Detection and characterization of MuSK antibodies in seronegative myasthenia gravis.код для вставкиСкачать
7. Sullivan PG, Dube C, Dorenbos K, et al. Mitochondrial uncoupling protein-2 protects the immature brain from excitotoxic neuronal death. Ann Neurol 2003;53:711–717. 8. Richard D, Rivest R, Huang Q, et al. Distribution of the uncoupling protein 2 mRNA in the mouse brain. J Comp Neurol 1998;397:549 –560. 9. Rho JM, Kim DW, Robbins CA, et al. Age-dependent differences in flurothyl seizure sensitivity in mice treated with a ketogenic diet. Epilepsy Res 1999;37:233–240. 10. Sullivan PG, Geiger JD, Mattson MP, Scheff SW. Dietary supplement creatine protects against traumatic brain injury. Ann Neurol 2000;48:723–729. 11. Echtay KS, Roussel D, St-Pierre J, et al. Superoxide activates mitochondrial uncoupling proteins. Nature 2002;415:96 – 99. 12. Sullivan PG, Keller JN, Bussen WL, Scheff SW. Cytochrome c release and caspase activation after traumatic brain injury. Brain Res 2002;949:88 –96. 13. Todorova MT, Tandon P, Madore RA, et al. The ketogenic diet inhibits epileptogenesis in EL mice: a genetic model for idiopathic epilepsy. Epilepsia 2000;41:933–940. 14. Szot P, Weinshenker D, Rho JM, et al. Norepinephrine is required for the anticonvulsant effect of the ketogenic diet. Dev Brain Res 2001;129:211–214. 15. Skulachev VP. Role of uncoupled and non-coupled oxidations in maintenance of safely low levels of oxygen and its oneelectron reductants. Q Rev Biophys 1996;29:169 –202. 16. Mattiasson G, Shamloo M, Gido G, et al. Uncoupling protein-2 prevents neuronal death and diminishes brain dysfunction after stroke and brain trauma. Nat Med 2003;9:1062–1068. 17. Li LX, Skorpen F, Egeberg K, et al. Uncoupling protein-2 participates in cellular defense against oxidative stress in clonal beta-cells. Biochem Biophys Res Commun 2001;282:273–277. 18. Kim-Han JS, Reichert SA, Quick KL, Dugan LL. BMCP1: a mitochondrial uncoupling protein in neurons which regulates mitochondrial function and oxidant production. J Neurochem 2001;79:658 – 668. 19. Votyakova TV, Reynolds IJ. DeltaPsi(m)-dependent and -independent production of reactive oxygen species by rat brain mitochondria. J Neurochem 2001;79:266 –277. 20. Noh HS, Kim YS, Lee HP, et al. The protective effect of a ketogenic diet on kainic acid-induced hippocampal cell death in the male ICR mice. Epilepsy Res 2003;53:119 –128. Detection and Characterization of MuSK Antibodies in Seronegative Myasthenia Gravis John McConville, MRCP, DPhil,1 Maria Elena Farrugia, MRCP,1 David Beeson, PhD,1 Uday Kishore, PhD,1 Richard Metcalfe, FRCP,2 John Newsom-Davis, FRS,1 and Angela Vincent, FRCPath1 Antibodies to rat muscle specific kinase, MuSK, have recently been identified in some generalized “seronegative” myasthenia gravis (SNMG) patients, who are often females with marked bulbar symptoms. Using immunoprecipitation of 125I-labelled-human MuSK, 27 of 66 (41%) seronegative patients were positive, but 18 ocular SNMG patients, 105 AChR antibody positive MG patients, and 108 controls were negative. The antibodies are of high affinity (Kds around 100 pM) with titers between 1 and 200 nM. They bind to the extracellular Ig-like domains of soluble or native MuSK. Surprisingly they are predominantly in the IgG4 subclass. MuSK-antibody associated MG may be differnet in etiological and pathological mechanisms. Ann Neurol 2004;55:580 –584 Eighty-five percent of myasthenia gravis (MG) patients have autoantibodies to the muscle acetylcholine receptor (AChR).1 Patients without AChR antibodies (seronegative MG [SNMG]) may have more frequent bulbar involvement and less thymic pathology and often are resistant to conventional immunosuppression.2–5 Recently, IgG autoantibodies to the musclespecific kinase, MuSK, were identified in 70% of patients with generalized SNMG,6 and antibodies to a 110kDa protein, identified as MuSK, were found to From the 1Weatherall Institute of Molecular Medicine and Department of Clinical Neurology, Oxford; and 2Department of Clinical Neurology, Southern General Hospital, Glasgow, Scotland, United Kingdom. Received Sep 25, 2003, and in revised form Jan 13, 2004. Accepted for publication Jan 13, 2004. Current address for Dr McConville: Department of Neurology, Royal Victoria Hospitals Trust, Grosvenor Road, Belfast BT12, Northern Ireland, United Kindgom. Published online Mar 22, 2004, in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/ana.20061 Address correspondence to Dr Vincent, Neurosciences Group, Weatherall Institute of Molecular Medicine, John Radcliffe Hospital, Oxford OX3 9DS, United Kingdom. E-mail: firstname.lastname@example.org 580 © 2004 American Neurological Association Published by Wiley-Liss, Inc., through Wiley Subscription Services be associated with oculobulbar symptoms.7 Here, we describe new assays for detecting MuSK antibodies, demonstrate their high disease specificity, and investigate their epitope specificity, affinity, and IgG subclass. MuSK, or its polypeptide fragments, or soluble agrin as a control, were added to the serum before addition of 125IMuSK. MuSK Antibody IgG Subclasses Subjects and Methods Patients and Controls Generalized SNMG patients (n ⫽ 66) all had attended the Oxford clinic between 1987 and 2002 with welldefined MG (decrement on repetitive stimulation or jitter on single fiber electromyography and/or a clear response to edrophonium) and repeated negative AChR antibody titers. Sex, age, and clinical features at first clinic presentation were obtained from their records. Ocular SNMG patients (n ⫽ 18) were diagnosed in Oxford, or by R.M. in Glasgow. AChR-Ab–positive sera were taken either from the Oxford clinic or the routine AChR-Ab assay service. Control sera were from 38 volunteers, 30 community controls, and 40 patients with other immune-mediated neurological diseases. Serum (0.5l) was added to 5fmol 125I-human MuSK and incubated overnight with 10l anti–human IgG1, 5l anti–human IgG2, 5l anti–human IgG3, 2.5l anti–human IgG4 (monoclonal antibodies from Binding Site, Birmingham, UK), or no antibody. The complexes were precipitated with sheep anti–mouse IgG (Binding Site; preabsorbed with 10% human control serum), with normal mouse serum added to aid precipitation, and precipitated 125 I MuSK washed and counted as above. The enzymelinked immunosorbent assay (ELISA) for anti–MuSK IgG subclasses6 used biotinylated monoclonal mouse anti–human IgG1 (1:100; Calbiochem, San Diego, CA); anti–human IgG2 (1:500; Sigma, St. Louis, MO); anti–human IgG3 (1:1,500; Sigma); or anti–human IgG4 (1:1,000; Calbiochem), as second layers, and detected their binding with streptavidin-horseradish peroxidase (Sigma) at 1 to 4,000. Cloning of MuSK Ectodomains Human MuSK was cloned by reverse transcription polymerase chain reaction from poly(A)⫹ RNA isolated from the TE671 human rhabdomyosarcoma cell line.8 Expression constructs were generated in pSecTag (Invitrogen, La Jolla, CA), with coding sequences inserted between an Ig signal sequence at the N terminus and a His6 tag at the C terminus. MuSK1-4 encoded the entire extracellular domains (nucleotides 107–1526, numbering as in Valenzuela and colleagues9), MuSK 1-2 contained nucleotides 107 to 715, and MuSK 3-4 contained nucleotides 711 to 1526. Human embryonic kidney (HEK) 293 cells were transiently transfected with the constructs. Recombinant proteins were harvested from serum-free culture supernatants (UltraCHO; Cambrex, Berkshire, UK) and purified using nickel affinity chromatography (Probond; Invitrogen). Human muscle agrin was used as a control antigen.6 MuSK Antibody Radioimmunoprecipitation Assay and Epitope Mapping Ten micrograms of purified human MuSK was labeled with 125 I, as described for other proteins,10 to a specific activity of 5 to 10 ⫻ 103 cpm/fmol. Five microliters of human plasma or serum was added to 5fmol of MuSK and incubated for 1 hour at room temperature or overnight at 4°C in 0.02M phosphate buffer/0.1% Triton X-100/5% fetal calf serum, and IgG–MuSK complexes were precipitated with 50l sheep anti–human IgG (Binding Site, Birmingham, UK). The precipitates were centrifuged, washed, and counted for 125 I. All positive sera were titrated and the results were expressed as nanomoles of 125I-MuSK precipitated per liter of serum. To determine the affinities, we incubated a limiting amount of each serum overnight with different concentrations of 125I-MuSK; Kds were determined using GraphPad Prism (GraphPad Software, San Diego, CA.). For epitope mapping, culture supernatants containing 10-fold excess of MuSK Antibodies Binding to Native MuSK on MuSK–Green Fluorescent Protein Transfected Human Embryonic Kidney 293 Cells The sequence encoding enhanced green fluorescent protein (EGFP; derived from EGFP-N1; Clontech, Palo Alto, CA), was ligated into the naturally occurring restriction site EcoRI at nucleotide position 2607 in the MuSK cDNA sequence, so that the expressed product contained all but the C-terminal 27 amino acids of MuSK followed by EGFP. HEK 293 cells were transiently transfected using PEI.11 Trypsin-suspended cells (approximately 5 ⫻ 106) were incubated for 40 minutes in serum (1:40 dilution in phosphatebuffered saline/5% fetal calf serum). After washing, cells were incubated with anti–human IgG PE (Sigma) at 1 to 30 and fluorocytometry was conducted using a Becton Dickinson (San Jose, CA) FACScan 2. Results Antibodies to Human MuSK Overall, radioimmunoprecipitation of 125I-MuSK extracellular domains detected IgG antibodies in 27/66 SNMG patients (eg, Fig 1A). MuSK-Abs were not found in 18 patients with ocular SNMG (Ocular). Low levels of 125I precipitation were found in a few MG or control samples (see Fig 1A), but in these cases, in contrast with the clear MuSK-Ab–positive samples (Fig 2), 125I precipitation was nonspecific because it was not blocked by an excess of unlabeled MuSK (data not shown). Titers ranged from 0.3 to 200nM (median, 20nM). All samples previously positive with the rat MuSK ELISA6 were strongly positive by immunoprecipitation of 125I-human MuSK McConville et al: MuSK Antibody Characterization 581 Fig 1. Radioimmunoprecipitation assay for MuSK IgG antibodies. (A) Scatterplot of counts precipitated (cpm) by 5l serum or plasma. The line is drawn at the mean ⫹ 3 SDs of the healthy control values; the neurological controls gave a similar range. The few sera with low positive values were shown to bind nonspecifically. Antibody affinities (inset) ranged between 55 and 108pM. (B) Correlation between binding of IgG antibodies to human MuSK-GFP expressed by human embryonic kidney (HEK) cells, measured by FACS, and the immunoprecipitation results. Examples of the FACS analysis, for one MuSK antibody–positive serum and one control serum, are shown in the inset. SNMG ⫽ seronegative myasthenia gravis; AChR ⫽ acetylcholine receptor; GFP ⫽ green fluorescent protein. (r2⫽0.88; p ⬍ 0.0001 for accurate titers). All four sera tested showed high-affinity binding with Kd values of less than 1nM (eg, see Fig 1A inset). Patients with MuSK antibodies were usually female (M:F, 4:23) and presented between 5 and 56 years (median, 24 years), and 11 of 27 had prominent bulbar symptoms, whereas MuSK-Ab–negative patients were often male (M:F, 14:25; p ⫽ 0.03, Fisher’s exact test) and presented between 1 and 78 years (median, 37), and only 6 of 39 ( p ⫽ 0.03) had prominent bulbar symptoms. Antibodies Binding to Native MuSK We used the fluorescence activated cell sorter to detect human IgG binding to the surface of HEK 293 cells expressing human MuSK tagged with EGFP (MuSKGFP, see Fig 1B, inset). Clearly positive results were obtained with each of the MuSK-Ab–positive sera, the 582 Annals of Neurology Vol 55 No 4 April 2004 results correlating with the precipitation assay (see Fig 1B). None of the healthy control sera, or five SNMG sera negative for binding to MuSK by radioimmunoprecipitation, bound to the MuSK-GFP–expressing cells. Epitopes in the Extracellular Domains of Human MuSK Preincubation of sera with control protein (agrin) did not alter precipitation. Preincubation with MuSK1-2 (see Fig. 2A) reduced the precipitation of all nine sera tested (for examples, see Fig 2B), whereas MuSK 3-4 reduced precipitation of only five of nine sera. As expected, addition of both MuSK 1-2 and MuSK 3-4, or of MuSK 1-4, reduced precipitation to control levels in all cases (see Fig 2B). Fig 2. (A) Diagramatic representation of MuSK and its domains that were expressed as soluble proteins. (B) Binding of MuSK antibodies to 125I-MuSK in the presence of unlabeled MuSK domains. Precipitation was substantially reduced in all sera tested by incubation with forms that contained MuSK1-4, whereas MuSK3-4 reduced precipitation in only some sera. Precipitation in the presence of agrin, a control recombinant protein, was similar to that by serum alone. We found that MuSK antibodies bind to extracellular IgG-like domains of MuSK, both when expressed in HEK 293 cells, or in solution. MuSK has a major role during development of the neuromuscular junction (see Liyanage and colleagues13 and Hopf and Hoch14), and MuSK antibodies reduce agrin-induced clustering of AChRs in mouse C2C12 myotubes in vitro,6 but it is not yet clear how they affect the neuromuscular junction in vivo. Although pathogenic effects were observed after transfer of SNMG plasmas to mice,4,15,16 and most of the plasmas tested are now known to be MuSK-Ab positive (A. Vincent, unpublished observations), there was little effect on total AChR numbers.15,16 The results are complicated by the existence of a non-IgG plasma factor in SNMG that reduces AChR function17; although distinct, this factor is present in many MuSK-Ab–positive SNMG patients.17 IgG MuSK antibodies do not appear to affect AChR function directly.17 One hypothetical pathogenic mechanism would be that MuSK antibodies activate complement which induces lysis of the AChR-containing postsynaptic membrane. However, in contrast with AChR-Ab MG, where complement-fixing IgG1 and IgG3 subclasses predominate,18,19 MuSK antibodies were mainly IgG4 which is strong complement activators. Together with the relative lack of thymic pathology,2,3 and the tendency to marked bulbar12 and neck MuSK Antibody IgG Subclasses Surprisingly, MuSK antibodies were predominantly IgG4, with some IgG2 (Fig 3A). To confirm these findings, we used a different set of mouse monoclonal subclass–specific antibodies in an ELISA assay (see Fig 3B) with very similar results. Discussion We confirmed, using two novel assays, that a proportion of patients with generalized MG without AChR antibodies have increased levels of MuSK antibodies and these antibodies are not present in patients with AChR antibody–positive MG, in patients with purely ocular SNMB, or in controls. The proportion of MuSK antibody–positive patients was lower than our previous report on a smaller number of SNMG cases,6 probably because the first study inadvertently included more patients with severe or refractory disease, who we now know are more likely to be MuSK antibody positive.12 The simplicity, high specificity, and sensitivity of the radioimmunoprecipitation assay should make it useful in the routine diagnosis of this form of MG. Fig 3. IgG subclasses of MuSK antibodies. (A) IgG subclasses measured by immunoprecipitation. (B) Results confirmed by ELISA (below). In both cases, the MuSK antibodies were largely in the IgG4 and IgG2 subclasses. MuSK-Ab Patients 1and 2 were tested in both assays. McConville et al: MuSK Antibody Characterization 583 or respiratory weakness,20 these findings imply differences between MuSK-Ab MG and AChR-Ab MG in cause and pathological mechanisms. Because IgG2 antibodies frequently are directed at carbohydrate antigens on bacteria, we tried to inhibit the antibodies with appropriate sugars, but without success (L. Jacobson, U. Kishore, and A. Vincent, unpublished observations). Moreover, the MuSK antibodies bind with high affinity (similar to that of AChR antibodies18) to the native MuSK molecule. Thus, our current evidence favors affinity maturation of an immune response directed against MuSK itself, rather than cross-reactivity with a bacterial glycoprotein or glycolipid. Note Added in Proof Using sheep polyclonal antisera to IgG subclasses, we have recently found up to 30% of the MuSK antibody to be in the IgG1 subclass. However, IgG4 predominates in all sera examined so far. J.M. was supported by a Clinical Training Fellowship from The Wellcome Trust. We are very grateful to Dr W. Hoch for his advice and the rat MuSK constructs, and the ongoing support of the Myasthenia Gravis Association, the Muscular Dystrophy Campaign, and the French Association against myopathies. References 1. Vincent A, Palace J, Hilton-Jones D. Myasthenia gravis. Lancet 2001;357:2122–2128. 2. Willcox N, Schluep M, Ritter MA, Newsom-Davis J. The thymus in seronegative myasthenia gravis patients. J Neurol 1991; 238:256 –261. 3. Verma PK, Oger JJ. Seronegative generalized myasthenia gravis: low frequency of thymic pathology. Neurology 1992;42: 586 –589. 4. Birmanns B, Brenner T, Abramsky O, Steiner I. Seronegative myasthenia gravis: clinical features, response to therapy and synthesis of acetylcholine receptor antibodies in vitro. J Neurol Sci 1991;102:184 –189. 5. Evoli A, Batocchi AP, Lo Monaco M et al. Clinical heterogeneity of seronegative myasthenia gravis. Neuromusc Disord 1996;6:155–161. 6. Hoch W, McConville J, Helms S, et al. Auto-antibodies to the receptor tyrosine kinase MuSK in patients with myasthenia gravis without acetylcholine receptor antibodies. Nat Med 2001;7: 365–368. 7. Scuderi F, Marino M, Colonna L, et al. Anti-p110 autoantibodies identify a subtype of “seronegative” myasthenia gravis with prominent oculobulbar involvement. Lab Invest 2002;82: 1139 –1146. 8. Stratton MR, Darling J, Pilkington GJ, et al. Characterization of the human cell line TE671. Carcinogenesis 1989;10: 899 –905. 9. Valenzuela DM, Stitt TN, DiStefano PS, et al. Receptor tyrosine kinase specific for the skeletal muscle lineage: expression in embryonic muscle, at the neuromuscular junction, and after injury. Neuron 1995;15:573–584. 584 Annals of Neurology Vol 55 No 4 April 2004 10. Palace J, Nairne A, Hyman N, et al. A radioimmunoprecipitation assay for antibodies to botulinum A. Neurology 1998;50:1463–1466. 11. Boussif O, Lezoualc’h F, Zanta MA, et al. A versatile vector for gene and oligonucleotide transfer into cells in culture and in vivo: polyethylenimine. Proc Natl Acad Sci USA 1995;92: 7297–7301. 12. Evoli A, Tonali PA, Padua L, et al. Clinical correlates with antiMuSK antibodies in generalized seronegative myasthenia gravis. Brain 2003;126:2304 –2311. 13. Liyanage Y, Hoch W, Beeson D, Vincent A. The agrin/musclespecific kinase pathway: new targets for autoimmune and genetic disorders at the neuromuscular junction. Muscle Nerve 2002;25:4 –16. 14. Hopf C, Hoch W. Dimerization of the muscle-specific kinase induces tyrosine phosphorylation of acetylcholine receptors and their aggregation on the surface of myotubes. J Biol Chem 1998;273:6467– 6473. 15. Mossman S, Vincent A, Newsom-Davis J. Myasthenia gravis without acetylcholine-receptor antibody: a distinct disease entity. Lancet 1986;1:116 –119. 16. Burges J, Vincent A, Molenaar PC, et al. Passive transfer of seronegative myasthenia gravis to mice. Muscle Nerve 1994;17: 1393–1400. 17. Plested CP, Tang T, Spreadbury I, et al. AChR phosphorylation and indirect inhibition of AChR function in seronegative MG. Neurology 2002;59:1682–1688. 18. Vincent A, Newsom-Davis J. Acetylcholine receptor antibody characteristics in myasthenia gravis. 1. Patients with generalized myasthenia or disease restricted to ocular muscles. Clin Exp Immunol 1982;49:257–265. 19. Rodgaard A, Nielsen FC, Djurup R, et al. Acetylcholine receptor antibody in myasthenia gravis: predominance of IgG subclasses 1 and 3. Clin Exp Immunol 1987;67:82– 88. 20. Sanders DB, El-Salem K, Massey JM, et al. Clinical aspects of MuSK antibody positive seronegative MG. Neurology 2003;60: 1978 –1980.