High anti-ebna-1 IgG levels are associated with early-onset myasthenia gravis

European Journal of Neurology 2012, 19: 842 846 doi:10.1111/j.1468-1331.2011.03636.x High anti-ebna-1 IgG levels are associated with early-onset myasthenia gravis D. Csuka a, M. Banati b, C. Rozsa c,g.füst a and Z. Illes b a 3 rd Department of Internal Medicine, Semmelweis University, Budapest; b Division of Clinical and Experimental Neuroimmunology, Department of Neurology, University of Pecs, Pecs; and c Department of Neurology, Jahn Ferenc Teaching Hospital, Budapest, Hungary Keywords: AchR, EBNA, EBV, myasthenia gravis, thymoma Received 22 June 2011 Accepted 17 November 2011 Background: Myasthenia gravis (MG) is an autoimmune disorder mediated by antibodies against the acethylcholine receptor (AchR) of the neuromuscular junction in the majority of patients. Methods: Here, we examined IgG antibodies against the type 1 nuclear antigen of Epstein-Barr virus (EBNA-1) in the sera of 158 patients with MG compared to 184 healthy controls. Results: Although serum concentration in the sera was not different, high anti- EBNA-1 IgG titers (above 90th percentile of the normal values) were more common in the patients (26.6 vs. 16.3%, P = 0.024). In addition, high EBNA-1 IgG levels occurred more frequently amongst the 94 patients with early-onset myasthenia gravis (EOMG, 30.8%) as compared to the 64 patients with late-onset disease (LOMG, 14.1%) (P = 0.021). Using multiple logistic regression, high serum concentration of the anti-ebna-1 IgG antibodies was significantly associated with EOMG (OR: 3.17, P = 0.027), even after adjustment for sex, presence/absence of anti-achr antibodies and presence/absence of anti-titin antibodies. Out of 39 patients with EOMG, who underwent thymectomy, 18 patients (46%) had thymoma, 6 had thymic hyperplasia (15%), and 15 patients had thymic atrophy (39%); there was no difference comparing EBNA-1 antibody titers in the sera. As no correlation was found between the titers of anti-achr, anti-titin, and EBNA-1 antibodies, a dysregulated heterogeneous B-cell response was unlikely to be responsible for the elevated levels of EBV-associated antibody in patients. Conclusions: In summary, our data suggest that high levels of EBNA-1 antibodies are more common in MG compared to healthy controls and are especially associated with EOMG. Introduction Epstein-Barr virus (EBV) is a ubiquitous virus, which infects practically the whole human population in childhood, adolescent age or early adulthood, therefore almost every adult carries the virus [1]. In early childhood, EBV infection mostly occurs symptom-free, whilst in later ages it is frequently associated with infectious mononucleosis. EBV replicates in B lymphocytes and epithelial cells of pharynx and tonsils. Shortly after infection, EBV in B lymphocytes switches to so-called latency III program, i.e. no virus proteins are expressed on the surface of the cells except during Correspondence: Z. Illes, Department of Neurology, University of Pecs, Ret u. 2., Pecs, 7623 Hungary (tel.: +(36) 72 535 900/5963; fax: +(36) 72 535 911; e-mail: zsolt.illes@aok.pte.hu). division, when one of the nuclear antigens, Epstein-Barr virus nuclear antigen (EBNA-1) is shortly presented [1]. Different types of humoral immune responses occur after EBV infection. IgM antibodies against the viral capsid antigen (VCA) develop first, which disappear shortly thereafter followed by the development of IgG anti-vca antibodies. These antibodies peak after 2 4 weeks and although decline thereafter, can be detected life-long. Antibodies against the early antigen (EA-D) also appear relatively early but only after the anti-vca antibodies, and gradually disappear from the majority of infected people. Antibodies against the Epstein-Barr virus nuclear antigen (EBNA-1, EBNA-2, etc.) develop relatively late, usually 2 4 months after infection, but they also can be detected life-long [2]. Myasthenia gravis is an autoimmune disorder of the neuromuscular junction mediated by antibodies against the acethylcholine receptor (AchR) in the majority of 842 European Journal of Neurology Ó 2012 EFNS

High anti-ebna-1 titers and EOMG 843 patients. Early-onset MG (EOMG) is commonly associated with hyperplastic alteration of the thymus characterized by the presence of germinative centers and production of anti-achr antibodies. In contrast, anti- Titin antibodies are more common in late-onset myasthenia gravis (LOMG) and hyperplastic alteration of the thymus is scarce [3,4]. Recently, a connection between EBV and MG has been indicated based on the presence of EBV DNA in hyperplastic thymus and suggesting that dysregulated EBV infection may contribute to the immunological alterations initiating and/ or perpetuating the disease [5]. Here, we measured the anti-ebna-1 IgG levels in patients with both earlyand late-onset MG. Materials and methods Patients Patients were consecutively recruited from the two main MG centers of Hungary (Department of Neurology, University of Pecs at Pecs and Department of Neurology, Jahn Ferenc Teaching Hospital, Budapest). Tissue banks (sera, DNA) along with clinical data have been established and used also for this study. Blood was taken for the measurement of anti-achr antibodies soon after establishing the clinical diagnosis. Altogether sera from 158 patients with MG were examined (mean age: 54.3 ± 18.7 years): 103 female (65%; disease onset 35.5 ± 17.4 years) and 55 male patients (35%, disease onset 53.2 ± 16.6 years). Control subjects Serum samples were taken from 184 healthy volunteers (89 men, 95 women, mean age 35.5 ± 9 years) as control. Control subjects were located at the 3rd Department of Internal Medicine, Semmelweis University, Budapest, Hungary. Serum samples both from patients and controls were stored at )80 C. The study protocol was approved by the Regional Local Ethics Committee. Written informed consent was obtained from each patients and control subjects. Quantitation of anti-achr an anti-titin antibodies Anti-Titin and anti-achr antibodies were measured using an anti-titin antibody enzyme-linked immunosorbent assay (DLD Diagnostika GmbH, Hamburg, Germany) and an anti-achr antibody radioimmunoassay (IBL GmbH, Hamburg, Germany), respectively, according to the manufacturerõs instructions. Cutoff values for the anti-titin enzyme-linked immunosorbent assay were two for the quotient OD sample/od calibrator and 0.4 nm for AchR antibody radioimmunoassay (RIA). Quantitation of the IgG antibodies to Epstein-BarrVirus Nuclear Antigen type 1 In brief: 1:1000 diluted serum samples and calibrators (5, 20, 110, 200 AU/ml) were applied to the plate coated with EBNA-1 synthetic peptide (37 C, 1 h). After rinsing, affinity-purified goat IgG to human IgG conjugated to horseradish peroxidase was dispensed into all wells except for the blank well (37 C, 1 h). After rinsing, tetramethylbenzidine/hydrogen peroxide system was added as chromogen/substrate into all wells (RT, 30 min), and the reaction was stopped with sulfuric acid solution. The enzyme activity was proportional to the concentration of IgG to EBNA-1 present in samples and calibrators. Absorbance of specimens was measured with a photometer at 450/630 nm. The specificity of measurements by the Kit is 99.4%, and sensitivity is 96.8%. Statistical analysis Statistical analysis was performed using the GraphPad Prism 3.0 (GraphPad Software Inc, San Diego, CA, USA, http://www.graphpad.com) software. Betweengroup differences were evaluated by the Mann Whitney test and categorical differences with the FischerÕs exact test. Multiple logistic regression analysis was made by the SPSS 13.0 (SPSS Inc., Chicago, IL, USA) software. All tests were two-tailed, and P < 0.05 was considered significant. Results Clinical data of patients The diagnosis of MG was confirmed according to international guidelines by clinical and EMG examination (routinely by RNS and in the majority of the patients also by sfemg). Patients with uncertain diagnosis because of negative or atypical ENG/EMG findings were excluded from the study. All patients were examined for the presence of anti-achr antibodies. Anti-AchR antibodies were found in 78% and anti- Titin antibodies in 19% of patients. Using age of 50 years as cut-off [6], 58% of patients had early-onset myasthenia gravis (EOMG, onset < 50 years, current age: 40.3 ± 11.8 years) and 42% had (LOMG, onset 50 years, current age: 71.5 ± 8.7 years). Eighty percentages of EOMG and 43% of LOMG were women. Forty-five patients underwent thymectomy (39 EOMG): 21 patients had thymoma (17 women, 18

844 D. Csuka et al. EOMG); six patients had thymic hyperplasia (six women, six EOMG), and 18 patients had thymic atrophy (13 women, 15 EOMG). None of the patients had myasthenic crisis or were treated with plasma exchange within 3 months. Of 158 patients, 90 patients received chronic immunosuppressive therapy, corticosteroid and/or azathiporine (53 women, 42 EOMG). Distribution of antibody titers in patients and control subjects First, we compared the serum concentration of the anti- EBNA-1 antibodies in the sera of 158 MG patients [163.4 (58.5 598.5) AU/ml] and 184 healthy controls [178.1 (76.1 431.5) AU/ml]: the difference was not significant (P = 0.477). However, as the distribution of the anti-ebna-1 IgG titers was strongly skewed toward high values, we divided the patients and controls according to the titers obtained in the control group: 90th percentile (600 AU/ml) of the normal values was considered as the cut-off of the high/low anti-ebna-1 IgG titers. High anti-ebna-1 IgG titers were found in 30/184 (16.3%) and 42/158 (26.6%) of the control persons and MG patients, respectively (P = 0.024), indicating that high anti-ebna-1 IgG titers are more common in the patients. Next, we examined the frequency of high anti- EBNA-1 IgG titers in different subgroups of the patients. High EBNA-1 IgG levels occurred significantly more frequently amongst the 94 patients with EOMG (30.8%) as compared to the 64 patients with LOMG (14.1%) (P = 0.021) (Fig. 1). When we separately compared EOMG and LOMG to the controls, subjects with high anti-ebna-1 IgG levels had an odds ratio of 2.28 to have EOMG (1.26 4.13, P < 0.007), but a nonsignificant OR of 0.84 to have LOMG (0.37 18.81, P = 0.84). In contrast to the age of onset of disease, there was no significant difference in EBNA-1 antibody titers according to the sex of the patients (P = 0.177), presence/absence of thymoma (P = 0.767), presence/ absence of anti-achr antibodies (P = 0.809), or presence/absence of anti-titin antibodies (P = 0.317). There was no correlation between EBNA-1 antibody titers and immunosuppressive treatment. We did not find any correlation between antibody titers against EBNA-1 and AchR or anti-titin antibodies in total, and in subgroups of EOMG and LOMG. As many variables, such as sex (P < 0.001), presence/absence of anti-achr antibodies (P = 0.002) and presence/absence of anti-titin antibodies (P < 0.001), significantly differed between EOMG and LOMG, next we calculated the correlation between high anti-ebna- 1 IgG titers and EOMG by multiple logistic regression analysis adjusted to these variables (Table 1). High Figure 1 Distribution of the patients with high and low anti- EBNA-1 titers amongst patients with early-onset and late-onset myasthenia gravis. High anti-ebna-1 IgG titers (> 600 AU/ml) were defined as 90th percentile (600 AU/ml) of the values obtained in the control group. FisherÕs exact test was significant (P < 0.05). Table 1 Correlation between high anti-ebna-1 IgG antibody levels and early-onset myasthenia gravis calculated by multiple logistic regression analysis Variable serum concentration of the anti-ebna-1 IgG antibodies was significantly associated with EOMG even after adjustment with an OR of almost 3. Discussion B SE of B P value Odds ratio of early/ late-onset disease OR 95% CI Anti-EBNA1 1.155 0.524 0.027 3.174 1.137 8.862 IgGhigh a /low b Sex female/male 1.316 0.431 0.002 3.729 1.602 8.609 Anti-TitinAb, )1.772 0.587 0.003 0.1704 0.054 0.536 yes/no Anti-AchRAb )1.380 0.609 0.023 0.252 0.076 0.829 yes/no Constant )0.554 0.889 0.533 0.575 a > 600 AU/ml. b 600 AU/ml. Calculated by multiple logistic regression analysis. AchR, antibodies against the acethylcholine receptor; EBNA-1, Epstein-Barr virus nuclear antigen. The strong association we observed between EOMG and the high serum concentration of the anti-ebna-1 IgG levels is a novel observation, and similar to some other autoimmune diseases. Besides epidemiological

High anti-ebna-1 titers and EOMG 845 data [7], a strong association with high titer of anti- EBNA-1 IgG indicates connection of EBV infection and MS [8]. Such high titers may precede clinical symptoms by several years or decades [9] and seem to be associated with the more severe clinical course of the disease as well [10]. High anti-ebna-1 IgG titer is a strong risk factor of MS, independent of and additive to the strongest genetic risk factor, carrier state of the HLA-DRB1*1501 [11]. The 90th percentile of the control values as an absolute value of 600 U/ml selected for the limit of significantly elevated EBNA-1 IgG titers is comparable to the limit used in a longitudinal followup study for distinguishing subjects with an increased risk for MS [12]. Besides MS, high anti-ebna-1 IgG titers were found in systemic lupus erythematosus [13] and rheumatoid arthritis (RA) [14]. Although elevated level of anti-ebna-1 antibodies is considered as a risk factor for multiple sclerosis [8,9], RA [12], and SLE [13], and it seems to also be the case for MG, especially EOMG, little is known on its causes. Increased titers to EBNA-1 may suggest a more severe or more recent primary infection or reactivation of infection [9]. Another possibility is the occurrence of coinfection with a strain of EBV different from that originally carried by the host, or infection with another microorganism, which alters the immune response to EBV. Data about EBV-associated antibodies in the sera of patients with MG are scarce; an early study using different methodology could not establish a link between EBV and MG [14]. Nevertheless, development of MG with anti-musk antibodies following infectious mononucleosis has been recently reported [15]. Association with other herpes visuses, like CMV, is also contradicting [14,16]. As we did not find any correlation between titers of EBNA-1 and anti-achr/anti-titin antibodies, it seems unlikely that polyclonal B-cell activation would be responsible for the elevation of EBV-specific antibodies. Similarly, no correlation was found between immunosuppressive treatment and EBNA antibody titers arguing against the role of immunosuppressive therapy as a consequence of heterogeneous EBNA titers. We found that high titers of EBNA-1 antibodies were more likely to be associated with EOMG than LOMG. In a recent paper, infection of the thymus by EBV has been indicated in patients with EOMG and thymic hyperplasia. EBV actively replicated in all thymuses from MG patients but not in control thymuses [5]. It can be assumed that similarly to SLE [17], the control of the EBV infection is defective in MG, and the active infection leads to raise of production of specific anti- EBV antibodies. According to the data of Cavalcante et al. [5] and our present findings, it can be assumed that active EBV infection in the thymus may increase the risk of the EOMG. Alternatively, if we consider that the levels of the anti-ebna-1 IgG antibodies is constant for years [9], it is possible that such high antibody levels may be a marker of a factor(s), which contribute to the development of the EOMG, similarly to that of multiple sclerosis [9]. Of note, another recent report was unable to support a direct role of thymic EBV infection in the pathogenesis of myasthenia gravis. [18]. Unfortunately, because thymectomy in our series was performed mainly in cases with thymoma, the number of cases with hyperplastic thymus was low to examine correlations with EBNA-1 titers. In summary, although serum concentration of the anti-ebna-1 antibodies in the sera of 158 MG patients were not different from 184 healthy controls, high serum concentration of the anti-ebna-1 IgG was significantly associated with MG even after adjustment. Considering these and previous data indicating persistence and reactivation of EBV in hyperplastic thymus of patients with EOMG, the role of EBV in autoimmune MG should be similarly considered to other autoimmune disorders like MS, RA, and SLE. Acknowledgements This work was supported by grants from Hungarian National Research Fund (OTKA 77892 to ZI and OTKA-NKTH 80842 to GF) and the Hungarian Neuroimaging Foundation (to ZI). Disclosure of conflicts of interest The authors declare no financial or other conflict of interests. References 1. Niller HH, Wolf H, Minarovits J. Epigenetic dysregulation of the host cell genome in Epstein-Barr virus-associated neoplasia. Semin Cancer Biol 2009; 19: 158 164. 2. Tsuchiya S. Diagnosis of Epstein-Barr virus-associated diseases. Crit Rev Oncol Hematol 2002; 44: 227 238. 3. Meriggioli MN, Sanders DB. Autoimmune myasthenia gravis: emerging clinical and biological heterogeneity. Lancet Neurol 2009; 8: 475 490. 4. Pal J, Rozsa C, Komoly S, et al. Clinical and biological heterogeneity of autoimmune myasthenia gravis. J Neuroimmunol 2011; 231: 43 54. 5. Cavalcante P, Serafini B, Rosicarelli B, et al. Epstein-Barr virus persistence and reactivation in myasthenia gravis thymus. Ann Neurol 2010; 67: 726 738. 6. Gilhus NE, Owe JF, Hoff JM, et al. Myasthenia gravis: a review of available treatment approaches. Autoimmune Dis 2011; 2011: 847393. 7. Ascherio A, Munch M. Epstein-Barr virus and multiple sclerosis. Epidemiology 2000; 11: 220 224.

846 D. Csuka et al. 8. Ascherio A, Munger KL, Lennette ET, et al. Epstein-Barr virus antibodies and risk of multiple sclerosis: a prospective study. JAMA 2001; 286: 3083 3088. 9. Levin LI, Munger KL, Rubertone MV, et al. Temporal relationship between elevation of epstein-barr virus antibody titers and initial onset of neurological symptoms in multiple sclerosis. JAMA 2005; 293: 2496 2500. 10. Farrell RA, Antony D, Wall GR, et al. Humoral immune response to EBV in multiple sclerosis is associated with disease activity on MRI. Neurology 2009; 73: 32 38. 11. De Jager PL, Simon KC, Munger KL, et al. Integrating risk factors: HLA-DRB1*1501 and Epstein-Barr virus in multiple sclerosis. Neurology 2008; 70: 1113 1118. 12. Munger K, Levin L, OÕReilly E, et al. Anti-Epstein-Barr virus antibodies as serological markers of multiple sclerosis: a prospective study among United States military personnel. Mult Scler 2011; 17: 1185 1193. 13. McClain MT, Rapp EC, Harley JB, et al. Infectious mononucleosis patients temporarily recognize a unique, cross-reactive epitope of Epstein-Barr virus nuclear antigen-1. J Med Virol 2003; 70: 253 257. 14. Ferrell PB, Aitcheson CT, Pearson GR, et al. Seroepidemiological study of relationships between Epstein-Barr virus and rheumatoid arthritis. J Clin Invest 1981; 67: 681 687. 15. Bhibhatbhan A, Kline G, Vincent A, et al. Anti-MuSK myasthenia gravis presenting with Epstein-Barr virusassociated mononucleosis and immune-mediated diabetes mellitus. Muscle Nerve 2007; 36: 264 266. 16. Tindall RS, Cloud R, Luby J, et al. Serum antibodies to cytomegalovirus in myasthenia gravis: effects of thymectomy and steroids. Neurology 1978; 28: 273 277. 17. Kang I, Quan T, Nolasco H, et al. Defective control of latent Epstein-Barr virus infection in systemic lupus erythematosus. J Immunol 2004; 172: 1287 1294. 18. Meyer M, Hols AK, Liersch B, et al. Lack of evidence for Epstein-Barr virus infection in myasthenia gravis thymus. Ann Neurol 2011; 70: 515 518.