Key points

Introduction

Sjögren’s syndrome is characterized by the presence of a plethora of autoantibodies ( Table 1 ). We herein review major recent development in autoantibodies associated with Sjögren syndrome. This includes description of new antibodies found in the sera of patients with the disease as well as clinical associations and new insights into the process by which autoantibodies are produced and are involved in pathogenicity. Animal models of Sjögren syndrome, including autoantibodies, have been reviewed exhaustively in the recent past, and are not considered herein.

Introduction

Sjögren’s syndrome is characterized by the presence of a plethora of autoantibodies ( Table 1 ). We herein review major recent development in autoantibodies associated with Sjögren syndrome. This includes description of new antibodies found in the sera of patients with the disease as well as clinical associations and new insights into the process by which autoantibodies are produced and are involved in pathogenicity. Animal models of Sjögren syndrome, including autoantibodies, have been reviewed exhaustively in the recent past, and are not considered herein.

Autoantibodies precede disease

Almost all autoimmune diseases are associated with circulating autoantibodies directed against self-protein. Interestingly many of these disease antibodies are detected years before the clinical manifestations of the disease. Passively acquired autoimmunity substantiated in neonates born to asymptomatic anti-Ro/SSA antibody-positive mothers is one such demonstration. The various manifestations of neonatal lupus syndrome provide a remarkable possibility to explore disease evolution because asymptomatic mothers are brought to medical attention solely on identification of heart block or rash in a neonate. Almost half of these asymptomatic mothers ultimately progress to develop some autoimmune ailment with a higher probability toward Sjögren syndrome.

Theander and colleagues recently performed a nested case control study linking data from the Malmo primary Sjögren registry and 3 Swedish health care biobanks. Serum samples were analyzed for antinuclear antibody (ANA), rheumatoid factor (RF), and antibodies against Ro60, Ro52, and La/SSB. Of 117 patients with Sjögren syndrome in the registry who had predisease sera samples available in the biobanks, 81% had autoantibodies before symptom onset. Furthermore, many of them had antibodies present in the very earliest available serum sample. Thus, antibodies appeared at least 18 to 20 years before the diagnosis was made. The positive predictive value for anti-Ro60 and anti-Ro52 was high for ultimate development of the disease.

Interestingly, these results are consistent with similar studies carried out for systemic lupus erythematosus (SLE) and rheumatoid arthritis (RA). In particular, in a study of autoantibodies preceding the diagnosis of SLE in 130 patients diagnosed while in the US Armed Forces, anti-Ro/SSA was among the earliest of antiantibodies to appear with lupus-specific antibodies found closer to the onset of the disease. Congruently, patients with RA have antibodies present years before the diagnosis. Similar data have been generated for other autoimmune diseases, including type 1 diabetes mellitus and primary biliary cirrhosis. Sjögren syndrome, then, falls into a large group of autoimmune diseases in which antibodies precede disease, but the risk of disease in a given antibody-positive individual is not well characterized, except in type 1 diabetes.

Anti-Ro/Sjögren’s Syndrome A (SSA) and anti-La/Sjögren’s syndrome B (SSB)

The most common ANAs in patients with Sjögren syndrome are those directed against the autoantigens Ro/SSA and La/SSB. The Ro/La particle is a protein-RNA complex formed by the association of the Ro60, and La/SSB proteins with small cytoplasmic RNA (hyRNA). Various methods have been used for the detection of anti-Ro/SSA and anti-La/SSB antibodies. RNA precipitation is considered to be the gold standard method, but other techniques, such as counter-immunoelectrophoresis, immunodiffusion, and enzyme linked immunosorbent assay (ELISA) are more frequently used. It is well to remember that most of the clinical association data for these antibodies in patients with Sjögren syndrome were performed with older assays, such as immunodiffusion, whereas most commercially available assays today are high-throughput, easily automated assays such as ELISA or multiplex bead. These newer assays generally have higher sensitivity and lower specificity; and, thus, associations found with higher specificity, lower sensitivity older assays may not hold true. One study found that the presence of peripheral neuropathy is related to anti-Ro/SSA and anti-La/SSB when determined by precipitation in immunodiffusion but not when determined by ELISA or multiplex bead assay.

Anti-Ro/SSA and anti-La/SSB antibodies are detected in 50% to 70% of patients with primary Sjögren syndrome, depending on the method applied. Anti-Ro/SSA is independent of anti-La/SSB antibody but the contrary is rare, with a recent analysis of the Sjögren’s Syndrome International Clinical Alliance cohort concluding that individuals (n = 74) with anti-La/SSB but no anti-Ro/SSA did not have the disease. Data from the Oklahoma Sjögren syndrome cohort support this conclusion (Danda, Kurien, Scofield, manuscript submitted). On the other hand, anti-La/SSB in the presence of anti-Ro/SSA tends to identify patients with Sjögren syndrome. Venables and colleagues found 29 of 35 patients with both anti-Ro/SSA and anti-La/SSB had Sjögren syndrome, whereas among 53 with only anti-Ro/SSA, 23 had Sjögren syndrome, 25 had SLE, and 13 had another disease.

Anti-Ro/SSA and anti-La/SSB antibodies are correlated with younger age at diagnosis, longer disease duration, more severe dysfunction of the exocrine glands, recurrent parotid gland enlargement, and higher intensity of the lymphocytic infiltrates in the minor salivary glands. Large studies demonstrate a higher prevalence of extraglandular manifestations in patients with primary Sjögren syndrome (pSS), including splenomegaly, lymphadenopathy, vasculitis, and Raynaud phenomenon. Sicca-limited disease was found in 292 (29%) of 1010 patients and was associated with an absence of anti-Ro/SSA. This same study found that patients with anti-Ro/La had a lower age at diagnosis, and were statistically more likely to have a host of manifestations. These included parotid enlargement, Raynaud phenomenon, arthritis, vasculitis, renal tubular acidosis, peripheral neuropathy, cytopenias, and RF. A decade long-term study of 100 patients found that only patients with anti-Ro/SSA developed systemic, extraglandular complications.

Patients with Sjögren syndrome are at a marked increased risk for lymphoma, especially mucosal-associated lymphoid tissue (MALT) lymphoma. Despite the increase in incidence, lymphoma is an uncommon event among patients with Sjögren syndrome such that most studies contain a small number of patients. The data concerning the risk of lymphoma in relationship to anti-Ro/SSA and anti-La/SSB are mixed. Studying patients in the north of England with 10-year follow-up, Davidson and colleagues found patients positive for both these antibodies were at high risk for non-Hodgkin lymphoma. Meanwhile, another long-term study with an average of 7 years of follow-up found no difference in serologic testing among the 6 patients who developed lymphoma compared with the 74 who did not. A recent comprehensive review concluded that anti-Ro/La was not a risk factor for lymphoma.

Pulmonary involvement is another complication that may not be associated with the presence of anti-Ro/SSA. A total of 507 patients with Sjögren syndrome underwent chest computed tomography (CT) scan and 50 had bronchiectasis. Only 27% of those with this pulmonary manifestation had anti-Ro/SSA, compared with 54% of those without bronchiectasis. Thus, this serious complication of Sjögren syndrome, unlike most other extraglandular manifestations, is not associated with anti-Ro/SSA. However, this was a study in which all patients underwent a CT scan, and associations may be different when considering patients diagnosed clinically with lung involvement.

The latest biomedical technology continues to be applied to the autoantigenic response to the Ro/SSA and La/SSB proteins. For example, anti-Ro/SSA is strongly associated with increased expression of interferon-regulated genes in peripheral blood mononuclear cells of patients with Sjögren syndrome. Proteomic-based studies to determine V region structures of autoantibodies directed against Ro52 or La/SSB concluded that a set of public clonotypes are used to produce these autoantibodies. This same Australian group has conducted proteomic studies of anti-Ro60 that challenge the long-held assumption that these antibodies are produced by long-lived plasma cells. Anti-Ro60 clonotypes in the peripheral blood were determined in 4 patients with Sjögren syndrome over a 7-year period. Immunoglobulin variable gene analysis showed clonotype turnover at approximately 3-month intervals despite long-term high-titer anti-Ro60. Thus, anti-Ro60 was produced by short-lived B cells with rapid turnover. Maier-Moore and colleagues produced recombinant monoclonal antibody from antibody-producing B cells infiltrating the salivary glands of patients with Sjögren syndrome. These studies showed the autoantibody repertoire of the monoclonal antibodies produced from a given subject reflected the antibodies found in the peripheral blood. Thus, this study confirms that the salivary glands are a site for the production of autoantibodies in Sjögren syndrome. The peripheral proteomic study and the salivary gland monoclonal antibody study have not determined the extent to which salivary autoantibody production contributes to circulating anti-Ro60, however.

Anti-Ro/SSA in neonatal lupus

Isolated congenital heart block (CHB) in concert with neonatal lupus dermatitis, hepatitis, and hematologic abnormalities are clinical manifestations of passively acquired autoimmune injury in a neonate; namely, the neonatal lupus syndrome. Third-degree heart block or complete heart block presents a potentially fatal outcome in comparison with other manifestations, which resolve as the maternal antibodies are cleared from the circulatory system of the infant.

Autoimmune CHB occurs in 1% to 2% of anti-Ro/SSA antibody-positive pregnancies, develops in absence of cardiac structural abnormality, and has a recurrence rate of 12% to 20% in subsequent pregnancies. CHB is usually diagnosed between weeks 18 and 24 of gestation by fetal echocardiography. The association of CHB with maternal autoantibodies to the Ro/SSA autoantigen, which comprises the 2 unrelated proteins, Ro52 and Ro60, is well appreciated. A crucial point left unexplained is the low penetrance and recurrence rate of the disease in children of anti-Ro/SSA–positive women, despite persistent maternal antibodies.

The molecular mechanisms underlying CHB pathogenesis are not fully understood. Recently efforts are being invested in delineating a more refined molecular mechanism. CHB is characterized by the presence of immune complex deposits, calcification, and fibrosis at the atrioventricular node in the fetal heart. It has to be discerned whether maternal antibodies exert their pathogenic effect by binding their cognate antigen in the heart or whether they cross-react with another molecule on the surface of fetal cardiac cells to directly affect the electrophysiology of the developing heart. Still many reasons have been placed forward to outline a tentative mechanism contributing toward this serious outcome of neonatal lupus.

Ro52 and Ro60 are intracellular proteins. Ro52, which is also known as TRIM21, is a ubiquitin E3 ligase, involved in the regulation of interferon regulatory factor–mediated immune responses, mainly expressed in immune cells. Ro60 contributes to RNA quality control. Furthermore, Ro60/TROVE2 promotes cell survival after ultraviolet (UV) irradiation, possibly by assisting in the decay of UV-induced damaged RNA. The mechanism suggested for anti-Ro/SSA antibody binding to its cognate antigen relates to the relocation of Ro60 antigen to the cell surface during apoptosis. This observation led to the apoptosis-inflammation hypothesis, which postulates that maternal anti-Ro/SSA antibodies bind to apoptotic cardiac cells during the normal remodeling in the developing heart. This leads to diversion from a noninflammatory to an inflammatory pathway. In vitro studies have demonstrated that opsonized apoptotic cardiocytes can eventually activate phagocytic cells to produce proinflammatory and profibrotic cytokines that ultimately result in atrioventricular (AV) node scarring. Translocation of the Ro/SSA and La/SSB antigens to the cell surface in the salivary gland during apoptosis also has been invoked as a mechanism of Sjögren syndrome in general, which involves activation of epithelial cells.

Anti-Ro52 antibodies induce AV block in several animal models of CHB. These antibodies have been found in the vast majority of mothers giving birth to children with CHB. Boutjdir and colleagues in the late 1990s demonstrated a direct effect of maternal autoantibodies on the heart conduction system. By perfusing rat hearts with isolated fractions of immunoglobulin (Ig)G from the maternal serum, and dissecting atrial and AV nodal areas of rat heart, this study demonstrated bradycardia with AV nodal block. Most recently, the fine specificity of the anti-Ro52 response has been shown to correlate with complete CHB by Salomonsson and colleagues. By study of maternal antibodies directed to a specific epitope within the leucine zipper amino acid sequence 200 to 239 (p200) of the Ro52 protein, this group showed prolongation of fetal AV nodal time and AV block, while antibodies targeting other domains of Ro52 did not lead to such changes. The mechanism delineated by these experiments showed anti-Ro/SSA binding cell surface type E calcium channels with resultant effect on Ca ²⁺ oscillations, leading to accumulating levels and overload of intracellular Ca ²⁺ levels with subsequent loss of contractility and ultimately apoptosis. But, the role of anti-Ro52 antibodies as the sole drivers in the pathogenesis of CHB has yet to be proven.

CHB recurrence rates are at most 20% in a subsequent pregnancy, indicating that factors beyond maternal antibody profile are involved in CHB. Reed and colleagues demonstrated that β-2 glycoprotein 1 (β2GP1) prevented opsonization of apoptotic cardiomyocytes by maternal anti-Ro60 IgG. This can be considered in relation to the “apoptotic” theory of cardiac neonatal lupus, in which the formation of pathogenic antibody-apoptotic cell immune complexes promotes proinflammatory and profibrotic responses. Briassouli and colleagues demonstrated that transforming growth factor (TGF)-beta is triggered by immune reactions leading to amplification of TGF inducing a cascade of events that promotes myofibroblast transdifferentiation and scar formation. Finally, genetic association has been demonstrated for CHB.

Rheumatoid factor

RF is commonly found in the sera of patients with Sjögren syndrome and is associated with serologic positivity for anti-Ro/SSA and anti-La/SSB as well as systemic disease. Recent reports continue to support these associations. In a study of 212 patients with Sjögren syndrome, only anti-Ro/SSA, anti-La/SSB, hypergammaglobulinemia, and RF were associated with systemic disease and use of corticosteroid. Similarly, a large Italian study found RF was one of only a few markers for severe disease. The relationship of RF to systemic disease is found in all racial and ethnic groups studied. One small study found excess pulmonary disease among patients with RF, whereas IgA RF was associated with renal disease in another cohort. The presence of RF was a marker of more severe exocrine gland manifestations (keratoconjunctivitis sicca) among 121 patients with Sjögren syndrome followed for at least 1 year. Thus, RF is a prognostic finding in Sjögren syndrome but may not be useful for clinical diagnosis or research classification purposes because this antibody is found commonly in other diseases.

Anti-muscarinic receptor autoantibodies

Salivary flow is a result of neural stimulation of the acinar and ductal cells of the glands, specifically in response to muscarinic/cholinergic receptor agonists. The presence of functional autoantibodies against glandular M3 muscarinic acetylcholine receptors (M3R) has been reported in pSS. However, the pathogenic role of these autoantibodies in Sjögren syndrome development remains to be unraveled. Robinson and colleagues in 1998 transferred immunoglobulin from anti-Ro/SSA–positive patients with Sjögren syndrome into mice that lacked native immunoglobulin. This study showed reduction of salivary flow in the mice receiving patient immunoglobulin. Their study demonstrated antibody that affected exocrine gland function and that this affect potentially was mediated by binding muscarinic receptors. Hence, this study supports the concept that antibodies directed against autonomic nervous system receptors play a central role in the clinical manifestation of Sjögren syndrome.

Borda and colleagues have been working over many years to elucidate the pathogenic mechanisms associated with anti-M3R autoantibodies contributing to the clinical manifestation of Sjögren syndrome. These investigators demonstrated that primary patients with Sjögren syndrome produce functional IgG autoantibodies that interact with M3R. IgG from patients has 2 effects on the submandibular gland. The antibodies can act as an inducer of the proinflammatory molecule prostaglandin E ₂ (PGE ₂₎ , which in turn inhibits Na ⁺ /K ⁺ -ATPase activity. Antibodies may also have a role in the pathogenesis of dry mouth by Na ⁺ /K ⁺ -ATPase inhibition and the net K ⁺ efflux stimulation of the salivary gland in response to the authentic agonist pilocarpine, decreasing salivary fluid production. The same group demonstrated serum IgG from patients with Sjögren syndrome, interact with the second extracellular loop of human glandular M3R, triggering the production of matrix metalloproteinase-3 (MMP-3) and prostaglandin E ₂ . Thus, Borda and colleagues propose that PGE2 and MMP-3 are generated by autoantibody activation of COX-2. Hence, an imbalance in the expression and activity of PGE ₂ and MMP-3 may lead to severe dysfunction of the salivary glands.

In a recent study involving 24 Sjögren syndrome subjects, Kim and colleagues investigated the pathologic role of autoantibodies associated with downregulation of the major histocompatibility complex I (MHC I) molecule through M3R internalization. The study implicated this action as an important mechanism contributing to the impaired salivation seen in Sjögren syndrome. This study also showed that MHC I did not directly interact with Sjögren syndrome IgG, validating the presence of specific anti-M3R autoantibodies in pSS. The Sjögren syndrome IgG-induced internalization of M3R with MHC I was significantly inhibited by the cholesterol-sequestering drug filipin. The fact that filipin significantly inhibited autoantibody-induced internalization of M3R with MHC I suggests a potential therapeutic for patients with pSS.

Park and colleagues studied the role of functional anti-M3R antibodies in gastrointestinal (GI) dysfunction associated with pSS. Using muscle strip and whole-organ functional assays, this study determined whether anti-M3R antibodies disrupted neurotransmission in tissue throughout the mouse GI tract. The effect of the IgG on GI tissue was dependent on expression of the M3R, demonstrating for the first time a role for autoantibodies specific for this receptor mediate autonomic dysfunction in pSS.

However, there is no consensus regarding the presence or role of anti-M3R in Sjögren syndrome or method of detection. In 2001, Bacman and colleagues used synthetic 25-mer peptide, corresponding to the second extracellular loop of human M3R, as an antigen to demonstrate molecular interaction with autoantibodies from sera of primary and secondary patients with Sjögren syndrome. However, Cavill and colleagues challenged the method of determination in Bacman’s study. Waterman and colleagues used functional assays to investigate autoantibody-mediated effects on parasympathetic neurotransmission and smooth muscle contraction in their studies, and suggest that synthetic peptide is a linear structure that does not reproduce the actual in vivo configuration of the epitope. These investigators advocate that simple peptide-based immunoassays cannot replace complex functional assays for detection of anti-M3R antibodies.