- Email: [email protected]
The role of complement in myasthenia gravis: serological evidence of complement consumption in vivo
Journal of Neuroimmunology 158 (2005) 191 – 194 www.elsevier.com/locate/jneuroim
The role of complement in myasthenia gravis: serological evidence of complement consumption in vivo Fredrik Romia,*, Einar K. Kristoffersenb, Johan A. Aarlia,c, Nils Erik Gilhusa,c a Department of Neurology, Haukeland University Hospital, Bergen N-5021, Norway Microbiology and Immunology (The Bloodbank), Haukeland University Hospital, Bergen, Norway c University of Bergen, Bergen N-5021, Norway
b
Received 7 June 2004; received in revised form 5 August 2004; accepted 5 August 2004
Abstract Background: Antibodies to the acetylcholine receptor (AChR) titin and the ryanodine receptor (RyR) occur in myasthenia gravis (MG). These antibodies are capable of complement activation in vitro. The involvement of the complement system should cause consumption of complement components such as C3 and C4 in vivo. Materials and methods: Complement components C3 and C4 were assayed in sera from 78 AChR antibody-positive MG patients and 52 healthy controls. Forty-eight of the patient sera contained titin antibodies as well, and 20 were also RyR antibody-positive. Results: MG patients with AChR antibody concentrations above the median (11.2 nmol/l) had significantly lower mean C3 and C4 concentrations in serum compared to those with AChR antibody concentrations below the median. Titin antibody-positive MG patients, titin antibody-negative early-onset MG patients, titin antibody-negative late-onset MG patients, and controls had similar C3 and C4 concentrations. Nor did mean C3 and C4 concentrations differ in MG patients with RyR antibodies. Patients with severe MG (grades 4 and 5) had similar C3 and similar C4 levels compared to those with mild MG (grades 1 and 2). Conclusion: An increased in vivo complement consumption was detected in MG patients with high AChR antibody concentrations, unrelated to MG severity and non-AChR muscle antibodies. D 2004 Elsevier B.V. All rights reserved. Keywords: Myasthenia gravis; Acetylcholine receptor antibodies; Titin antibodies; Ryanodine receptor antibodies; C3; C4; Complement activation
1. Introduction Myasthenia gravis (MG) is an autoimmune disease affecting the neuromuscular junction of skeletal muscles. Acetylcholine receptor (AChR) antibodies are present in sera from 80% to 90% of patients with generalized MG, and occur only very infrequently in healthy individuals (Lindstro¨m, 1985; Lindstro¨m et al., 1998). Non-AChR muscle autoantibodies, such as against titin (Aarli et al., 1990), and ryanodine receptor (RyR) (Mygland et al., 1992a) are present in many MG sera, mainly from patients with thymoma or late-onset MG, and they occur * Corresponding author. Tel.: +47 55975000; fax: +47 55975165. E-mail address: [email protected] (F. Romi). 0165-5728/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.jneuroim.2004.08.002
more frequently in severe MG (Aarli et al., 1987; Mygland et al., 1994; Skeie et al., 1995, 1996; Voltz et al., 1997; Romi et al., 2000a,b). Antibodies to musclespecific kinase (MuSK) occur in 41% of sera from AChR antibody-negative generalized MG patients (McConville et al., 2004). AChR antibodies impair neuromuscular transmission by complement-mediated focal muscle membrane damage and accelerated degradation of AChR (Lindstro¨m, 1985; Lindstro¨m et al., 1998). Titin and RyR antibodies are capable of complement activation in vitro and could, if binding in vivo, cause complementmediated muscle cell damage (Romi et al., 2000c). The involvement of the complement system should cause an increased consumption of complement components such as C3 and C4.
192
F. Romi et al. / Journal of Neuroimmunology 158 (2005) 191–194
Our objective in this study was to measure the levels of C3 and C4 in MG sera and to correlate these concentrations to the presence of muscle antibodies and to MG severity.
sample from each patient was used for assessment of all antibody assays and complement components. 2.2. Complement components C3 and C4
2. Patients and methods 2.1. Patients Forty-eight (male/female: 21/27) consecutive titin antibody-positive MG patients, of whom 20 also had RyR antibodies, were included. Nineteen had thymoma, 10 had thymic atrophy, 4 had thymic hyperplasia, and the rest were not thymectomized. Thirty (male/female: 10/20) randomly chosen nonthymoma titin and RyR antibodynegative MG patients, of whom 15 had early-onset MG and 15 late-onset MG, were also included as MG controls (Romi et al., 2000a). All 78 MG patients had AChR antibodies and generalized MG, and they were all treated at Haukeland University Hospital (Bergen, Norway). Five patients had additional autoimmune diseases: two female patients had thyroiditis, one female patient had systemic lupus erythematosus, one female patient had polymyalgia rheumatica, and one male patient had rheumatoid arthritis. Fifty-two healthy controls, comparable regarding age and sex, were also included. None of the patients or controls had systemic infection when obtaining the blood sample. The diagnosis of MG was based on a typical clinical pattern, a positive edrophonium test, and neurophysiological investigations with decrement at repetitive stimulation and increased jitter at single-fiber EMG. None of the patients received immunosuppressive drug treatment at the time when the blood sample was obtained. The diagnosis of thymoma was based on histopathology. In patients not thymectomized, a thymoma was excluded on computed tomography (CT) or magnetic resonance imaging (MRI) of the mediastinum. The severity of MG was graded using a modified Osserman’s scale on the day the blood sample was obtained (Osserman and Genkins, 1971; Romi et al., 2000b): 1. 2. 3.
4. 5.
MG with purely ocular muscle weakness. MG with mild generalized weakness, without bulbar involvement. MG with mild generalized weakness including bulbar involvement with dysarthria, dysphagia, and poor mastication. MG with moderate generalized weakness. MG with severe generalized, bulbar, and respiratory muscle weakness.
The severity groups were also rearranged in mild (severity groups 1 and 2), moderate (severity group 3), and severe (severity groups 4 and 5) MG. All sera were stored at 80 8C and, at 20 8C prior to testing, were treated in the same way. The same serum
Sera were allowed to thaw in room temperature prior to analysis for complement factors C3 and C4 by nephelometry on a Behring Nephelometry analyser II (Dade Behring, Marburg, Germany), using rabbit polyclonal antiserum to highly purified human C3c and C4 as detailed by the manufacturer (Dade Behring). 2.3. Muscle autoantibody assays 2.3.1. AChR antibodies AChR antibodies were measured by a standard radioimmunoassay (RIA) method with human 125I AChR as the antigen and using AChR RRA kits (IBL, Hamburg, Germany). (Lefvert et al., 1978; Romi et al., 2000a). 2.3.2. Titin antibodies Titin antibodies were detected by ELISA using purified titin fragment MGT-30 as the antigen at a concentration of 3.5 Ag/ml in PBS-S (Skeie et al., 1995; Romi et al., 2000c). 2.3.3. RyR antibodies RyR antibodies were assayed by Western blot using pc2 Ry1 fusion protein as the antigen (Mygland et al., 1992b; Skeie et al., 2003). RyR antibodies were also detected by ELISA using the same antigen at 2 Ag/ml in PBS-S (Skeie et al., 1995; Romi et al., 2000c). 2.4. Statistical analysis Mean C3 and C4 concentrations and mean antibody concentrations and titers were compared for the various groups using the t test for difference between population means with unequal variances and the Mann–Whitney test. MG severity groups were compared using the chi-square test of independence. p values less than 0.05 were considered significant.
3. Results 3.1. C3, C4, and AChR antibodies MG patients with AChR antibody concentrations above the median (11.2 nmol/l) had lower mean C3 (1.73F0.74 g/l) and C4 (0.48F0.56 g/l) concentrations than those with AChR antibody concentrations below the median (C3, 2.11F0.95 g/l; C4, 0.57+0.34 g/l) (0.025bpb0.05 for C3; 0.01bpb0.05 for C4). Only one of the five patients with additional autoimmune diseases had AChR antibody concentration above the median value.
F. Romi et al. / Journal of Neuroimmunology 158 (2005) 191–194
193
Table 1 The immunological characteristics of patients and controls, showing mean C3, C4, and antibody concentrations, and mean MG severity in titin antibodypositive MG, titin antibody-negative MG, and healthy controls Titin antibody-positive MG group
Mean C3 concentration (g/l) Mean C4 concentration (g/l) Mean AChR antibody concentration (nmol/l) Titin antibodies n positive Mean titer RyR antibodies n positive Mean titer Mean MG severity
Titin antibody-negative MG group
Healthy controls
Early-onset MG
Late-onset MG
1.89 (F0.96) 0.51 (F0.44) 48.7 (F139.6)
1.76 (F0.62) 0.39 (F0.17) 77.8 (F161.9)
1.93 (F0.82) 0.55 (F0.27) 43.3 (F78.9)
1.91 (F0.75) 0.45 (F0.20) 0
48 5417 (F6571)
0 0
0 0
0 0
20 2838 (F3022) 3.0 (F0.9)
0 0 2.7 (F0.5)
0 0 3.0 (F1)
0 0 0
There was no difference between the total MG patient group and the healthy controls (Table 1). 3.2. C3, C4, titin, and RyR antibodies The mean concentration of C3 and C4 was similar in titin antibody-positive MG patients (C3, 1.89F0.96 g/l; C4, 0.51F0.44 g/l) and titin antibody-negative MG patients (C3, 1.84F0.72 g/l; C4, 0.47F0.24 g/l). Nor did the concentration differ in MG patients with RyR antibodies (C3, 1.79F0.92 g/l; C4, 0.54F0.54 g/l) (Table 1). Mean C3 (1.91F1.06 g/l) and C4 (0.54F0.51 g/l) concentrations in patients with titin antibody titers above the median (3200) did not differ significantly from those in patients with mean titin antibody titers below the median (C3, 1.79F0.89 g/l; C4, 0.44F0.34 g/l). 3.3. C3, C4, and MG severity Patients with severe MG (C3, 1.96F1.15 g/l; C4, 0.59F0.58 g/l) did not have lower C3 and C4 levels compared to those with mild MG (C3, 1.77F0.76 g/l; C4, 0.39F0.17 g/l). There were no significant differences in C3 and C4 levels between any severity groups in this study (Table 1).
4. Discussion This study demonstrates serological evidence for the involvement of the complement system in the pathophysiology of MG in vivo. The lower concentration of C3 and C4 in patients with high AChR antibody concentration can best be explained by an increased consumption of C3 and C4: (1) there is no evidence for reduced C3 and C4 production into the serum among MG patients since patients and controls had similar C3 and C4 levels; (2) increased C3 and C4 consumption has been demonstrated for other autoimmune diseases, although in this study, only one of the five patients with additional autoimmune disease was in the group with low C3 and C4 levels; and (3) none of the patients had
infectious disease explaining an increased complement consumption. There is, therefore, circumstantial evidence for increased consumption of the complement factors C3 and C4 in patients with high AChR antibody concentration. This consumption was statistically higher for C4 than for C3, indicating involvement of the classical fast and efficient complement activation pathway that leads to immunological damage through the formation of membrane attack complexes. Such membrane attack complexes on the AChR in MG muscle cells has been demonstrated electronmicroscopically (Engel and Arahata, 1987). AChR antibodies consist mainly of IgG 1 and IgG 3, which are both efficient activators of the classical complement activation pathway (Lindstro¨m, 1985; Lindstro¨m et al., 1998; Rbdgaard et al., 1987). Although IgG 1 is highly represented in both titin and RyR antibodies (Mygland et al., 1993; Romi et al., 2000c) and both antibodies are capable of complement activation in vitro (Romi et al., 2000c), this study did not show any reduction in the concentrations of C3 or C4 in patients with high titin and RyR antibody concentrations. Nor was any C3 and C4 reduction evident in patients with AChR antibody concentrations below the median value. This does not mean that patients with low AChR antibody concentrations have less severe disease, since MG severity was equal in these patients and in patients with high AChR antibody concentrations. Additional mechanisms may therefore contribute to AChR antibody pathogenicity. Similarly, we cannot exclude the in vivo pathogenicity of titin and RyR antibodies, including C3 and C4 activation. MG severity does not correlate with the overall serum concentration of AChR antibodies (Oosterhuis et al., 1983; Kuks et al., 1993). Previous studies have shown that titin and RyR antibodies are present more often in severe MG but their concentration is not correlated to MG severity, although such correlation is seen in individual patients over time (Aarli et al., 1987; Mygland et al., 1994; Skeie et al., 1995, 1996; Romi et al., 2000b). The reason why the AChR antibody concentration does not correlate with MG severity despite reduced C3 and C4 concentrations in sera from patients with a high antibody concentration is not clear.
194
F. Romi et al. / Journal of Neuroimmunology 158 (2005) 191–194
Our results are consistent with a recent study showing that the induction of EAMG was dependent on the classical complement pathway (Tuzun et al., 2003). The serum concentration of complement components falls during MG exacerbations, probably as a result of complement binding to muscle antigens (Plescia et al., 1966), and varies according to the severity of the disease (Nastuk et al., 1960). In summary, this study gives serological evidence for the involvement of the complement system in the pathophysiology of MG in vivo through increased consumption of the complement factors C3 and C4 in patients with high AChR antibody concentrations, although this complement consumption was not related to MG severity or to the presence of other muscle antibodies.
Acknowledgement This work was supported by EU grant QL61-CT-200101918 and the Norwegian Association for Muscle Disease.
References Aarli, J.A., Gilhus, N.E., Hofstad, H., 1987. CA-antibody: an immunological marker of thymic neoplasia in myasthenia gravis? Acta Neurol. Scand. 63, 55 – 57. Aarli, J.A., Stefansson, K., Marton, L.S.G., Wollmann, R.L., 1990. Patients with myasthenia gravis and thymoma have in their sera IgG autoantibodies against titin. Clin. Exp. Immunol. 82, 284 – 288. Engel, A.G., Arahata, K., 1987. The membrane attack complex of complement at the endplate in myasthenia gravis. Ann. N.Y. Acad. Sci. 505, 326 – 332. Kuks, J.B., Limburg, P.C., Horst, G., Oosterhuis, H.J., 1993. Antibodies to skeletal muscle in myasthenia gravis: Part 3. Relation with clinical course and therapy. J. Neurol. Sci. 120, 168 – 173. Lefvert, A.K., Bergstrbm, K., Matell, G., Osterman, P.O., Pirskanen, R., 1978. Determination of acetylcholine receptor antibody in myasthenia gravis—clinical usefulness and pathogenic implications. J. Neurol. Neurosurg. Psychiatry 41, 394 – 403. Lindstrfm, J., 1985. Immunobiology of myasthenia gravis, experimental autoimmune myasthenia gravis, and Lambert–Eaton syndrome. Annu. Rev. Immunol. 3, 109 – 131. Lindstrfm, J.M., Seybold, M.E., Lennon, V.A., Whittingham, S., Duane, D.D., 1998. Antibody to acetylcholine receptor in myasthenia gravis: prevalence, clinical correlates, and diagnostic value. 1975 Classical article. Neurology 51, 933 – 939. McConville, J., Farrugia, M.E., Beeson, D., Kishore, U., Metcalfe, R., Newsom-Davis, J., Vincent, A., 2004. Detection and characterization of MuSK antibodies in seronegative myasthenia gravis. Ann. Neurol. 55, 580 – 584. Mygland, 2., Tysnes, O.B., Aarli, J.A., Flood, P.R., Gilhus, N.E., 1992a. Myasthenia gravis patients with a thymoma have antibodies against a
high molecular weight protein in sarcoplasmatic reticulum. J. Neuroimmunol. 37, 1 – 7. Mygland, 2., Tysnes, O.B., Matre, R., Volpe, P., Aarli, J.A., Gilhus, N.E., 1992b. Ryanodine receptor autoantibodies in myasthenia gravis patients with thymoma. Ann. Neurol. 32, 589 – 591. Mygland, 2., Tysnes, O.B., Aarli, J.A., Matre, R., Gilhus, N.E., 1993. IgG subclass distribution of ryanodine receptor autoantibodies in patients with myasthenia gravis and thymoma. J. Autoimmun. 6, 507 – 515. Mygland, 2., Aarli, J.A., Matre, R., Gilhus, N.E., 1994. Ryanodine receptor antibodies related to severity of thymoma associated myasthenia gravis. J. Neurol. Neurosurg. Psychiatry 57, 843 – 846. Nastuk, W.L., Plescia, O.J., Osserman, K.E., 1960. Changes in serum complement activity in patients with myasthenia gravis. Proc. Soc. Exp. Biol. Med. 105, 177 – 184. Oosterhuis, H.J.G.H., Limburg, P.C., Hummel-Tappel, E., 1983. Antiacetylcholine receptor antibodies in myasthenia gravis: Part 2. Clinical and serological follow-up of individual patients. J. Neurol. Sci. 58, 371 – 385. Osserman, K., Genkins, G., 1971. Studies in myasthenia gravis: review of a twenty-year experience in over 1200 patients. Mt. Sinai J. Med. 38, 497 – 537. Plescia, O.J., Segovia, J.M., Strampp, A., 1966. An assessment of changes in the complement level of myasthenic sera. Ann. N.Y. Acad. Sci. 135, 580 – 587. Rbdgaard, A., Nielsen, F.C., Djurup, R., Somnier, F., Gammeltoft, S., 1987. Acetylcholine receptor antibody in myasthenia gravis: predominance of IgG subclass 1 and 3. Clin. Exp. Immunol. 67, 82 – 88. Romi, F., Skeie, G.O., Aarli, J.A., Gilhus, N.E., 2000a. Muscle autoantibodies in subgroups of myasthenia gravis patients. J. Neurol. 247, 369 – 375. Romi, F., Skeie, G.O., Aarli, J.A., Gilhus, N.E., 2000b. The severity of myasthenia gravis correlates with the serum concentration of titin and ryanodine receptor antibodies. Arch. Neurol. 57, 1596 – 1600. Romi, F., Skeie, G.O., Vedeler, C., Aarli, J.A., Zorzato, F., Gilhus, N.E., 2000c. Complement activation by titin and ryanodine receptor autoantibodies in myasthenia gravis. A study of IgG subclasses and clinical correlations. J. Neuroimmunol. 111, 169 – 176. Skeie, G.O., Mygland, 2., Aarli, J.A., Gilhus, N.E., 1995. Titin antibodies in patients with late onset myasthenia gravis: clinical correlations. Autoimmunity 20, 99 – 104. Skeie, G.O., Bartoccioni, E., Evoli, A., Aarli, J.A., Gilhus, N.E., 1996. Ryanodine receptor antibodies are associated with severe myasthenia gravis. Eur. J. Neurol. 3, 136 – 140. Skeie, G.O., Mygland, A., Treves, S., Gilhus, N.E., Aarli, J.A., Zorzato, F., 2003. Ryanodine receptor antibodies in myasthenia gravis: epitope mapping and effect on calcium release in vitro. Muscle Nerve 27, 81 – 89. Tuzun, E., Scott, B.G., Goluszko, E., Higgs, S., Christadoss, P., 2003. Genetic evidence for involvement of classical complement pathway in induction of experimental autoimmune myasthenia gravis. J. Immunol. 171, 3847 – 3854. Voltz, R.D., Albrich, W.C., N7gele, A., Schumm, F., Wick, M., Freiburg, A., Gautel, M., Thaler, H.T., Aarli, J.A., Kirchner, T., Hohlfeld, R., 1997. Paraneoplastic myasthenia gravis: detection of anti-MGT30 (titin) antibodies predicts thymic epithelial tumor. Neurology 49, 1454 – 1457.
The role of complement in myasthenia gravis: serological evidence of complement consumption in vivo Fredrik Romia,*, Einar K. Kristoffersenb, Johan A. Aarlia,c, Nils Erik Gilhusa,c a Department of Neurology, Haukeland University Hospital, Bergen N-5021, Norway Microbiology and Immunology (The Bloodbank), Haukeland University Hospital, Bergen, Norway c University of Bergen, Bergen N-5021, Norway
b
Received 7 June 2004; received in revised form 5 August 2004; accepted 5 August 2004
Abstract Background: Antibodies to the acetylcholine receptor (AChR) titin and the ryanodine receptor (RyR) occur in myasthenia gravis (MG). These antibodies are capable of complement activation in vitro. The involvement of the complement system should cause consumption of complement components such as C3 and C4 in vivo. Materials and methods: Complement components C3 and C4 were assayed in sera from 78 AChR antibody-positive MG patients and 52 healthy controls. Forty-eight of the patient sera contained titin antibodies as well, and 20 were also RyR antibody-positive. Results: MG patients with AChR antibody concentrations above the median (11.2 nmol/l) had significantly lower mean C3 and C4 concentrations in serum compared to those with AChR antibody concentrations below the median. Titin antibody-positive MG patients, titin antibody-negative early-onset MG patients, titin antibody-negative late-onset MG patients, and controls had similar C3 and C4 concentrations. Nor did mean C3 and C4 concentrations differ in MG patients with RyR antibodies. Patients with severe MG (grades 4 and 5) had similar C3 and similar C4 levels compared to those with mild MG (grades 1 and 2). Conclusion: An increased in vivo complement consumption was detected in MG patients with high AChR antibody concentrations, unrelated to MG severity and non-AChR muscle antibodies. D 2004 Elsevier B.V. All rights reserved. Keywords: Myasthenia gravis; Acetylcholine receptor antibodies; Titin antibodies; Ryanodine receptor antibodies; C3; C4; Complement activation
1. Introduction Myasthenia gravis (MG) is an autoimmune disease affecting the neuromuscular junction of skeletal muscles. Acetylcholine receptor (AChR) antibodies are present in sera from 80% to 90% of patients with generalized MG, and occur only very infrequently in healthy individuals (Lindstro¨m, 1985; Lindstro¨m et al., 1998). Non-AChR muscle autoantibodies, such as against titin (Aarli et al., 1990), and ryanodine receptor (RyR) (Mygland et al., 1992a) are present in many MG sera, mainly from patients with thymoma or late-onset MG, and they occur * Corresponding author. Tel.: +47 55975000; fax: +47 55975165. E-mail address: [email protected] (F. Romi). 0165-5728/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.jneuroim.2004.08.002
more frequently in severe MG (Aarli et al., 1987; Mygland et al., 1994; Skeie et al., 1995, 1996; Voltz et al., 1997; Romi et al., 2000a,b). Antibodies to musclespecific kinase (MuSK) occur in 41% of sera from AChR antibody-negative generalized MG patients (McConville et al., 2004). AChR antibodies impair neuromuscular transmission by complement-mediated focal muscle membrane damage and accelerated degradation of AChR (Lindstro¨m, 1985; Lindstro¨m et al., 1998). Titin and RyR antibodies are capable of complement activation in vitro and could, if binding in vivo, cause complementmediated muscle cell damage (Romi et al., 2000c). The involvement of the complement system should cause an increased consumption of complement components such as C3 and C4.
192
F. Romi et al. / Journal of Neuroimmunology 158 (2005) 191–194
Our objective in this study was to measure the levels of C3 and C4 in MG sera and to correlate these concentrations to the presence of muscle antibodies and to MG severity.
sample from each patient was used for assessment of all antibody assays and complement components. 2.2. Complement components C3 and C4
2. Patients and methods 2.1. Patients Forty-eight (male/female: 21/27) consecutive titin antibody-positive MG patients, of whom 20 also had RyR antibodies, were included. Nineteen had thymoma, 10 had thymic atrophy, 4 had thymic hyperplasia, and the rest were not thymectomized. Thirty (male/female: 10/20) randomly chosen nonthymoma titin and RyR antibodynegative MG patients, of whom 15 had early-onset MG and 15 late-onset MG, were also included as MG controls (Romi et al., 2000a). All 78 MG patients had AChR antibodies and generalized MG, and they were all treated at Haukeland University Hospital (Bergen, Norway). Five patients had additional autoimmune diseases: two female patients had thyroiditis, one female patient had systemic lupus erythematosus, one female patient had polymyalgia rheumatica, and one male patient had rheumatoid arthritis. Fifty-two healthy controls, comparable regarding age and sex, were also included. None of the patients or controls had systemic infection when obtaining the blood sample. The diagnosis of MG was based on a typical clinical pattern, a positive edrophonium test, and neurophysiological investigations with decrement at repetitive stimulation and increased jitter at single-fiber EMG. None of the patients received immunosuppressive drug treatment at the time when the blood sample was obtained. The diagnosis of thymoma was based on histopathology. In patients not thymectomized, a thymoma was excluded on computed tomography (CT) or magnetic resonance imaging (MRI) of the mediastinum. The severity of MG was graded using a modified Osserman’s scale on the day the blood sample was obtained (Osserman and Genkins, 1971; Romi et al., 2000b): 1. 2. 3.
4. 5.
MG with purely ocular muscle weakness. MG with mild generalized weakness, without bulbar involvement. MG with mild generalized weakness including bulbar involvement with dysarthria, dysphagia, and poor mastication. MG with moderate generalized weakness. MG with severe generalized, bulbar, and respiratory muscle weakness.
The severity groups were also rearranged in mild (severity groups 1 and 2), moderate (severity group 3), and severe (severity groups 4 and 5) MG. All sera were stored at 80 8C and, at 20 8C prior to testing, were treated in the same way. The same serum
Sera were allowed to thaw in room temperature prior to analysis for complement factors C3 and C4 by nephelometry on a Behring Nephelometry analyser II (Dade Behring, Marburg, Germany), using rabbit polyclonal antiserum to highly purified human C3c and C4 as detailed by the manufacturer (Dade Behring). 2.3. Muscle autoantibody assays 2.3.1. AChR antibodies AChR antibodies were measured by a standard radioimmunoassay (RIA) method with human 125I AChR as the antigen and using AChR RRA kits (IBL, Hamburg, Germany). (Lefvert et al., 1978; Romi et al., 2000a). 2.3.2. Titin antibodies Titin antibodies were detected by ELISA using purified titin fragment MGT-30 as the antigen at a concentration of 3.5 Ag/ml in PBS-S (Skeie et al., 1995; Romi et al., 2000c). 2.3.3. RyR antibodies RyR antibodies were assayed by Western blot using pc2 Ry1 fusion protein as the antigen (Mygland et al., 1992b; Skeie et al., 2003). RyR antibodies were also detected by ELISA using the same antigen at 2 Ag/ml in PBS-S (Skeie et al., 1995; Romi et al., 2000c). 2.4. Statistical analysis Mean C3 and C4 concentrations and mean antibody concentrations and titers were compared for the various groups using the t test for difference between population means with unequal variances and the Mann–Whitney test. MG severity groups were compared using the chi-square test of independence. p values less than 0.05 were considered significant.
3. Results 3.1. C3, C4, and AChR antibodies MG patients with AChR antibody concentrations above the median (11.2 nmol/l) had lower mean C3 (1.73F0.74 g/l) and C4 (0.48F0.56 g/l) concentrations than those with AChR antibody concentrations below the median (C3, 2.11F0.95 g/l; C4, 0.57+0.34 g/l) (0.025bpb0.05 for C3; 0.01bpb0.05 for C4). Only one of the five patients with additional autoimmune diseases had AChR antibody concentration above the median value.
F. Romi et al. / Journal of Neuroimmunology 158 (2005) 191–194
193
Table 1 The immunological characteristics of patients and controls, showing mean C3, C4, and antibody concentrations, and mean MG severity in titin antibodypositive MG, titin antibody-negative MG, and healthy controls Titin antibody-positive MG group
Mean C3 concentration (g/l) Mean C4 concentration (g/l) Mean AChR antibody concentration (nmol/l) Titin antibodies n positive Mean titer RyR antibodies n positive Mean titer Mean MG severity
Titin antibody-negative MG group
Healthy controls
Early-onset MG
Late-onset MG
1.89 (F0.96) 0.51 (F0.44) 48.7 (F139.6)
1.76 (F0.62) 0.39 (F0.17) 77.8 (F161.9)
1.93 (F0.82) 0.55 (F0.27) 43.3 (F78.9)
1.91 (F0.75) 0.45 (F0.20) 0
48 5417 (F6571)
0 0
0 0
0 0
20 2838 (F3022) 3.0 (F0.9)
0 0 2.7 (F0.5)
0 0 3.0 (F1)
0 0 0
There was no difference between the total MG patient group and the healthy controls (Table 1). 3.2. C3, C4, titin, and RyR antibodies The mean concentration of C3 and C4 was similar in titin antibody-positive MG patients (C3, 1.89F0.96 g/l; C4, 0.51F0.44 g/l) and titin antibody-negative MG patients (C3, 1.84F0.72 g/l; C4, 0.47F0.24 g/l). Nor did the concentration differ in MG patients with RyR antibodies (C3, 1.79F0.92 g/l; C4, 0.54F0.54 g/l) (Table 1). Mean C3 (1.91F1.06 g/l) and C4 (0.54F0.51 g/l) concentrations in patients with titin antibody titers above the median (3200) did not differ significantly from those in patients with mean titin antibody titers below the median (C3, 1.79F0.89 g/l; C4, 0.44F0.34 g/l). 3.3. C3, C4, and MG severity Patients with severe MG (C3, 1.96F1.15 g/l; C4, 0.59F0.58 g/l) did not have lower C3 and C4 levels compared to those with mild MG (C3, 1.77F0.76 g/l; C4, 0.39F0.17 g/l). There were no significant differences in C3 and C4 levels between any severity groups in this study (Table 1).
4. Discussion This study demonstrates serological evidence for the involvement of the complement system in the pathophysiology of MG in vivo. The lower concentration of C3 and C4 in patients with high AChR antibody concentration can best be explained by an increased consumption of C3 and C4: (1) there is no evidence for reduced C3 and C4 production into the serum among MG patients since patients and controls had similar C3 and C4 levels; (2) increased C3 and C4 consumption has been demonstrated for other autoimmune diseases, although in this study, only one of the five patients with additional autoimmune disease was in the group with low C3 and C4 levels; and (3) none of the patients had
infectious disease explaining an increased complement consumption. There is, therefore, circumstantial evidence for increased consumption of the complement factors C3 and C4 in patients with high AChR antibody concentration. This consumption was statistically higher for C4 than for C3, indicating involvement of the classical fast and efficient complement activation pathway that leads to immunological damage through the formation of membrane attack complexes. Such membrane attack complexes on the AChR in MG muscle cells has been demonstrated electronmicroscopically (Engel and Arahata, 1987). AChR antibodies consist mainly of IgG 1 and IgG 3, which are both efficient activators of the classical complement activation pathway (Lindstro¨m, 1985; Lindstro¨m et al., 1998; Rbdgaard et al., 1987). Although IgG 1 is highly represented in both titin and RyR antibodies (Mygland et al., 1993; Romi et al., 2000c) and both antibodies are capable of complement activation in vitro (Romi et al., 2000c), this study did not show any reduction in the concentrations of C3 or C4 in patients with high titin and RyR antibody concentrations. Nor was any C3 and C4 reduction evident in patients with AChR antibody concentrations below the median value. This does not mean that patients with low AChR antibody concentrations have less severe disease, since MG severity was equal in these patients and in patients with high AChR antibody concentrations. Additional mechanisms may therefore contribute to AChR antibody pathogenicity. Similarly, we cannot exclude the in vivo pathogenicity of titin and RyR antibodies, including C3 and C4 activation. MG severity does not correlate with the overall serum concentration of AChR antibodies (Oosterhuis et al., 1983; Kuks et al., 1993). Previous studies have shown that titin and RyR antibodies are present more often in severe MG but their concentration is not correlated to MG severity, although such correlation is seen in individual patients over time (Aarli et al., 1987; Mygland et al., 1994; Skeie et al., 1995, 1996; Romi et al., 2000b). The reason why the AChR antibody concentration does not correlate with MG severity despite reduced C3 and C4 concentrations in sera from patients with a high antibody concentration is not clear.
194
F. Romi et al. / Journal of Neuroimmunology 158 (2005) 191–194
Our results are consistent with a recent study showing that the induction of EAMG was dependent on the classical complement pathway (Tuzun et al., 2003). The serum concentration of complement components falls during MG exacerbations, probably as a result of complement binding to muscle antigens (Plescia et al., 1966), and varies according to the severity of the disease (Nastuk et al., 1960). In summary, this study gives serological evidence for the involvement of the complement system in the pathophysiology of MG in vivo through increased consumption of the complement factors C3 and C4 in patients with high AChR antibody concentrations, although this complement consumption was not related to MG severity or to the presence of other muscle antibodies.
Acknowledgement This work was supported by EU grant QL61-CT-200101918 and the Norwegian Association for Muscle Disease.
References Aarli, J.A., Gilhus, N.E., Hofstad, H., 1987. CA-antibody: an immunological marker of thymic neoplasia in myasthenia gravis? Acta Neurol. Scand. 63, 55 – 57. Aarli, J.A., Stefansson, K., Marton, L.S.G., Wollmann, R.L., 1990. Patients with myasthenia gravis and thymoma have in their sera IgG autoantibodies against titin. Clin. Exp. Immunol. 82, 284 – 288. Engel, A.G., Arahata, K., 1987. The membrane attack complex of complement at the endplate in myasthenia gravis. Ann. N.Y. Acad. Sci. 505, 326 – 332. Kuks, J.B., Limburg, P.C., Horst, G., Oosterhuis, H.J., 1993. Antibodies to skeletal muscle in myasthenia gravis: Part 3. Relation with clinical course and therapy. J. Neurol. Sci. 120, 168 – 173. Lefvert, A.K., Bergstrbm, K., Matell, G., Osterman, P.O., Pirskanen, R., 1978. Determination of acetylcholine receptor antibody in myasthenia gravis—clinical usefulness and pathogenic implications. J. Neurol. Neurosurg. Psychiatry 41, 394 – 403. Lindstrfm, J., 1985. Immunobiology of myasthenia gravis, experimental autoimmune myasthenia gravis, and Lambert–Eaton syndrome. Annu. Rev. Immunol. 3, 109 – 131. Lindstrfm, J.M., Seybold, M.E., Lennon, V.A., Whittingham, S., Duane, D.D., 1998. Antibody to acetylcholine receptor in myasthenia gravis: prevalence, clinical correlates, and diagnostic value. 1975 Classical article. Neurology 51, 933 – 939. McConville, J., Farrugia, M.E., Beeson, D., Kishore, U., Metcalfe, R., Newsom-Davis, J., Vincent, A., 2004. Detection and characterization of MuSK antibodies in seronegative myasthenia gravis. Ann. Neurol. 55, 580 – 584. Mygland, 2., Tysnes, O.B., Aarli, J.A., Flood, P.R., Gilhus, N.E., 1992a. Myasthenia gravis patients with a thymoma have antibodies against a
high molecular weight protein in sarcoplasmatic reticulum. J. Neuroimmunol. 37, 1 – 7. Mygland, 2., Tysnes, O.B., Matre, R., Volpe, P., Aarli, J.A., Gilhus, N.E., 1992b. Ryanodine receptor autoantibodies in myasthenia gravis patients with thymoma. Ann. Neurol. 32, 589 – 591. Mygland, 2., Tysnes, O.B., Aarli, J.A., Matre, R., Gilhus, N.E., 1993. IgG subclass distribution of ryanodine receptor autoantibodies in patients with myasthenia gravis and thymoma. J. Autoimmun. 6, 507 – 515. Mygland, 2., Aarli, J.A., Matre, R., Gilhus, N.E., 1994. Ryanodine receptor antibodies related to severity of thymoma associated myasthenia gravis. J. Neurol. Neurosurg. Psychiatry 57, 843 – 846. Nastuk, W.L., Plescia, O.J., Osserman, K.E., 1960. Changes in serum complement activity in patients with myasthenia gravis. Proc. Soc. Exp. Biol. Med. 105, 177 – 184. Oosterhuis, H.J.G.H., Limburg, P.C., Hummel-Tappel, E., 1983. Antiacetylcholine receptor antibodies in myasthenia gravis: Part 2. Clinical and serological follow-up of individual patients. J. Neurol. Sci. 58, 371 – 385. Osserman, K., Genkins, G., 1971. Studies in myasthenia gravis: review of a twenty-year experience in over 1200 patients. Mt. Sinai J. Med. 38, 497 – 537. Plescia, O.J., Segovia, J.M., Strampp, A., 1966. An assessment of changes in the complement level of myasthenic sera. Ann. N.Y. Acad. Sci. 135, 580 – 587. Rbdgaard, A., Nielsen, F.C., Djurup, R., Somnier, F., Gammeltoft, S., 1987. Acetylcholine receptor antibody in myasthenia gravis: predominance of IgG subclass 1 and 3. Clin. Exp. Immunol. 67, 82 – 88. Romi, F., Skeie, G.O., Aarli, J.A., Gilhus, N.E., 2000a. Muscle autoantibodies in subgroups of myasthenia gravis patients. J. Neurol. 247, 369 – 375. Romi, F., Skeie, G.O., Aarli, J.A., Gilhus, N.E., 2000b. The severity of myasthenia gravis correlates with the serum concentration of titin and ryanodine receptor antibodies. Arch. Neurol. 57, 1596 – 1600. Romi, F., Skeie, G.O., Vedeler, C., Aarli, J.A., Zorzato, F., Gilhus, N.E., 2000c. Complement activation by titin and ryanodine receptor autoantibodies in myasthenia gravis. A study of IgG subclasses and clinical correlations. J. Neuroimmunol. 111, 169 – 176. Skeie, G.O., Mygland, 2., Aarli, J.A., Gilhus, N.E., 1995. Titin antibodies in patients with late onset myasthenia gravis: clinical correlations. Autoimmunity 20, 99 – 104. Skeie, G.O., Bartoccioni, E., Evoli, A., Aarli, J.A., Gilhus, N.E., 1996. Ryanodine receptor antibodies are associated with severe myasthenia gravis. Eur. J. Neurol. 3, 136 – 140. Skeie, G.O., Mygland, A., Treves, S., Gilhus, N.E., Aarli, J.A., Zorzato, F., 2003. Ryanodine receptor antibodies in myasthenia gravis: epitope mapping and effect on calcium release in vitro. Muscle Nerve 27, 81 – 89. Tuzun, E., Scott, B.G., Goluszko, E., Higgs, S., Christadoss, P., 2003. Genetic evidence for involvement of classical complement pathway in induction of experimental autoimmune myasthenia gravis. J. Immunol. 171, 3847 – 3854. Voltz, R.D., Albrich, W.C., N7gele, A., Schumm, F., Wick, M., Freiburg, A., Gautel, M., Thaler, H.T., Aarli, J.A., Kirchner, T., Hohlfeld, R., 1997. Paraneoplastic myasthenia gravis: detection of anti-MGT30 (titin) antibodies predicts thymic epithelial tumor. Neurology 49, 1454 – 1457.