5 gene influences the devel-opment of murine myasthenia gravis. J. Immunol.140:25892592.

26. Karachunski, P.I., Ostlie, N.S., Monfardini, C., andConti-Fine, B.M. 2000. Absence of IFN-gammaor IL-12 has different effects on experimentalmyasthenia gravis in C57BL/6 mice. J. Immunol.164:52365244.

27. Medof, M.E., Walter, E.I., Rutgers, J.L., Knowles,D.M., and Nussenzweig, V. 1987. Identification ofthe complement decay-accelerating factor (DAF)on epithelium and glandular cells and in body flu-ids.J. Exp. Med.165:848864.

28. McNearney, T., Ballard, L., Seya, T., and Atkinson, J.P. 1989. Membrane cofactor protein of comple-ment is present on human fibroblast, epithelial,and endothelial cells.J. Clin. Invest.84:538545.

29. Meri, S., Waldmann, H., and Lachmann, P.J. 1991.Distribution of protectin (CD59), a complementmembrane attack inhibitor, in normal human tis-sues.Lab. Invest.65:532537.

30. Lin, F., et al. 2002. Markedly enhanced susceptibil-ity to experimental autoimmune myasthenia gravisin the absence of decay-accelerating factor protec-tion. J. Clin. Invest.110:12691274. doi:10.1172/

JCI200216086.31. Drachman, D.B., Angus, C.W., Adams, R.N., Michel-

son, J.D., and Hoffman, G.J. 1978. Myasthenic anti-bodies cross-link acetylcholine receptors to acceler-ate degradation.N. Engl. J. Med.298:11161122.

32. Engel, A.G., and Fumagalli, G. 1982. Mechanismsof acetylcholine receptor loss from the neuromus-

cular junction. Ciba Found. Symp.90:197224.33. Conti-Tronconi, B., Tzartos, S., and Lindstrom,

J. 1981. Monoclonal antibodies as probes of ace-tylcholine receptor structure. 2. Binding to nativereceptor.Biochemistry.20:21812191.

34. Gomez, C.M., and Richman, D.P. 1983. Anti-ace-tylcholine receptor antibodies directed against the

alpha-bungarotoxin binding site induce a uniqueform of experimental myasthenia.Proc. Natl. Acad.Sci. U. S. A.80:40894093.

35. Whiting, P.J., Vincent, A., and Newsom-Davis, J.1983. Acetylcholine receptor antibody characteris-tics in myasthenia gravis. Fractionation of alpha-bungarotoxin binding site antibodies and their rela-tionship to IgG subclass.J. Neuroimmunol. 5:19.

36. Eng, H., Lefvert, A .K., Mellstedt, H., and Osterborg,A. 1987. Human monoclonal immunoglobulinsthat bind the human acetylcholine receptor.Eur. J.

Immunol.17:18671869.37. Hohlfeld, R., Toyka, K.V., Heininger, K., Grosse-

Wilde, H., and Kalies, I. 1984. Autoimmune humanT lymphocytes specific for acetylcholine receptor.

Nature.310:244246.38. Morgutti, M., Conti-Tronconi, B.M., Sghirlan-

zoni, A., and Clementi, F. 1979. Cellular immuneresponse to acetylcholine receptor in myastheniagravis: II. Thymectomy and corticosteroids.Neurol-ogy.29:734738.

39. Ahlberg, R., et al. 1994. Treatment of myastheniagravis with anti-CD4 antibody: improvement cor-relates to decreased T-cell autoreactivity.Neurology.44:17321737.

40. Nath, A., Kerman, R.H., Novak, I.S., and Wolinsky,J.S. 1990. Immune studies in human immunode-ficiency virus infection with myasthenia gravis: acase report.Neurology.40:581583.

41. Wang, Z.Y. 1999. Myasthenia in SCID mice graftedwith myasthenic patient lymphocytes: role of CD4+and CD8+ cells.Neurology.52:484497.

42. Kaul, R., Shenoy, M., Goluszko, E., and Chris-tadoss, P. 1994. Major histocompatibility com-plex class II gene disruption prevents experimen-

tal autoimmune myasthenia gravis. J. Immunol.152:31523157.

43. Wang, Z.Y., Okita, D.K., Howard, J., Jr., and Conti-Fine, B.M. 1998. T-cell recognition of muscle acetyl-choline receptor subunits in generalized and ocularmyasthenia gravis.Neurology.50:10451054.

44. Wang, Z.Y., et al. 2000. T cell recognition of muscleacetylcholine receptor in ocular myasthenia gravis.

J. Neuroimmunol.108:2939.45. Protti, M.P., Manfredi, A.A., Straub, C., Howard,

J.F., Jr., and Conti-Tronconi, B.M. 1990. Immuno-dominant regions for T helper-cell sensitizationon the human nicotinic receptor alpha subunitin myasthenia gravis. Proc. Natl. Acad. Sci. U. S. A.87:77927796.

46. Kaul, R., Shenoy, M., and Christadoss, P. 1994. Therole of major histocompatibility complex genes in

myasthenia gravis and experimental autoimmunemyasthenia gravis pathogenesis.Adv. Neuroimmunol.4:387402.

47. Christadoss, P., Lindstrom, J.M., Melvold, R.W.,and Talal, N. 1985. Mutation at I-A beta chainprevents experimental autoimmune myastheniagravis.Immunogenetics.21:3338.

48. Poussin, M.A., Goluszko, E., David, C.S., Franco,J.U., and Christadoss, P. 2001. HLA-DQ6 trans-genic mice resistance to experimental autoimmunemyasthenia gravis is linked to reduced acetylcho-line receptor-specific IFN-gamma, IL-2 and IL-10production.J. Autoimmun.17:175180.

49. Yang, H., et al. 2003. Induction of myasthenia gra-vis in HLA transgenic mice by immunization withhuman acetylcholine receptors.Ann. N. Y. Acad. Sci. 998:375378.

50. Yates, A., Bergmann, C., Van Hemmen, J.L., Stark,J., and Callard, R. 2000. Cytokine-modulated regu-

lation of helper T cell populations. J. Theor. Biol.206:539560.

51. Weiner, H.L. 2001. Induction and mechanism ofaction of transforming growth factor-beta-secretingTh3 regulatory cells.Immunol. Rev.182:207214.

52. Saoudi, A., et al. 1999. Experimental autoimmunemyasthenia gravis may occur in the context of a

polarized Th1- or Th2-type immune response inrats.J. Immunol.162:71897197.

53. Moiola, L., Karachunski, P., Protti, M.P., Howard,J.F., Jr., and Conti-Tronconi, B.M. 1994. Epitopeson the subunit of human muscle acetylcholinereceptor recognized by CD4+ cells of myastheniagravis patients and healthy subjects. J. Clin. Invest.93:10201028.

54. Yi, Q., Ahlberg, R., Pirskanen, R., and Lefvert, A.K.1994. Acetylcholine receptor-reactive T cells inmyasthenia gravis: evidence for the involvement ofdifferent subpopulations of T helper cells.J. Neuro-immunol.50:177186.

55. Im, S.H., et al. 2001. Suppression of experimentalmyasthenia gravis, a B cell-mediated autoimmune dis-ease, by blockade of IL-18.FASEB J.15:21402148.

56. Goluszko, E., Deng, C., Poussin, M.A., and Chris-

tadoss, P. 2002. Tumor necrosis factor receptor p55and p75 deficiency protects mice from developingexperimental autoimmune myasthenia gravis.

J. Neuroimmunol.122:8593.57. Wang, H.B., et al. 2000. Tumor necrosis factor

receptor-1 is critically involved in the developmentof experimental autoimmune myasthenia gravis.

Int. Immunol.12:13811388.58. Duan, R.S., et al. 2002. Anti-TNF-alpha antibodies

suppress the development of experimental autoim-mune myasthenia gravis.J. Autoimmun.19:169174.

59. Christadoss, P., and Goluszko, E. 2002. Treatmentof experimental autoimmune myasthenia graviswith recombinant human tumor necrosis factorreceptor Fc protein.J. Neuroimmunol. 122:186190.

60. Delpy, L., et al. 2005. Estrogen enhances suscep-tibility to experimental autoimmune myastheniagravis by promoting type 1-polarized immune

responses.J. Immunol.175:50505057.61. Hohlfeld, R., and Engel, A.G. 1990. Induction of

HLA-DR expression on human myoblasts withinterferon-gamma.Am. J. Pathol. 136:503508.

62. Feferman, T., et al. 2005. Overexpression of IFN-induced protein 10 and its receptor CXCR3 inmyasthenia gravis.J. Immunol.174:53245331.

63. Ostlie, N., Milani, M., Wang, W., Okita, D., andConti-Fine, B.M. 2003. Absence of IL-4 facilitates thedevelopment of chronic autoimmune myastheniagravis in C57BL/6 mice.J. Immunol.170:604612.

64. Balasa, B., Deng, C., Lee, J., Christadoss, P., andSarvetnick, N. 1998. The Th2 cytokine IL-4 is notrequired for the progression of antibody-depen-dent autoimmune myasthenia gravis. J. Immunol.161:28562862.

65. Poussin, M.A., Goluszko, E., Hughes, T.K., Duchi-

cella, S.I., and Christadoss, P. 2000. Suppression ofexperimental autoimmune myasthenia gravis inIL-10 gene-disrupted mice is associated with reducedB cells and serum cytotoxicity on mouse cell lineexpressing AChR.J. Neuroimmunol.111:152160.

66. Poussin, M.A., Goluszko, E., Franco, J.U., andChristadoss, P. 2002. Role of IL-5 during primaryand secondary immune response to acetylcholinereceptor.J. Neuroimmunol.125:5158.

67. Deng, C., Goluszko, E., Tuzun, E., Yang, H., andChristadoss, P. 2002. Resistance to experimen-tal autoimmune myasthenia gravis in IL-6-defi-cient mice is associated with reduced germinalcenter formation and C3 production. J. Immunol.169:10771083.

68. Balandina, A., Lecart, S., Dartevelle, P., Saoudi, A.,and Berrih-Aknin, S. 2005. Functional defect ofregulatory CD4(+)CD25+ T cells in the thymusof patients with autoimmune myasthenia gravis.


12/12

science in medicine

Blood.105:735741.69. Sun, Y., et al. 20 04. Increa se of circul ating

CD4+CD25+ T cells in myasthenia gravis patientswith stability and thymectomy. Clin. Immunol.112:284289.

70. Liu, R., et al. 2005. Cooperation of invariant NKTcells and CD4+CD25+ T regulatory cells in the pre-

vention of autoimmune myasthenia. J. Immunol.175:78987904.

71. Shi, F.D., et al. 2000. Natural killer cells determinethe outcome of B cell-mediated autoimmunity.

Nat. Immunol. 1:245251.72. Jander, S., and Stoll, G. 2002. Increased serum

levels of the interferon-gamma-inducing cytokineinterleukin-18 in myasthenia gravis. Neurology.59:287289.

73. Vincent, A., McConville, J., Farrugia, M.E., andNewsom-Davis, J. 2004. Seronegative myastheniagravis. Semin. Neurol.24:125133.

74. Vincent, A., and Leite, M.I. 2005. Neuromuscularjunction autoimmune disease: muscle specifickinase antibodies and treatments for myastheniagravis. Curr. Opin. Neurol.18:519525.

75. Yeh, J.H., Chen, W.H., Chiu, H.C., and Vincent, A.

2004. Low frequency of MuSK antibody in gener-alized seronegative myasthenia gravis among Chi-nese.Neurology.62:21312132.

76. Ohta, K., et al. 2004. MuSK antibodies in AChRAb-seropositive MG vs AChR Ab-seronegative MG.Neurology.62:21322133.

77. Shigemoto, K., et al. 2006. Induction of myasthe-nia gravis by immunization against muscle-specifickinase.J. Clin. Invest.116:10161024. doi:10.1172/

JCI21545.78. Shiraishi, H., et al. 2005. Acetylcholine receptors loss

and postsynaptic damage in MuSK antibody-posi-tive myasthenia gravis.Ann. Neurol. 57:289293.

79. McConville, J., et al. 2004. Detection and character-ization of MuSK antibodies in seronegative myas-thenia gravis.Ann. Neurol.55:580584.

80. Boneva, N., Frenkian-Cuvelier, M., Bidault, J.,Brenner, T., and Berrih-Aknin, S. 2006. Major

pathogenic effects of anti-MuSK antibodies inmyasthenia gravis.J. Neuroimmunol.177:119131.

81. Plested, C.P., et al. 2002. AChR phosphorylationand indirect inhibition of AChR function in sero-negative MG.Neurology.59:16721673.

82. Romi, F., Skeie, G.O., Gihus, N.E., and Aarli, J.A.2005. Striational antibodies in myasthenia gravis:reactivity and possible clinical significance. Arch.

Neurol.62:442446.83. Mohan, S., Barohn, R.J., Jackson, C.E., and Krolick,

K.A. 1994. Evaluation of myosin-reactive antibod-ies from a panel of myasthenia gravis patients. Clin.

Immunol. Immunopathol. 70:266273.84. Aarli, J.A. 2001. Titin, thymoma, and myasthenia

gravis.Arch. Neurol.58:869870.85. Baggi, F., et al. 1998. Anti-titin and antiryanodine

receptor antibodies in myasthenia gravis patients

with thymoma.Ann. N. Y. Acad. Sci.841:538541.86. Keesey, J.C. 2004. Clinical evaluation and manage-

ment of myasthenia gravis.Muscle Nerve.29:484505.87. Katirji, B., and Kaminski, H.J. 2002. Electrodiagnostic

approach to the patient with suspected neuromuscu-lar junction disorder.Neurol. Clin.20:557586, viii.

88. [Anonymous]. 1992. Guidelines in electrodiagnos-tic medicine. American Association of Electrodiag-nostic Medicine.Muscle Nerve.15:229253.

89. Soreq, H., and Seidman, S. 2001. Acetylcholines-terasenew roles for an old actor.Nat. Rev. Neurosci.2:294302.

90. Brenner, T., et al. 2003. The role of readthroughacetylcholinesterase in the pathophysiology ofmyasthenia gravis.FASEB J.17:214222.

91. Bedlack, R.S., and Sanders, D.B. 2002. Steroid

treatment for myasthenia gravis: steroids have animportant role.Muscle Nerve.25:117121.

92. Lindner, A., Schalke, B., and Toyka, K.V. 1997.Outcome in juvenile-onset myasthenia gravis: aretrospective study with long-term follow-up of 79patients.J. Neurol.244:515520.

93. Palace, J., Newsom-Davis, J., and Lecky, B. 1998.

A randomized double-blind trial of prednisolonealone or with azathioprine in myasthenia gra-

vis. Myasthenia Gravis Study Group. Neurology.50:17781783.

94. Spring, P.J., and Spies, J.M. 2001. Myasthenia gravis: options and timing of immunomodulatorytreatment.BioDrugs.15:173183.

95. Tindall, R.S., et al. 1987. Preliminary results of adouble-blind, randomized, placebo-controlled trialof cyclosporine in myasthenia gravis.N. Engl. J. Med. 316:719724.

96. Tindall, R.S., Phillips, J.T., Rollins, J.A., Wells, L.,and Hall, K. 1993. A clinical therapeutic trial ofcyclosporine in myasthenia gravis. Ann. N. Y. Acad.Sci.681:539551.

97. Ponseti, J.M., et al. 2005. Benefits of FK506 (tacro-limus) for residual, cyclosporin- and prednisone-

resistant myasthenia gravis: one-year follow-upof an open-label study. Clin. Neurol. Neurosurg.107:187190.

98. Ciafaloni, E. 2005. Mycophenolate mofetil andmyasthenia gravis.Lupus.14(Suppl. 1):s46s49.

99. Chaudhry, V., Cornblath, D.R., Griffin, J.W., OBrien,R., and Drachman, D.B. 2001. Mycophenolatemofetil: a safe and promising immunosuppressantin neuromuscular diseases.Neurology.56:9496.

100. Drachman, D.B., Jones, R.J., and Brodsky, R.A. 2003.Treatment of refractory myasthenia: rebootingwith high-dose cyclophosphamide. Ann. Neurol.53:2934.

101. Pescovitz, M.D. 2 006. Rituximab, an anti-cd2 0monoclonal antibody: history and mechanism ofaction.Am. J. Transplant. 6:859866.

102. Tuzun, E., Meriggioli, M.N., Rowin, J., Yang, H., andChristadoss, P. 2005. Myasthenia gravis patients

with low plasma IL-6 and IFN-gamma benefit frometanercept treatment.J. Autoimmun.24:261268.

103. Newsom-Davis, J. 2003 . Therapy in myastheniagravis and Lambert-Eaton myasthenic syndrome.Semin. Neurol.23:191198.

104. Samuelsson, A ., Towers, T.L., and Ravetch, J.V.2001. Anti-inflammatory activity of IVIG medi-ated through the inhibitory Fc receptor. Science.291:484486.

105. Psaridi-Linardaki, L., Trakas, N., Mamalaki, A., andTzartos, S.J. 2005. Specific immunoadsorption ofthe autoantibodies from myasthenic patients usingthe extracellular domain of the human muscle ace-tylcholine receptor alpha-subunit. Development ofan antigen-specific therapeutic strategy.J. Neuroim-munol.159:183191.

106. Gronseth, G.S., and Barohn, R.J. 2000. Practice

parameter: thymectomy for autoimmune myasthe-nia gravis (an evidence-based review): report of theQuality Standards Subcommittee of the American

Academy of Neurology. Neurology.55:715.107. Jaretzki, A., Steinglass, K.M., and Sonett, J.R. 2004.

Thymectomy in the management of myastheniagravis. Semin. Neurol.24:4962.

108. Wolfe, G.I., Kaminski, H.J., Jaretzki, A., 3rd, Swan,A., and Newsom-Davis, J. 2003. Development of athymectomy trial in nonthymomatous myasthe-nia gravis patients receiving immunosuppressivetherapy.Ann. N. Y. Acad. Sci.998:473480.

109. Jaretzki, A., 3rd, et al. 2000. Myasthenia gravis:recommendations for clinical research standards.Task Force of the Medical Scientific Advisory Boardof the Myasthenia Gravis Foundation of America.

Neurology.55:1623.110. Patel, M.R., and Granger, C.B. 2005. Pexelizumab: a

novel therapy for myocardial ischemia-reperfusion.Drugs Today (Barc.).41:165170.

111. Fitch, J. C., et al. 1999. Phar macology and bio-logical efficacy of a recombinant, humanized,single-chain antibody C5 complement inhibitor in

patients undergoing coronary artery bypass graftsurgery with cardiopulmonary bypass. Circulation.100:24992506.

112. Keshavjee, S., Davis, R.D., Zamora, M.R., de Perrot,M., and Patterson, G.A. 2005. A randomized, pla-cebo-controlled trial of complement inhibition inischemia-reperfusion injury after lung transplan-tation in human beings.J. Thorac. Cardiovasc. Surg.129:423428.

113. Drachman, D.B., McIntosh, K.R., Reim, J., and Bal-cer, L. 1993. Strategies for treatment of myastheniagravis.Ann. N. Y. Acad. Sci.681:515528.

114. Yarilin, D., Duan, R., Huang, Y.M., and Xiao, B.G.2002. Dendritic cells exposed in vitro to TGF-beta1ameliorate experimental autoimmune myastheniagravis. Clin. Exp. Immunol.127:214219.

115. Adikari, S.B., Lian, H., Link, H., Huang, Y.M., and

Xiao, B.G. 2004. Interferon-gamma-modified den-dritic cells suppress B cell function and amelioratethe development of experimental autoimmunemyasthenia gravis. Clin. Exp. Immunol.138:230236.

116. Duan, R.S., Adikari, S.B., Huang, Y.M., Link, H.,and Xiao, B.G. 2004. Protective potential of experi-mental autoimmune myasthenia gravis in Lewisrats by IL-10-modified dendritic cells. Neurobiol.Dis.16:461467.

117. Karachunski, P.I., Ostlie, N.S., Okita, D.K., andConti-Fine, B.M. 1997. Prevention of experimentalmyasthenia gravis by nasal administration of syn-thetic acetylcholine receptor T epitope sequences.J. Clin. Invest.100:30273035.

118. Karachunski, P.I., Ostlie, N.S., Okita, D.K., Gar-man, R., and Conti-Fine, B.M. 1999. Subcutaneousadministration of T-epitope sequences of the ace-tylcholine receptor prevents experimental myasthe-

nia gravis.J. Neuroimmunol.93:108121.119. Baggi, F., et al. 1999. Oral administration of an

immunodominant T-cell epitope downregulatesTh1/Th2 cytokines and prevents experimentalmyasthenia gravis.J. Clin. Invest.104:12871295.

120. Xiao, B.G., and Link, H. 1997. Mucosal tolerance: atwo-edged sword to prevent and treat autoimmunediseases. Clin. Immunol. Immunopathol.85:119128.

121. Urbatsch, I.L., Sterz, R.K., Peper, K., and Trommer,W.E. 1993. Antigen-specific therapy of experimentalmyasthenia gravis with acetylcholine receptor-gelo-nin conjugates in vivo.Eur. J. Immunol.23:776779.

122. Wu, J.M., Wu, B., Mia gkov, A., Ada ms, R.N., andDrachman, D.B. 2001. Specific immunotherapy ofexperimental myasthenia gravis in vitro: the guid-ed missile strategy. Cell. Immunol.208:137147.

123. Vandenbark, A.A., Vainiene, M., Celnik, B., Hashim,

G., and Offner, H. 1992. TCR peptide therapydecreases the frequency of encephalitogenic T cellsin the periphery and the central nervous system.

J. Neuroimmunol.39:251260.124. Cohen-Kaminsky, S., and Jambou, F. 2005. Pros-

pects for a T-cell receptor vaccination against myas-thenia gravis.Expert Rev. Vaccines. 4:473492.

125. Dayan, M., et a l. 2004. Immun omodulatio n by adual altered peptide ligand of autoreactive respons-es to the acetylcholine receptor of peripheral bloodlymphocytes of patients with myasthenia gravis.

Hum. Immunol. 65:571577.126. Bromberg, M.B. 2005. Myasthenia gravis and myas-

thenic syndromes. InMotor disorders. D.S. Younger,editor. Lippincott Williams & Wilkins. Philadel-phia, Pennsylvania, USA. 231246.

myastenia gravis- past, present and future

Documents