Antibodies Affecting Ion Channel Function in Acquired Neuromyotonia, in Seropositive and...

PDFlib PLOP: PDF Linearization, Optimization, Protection

Page inserted by evaluation versionwww.pdflib.com – [email protected]

Antibodies Affecting Ion Channel Functionin Acquired Neuromyotonia, in Seropositive

and Seronegative Myasthenia Gravis,and in Antibody-mediated

Arthrogryposis Multiplex Congenita

ANGELA VINCENT, LESLIE JACOBSON, PAUL PLESTED,AGATA POLIZZI, TERESA TANG, SIETSKE RIEMERSMA,

CLAIRE NEWLAND, SARA GHORAZIAN, JEREMY FARRAR,CAL MACLENNAN, NICHOLAS WILLCOX, DAVID BEESON,

AND JOHN NEWSOM-DAVIS

Neurosciences GroupInstitute of Molecular Medicine

John Radcliffe HospitalUniversity of Oxford

Headington, Oxford OX3 9DS, United Kingdom

INTRODUCTION

Three autoimmune ion channel disorders are now recognized that affect neuro-muscular transmission. Antibodies to acetylcholine receptors (AChR) are well recog-nized as causative in myasthenia gravis, but much remains to be understood concern-ing the etiology and pathology of the different forms of the disease. The Lambert-Eaton myasthenic syndrome is caused by antibodies to voltage-gated calcium chan-nels and is the subject of several other contributions in this volume. Acquiredneuromyotonia is associated with antibodies to voltage-gated potassium channels(VGKC) that are thought to be pathogenic. This chapter first summarizes the role ofanti-VGKC antibodies in neuromyotonia and then discusses findings from our labo-ratory that relate to antibody specificity in myasthenia gravis (MG), seronegativemyasthenia, and a related fetal/neonatal disorder, arthrogryposis multiplex congenita.

AUTOIMMUNITY IN ACQUIRED NEUROMYOTONIA(ISAACS’ SYNDROME)

Acquired neuromyotonia or Isaacs’ syndrome is a disorder that usually presents inadult life and consists of spontaneous muscle activity with twitching, painful cramps,and increased sweating. An autoimmune basis for neuromyotonia was suggested bythe association with other autoimmune disorders or with tumors such as thymomaand small cell lung cancer that are frequently associated with immune-mediated neu-rological diseases, by induction in rare cases by penicillamine treatment, and by the

482

VINCENT et al.: ANTIBODIES 483

presence of oligoclonal bands in cerebrospinal fluid from some patients. For a re-view, see reference 1. Proof of the autoimmune nature of the disease came from thesame approaches that had earlier demonstrated the role of antibodies in myastheniagravis and the Lambert-Eaton myasthenic syndrome. Some patients improve tem-porarily following plasma exchange2–4 and, in a few cases, immunosuppression or in-travenous immunoglobulin has been found effective.1 Injection of neuromyotoniaIgG into mice did not produce spontaneous muscle activity, but increased the resis-tance to curare-induced paralysis,2 increased the number of packets of ACh releasedper nerve impulse, increased the duration of sensory action potentials, and inducedspontaneous activity in dorsal root ganglia cell cultures.5 The latter effects were sim-ilar to those resulting from application of low doses of the potassium channel block-er, 3,4-diaminopyridine, suggesting that antibodies to VGKC expressed by the motornerve might be involved. Because VGKC are responsible for resetting the membranepotential after each nerve impulse, reduced VGKC activity would lead to neuronalhyperexcitability.

Antibodies to VGKC were first detected in a proportion of neuromyotonia pa-tients using 125I-�-dendrotoxin-labeled VGKC extracted from human frontal cortex.5

This assay is now in routine use, although only about 65% of patients are positive,and the titers are low (FIGURE 1).

A novel immunohistochemical method has also been used to detect anti-VGKC

FIGURE 1. Anti-VGKC antibodies measured in serum from patients with acquired neuromy-otonia by immunoprecipitation of 125I-�-dendrotoxin-labeled voltage-gated potassium chan-nels extracted from human frontal cortex.5,6 The dotted line denotes the cutoff based on themean ± 2SD of results from healthy and other neurological controls. Sera were also tested foranti-AChR antibodies; positive values were found in 12 of 23 sera and did not correlate withanti-VGKC antibody levels. Eight of these patients had a thymoma.

antibodies.6 In this technique, VGKC subunit cRNAs are expressed in Xenopusoocytes; frozen sections of the oocytes are tested for binding of patients’ antibodiesby immunohistochemistry. Using this approach, anti-VGKC antibodies were found inall neuromyotonia patients. The discrepancy between the two assays may relate todiffering sensitivity or to the fact that, in the immunohistochemical assay, relativelyhigh concentrations of individual VGKC subunits can be tested.

It has been recognized that neuromyotonia may occur in association with myas-thenia gravis and thymoma (e.g., see reference 7). Interestingly, several neuromyoto-nia patients with or without thymoma proved to have anti-AChR antibodies (FIGURE

1). There is no clear correlation between the presence of the two antibodies, but bothVGKC and AChR subunits can be detected in thymoma tissue by PCR (see Liyanageet al. and MacLennan et al. elsewhere in this volume).

ANTI-ACHR ANTIBODY REACTIVITY WITHADULT AND FETAL ACHR

It has long been recognized that antibodies to AChR in MG sera are heteroge-neous in their immunochemistry and in their specificity for different epitopes on theAChR. In particular, several studies have demonstrated variability in the competitionbetween monoclonal anti-AChR antibodies and anti-AChR antibodies in MG sera.8,9

In addition, there is evidence for antibodies that react differentially with fetal or adultAChR.10

It has, however, been difficult to compare antibody specificity for human adultand fetal AChR because of the small amounts of the former in human muscle. Toovercome this problem and to provide a better source of adult AChR for diagnosticassays, we have stably transfected the TE671 cells, which normally produce fetalAChR, with cDNA for the � subunit. These cells, called TE671-� to distinguish themfrom TE671-� or wild type, produce large amounts of AChR, of which >70% is ofthe adult form.11 Use of the TE671-� and TE671-� cells has allowed us to comparethe binding of patients’ sera to the two forms. Sera from some low-titer patients showa preference for the adult AChR or fetal AChR,11 and a few cases have antibodies thatare specific for adult AChR (FIGURE 2). In most cases with moderate or high titers ofanti-AChR, the sera recognize each form (presumably because most of the antibodiesare directed equally to the � and � subunits, or to the other subunits), although it iscommon for antibodies to bind slightly better to the fetal form (see below).

THYMOMA-ASSOCIATED MG

In thymoma-associated MG, the thymic tumor does not appear to express the in-tact AChR,12,13 but individual AChR subunits have been identified by RNase protec-tion assays (MacLennan et al., in preparation). Although all the subunits can be de-tected by PCR techniques, in a high proportion of thymomas the � subunit predomi-nates when the less sensitive, but more quantitative technique of RNase protection isemployed. This suggests that the � subunit may be involved in the initiation of the au-

ANNALS NEW YORK ACADEMY OF SCIENCES484


toimmunity (see also Willcox et al. elsewhere in this volume). Thus, it was of interestto see whether thymoma/MG patients’ antibodies bound preferentially to the adultform of the AChR. The reverse was found in most thymoma cases; only two serabound more strongly to adult than to fetal AChR, and these displayed high � subunitexpression and low total anti-AChR titers (MacLennan et al., in preparation).

OCULAR MG

Ocular weakness predominates in many patients at presentation and purely ocularsymptoms may persist after other symptoms have responded to treatment. It has beendemonstrated that multiply innervated muscle fibers, which represent about 20% ofthe fibers in ocular muscle, express the � subunit, suggesting that antibodies specificfor fetal AChR might be an important cause of ocular muscle weakness14,15 (see alsoKaminski elsewhere in this volume). However, by RNase protection, we found a fargreater amount of � subunit than � subunit overall in ocular muscle.16 Thus, althoughthe multiply innervated slow-twitch fibers contain � subunit, they probably also con-tain � subunit and the extent to which antibodies that are fetal AChR–specific are re-sponsible for ocular muscle weakness is debatable (see also reference 17 and com-ments below).

FIGURE 2. Immunoprecipitation by an MG serum (circles) of 125I-�-BuTx-labeled AChR ex-tracted from TE671-� (open symbols) and TE671-� (closed symbols) cells. This serum, unusu-ally, binds quite strongly to adult AChR and does not show any binding to fetal AChR. Thismale patient had generalized MG. For comparison, results are shown from a typical female pa-tient (triangles) whose antibodies recognized both forms equally.


It was noteworthy that about 50% of sera from ocular MG patients had highertiters of antibody against adult AChR than against fetal AChR and that inclusion ofadult AChR improved the yield of positive results in diagnostic assays.16,18 However,it should be noted that when total anti-AChR levels are very low, as in many cases ofocular MG, the pathogenic antibodies may be underrepresented in the serum as a re-sult of their absorption by the muscle AChR (see reference 19).

DEFINITION OF BINDING SITES FOR MONOCLONAL ANTIBODIESON HUMAN ACHR

Ten mAbs raised against human AChR were shown to bind to five regions (TABLE

1; see also Jacobson et al. elsewhere in this volume), none of which overlapped the�-bungarotoxin (�-BuTx) binding site.8,20 One region was specific for fetal AChRand was therefore presumed to lie on the � subunit. Two regions appeared to overlapthe main immunogenic region, originally defined by Tzartos and Lindstrom,21 andone of the mAbs (D6) showed very similar binding characteristics to anti-MIR mAb35.22 However, we were unable to demonstrate binding of any of the mAbs to syn-thetic peptides of the AChR � subunit (Jacobson and Vincent, unpublished data).

Now, each of the mAbs has been mapped by Western blotting of recombinantAChR subunits expressed in E. coli. The results confirm the � subunit binding ofmAbs D6, C3, and G10 and the � subunit binding of C2, C9, B8, and F8, which arefetal-specific. We have also found that B3 binds the � subunit and that C7 and G3bind the � subunit (Jacobson et al., in preparation). Thus, there are binding sites oneach of the four subunits of fetal AChR, and the subunit specificities correspond re-markably well to the regions defined in our previous studies, which suggested that re-gion 1 was close to the �-BuTx binding site near the �/� interface, whereas region 5was close to the site near the �/� interface22 (TABLE 1).

TABLE 1. AChR Subunit Binding Sites for Monoclonal Antibodies Raised againstHuman AChR

Monoclonal Antibody

B8, C2, C9, F8 B3 C3, G10 D6 C7, G3

Binding site defined 1 2 3 4 5by competition studies22

Specificity Fetal AChR Overlapping MIR �/� interface�/� interface MIR

Binding site defined �1–174 �1–161 �3–181 �3–181 �1–194by immunoblotting on recombinant AChR subunits (Jacobson et al., in preparation)


CLONING OF ANTI-ACHR ANTIBODIES FROMA COMBINATORIAL LIBRARY

It would clearly be useful to obtain monoclonal antibodies representative of theMG patients’ repertoires in order to study the specificity and pathogenicity of indi-vidual human anti-AChR antibodies. We have cloned 4 Fabs using a kappa lightchain/IgG1 heavy chain library created from mRNA obtained from the thymus of oneearly-onset female patient.23 Two of the Fabs (e.g., ABk1) compete with anti-MIRmAbs and bind to both �� and �� dimers, whereas another (ABk5) binds only to ��dimers (FIGURE 3) and is specific for the fetal isoform (see also Matthews et al. else-where in this volume). Because anti-MIR and fetal-specific antibodies were seen inthe patient’s serum,23 these results suggest that combinatorial libraries may be a use-ful approach to studying individual human autoantibodies. Over 20 Fabs have beenobtained from the lambda light chain/IgG1 library from this patient and are beingcharacterized (Matthews et al., this volume). Graus and colleagues have also clonedspecific Fabs from MG thymus samples.24

ANTIBODIES SPECIFIC FOR FETAL ACHR CAUSINGARTHROGRYPOSIS MULTIPLEX CONGENITA

Antibodies binding preferentially to fetal forms of the AChR could be particularlydamaging to the fetus in utero. Recently, antibodies that specifically inhibit the func-tion of fetal AChR have been found to associate with a severe developmental abnor-

FIGURE 3. Immunoprecipitation of AChR subunit dimers by Fabs cloned from a combinator-ial library. Fab ABk1 binds equally to �� and �� dimers, consistent with its specificity for theMIR.23 By contrast, Fab ABk5 binds to ��, but not to �� dimers, and was inhibited by mono-clonal antibodies specific for the � subunit.23


mality, arthrogryposis multiplex congenita (AMC). AMC is a syndrome consisting ofjoint contractures at birth or detected in utero, often associated with hydramnios,lung hypoplasia, and micrognathia. These features appear to result from lack of fetalmovement in utero and, in some cases, they can be severe enough to cause fetal deathor stillbirth. AMC sometimes reflects a genetic disorder of muscle or the central ner-vous system, but most often the cause is unknown, or environmental factors are im-plicated. For instance, joint contractures were demonstrated in chick embryos inject-ed with curare.25 Arthrogryposis has also been reported in association with typicalmaternal myasthenia gravis with transient neonatal MG.26 For a review of arthrogry-posis, see reference 27.

We studied two women with histories of severe AMC in their offspring.28–30 Onewoman was completely asymptomatic despite high levels of anti-AChR antibodies,29

and the other was only diagnosed as having MG after the stillbirth of her fourth af-fected child.28 The severe fetal paralysis (demonstrated by complete lack of fetalmovement on ultrasound), combined with the relative lack of maternal involvement,suggested that placental transfer of antibodies specific for fetal AChR might be in-volved in causing the fetal damage.

There were no differences in reactivity with adult or fetal AChR when sera fromthese two mothers were titrated in standard immunoprecipitation assays.30 However,the sera inhibited �-BuTx binding to about 50% of its binding sites on fetal AChR(FIGURE 4a). Moreover, when the sera were tested for their ability to inhibit the func-

FIGURE 4a. Serum from mothers whose babies suffer from severe arthrogryposis multiplexcongenita contains antibodies that specifically inhibit the function of fetal AChR. Inhibition of�-BuTx binding to AChR in human muscle extracts (predominantly of fetal type) by sera fromtwo mothers (AMC-M) of AMC babies. In each case, the inhibition plateaus at about 50%, pre-sumably because the antibodies bind to only one of the two �-BuTx binding sites (see FIGURE

4d).


tion of the AChRs, a striking difference was observed (FIGURE 4b). IgG from the twopatients was applied to fetal and adult AChRs expressed in Xenopus oocytes. IgG di-luted 1:50 markedly inhibited the ACh responses of fetal AChR, but had no effect onthose of adult AChR.28,30 In addition, the two sera, and three others kindly donated byBeatrice Vernet der Garabedian and Bruno Eymard, inhibited carbchhol-induced Na+

influx through fetal AChRs in TE671 cells, even when diluted 1:100, and had little orno effect on adult AChRs in TE671-� cells.30

Further confirmation of the fetal specificity of these antibodies was shown by theprotection of fetal AChR function by preincubation with the mAbs that bind to the �subunit (FIGURE 4c), but not by one that binds to the � subunit. Thus, we hypothesizethat antibodies from AMC mothers bind to a site on the � subunit that partially over-laps the ACh-binding site on the adjacent � subunit (FIGURE 4d).

These results indicate a role for antibodies against fetal AChR in causing somecases of AMC. It is noteworthy that neither mother had clear signs of ocular muscleweakness in spite of high levels of serum antibodies to fetal AChR (see ocular myas-thenia, above). To confirm the role of these antibodies in AMC, and to establish amodel that can be used to investigate other putative antibody-mediated conditions,we have injected serum from mothers of AMC babies into pregnant mice. When 0.5mL of plasma was injected daily from E13 till E20, the pups were stillborn andshowed signs of lack of fetal movement such as fixed limbs and muscle atrophy (seeJacobson et al. elsewhere in this volume). This model can now be employed to inves-tigate the effects of sera from other neurological disorders.

FIGURE 4b. Inhibition of ACh-induced currents of fetal AChR ex-pressed in Xenopus oocytes byIgG purified from AMC-M sera.No effect was seen when the IgGwas tested on adult AChR. Taken,with permission, from reference30.


CHARACTERIZING SERUM FACTORS INSERONEGATIVE MG (SNMG)

At least 10–15% of patients with typical signs of generalized MG do not have de-tectable antibody by standard tests, including those that look at “blocking” or “modu-lating” antibodies.31 We have previously reviewed the evidence for the existence ofimmunological factors, possibly IgM, in plasma from SNMG patients.32 Briefly,SNMG patients responded to plasma exchange and immunosuppression,33 and invivo injection of their Ig fractions induced changes in neuromuscular transmission in

FIGURE 4c. The inhibition of AChR function seen in TE671-� cells was prevented by prein-cubation of the cells in monoclonal antibodies that bind to the � subunit (mAb C2 and C9), butnot by a monoclonal antibody to the � subunit (B3).

FIGURE 4d. Proposed binding site for antibodies from AMC-M sera. The antibodies bind tothe � subunit, but in such a way that ACh binding, or ACh-induced channel opening, is blocked.


the mouse hemidiaphragm preparation.33,34 SNMG plasmas inhibited the function ofAChR in TE671 cells, and the active factor was in the non-IgG fraction and copuri-fied with IgM;35 the effect was partially dependent on intracellular calcium,36 sug-gesting that the antibodies might act indirectly by altering intracellular signaling.37

Moreover, attempts to demonstrate that SNMG plasma antibodies bound directly tothe AChR were unsuccessful.33,35,36

We have now found additional evidence consistent with an indirect effect ofSNMG plasmas. We used TE671-� cells to ensure that the effects seen are applicableto the neuromuscular junction where adult AChR is expressed. SNMG plasmas tran-siently reduced AChR function (as measured by carbachol-induced 22Na+ influx) byup to 85% at 1:20 dilution. The inhibition of function was greatest after about 10 minof incubation and thereafter reversed even in the continuing presence of the plasma(see Plested et al. elsewhere in this volume). Control plasma showed little or no inhi-bition. The effects of SNMG plasmas varied between individuals, but overall werehighly significant (FIGURE 5a).

One possibility is that the SNMG plasma factor may act to stimulate an intracellu-lar pathway (perhaps calcium-dependent) that leads to AChR phosphorylation and re-duced function. It is well accepted, for instance, that AChR phosphorylation bycAMP-dependent protein kinase (PKA) or tyrosine kinases leads to increased AChRdesensitization.38 The effect would be expected to be rapid (as seen in TE671-� cells)and would probably be transient because many of the metabotropic receptors them-selves desensitize over a period of minutes.

These results led us to measure PKA activity in TE671 cells and to investigate theresults of stimulating PKA activity on AChR function. PKA activity is low endoge-nously in TE671 cells, but can be stimulated by CGRP, salbutamol, dibutyryl cAMP,and cholera toxin. Each of these compounds applied for 30–60 min moderately re-

FIGURE 5a. SNMG plasmas transiently inhibit AChR function in TE671-� cells. Carbachol-induced 22Na+ flux was measured after 10 min of incubation in SNMG plasmas at 1:20 dilu-tion. The mean values, 59.1 ± 23.7 (SD; n = 12), were significantly different from controls,100.7 ± 15.9 (SD; n = 3, p < 0.005).


duces AChR function in TE671 cells,37 and CGRP shows a marked transient inhibi-tion similar to that found with SNMG plasma (Plested et al., in preparation). Howev-er, SNMG plasmas did not have any demonstrable effect on PKA activity in TE671-�cells, suggesting that the putative target for the SNMG plasma factor acts via anotherpathway.

If SNMG plasma contains a factor that transiently affects AChR function, it mightbe possible to see this during short incubations of the mouse hemidiaphragm. Appli-cation of SNMG plasmas at room temperature reduced the twitch tension of themouse hemidiaphragm when a low concentration of d-tubocurarine was included inthe bath (FIGURE 5b), whereas control plasmas under the same conditions had nodemonstrable effect. To see if there was a transient effect on AChR function, MEPPamplitudes were measured before and after application of SNMG plasmas and con-trol plasma (diluted 1:2 in Krebs solution; coded samples). SNMG plasmas produceda small reduction in MEPP amplitudes that tended to reverse over a period of 15 min(FIGURE 5c), whereas the control plasma–treated muscles showed a small increase inMEPP amplitude (FIGURE 5c). However, these results were not significant because ofthe wide interfiber variation in MEPP amplitudes. To overcome this problem, theplasmas were applied (coded) to strips of hemidiaphragm during impalement of sin-gle muscle fibers; in this case, although the changes were small, the MEPP ampli-tudes were significantly reduced by two SNMG plasmas compared to control (Tanget al., in preparation).

In conclusion, SNMG plasma appears to contain a factor that is not IgG (and maybe IgM35) and that causes a rapid reduction in AChR function both in TE671-� andTE671-� cells and at the mouse neuromuscular junction; there may also be longer-term effects as seen in FIGURE 5b. However, the target or targets for these actions of

FIGURE 5b. SNMG plasmas reduce nerve-evoked twitch tension in mouse hemidiaphragmpreparations. The twitch tension was measured before and after a 3-hour application of threecontrol and three SNMG plasmas in the presence of curarized Krebs solution with 0.6 �g/mLof d-tubocurarine (see also reference 33).


SNMG plasmas are still unknown. Further studies are in progress to investigate othersecond messenger systems that might be involved in the SNMG action.

SUMMARY AND CONCLUSIONS

A new autoimmune disease affecting the neuromuscular junction has been de-fined. Acquired neuromyotonia is associated with antibodies to voltage-gated potas-sium channels that act, at least in part, by reducing potassium channel function withresulting neuronal hyperactivity. This condition is quite frequently associated withthymoma and, in many cases, antibodies to acetylcholine receptors are present aswell as antibodies to VGKC.

Improvements in techniques and the availability of cloned DNA and recombinantforms of the AChR subunits have led to new observations concerning the specificityand roles of antibodies in myasthenia gravis. The transfection of a cell line with the �subunit means that we can now accurately compare antibodies reactive with adult andfetal human AChR. This may help to determine the relationship between AChR sub-unit expression in different tissues and the induction of antibodies that bind specifi-cally to the two forms, as well as to clarify the role of antibodies to fetal or adultAChR in causing ocular muscle symptoms. Serum antibodies from a few motherswith obstetric histories of recurrent arthrogryposis multiplex congenita in their ba-bies specifically inhibit the function of fetal AChR. These observations not only ex-

FIGURE 5c. SNMG plasmas transiently reduce AChR function at the mouse neuromuscularjunction. MEPP amplitudes (a measure of AChR function) were measured before and followingapplication of four SNMG plasmas and one control plasma. The results for each 10-min timeperiod were pooled. There was an apparent rapid, but slight, inhibitory effect of SNMG plas-mas; however, because of the relatively large interfiber variation in the MEPP amplitudes, thevalues were not statistically different.

plain the cause of some cases of arthrogryposis multiplex congenita, but also suggestthat other fetal-specific antibodies might be responsible for other fetal or neonatalconditions. An animal model has been established to enable us to investigate the roleof maternal serum factors in causing such disorders.

Seronegative MG has been the subject of many studies from our laboratory overthe last ten years. The transience of the effects of SNMG plasmas on AChR functionstrongly suggests that the plasma antibodies do not bind directly to the AChR, but in-hibit function by some indirect mechanism. They do not appear to act via the cAMP-dependent protein kinase pathway, and studies are in progress to investigate the in-volvement of other second messenger systems.

REFERENCES

1. NEWSOM-DAVIS, J. & K. R. MILLS. 1993. Immunological associations of acquired neu-romyotonia (Isaacs’ syndrome): report of five cases and literature review. Brain 116:453–469.

2. SINHA, S., J. NEWSOM-DAVIS, K. MILLS et al. 1991. Autoimmune aetiology for acquiredneuromyotonia (Isaacs’ syndrome). Lancet 338: 75–77.

3. BADY, B., G. CHAUPLANNAZ, C. VIAL et al. 1991. Autoimmune aetiology for acquired neu-romyotonia. Lancet 338: 1330.

4. WINTZEN, A. R., J. G. VAN DIJK & A. BRAND. 1994. Neuromyotonia with early response toplasmapheresis associated with proximal action myoclonus with late response to plasma-pheresis. Muscle Nerve 17(suppl. 1): S221.

5. SHILLITO, P., P. C. MOLENAAR, A. VINCENT et al. 1995. Acquired neuromyotonia: evidencefor autoantibodies against K+ channels of peripheral nerves. Ann. Neurol. 38: 714–722.

6. HART, I. K., C. WATERS, A. VINCENT et al. 1997. Autoantibodies detected to expressed K+

channels are implicated in neuromyotonia (Isaacs’ syndrome). Ann. Neurol. 41:238–246.

7. HALBACH, M., V. HOMBERG & H-J. FREUND. 1987. Neuromuscular, autonomic, and centralcholinergic hyperactivity associated with thymoma and acetylcholine receptor–bindingantibody. J. Neurol. 234: 433–436.

8. VINCENT, A., P. J. WHITING, M. SCHLUEP et al. 1987. Antibody heterogeneity and specifici-ty in myasthenia gravis. Ann. N.Y. Acad. Sci. 505: 106–120.

9. TZARTOS, S. J., T. BARKAS, M. T. CUNG et al. 1991. The main immunogenic region of theacetylcholine receptor: structure and role in myasthenia gravis. Autoimmunity 8:259–270.

10. VINCENT, A. & J. NEWSOM-DAVIS. 1982. Acetylcholine receptor antibody characteristics inmyasthenia gravis. 1. Patients with generalized myasthenia or disease restricted to ocularmuscles. Clin. Exp. Immunol. 49: 257–265.

11. BEESON, D., M. AMAR, I. BERMUDEZ et al. 1996. Stable high level expression of the adultsubtype of human muscle acetylcholine receptor in the TE671 cell line following trans-fection with cDNA encoding the � subunit. Neurosci. Lett. 207: 57–60.

12. GEUDER, K. I., A. MARX, V. WITZEMANN et al. 1992. Genomic organization and lack oftranscription of the nicotinic acetylcholine receptor subunit genes in myastheniagravis–associated thymoma. Lab. Invest. 66: 452–458.

13. KAMINSKI, H. J., R. A. FENSTERMAKER, F. W. ABDUL KARIM et al. 1993. Acetylcholine re-ceptor subunit gene expression in thymic tissue. Muscle Nerve 16: 1332–1337.


14. HORTON, R. M., A. A. MANFREDI & B. M. CONTI-TRONCONI. 1993. The “embryonic” gam-ma subunit of the nicotinic acetylcholine receptor is expressed in adult extraocular mus-cle. Neurology 43(5): 983–985.

15. KAMINSKI, H., L. KUSNER & C. BLOCK. 1996. Expression of acetylcholine receptor iso-forms at extraocular endplates. Invest. Ophthalmol. Visual Sci. 321: 406–411.

16. MACLENNAN, C., D. BEESON, A-M. BUIJS et al. 1997. Acetylcholine receptor expression inhuman extraocular muscles and their susceptibility to myasthenia gravis. Ann. Neurol.41: 423–431.

17. KAMINSKI, H. J. & U. L. RUFF. 1997. Ocular muscle involvement by myasthenia gravis. Ed-itorial. Ann. Neurol. 41: 419–420.

18. BEESON, D., L. JACOBSON, J. NEWSOM-DAVIS et al. 1996. A transfected human muscle cellline expressing the adult subtype of the human muscle acetylcholine receptor for diag-nostic assays in myasthenia gravis. Neurology 47: 1552–1555.

19. VINCENT, A. 1980. Immunology of acetylcholine receptors in relation to myasthenia gravis.Physiol. Rev. 60: 756–824.

20. WHITING, P. J., A. VINCENT & J. NEWSOM-DAVIS. 1986. Myasthenia gravis: monoclonal an-tihuman acetylcholine receptor antibodies used to analyze antibody specificities and re-sponses to treatment. Neurology 36: 612–617.

21. TZARTOS, S. J. & J. L. LINDSTROM. 1980. Monoclonal antibodies to probe acetylcholine re-ceptor structure: localization of the main immunogenic region and detection of similari-ties between subunits. Proc. Natl. Acad. Sci. U.S.A. 77: 755–759.

22. HEIDENREICH, F., A. VINCENT, A. ROBERTS et al. 1988. Epitopes on human acetylcholine re-ceptor defined by monoclonal antibodies and myasthenia gravis sera. Autoimmunity 1:285–297.

23. FARRAR, J., S. PORTOLANO, N. WILLCOX et al. 1997. Diverse Fabs specific for acetylcholinereceptor epitopes from a myasthenia gravis thymus combinatorial library. Int. Immunol.In press.

24. GRAUS, Y. F., M. H. DE BAETS, P. W. H. PARREN et al. 1997. Human anti-nicotinic acetyl-choline receptor recombinant Fab fragments isolated from thymus-derived phage displaylibraries from myasthenia gravis patients reflect predominant specificities in serum andblock the action of pathogenic serum antibodies. J. Immunol. 158: 1919–1929.

25. DRACHMAN, D. B. & A. J. COULOMBRE. 1961. Experimental clubfoot and arthrogryposismultiplex congenita. Lancet ii: 523.

26. VERNET DER GARABEDIAN, B., M. LACOKOVA, B. EYMARD et al. 1994. Association ofneonatal myasthenia gravis with antibodies against the fetal acetylcholine receptor. J.Clin. Invest. 94: 555–559.

27. HALL, J. G. 1990. Arthrogryposes (multiple congenital contractures). In Principles andPractice of Medical Genetics. Churchill Livingstone. New York.

28. VINCENT, A., C. NEWLAND, L. BRUETON et al. 1995. Arthrogryposis multiplex congenitawith maternal autoantibodies specific for a fetal antigen. Lancet 346: 24–25.

29. BARNES, P., D. KANABAR, L. BRUETON et al. 1994. Recurrent congenital arthrogryposisleading to a diagnosis of myasthenia gravis in an initially asymptomatic mother. Neuro-musc. Dis. 5: 59–65.

30. RIEMERSMA, S., A. VINCENT, D. BEESON et al. 1996. Association of arthrogryposis multi-plex congenita with maternal antibodies inhibiting fetal acetylcholine receptor function.J. Clin. Invest. 98: 2358–2363.

31. SANDERS, D. B., I. ANDREWS, J. F. HOWARD et al. 1997. Seronegative myasthenia gravis.Neurology 48(suppl. 5): S40–S45.

32. VINCENT, A., Z. LI, A. HART et al. 1993. Seronegative myasthenia gravis: evidence for plas-ma factor(s) interfering with acetylcholine receptor function. Ann. N.Y. Acad. Sci. 681:529–538.


33. MOSSMAN, S., A. VINCENT & J. NEWSOM-DAVIS. 1986. Myasthenia gravis without acetyl-choline-receptor antibody: a distinct disease entity. Lancet i: 116–119.

34. BURGES, J., D. W. WRAY, S. PIZZIGHELLA et al. 1990. A myasthenia gravis plasma im-munoglobulin reduces miniature endplate potentials at human endplates in vitro. MuscleNerve 13: 407–413.

35. YAMAMOTO, T., A. VINCENT, T. A. CIULLA et al. 1991. Seronegative myasthenia gravis: aplasma factor inhibiting agonist-induced acetylcholine receptor function copurifies withIgM. Ann. Neurol. 30: 550–557.

36. BARRETT-JOLLEY, R., N. BYRNE, A. VINCENT et al. 1994. Seronegative myasthenia gravisplasmas reduce acetylcholine-induced currents in TE671 cells. Pflügers Arch. 428:492–498.

37. LI, Z., N. FORESTER & A. VINCENT. Modulation of acetylcholine receptor function inTE671 (rhabdomyosarcoma) cells by non-AChR ligands; a role in seronegative myasthe-nia gravis? J. Neuroimmunol. 64: 179–184.

38. HUGANIR, R. L. & K. MILES. 1989. Protein phosphorylation of nicotinic acetylcholine re-ceptors. Crit. Rev. Biochem. 24: 183–215.


Antibodies Affecting Ion Channel Function in Acquired Neuromyotonia, in Seropositive and...

Documents

Transcript of Antibodies Affecting Ion Channel Function in Acquired Neuromyotonia, in Seropositive and...