Anti-MJ/NXP-2 autoantibody specificity in a cohort of adult Italian patients with polymyositis/dermatomyositis

Fifty-eight consecutive adult Italian patients with PM/DM, who visited the Rheumatology Unit of Spedali Civili (Brescia, Italy) from 2009 to 2011, were studied. The diagnosis of PM/DM was based on Bohan and Peter criteria [13]. Clinical information was obtained from medical records. The study was approved by the Institutional Review Board of the hospital. This study meets, and is in compliance with, all ethical standards of medicine, and informed consent was obtained from all patients in accordance with the Declaration of Helsinki.

Autoantibodies in sera were screened by IP using ³⁵S-methionine-labeled K562 cell extract [14]. Specificities of autoantibodies were determined using reference sera.

Anti-MJ antibodies were also tested by antigen-capture ELISA [15]. Briefly, a 96-well plate Immobilizer Amino (Nalge Nunc International, Rochester, NY, USA) was incubated with 50 μl/well of 2 μg/ml mouse monoclonal antibody (mAb) to MORC3 (MBL International, Woburn, MA, USA). Wells were then incubated with K562 cell extract, followed by 1:500 diluted sera, and probed with alkaline phosphatase conjugated mouse mAb anti-human Immunoglobulin G (1:2000 dilution) (Sigma, St. Louis, MO, USA). Anti-Ro52, -La [16, 17], and -Jo-1 (Abazyme, Needham, MA, USA) antibodies were tested by ELISA using recombinant protein at 1:500 serum dilution. Optical density (OD) of the samples was converted into units using a standard curve created by a prototype positive serum [15].

Candidates for anti-MJ were selected based on immunoprecipitation of a 140 kDa protein that comigrates with MORC3 protein recognized by mAb. Identity of the 140 kDa proteins corresponding to MJ/MORC3/NXP-2 was verified by IP-Western [14]. Cell extract from 10⁷ K562 cells was immunoprecipitated by candidate sera. Proteins were fractionated by 8% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to a nitrocellulose filter. The filter was probed with 0.4 μg/ml of anti- MORC3 mouse mAb, followed by horseradish peroxidase (HRP) goat anti-mouse IgG (1: 7,500 dilution) (Southern Biotechnology, Birmingham, AL, USA) and developed using SuperSignal West Femto chemiluminescent substrate (Thermo Scientific, Barrington, IL, USA).

Immunofluorescent antinuclear/cytoplasmic antibodies (HEp-2 antinuclear antibody (ANA) slides) (INOVA Diagnostics, San Diego, CA, USA) were tested using a 1:80 dilution of human sera, or 4 μg/ml of anti-MORC3 mAb followed by DyLight 488 donkey IgG F(ab)'₂ anti-human IgG or goat anti-mouse IgG (1:100 dilution) (Jackson ImmunoResearch, West Grove, PA, USA). Double staining was also performed using rabbit anti-PML antiserum (1:300 dilution) [18] and Alexa 568 goat anti-rabbit IgG (Invitrogen, Carlsbad, CA, USA).

Fifty-eight patients (43 females, 15 males) with a diagnosis of PM (25 cases), DM (27 cases) or overlap syndrome (6 cases) were studied. In the whole cohort, the patients' age was 52 ± 16 (mean ± standard deviation (SD)) years, the age at disease onset was 43 ± 17.4 years, and the mean follow-up was 56 months (range 1 to 288 months).

Prevalence of myositis-related autoantibodies by IP in total PM/DM, DM, and PM is summarized (Table 1). To our surprise, anti-MJ antibodies were the most prevalent specificity (10/58; 17%) in our PM/DM patients, followed by anti-Jo-1 (10%), -PM-Scl (10%), -U1RNP (7%), -p155/140 (5%), -SRP (5%), -EJ (3%), -Mi-2 (2%), and -OJ (2%). Anti-Ro and anti-Su were found in two cases (3%) each. In 22 patients (38%), no myositis-related autoantibodies were detected. In DM, no patient had anti-Jo-1 antibodies, while in PM it was the most common autoantibody specificity (0% vs 24%; P = 0.0087). Anti-MJ was found only in two cases in PM vs 30% in DM (P = 0.078). Thus, anti-MJ was common in DM, whereas anti-Jo-1 was found only in PM in our cohort. Candidates positive for anti-MJ were initially selected based on the IP of ~140 kD proteins compared with molecular weight markers and the mobility of other known autoantigens (Figure 1A). One of the known myositis autoantigen CADM140/MDA5, which migrates close to MJ, is not detectable in our IP system using K562 cells (Figure 1A, lane MDA5). IP of candidates were then run in SDS-PAGE along with anti-MORC3 mAb IP. If the mobility of a ~140 kD protein immunoprecipitated by the serum is considered same as that of the mAb to MORC3, sera were then tested by IP-Western to verify their identity as MJ. The 10 anti-MJ (+) were confirmed by 1) identical mobility of the 140 kDa protein with MORC3 recognized by mAb (Figure 1B); 2) IP- Western Blot (Figure 1C); and 3) positive results in antigen-capture ELISA (data not shown). None of anti-MJ (+) patients had other MSAs, consistent with previous observations [11], though the study in the Argentinian pediatric cohort reported 2/18 anti-MJ (+) cases with an additional autoantibody specificity [6]. NXP-2/MORC3 localizes to nucleus and nuclear dots, known as PML bodies. Number and intensity of PML bodies stained by mAb appear to vary depending on cell type, cell cycle and other factors [10]. In HEp-2 ANA slides, mAb to MORC3 stained nuclei in a fine speckled pattern with a few nuclear dots (Figure 1D, panel a). Six (three strong, three weak) of ten anti-MJ (+) sera showed PML bodies nuclear dots in indirect immunofluorescence, (Figure 1D, panels b to d), but PML bodies staining was not clear by other sera (Figure 1D, panel e), indicating that PML bodies immunofluorescence cannot be a good method for screening or confirmation of anti-MJ specificity.

Since anti-MJ antibodies were the most common in PM/DM (17%) or in DM (30%), clinical features of 10 anti-MJ (+) vs 48 anti-MJ (-) cases were compared (Table 2). Anti-MJ (+) patients had younger age of disease onset (25.5 ± 13.8 vs 46.1 ± 15.9 years, mean ± SD, P = 0.0006 by Mann-Whitney) and age at initial visit (37.6 ± 12.0 vs 54.6 ± 14.8 years, P = 0.0023 by Mann-Whitney) compared with anti-MJ (-). Two anti-MJ (+) patients had pediatric onset DM. DM is more common in anti-MJ (+) vs (-) (80% vs 40%, P = 0.03) and no overlap syndrome patients were found in the anti-MJ group (0% vs 13%). In anti-MJ (+) patients, heliotrope rash (P = 0.01) and calcinosis (P = 0.09) were common, but none of them had heart involvement (0% vs 13%), interstitial lung disease (ILD) (0% vs 33%, P = 0.048), or malignancy (0% vs 8%). Myopathy in anti-MJ (+) patients responded well to therapy and elevated creatine phosphokinase (CPK) was not seen in any patient in the last visit (0% vs 25%). Thus, anti-MJ antibodies are associated with DM of young onset with heliotrope rash and calcinosis, without cardiac or lung involvement or overlapping features, and they respond well to therapy. Other clinical aspects, such as arthritis and Raynaud's phenomenon, were not significantly associated with anti-MJ antibodies (Table 2). Clinical characteristics of anti-MJ (+) DM compared to anti-MJ (-) DM were similar to those shown in Table 2, however, they did not reach statistical significance due to small numbers; heliotrope, 75% vs 47%; calcinosis 38% vs 21%; arthritis 25% vs 5%; heart involvement 0% vs 16%; and ILD 0% vs 32%.