Article Text

Extended report
Disease specificity of autoantibodies to cytosolic 5′-nucleotidase 1A in sporadic inclusion body myositis versus known autoimmune diseases
  1. Megan K Herbert1,
  2. Judith Stammen-Vogelzangs1,
  3. Marcel M Verbeek2,3,
  4. Anke Rietveld2,
  5. Ingrid E Lundberg4,
  6. Hector Chinoy5,
  7. Janine A Lamb6,
  8. Robert G Cooper7,
  9. Mark Roberts8,
  10. Umesh A Badrising9,
  11. Jan L De Bleecker10,
  12. Pedro M Machado11,
  13. Michael G Hanna11,
  14. Lenka Plestilova12,
  15. Jiri Vencovsky12,
  16. Baziel G van Engelen2,
  17. Ger J M Pruijn1
  1. 1Department of Biomolecular Chemistry, Radboud Institute for Molecular Life Sciences and Institute for Molecules and Materials, Radboud University, Nijmegen, The Netherlands
  2. 2Department of Neurology, Donders Institute for Brain, Cognition and Behaviour, Radboud University Medical Centre, Nijmegen, The Netherlands
  3. 3Department of Laboratory Medicine, Radboud University Medical Centre, Nijmegen, The Netherlands
  4. 4Rheumatology Unit, Department of Medicine, Karolinska Institutet/Karolinska University Hospital, Stockholm, Sweden
  5. 5Centre for Musculoskeletal Research, Manchester Academic Health Science Centre, The University of Manchester, Manchester, UK
  6. 6Centre for Integrated Genomic Medical Research, The University of Manchester, Manchester, UK
  7. 7Faculty of Health & Life Sciences, MRC/ARUK Institute of Ageing and Chronic Disease, University of Liverpool, Liverpool, UK
  8. 8Salford Royal NHS Foundation Trust, Manchester, UK
  9. 9Department of Neurology, Leiden University Medical Centre, Leiden, The Netherlands
  10. 10Department of Neurology, Neuromuscular Reference Centre, Ghent University Hospital, Ghent, Belgium
  11. 11MRC Centre for Neuromuscular Diseases, University College London, London, UK
  12. 12Department of Rheumatology, First Faculty of Medicine, Institute of Rheumatology, Charles University, Prague, Czech Republic
  1. Correspondence to Professor Ger J M Pruijn, Department of Biomolecular Chemistry 284, Radboud University Nijmegen, PO Box 9101, Nijmegen NL-6500 HB, The Netherlands; G.Pruijn{at}ncmls.ru.nl

Abstract

Objectives The diagnosis of inclusion body myositis (IBM) can be challenging as it can be difficult to clinically distinguish from other forms of myositis, particularly polymyositis (PM). Recent studies have shown frequent presence of autoantibodies directed against cytosolic 5′-nucleotidase 1A (cN-1A) in patients with IBM. We therefore, examined the autoantigenicity and disease specificity of major epitopes of cN-1A in patients with sporadic IBM compared with healthy and disease controls.

Methods Serum samples obtained from patients with IBM (n=238), PM and dermatomyositis (DM) (n=185), other autoimmune diseases (n=246), other neuromuscular diseases (n=93) and healthy controls (n=35) were analysed for the presence of autoantibodies using immunodominant cN-1A peptide ELISAs.

Results Autoantibodies directed against major epitopes of cN-1A were frequent in patients with IBM (37%) but not in PM, DM or non-autoimmune neuromuscular diseases (<5%). Anti-cN-1A reactivity was also observed in some other autoimmune diseases, particularly Sjögren's syndrome (SjS; 36%) and systemic lupus erythematosus (SLE; 20%).

Conclusions In summary, we found frequent anti-cN-1A autoantibodies in sera from patients with IBM. Heterogeneity in reactivity with the three immunodominant epitopes indicates that serological assays should not be limited to a distinct epitope region. The similar reactivities observed for SjS and SLE demonstrate the need to further investigate whether distinct IBM-specific epitopes exist.

  • Autoantibodies
  • Autoimmune Diseases
  • Dermatomyositis
  • Polymyositis

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

Idiopathic inflammatory myopathies (IIMs) comprise a diverse group of inflammatory muscle diseases with an insidious onset characterised by chronic muscle weakness, inflammation of skeletal muscle, electromyographic abnormalities and increases in muscle enzymes.1 ,2 Two of the three forms of IIM, polymyositis (PM) and dermatomyositis (DM), are well recognised as autoimmune diseases. They exhibit an array of autoantibodies directed against ubiquitous intracellular antigens that can be either myositis specific or myositis associated.3 In contrast, the pathogenesis of sporadic inclusion body myositis (IBM) has been a question of debate and many consider it to be a degenerative myopathy with secondary inflammation rather than a primary autoimmune disease.4 However, the recent discovery of an autoantibody (anti-Mup44) directed against cytosolic 5′-nucleotidase 1A (cN-1A) with frequent reactivity in the sera of patients with IBM5 ,6 has prompted a reconsideration of this paradigm.

IBM is supposedly the most common acquired muscle disease in adults aged >50 years,7 ,8 characterised clinically by an insidious onset of muscle weakness and muscle atrophy that slowly leads to severe disability.9 Unlike PM and DM, IBM is largely refractory to treatment with immunosuppressive, immunomodulatory or other therapies,10 ,11 which has contributed to the notion of IBM as a primarily degenerative disorder with a secondary immune component. Pathologically, IBM is characterised by a combination of degenerative features (rimmed vacuoles and abnormal protein accumulations) that partly resemble the histopathological features of other neurodegenerative diseases and which include aggregates of TAR DNA-binding protein 43, as found in frontotemporal dementia and amyotrophic lateral sclerosis, Aβ-42 and tau, as found in Alzheimer's dementia12–14 and the presence of p62 in rimmed vacuoles.15 However, muscle-specific autoimmune features such as cytotoxic T cell infiltration and clonal expansion of lymphocytes, which are common in IBM muscle biopsies, support a role for autoimmunity in these patients16 as does association of the autoimmune-prone human leukocyte antigen (HLA)-B8-DR3 ancestral haplotype with sporadic IBM.17 ,18 Antinuclear antibodies and antibodies against extractable nuclear antigens have been detected in up to 30% of patients with IBM.19 Detection of cN-1A-directed autoantibodies in patients with IBM in recent studies5 ,6 provides interesting new insights into IBM pathophysiology and which may lead to the development of a disease-specific serological diagnostic test for IBM.

The enzyme cN-1A catalyses the conversion of AMP into adenosine and phosphate, and it is highly expressed in skeletal muscle where it may be involved in muscle contraction.20 ,21 In an earlier study5 using immunoprecipitation techniques, we observed frequent anti-cN-1A reactivity in IBM sera and that was remarkably strong compared with other IIM autoantibodies, thus revealing the potential of anti-cN-1A antibodies as a disease-specific biomarker for IBM. Therefore, further characterisation of anti-cN-1A autoantibodies and their diagnostic role in IBM is clearly required. The aim of the current study was to determine the specificity of serum anti-cN-1A autoantibodies for IBM using an ELISA format with three synthetic peptides containing cN-1A autoepitopes previously identified by overlapping peptide microarray analyses.5 In addition, we investigated differences in cN-1A epitope recognition by IBM sera compared with sera from other patients with IIM and from a range of other neuromuscular and autoimmune diseases.

Methods

Patients and serum samples

We included serum samples from a large group of well-characterised patients with IBM (n=238) and 35 healthy controls with no known history of autoimmune or (neuro)muscular disease. The sera of the IBM group were gathered from the following participating European institutes: Radboud University Medical Centre, Nijmegen; Leiden University Medical Centre, Leiden (The Netherlands); University of Manchester on behalf of UKMYONET and University College London (UK); Ghent University Hospital, Ghent (Belgium); and Karolinska Institutet, Stockholm (Sweden). All patients fulfilled the European Neuromuscular Centre22 or Medical Research Council (MRC) 201023 criteria. Disease control sera were obtained from patients with DM and PM (n=185), scleroderma (n=44), PM/scleroderma overlap (n=12), multiple sclerosis (MS; n=40), Sjögren syndrome (SjS; n=22), systemic lupus erythematosus (SLE; n=44), rheumatoid arthritis (RA; n=44), type 1 diabetes (T1D; n=40) and other neuromuscular diseases (NMD; n=93).5 ,24 These were collected in Nijmegen except for PM/scleroderma overlap (Czech Republic) and PM/DM (Nijmegen and Manchester). Serum from healthy subjects was obtained from the Sanquin Blood Supply Foundation (Nijmegen, The Netherlands). Written or verbal informed consent was obtained from most of the patients from whom sera were used and all patient information was decoded to maintain confidentiality. The study protocol was in accordance with the Helsinki Declaration and all procedures were approved by the local ethics committees.

cN-1A peptides

Three 23 amino acid synthetic peptides derived from the sequence of cN-1A were used as target antigens in the ELISA. These peptides were identified as major epitope regions in our previous study5 and contained the sequences: peptide 1: PVWEEAKIFYDNLAPKKKPKSPK; peptide 2: SERIVKAHGLD RFFEHEKAHENK and peptide 3: AHVPYGVAQTPRRTAPAK QAPSA (all peptides contain aminohexanoic acid-Lys(biotin)-amide at the C terminus).

Peptide ELISA

Optimal ELISA conditions were determined using checkerboard titrations (data not shown) and the optical density (OD) 450 values for serially diluted serum samples at optimal dilutions were plotted (see online supplementary figure S1). Biotinylated peptides (∼30 ng per well) in phosphate buffered saline (PBS) containing 0.1% bovine serum albumin (BSA) were immobilised on preblocked Streptawell High Bind microplates (Roche, Mannheim, Germany) for 1 h at 37°C. After washing three times with 200 µL PBS/0.1% Tween-20 (PBST), 50 µL of 400-fold diluted patient serum in 1% BSA/PBS/0.05% Tween-20 was added, followed by incubation for 1 h at 37°C. Subsequently, wells were washed five times with 300 µL PBST and incubated with 50 µL 2000-fold diluted rabbit anti-human Ig (Dako, P0212) in 1% BSA/PBS/0.05% Tween-20 for 1 h at 37°C. Finally, after washing five times with 200 µL PBST, bound antibodies were visualised by adding 50 µL TMB (3,3′,5,5′-tetramethylbenzidine) substrate solution (Thermo Scientific) and the reaction was stopped by the addition of 50 µL 2 M H2SO4 after 5 min. Signals were quantified by determining ODs at 450 nm. Each plate contained at least one positive control and five negative control sera.

Statistical analysis

Statistical analyses were performed using GraphPad PRISM V.5 software (San Diego, California, USA) or SPSS software V.20.0 (Chicago, Illinois, USA).

Results

Based upon the reactivity of anti-cN-1A-positive IBM sera with a set of overlapping cN-1A peptides (15-mers, immobilised on microarrays),5 three 23-mer peptides corresponding to the most frequently targeted regions of cN-1A were synthesised for ELISA analyses (figure 1) and optimised using checkerboard titrations. Receiver operating characteristic curves were produced (see online supplementary figure S2) to compare IBM sera positive in immunoprecipitation experiments (reported previously5) against all disease controls excluding SLE and SjS. The OD450 value corresponding to the highest Youden Index (sensitivity+specificity−1) at which ≥98% specificity was achieved was chosen for each peptide. Sera were assessed as reactive if they were above the established cut-off value for at least one of the peptide antigens. Subsequently, the differences in reactivity between the different IBM cohorts (figure 2) and between IBM and other disease controls (figure 3) were investigated. No major differences were observed between the recognition of the cN-1A peptides among the IBM samples from different centres. The frequency of anti-cN-1A reactivity varied from 34% to 44% among the different IBM cohorts (figure 2).

Figure 1

Cytosolic 5′-nucleotidase 1A (cN-1A) peptide sequences of the synthetic peptides used in ELISA assays. Schematic representation of the cN-1A polypeptide (top) and amino acid sequence of cN-1A (bottom). The positions of the three peptide sequences used in ELISA are highlighted in blue, red and green, respectively.

Figure 2

Reactivity of inclusion body myositis (IBM) sera from various cohorts with cytosolic 5′-nucleotidase 1A (cN-1A) peptides. The presence of anti-cN-1A antibodies in samples from six IBM cohorts was analysed in ELISA using the three cN-1A peptides specified in figure 1. Lines represent cut-off values for each peptide (colour coding as in figure 1). Gen, Ghent, Belgium (n=9); Lei, Leiden, The Netherlands (n=32); Lon, London, England (n=24); Man, Manchester, England (n=89); Nijm, Nijmegen, The Netherlands (n=52); Sto, Stockholm, Sweden (n=32). Numbers following cohort abbreviation represent peptide numbers (1: PVWEEAKIFYDNLAPKKKPKSPK; 2: SERIVKAHGLDRFFEHEKAHENK and 3: AHVPYGVAQTPRRTAPAKQAPSA). Please note that samples from Manchester were provided on behalf of UKMyoNet.

Figure 3

Reactivity of patient sera with cytosolic 5′-nucleotidase 1A (cN-1A) peptides. The presence of anti-cN-1A antibodies in samples from different disease groups was analysed in ELISA using the three cN-1A peptides specified in figure 1. Lines represent cut-off values for each peptide (colour coding as in figure 1). Inclusion body myositis (IBM), sporadic IBM; MS, multiple sclerosis; NMD, other neuromuscular diseases; PM/DM, polymyositis/dermatomyositis; PM/Scl, polymyositis/scleroderma overlap syndrome; RA, rheumatoid arthritis; Scl, scleroderma; SjS, Sjögren's syndrome; SLE, systemic lupus erythematosus; T1D, type 1 diabetes. Numbers following disease abbreviation represent peptide numbers (1: PVWEEAKIFYDNLAPKKKPKSPK; 2: SERIVKAHGLDRFFEHEKAHENK and 3: AHVPYGVAQTPRRTAPAKQAPSA).

The frequency by which cN-1A peptides are recognised by patient sera is summarised in table 1. For the patients with IBM we observed 37% reactivity for at least one of the cN-1A peptides. Anti-cN-1A autoantibody reactivity was observed in just 4% of PM or DM patients (n=185) of which seven were PM patients and one a DM patient. In sera from other disease controls not >5% of sera from patients with PM/scleroderma overlap (n=12), other NMD (n=93), scleroderma (n=44), RA (n=44), MS (n=40) or type 1 diabetes (n=40) showed reactivity. However, we did observe frequent reactivity in sera from patients with SjS (36%; n=22) and SLE (20%; n=44). Anti-cN-1A reactivity correlates neither with the presence of other autoantibodies (see online supplementary tables S1 and S2) in IBM, SLE or SjS nor with IgG content (r2=0.07, p=0.23).

Table 1

Sensitivity and specificity of anti-cN-1A autoantibodies

There was also heterogeneity across the groups in terms of peptide specificity. Of the reactive IBM sera autoantibodies targeted single peptides—peptide 1 (23%), 2 (25%) or 3 (11%)—and also two (25%) or three peptides (16%; figure 4). Similarly, when we compared peptide reactivity in each of the other disease groups, we found high heterogeneity in reactivity for individual and combination peptides. However, reactivity to peptide 3 alone was only observed in IBM or PM sera, while the combination of peptides 2 and 3 was observed infrequently and only in patients with IBM (5%) and not in any of the disease control groups.

Figure 4

Reactivity to (combinations of) cytosolic 5′-nucleotidase 1A (cN-1A) peptides. Reactivity to either individual cN-1A peptides or combinations of peptides was calculated as a percentage of the total number of reactive sera for inclusion body myositis (IBM; blue), systemic lupus erythematosus (SLE; red), polymyositis/dermatomyositis (PM/DM; green) and Sjögren's syndrome (SjS; purple).

Among the patients with IBM, the average age at onset of disease did not differ significantly between the seronegative (median age 64 years, IQR=15, n=148) and seropositive (median age 66 years, IQR=17, n=84) groups. More male patients with IBM (41%) were reactive with one or more peptides than female patients with IBM (34%), but this difference was not statistically significant (p=0.18).

Discussion

In this study, we used a newly developed cN-1A peptide ELISA to investigate the prevalence and significance of anti-cN-1A autoantibodies in the sera of patients with IIM versus other autoimmune and NMD. We used synthetic peptides containing three immunodominant epitope regions of cN-1A to detect these autoantibodies. Screening of a large group of patients with IBM revealed that 37% had serum anti-cN-1A autoantibodies directed against at least one of the three epitopes. These results correlate well both with our own previous immunoprecipitation experiments5 showing high concentrations of cN-1A autoantibodies in 33% of IBM sera and with the results of a study showing cN-1A autoantibody reactivity in 34% of IBM sera using dot blot assays.6 Furthermore, we found that cN-1A autoantibodies were much more prominent in IBM than in PM or DM with only 4.3% of all PM/DM patients showing reactivity to cN-1A, seven with PM and one with DM. Although IBM may initially be misdiagnosed as PM, the clinical characteristics of three of the PM patients with cN-1A autoreactivity (clinical phenotype data for other patients was not available) showed that they did not exhibit the IBM phenotype.

In contrast to the frequent reactivity observed in 37% of IBM sera, <5% of sera from patients with scleroderma, PM/Scl overlap, MS, T1D and RA or non-autoimmune NMD contained anti-cN-1A autoantibodies. However, frequent anti-cN-1A autoantibody reactivity was detected in patients with SjS (36% of patients) or SLE (20% of patients). The significance of anti-cN-1A autoantibodies in these patient groups is unclear and will require further investigation. Interestingly, an increased incidence of SjS with IBM is supported by the literature.25 ,26

The peptides we designed for use as antigens in our ELISA are linear sequences containing 23 amino acids which were previously shown to contain the most frequently targeted regions of cN-1A by autoantibodies in microarray and in dot blot assays.5 ,6 In the peptide ELISAs, we observed high heterogeneity in serum reactivity within each disease group, with some showing reactivity to just a single peptide and others showing reactivity to two or all three peptides. Among the three peptides, peptide 3, located at the C terminus, was least frequently the single target of autoantibodies, particularly among the disease controls. We also noted that some of the sera testing positive for cN-1A autoantibodies during immunoprecipitation experiments5 were not reactive to any of the three epitopes using peptide ELISAs. Therefore, it is likely that cN-1A autoantibody reactivity is not restricted to the three linear epitopes used in the ELISA assays but that additional immunodominant epitopes are yet to be identified and will likely include conformational or discontinuous epitopes.

This notion was supported by the finding that reactivity to the combination of peptides 2 and 3 was observed only in some of the patients with IBM but not in any of the disease controls. In contrast, sera from patients with SLE predominantly targeted peptide 1 alone or in combination with peptide 2 and sera from patients with SjS predominantly targeted peptides 1 or 2 in isolation. However, the numbers of patients with SLE and SjS tested were relatively small so these results require confirmation in studies of larger groups. Furthermore, similar percentages of anti-cN-1A-positive patients were observed in our previous immunoprecipitation experiments and the current peptide ELISAs, but the correlation between the IP positive and peptide reactive sera was low (see online supplementary figure S3). This also strongly supports the existence of conformational autoepitopes in addition to the linear epitopes.

In summary, we developed an ELISA useful to measure anti-cN-1A autoantibodies in serum. Using this assay, we were able to confirm the results of earlier studies5 ,6 ,27 showing that cN-1A is a major autoantigen in IBM and that anti-cN-1A autoantibodies represent a new serological marker for IBM, particularly in differentiating IBM from other NMD, including PM. The clinical utility should not be severely reduced by the presence of anti-cN-1A autoantibodies in SLE and SjS, as in clinical practice there will generally be little difficulty differentiating a patient suffering with IBM from one suffering with SLE or SjS. However, in cases of comorbidity, this biomarker may be less informative. The results obtained here from larger patient cohorts have allowed confirmation of three distinct linear epitopes recognised by circulating antibodies. Thus, due to the heterogeneous anti-cN-1A response in established IBM, serological assays should not be limited to a distinct epitope region. Despite anti-cN-1A autoantibodies occurring in both IBM and in other autoimmune diseases, particularly SLE and SjS, the pattern of reactivity to distinct (combinations of) epitopes differed. Further research is now in progress to establish whether more IBM-specific cN-1A epitopes exist, and which could facilitate the future development of more IBM-specific serological assays.

Acknowledgments

Part of this study was supported by the Prinses Beatrix Spierfonds (project no. W.OR12-15). We would like to acknowledge Dr Maryam Dastmalchi and Snjolaug Arnardottir for collecting clinical data and blood samples and Eva Jemseby for handling of a serum biobank in the rheumatology research laboratory, Karolinska Institutet.

References

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:

Footnotes

  • Handling editor Tore K Kvien

  • Contributors All authors have substantially contributed to the conception and design of the work, the acquisition and analysis of the data and/or the interpretation of the data. All authors have contributed to drafting and/or revising the manuscript. The final version has been approved by all authors and they all agree to be accountable for all aspects of the work.

  • Funding The UKMYONET adult cohort was funded by Arthritis Research UK Programme Grant 18474. Further funding has been provided through the UK Myositis Support Group, Association Francaise Contre Les Myopathies, The European Union Sixth Framework Programme (project AutoCure; LSH-018661), the European Science Foundation in the framework of the Research Networking Programme European Myositis Network, the Swedish Research Council and The regional agreement on medical training and clinical research (ALF) between Stockholm County Council and Karolinska Institutet. This report includes independent research funded by the National Institute for Health Research Manchester Musculoskeletal Biomedical Research Unit. The views expressed in this publication are those of the author(s) and not necessarily those of the NHS, the National Institute for Health Research or the Department of Health, UK. The funding agencies played no role in study design, collection, analysis or interpretation of the data, preparation of the manuscript or the decision to submit the article for publishing.

  • Competing interests Patent and royalties pending, method of detecting autoantibodies from patients with sporadic inclusion-body myositis.

  • Ethics approval Local ethics committees.

  • Provenance and peer review Not commissioned; externally peer reviewed.