Introduction
Systemic sclerosis (SSc; scleroderma) is a disorder characterized by fibrosis of the skin and visceral organs. The pathogenesis of this disease is complex and remains incompletely understood. Nevertheless, autoantibodies represent a serologic hallmark of the disease and have proven value as diagnostic and prognostic biomarkers. Indeed, up to 95% of SSc patients [
1] have circulating autoantibodies directed against one or more autoantigens, including topoisomerase I (formerly called Scl-70), centromere proteins (CENPs), RNA polymerase III, and the PM/Scl complex, also known as the human exosome [
2]. In SSc, the major disease-related autoantibodies tend to be mutually exclusive [
3], suggesting unique pathways for the induction of the B-cell response in this condition. Evidence supporting this concept is extensive, given data indicating that each autoantibody is associated with specific demographic, clinical, genetic, and prognostic features [
4,
5]. In addition, a growing knowledge of the role of SSc autoantibodies in the pathogenesis of the disease is helping to gain a better understanding of potential novel modes of therapy [
6‐
9].
Two main types of SS-A/Ro antibodies have been described in SSc. One is directed at a 60-kDa protein known as SS-A/Ro60, which is part of a small cytoplasmic ribonucleoprotein (scRNP) multiprotein complex. Another, which often coexists with the former SS-A/Ro60 antibodies, is directed against a 52-kDa (Ro52) protein that is not normally part of the scRNP complex but is an E3 ubiquity ligase and a member of the tripartite motif (TRIM) family of proteins known as TRIM21 [
10]; hence, the preferred terminology of Ro52/TRIM21 is used in this report. The association of Ro60 antibodies with autoimmune conditions is well established, particularly in systemic lupus erythematosus (SLE), subacute cutaneous lupus, and Sjögren syndrome (SjS). Ro52/TRIM21 antibodies have also been reported in a wide variety of autoimmune diseases, although often overlapping with other autoantibodies. However, little is known of their clinical associations, and controversy still exists about whether they have an independent association with autoimmune diseases [
11].
The Canadian Scleroderma Research Group (CSRG) is a pan-Canadian, multicenter group of researchers that has, since 2004, recruited more than 1,200 SSc patients. The exact prevalence of SSc in Canada remains unknown, but estimates range from 70 to 440 cases/million [
12,
13]. Thus, by using the most conservative numbers, the CSRG currently captures up to 8% of all Canadian SSc cases. Detailed demographic, clinical, and serologic data have been obtained on these patients and entered into a central database.
Ro52/TRIM21 antibodies have been identified in 20% of the CSRG cohort, making it the second most common autoantibody in this cohort of SSc patients (Table
1). Given its high prevalence and the paucity of data on its clinical significance, we undertook this study to determine whether Ro52/TRIM21 antibodies in SSc are associated with distinct disease manifestations in SSc.
Table 1
Frequency of selected autoantibodies in the CSRG cohort
CENP-B | 285 | 36% | 162 |
Centromere by IIF | 334 | 35% | 2 |
CENP-A | 275 | 34% | 162 |
Ro52/TRIM21 | 194 | 20% | 0 |
RNA polymerase III | 144 | 19% | 188 |
Topoisomerase I | 150 | 16% | 0 |
PM/Scl | 53 | 7% | 188 |
Ro60 | 53 | 6% | 0 |
U1 RNP | 59 | 6% | 0 |
Chromatin | 44 | 5% | 0 |
Sm | 39 | 4% | 0 |
SS-B/La | 26 | 3% | 0 |
Jo-I | 6 | 1% | 0 |
Ribosomal P | 14 | 1% | 0 |
Results
This study included 963 SSc patients with complete serologic data on anti-Ro52/TRIM21 antibodies. Baseline characteristics of the study cohort are presented in Table
2. Mean age was 54.25 (± 12.04) years; 87% were female; 85% were White; mean disease duration (since the onset of the first non-Raynaud disease manifestation) was 11.04 (± 9.47) years; and 59% had limited, 37% had diffuse, and 3% had no skin involvement.
Table 2
Baseline characteristics of the study cohort, as a group and according to anti-Ro52/TRIM21 antibody status
Age, years | 55.42 | 12.04 | 57.16 | 11.45 | 54.98 | 12.16 | 1.02 (1.00, 1.03) | 0.0481 |
Female, % | 833 | 86.50% | 175 | 90.21% | 658 | 85.57% | NS | |
White, % | 816 | 84.74% | 160 | 82.47% | 656 | 85.31% | NS | |
Disease duration, years | 11.04 | 9.47 | 11.29 | 9.55 | 10.98 | 9.46 | NS | |
Disease subsets, % | | | | | | | NS | |
Limited disease | 563 | 59.26% | 121 | 62.69% | 442 | 58.39% | | |
Diffuse disease | 356 | 37.47% | 69 | 35.75% | 287 | 37.91% | | |
Sine scleroderma | 31 | 3.26% | 3 | 1.55% | 28 | 3.70% | | |
Modified Rodnan skin score (range, 0-51) | 10.02 | 9.47 | 9.68 | 8.53 | 10.11 | 9.69 | NS | |
Swollen-joint count (range, 0-28) | 0.91 | 3.34 | 1.07 | 4.26 | 0.87 | 3.07 | NS | |
Number of GI symptoms (range 0-14) | 4.17 | 3.13 | 4.27 | 3.30 | 4.14 | 3.09 | NS | |
Inflammatory myositis, % | 103 | 11.48% | 21 | 11.54% | 82 | 11.47% | NS | |
Thrombosis, % | 29 | 3.06% | 7 | 3.65% | 22 | 2.91% | NS | |
Scleroderma renal crisis, % | 42 | 4.44% | 7 | 3.65% | 35 | 4.64% | NS | |
Interstitial lung disease, % | 342 | 36.19% | 85 | 44.27% | 257 | 34.13% | 1.53 (1.11, 2.12) | 0.0091 |
Pulmonary hypertension,% | 96 | 11.72% | 26 | 15.38% | 70 | 10.77% | NS | |
Overlap with SLE, % | 30 | 3.19% | 8 | 4.23% | 22 | 2.93% | NS | |
Overlap with Sjögren, % | 63 | 6.70% | 18 | 9.52% | 45 | 5.98% | NS | |
Overlap with RA, % | 30 | 3.19% | 8 | 4.23% | 22 | 2.93% | NS | |
Overlap with MCTD, % | 20 | 2.13% | 7 | 3.70% | 13 | 1.73% | NS | |
Overlap with PM/DM, % | 36 | 3.83% | 12 | 6.35% | 24 | 3.19% | 2.06 (1.01, 4.19) | 0.0431 |
Overlap syndrome, % | 143 | 15.18% | 41 | 21.58% | 102 | 13.56% | 1.75 (1.17, 2.63) | 0.0059 |
Of the 963 SSc patients included in this study, 194 (20%) were positive for anti-Ro52/TRIM21 antibodies. This represents the second most common autoantibody in this cohort with anti-centromere, anti-RNA polymerase III, and anti-topoisomerase I antibodies present in approximately 35%, 19%, and 16% of the patients, respectively (Table
1). Anti-Ro52/TRIM21 antibody-positive patients were more likely to be older (57.16 versus 54.98 years;
P = 0.0481), to have interstitial lung disease (44% versus 34%;
P = 0.0091), and to have overlap syndrome (22% versus 14%;
P = 0.0059) compared with anti-Ro52/TRIM21 antibody-negative patients (Table
2). In particular, anti-Ro52/TRIM21 antibody-positive patients were significantly more likely to have overlap with polymyositis/dermatomyositis (6% versus 3%;
P = 0.0431) compared with anti-Ro52/TRIM21 antibody-negative patients. Also of note, two of 36 patients with polymyositis/dermatomyositis overlap had anti-Jo-1 antibodies (6%), and both of these had anti-Ro52/TRIM21 antibodies.
Disease severity measured by using the Medsger Disease Severity Scale was worse in two domains (range, 0 to 4): general (1.16 versus 0.82;
P = 0.0061) and lung (1.56 versus 1.34;
P = 0.0084) in anti-Ro52/TRIM21 antibody-positive compared with anti-Ro52/TRIM21 antibody-negative patients (Table
3).
Table 3
Association between Anti-Ro52/TRIM21 antibodies and disease severity (range, 0 to 4)
General | 0.89 | 1.22 | 1.16 | 1.41 | 0.82 | 1.16 | 1.23 (1.09, 1.39) | 0.0061 |
Skin | 1.25 | 0.68 | 1.23 | 0.60 | 1.25 | 0.69 | NS | |
Kidney | 0.12 | 0.61 | 0.12 | 0.61 | 0.12 | 0.61 | NS | |
Peripheral vascular | 1.54 | 1.25 | 1.58 | 1.26 | 1.53 | 1.24 | NS | |
Joint/tendon | 0.73 | 1.19 | 0.73 | 1.22 | 0.73 | 1.18 | NS | |
Muscle | 0.25 | 0.74 | 0.33 | 0.89 | 0.22 | 0.70 | NS | |
Gastrointestinal | 1.95 | 0.82 | 2.01 | 0.78 | 1.93 | 0.82 | NS | |
Lung | 1.38 | 1.11 | 1.56 | 1.08 | 1.34 | 1.12 | 1.19 (1.04, 1.37) | 0.0084 |
Heart | 0.50 | 1.00 | 0.60 | 1.12 | 0.48 | 0.97 | NS | |
Physician global assessments of severity (range, 0-10) | 2.77 | 2.26 | 2.99 | 2.31 | 2.71 | 2.24 | NS | |
Patient global assessments of severity (range, 0-10) | 3.64 | 2.61 | 3.66 | 2.67 | 3.63 | 2.60 | NS | |
Antibodies to Ro52/TRIM21 overlapped with many other autoantibodies (Table
4). However, the association between anti-Ro52/TRIM21 antibodies and other autoantibodies differed according to antibody. Indeed, whereas 16% of the overall cohort had anti-topoisomerase I antibodies, only 10% of anti-Ro52/TRIM21-positive patients had anti-topoisomerase I antibodies (
P = 0.0129), and the titers of anti-Ro52/TRIM21 antibodies were among the lowest measured (mean titer, 763.28 U/ml) in this subset of patients. Conversely, whereas 6% of the overall cohort had anti-Ro60 antibodies, 21% of anti-Ro52/TRIM21 antibody-positive patients were anti-Ro60 positive (
P < 0.0001), and the titers of anti-Ro52/TRIM21 antibodies were among the highest measured (mean titer, 5,530.66 U/ml). Similarly strong associations between anti-Ro52/TRIM21 antibodies and anti-Ro60 were noted between anti-Ro52/TRIM21 and anti-U1 RNP (
P = 0.0066; mean titer, 2,099.45 U/ml), anti-SS-B/La (
P < 0.0001; mean titer, 6,003.88 U/ml), and anti-Jo-1 (
P = 0.0171; mean titer, 4,412.83 U/ml), and to a lesser degree between anti-Ro52/TRIM21 and anti-CENP-B (
P = 0.0372; mean titer, 1,098.86 U/ml) and anti-Sm (
P = 0.0361; mean titer, 1,482.09 U/ml) antibodies.
Table 4
Association between anti-Ro52/TRIM21 antibodies and other SSc-related autoantibodies
Centromere
| 334 | 34.76% | 78 | 40.41% | 256 | 33.33% | NS | 1,098.64 |
Topoisomerase I
| 150 | 15.58% | 19 | 9.79% | 131 | 17.04% | 0.0129 | 763.28 |
RNApolIII
| 144 | 18.58% | 23 | 15.75% | 121 | 19.24% | NS | 1,164.47 |
Pm/Scl
| 53 | 6.84% | 10 | 6.85% | 43 | 6.84% | NS | 1,381.13 |
U1 RNP
| 59 | 6.13% | 20 | 10.31% | 39 | 5.07% | 0.0066 | 2,099.45 |
CENP-B
| 285 | 35.58% | 64 | 42.95% | 221 | 33.90% | 0.0372 | 1,098.86 |
CENP-A
| 275 | 34.33% | 61 | 40.94% | 214 | 32.82% | NS | 1,019.93 |
Ro60
| 53 | 5.50% | 40 | 20.62% | 13 | 1.69% | < 0.0001 | 5,530.66 |
SS-B/La
| 26 | 2.70% | 18 | 9.28% | 8 | 1.04% | < 0.0001 | 6,003.88 |
Jo-1
| 6 | 0.62% | 4 | 2.06% | 2 | 0.26% | 0.0171 | 4,412.83 |
Sm
| 39 | 4.05% | 13 | 6.70% | 26 | 3.38% | 0.0361 | 1,482.09 |
Ribosomal P
| 14 | 1.45% | 3 | 1.55% | 11 | 1.43% | NS | 2,088.14 |
Chromatin
| 44 | 4.57% | 10 | 5.15% | 34 | 4.42% | NS | 1,162.30 |
Finally, anti-Ro52/TRIM21 antibody titers were significantly higher in patients with compared to patients without overlap syndrome (mean titer, 1,330.40 versus 1,105.05 U/ml, respectively; P = 0.0026). A strong trend was noted toward higher titers in patients with compared to those without interstitial lung disease (mean titer, 1,542.65 versus 938.11 U/ml, respectively; P = 0.06).
Discussion
In this large, multicenter cohort study, we found that anti-Ro52/TRIM21 antibodies were present in 20% of 963 patients, making them the second most common autoantibodies in this SSc cohort, and they overlapped with all of the major SSc-related antibodies. In addition, in this cohort of SSc patients, we found strong suggestions of an association between anti-Ro52/TRIM21 antibodies and interstitial lung disease and overlap syndrome.
Several studies have demonstrated that anti-Ro52/TRIM21 antibodies are present in several systemic autoimmune rheumatic diseases [
11,
28‐
31]. In particular, anti-Ro52 antibodies are frequently found in the inflammatory myositides, often in the presence of anti-Jo1 and interstitial lung disease [
32‐
34].
The largest study to date of anti-Ro52 in SSc (measured using an ELISA), a recent British report of 1,010 SSc patients, found an overall frequency of anti-Ro52/TRIM21 antibodies of 27% [
35]. We have shown that the ALBIA anti-Ro52/TRIM21 antibody has high concordance (> 95%) with an anti-Ro52/TRIM21 ELISA (unpublished data), suggesting that the higher prevalence in the British study is not related to interlaboratory assay variation. Although the majority of our patients were White (85%), the British study did not report on the ethnic profile or disease duration of their patients. The British study also showed that anti-Ro52/TRIM21 antibodies overlapped with all of the major SSc-related antibodies, including anti-centromere, anti-topoisomerase I, and anti-RNA polymerase III antibodies (in their study, at frequencies of 28%, 19%, and 25%, respectively). In addition, they confirmed that anti-Ro52/TRIM21 antibody titers were particularly elevated in patients with anti-Ro60 and anti-aminoacyl-tRNA synthetase antibodies (of whom four were anti-Jo1 positive). Although the authors did not present actual clinical data, they nevertheless concluded that Ro52/TRIM21 antibodies appeared to be general serum markers with limited linkage to distinct clinical manifestations of SSc.
In another recent study by Ghillani
et al.
[
36], 155 patients positive for anti-Ro52/TRIM21 antibodies but negative for anti-Ro60 antibodies were analyzed. The authors found that anti-Ro52/TRIM21 antibodies were commonly found in the presence of other autoantibodies (in particular, Sm, chromatin, Jo-1, and CENP-B) and that 73% of patients had an autoimmune disease (in particular, polymyositis/dermatomyositis, SjS, SSc, and autoimmune hepatitis). Interestingly, the authors reported that the prevalence of pulmonary disease was particularly high among anti-Ro52/TRIM21 antibody-positive patients (22% of patients), including seven of 11 patients with SSc.
A recent article from the German Network for Systemic Sclerosis specifically examined the clinical correlates of a number of autoantibodies in SSc, including Ro52 [
37]. They found that the prevalence anti-Ro52 antibody in their cohort (
N = 863) was 21.7% and that 25% of patients with overlap syndrome were positive for anti-Ro52 antibody. Although they did not find any statistically significant association between anti-Ro52 antibodies and specific clinical features, it is interesting to note that their point estimate for an association with pulmonary fibrosis was 1.3. They also reported a statistically significant twofold increase in the rate of pulmonary fibrosis among patients who had anti-Ro60 antibodies (odds ratio (OR), 2.20; 95% confidence interval (CI), 1.29 to 3.75;
P = 0.004). In our study, the rate of interstitial lung disease among anti-Ro52/TRIM21 antibody-positive patients was significantly increased by 50% compared with anti-Ro52/TRIM21 antibody-negative patients (OR, 1.53; 95% CI, 1.11 to 2.12;
P = 0.0091). In addition, the association between anti-Ro60 antibody and interstitial lung disease was increased almost threefold (OR, 2.86; 95% CI, 1.62 to 5.04;
P = 0.0002; data not shown).
Thus, the findings of our study are generally consistent with the current literature. In addition to confirming the serologic associations between anti-Ro52/TRIM21 antibodies and other autoantibodies, we found important associations between anti-Ro52/TRIM21 antibodies, interstitial lung disease, and overlap syndrome.
Overlap syndrome is highly prevalent among patients with systemic autoimmune rheumatic diseases and is thought to suggest that these diseases share similar immunogenetic mechanisms [
38‐
40]. The concept of "overlap" syndrome in SSc, however, remains elusive, with some of the common manifestations of the disease, including sicca and myositis, being possibly related to SSc itself or to a separate autoimmune disease. We believe that serologic associations could help to shape a better definition of overlap in SSc, and that anti-Ro52/TRIM21 antibodies have the potential to be good biomarkers of overlap in this disease.
The Ro52 gene has been mapped to the end of the short arm of human chromosome 11 [
29]. Evidence suggests that this chromosome segment may harbor genes important in the development and progression of solid tumors. Given this, we reviewed our data and found that 12.04% of anti-Ro52/TRIM21 antibody-positive patients compared with 7.24% of anti-Ro52/TRIM21 antibody-negative patients had a physician-reported history of malignancy (
P = 0.03). These data have yet to be validated through chart review and remain preliminary.
Our understanding of how a number of SSc-related autoantibodies are generated (for example, through apoptosis, microbody release, molecular mimicry, or other mechanisms) and of their role in the pathogenesis of disease has increased remarkably over the past two decades (reviewed in [
41,
42]). The recently reported mechanism of intracellular immunity mediated by Ro52/TRIM21 in a cellular model of adenovirus infection has opened new perspectives for studying the effects of autoantibodies once they are inside cells [
10]. The association between Ro52/TRIM21 and other proteins such as TRIB2, implicated in driving tumorigenesis [
43], with Fas-associated death domain (FADD) and negatively regulating the IFN-α pathway in response to viral infection [
44] and its interaction with virus-like particles and high-affinity IgG receptors affecting the intracellular fate of viruses, including degradation by the proteasomes [
10,
45,
46] may provide important insights into the generation of the autoantibody response and its perpetuation in SSc. Further, the association of anti-Ro52/TRIM21 antibodies with cytokine responses such as IFN-α [
44] and Toll-like receptor-mediated NF-κB activation [
47] provides some insight into its regulation of cytokine responses. Thus, despite clearer understanding of the physiological function of Ro52/TRIM21
in vitro, it is still unclear how anti-Ro52/TRIM21 antibodies arise, particularly
in vivo, and whether they have any pathologic significance in SSc or other systemic autoimmune rheumatic diseases. Of note, though, is the fact that anti-Ro52/TRIM21 antibodies are seen across a number of autoimmune diseases with significant prevalence (ranging from 1% to 63%) but are only rarely seen in normal controls; this suggests that they are markers of autoimmune dysfunction [
29]. Further studies will be needed to shed greater insight into their precise role in autoimmune diseases.
Acknowledgements
We thank the Investigators of the Canadian Scleroderma Research Group: M. Baron, Montreal, Quebec; J. Pope, London, Ontario; J. Markland, Saskatoon, Saskatchewan; D. Robinson, Winnipeg, Manitoba; N. Jones, Edmonton, Alberta; N. Khalidi, Hamilton, Ontario; P. Docherty, Moncton, New Brunswick; E. Kaminska, Hamilton, Ontario; A. Masetto, Sherbrooke, Quebec; E. Sutton, Halifax, Nova Scotia; J-P. Mathieu, Montreal, Quebec; M. Hudson, Montreal, Quebec; S. LeClercq, Calgary, Alberta; S. Ligier, Montreal, Quebec; T. Grodzicky, Montreal, Quebec; C. Thorne, Newmarket, Ontario; M. Fritzler, Advanced Diagnostics Laboratory, Calgary, Alberta.
This study was funded in part by the Canadian Institutes of Health Research, the Scleroderma Society of Canada, and educational grants from Actelion Pharmaceuticals and Pfizer Inc. Autoantibody diagnostic kits were a gift of INOVA Diagnostics Inc. (San Diego, CA, USA), ImmunoConcepts (Sacramento, CA, USA), and Dr. Fooke Laboratorien GmbH (Neuss, Germany). Dr. Hudson is a New Investigator funded by the Canadian Institutes of Health Research and the Fonds de la Recherche en Santé du Québec, and Dr. Fritzler holds the Arthritis Society Research Chair at the University of Calgary. The funding sources had no role in the design of the study, analysis of the data, preparation of the manuscript, and decision to submit for publication.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
MH participated in the design of the study, the data collection, and the statistical analysis, and prepared the manuscript. JP participated in the design of the study, the data collection, the statistical analysis, and the interpretation of the data. MM participated in the design of the study and the interpretation of the data. ST and RS performed the statistical analysis. MB and MF participated in the design of the study, the data collection, and the interpretation of the data. All co-authors read and approved the final manuscript. The Investigators of the Canadian Scleroderma Research Group participated in the acquisition of the data and read and approved the final manuscript.