Autoantibodies and Autoantigens in Sjögren’s Syndrome

1

Background of Autoantibodies SSA/B

Initial pioneering studies in systemic lupus erythematosus (SLE) noticed that patient sera could react with the nucleus of acetone-fixed nucleated cells using an immunofluorescent assay (IFA) ( Fig. 16.1 ). A series of different patterns of reactivity (ie, diffuse, fine-speckled, large-speckled, etc.) were noted and associated with distinct clinical features. The autoantibodies that produced each pattern were named by investigators at Rockefeller Institute for the initials of the patient who donated that serum sample (ie, patient Ro or La). However, in a different laboratory, the antibodies that reacted with the same antigens were called SSA or SSB for Sjögren’s-associated antigen A or B. The use of specific terminology “Ro/La or SS-A/SS-B” is quite inconsistent in published papers and thus confusing to readers. For example, the name for the same antigen (ie, Ro or SS-A) may be used interchangeable in the same sentence or paragraph for no apparent reason except historical associations with particular reported symptoms.

The relationship with Sjögren’s syndrome (SS) was initially noted by Tan et al., leading to the terms SSA and SSB . The same antibodies were associated with a subset of SLE by Reichlin et al. and were termed Ro and La .

• For simplicity, I will use the terminology Ro/SSA and La/SSB .

• For the constituent molecules of Ro/SSA , I will use the terminology Ro62 and Ro50 .

The Ro/SSA antigen was initially identified by a line on Ouchterlony plates, and subsequently shown by immunoprecipitation or Western blot analysis to result from reactivity with a 60 kD (Ro60) and/or 52 kD (Ro52) molecules. Exchange of sera samples indicates La/SSB antibodies identified by Ro antibodies, or the identical molecules.

It has been shown that most anti-Ro60 positive sera also react with a structurally unrelated Ro52 (26, 34, 35). There has been report of association of Ro60 and Ro52 via direct protein–protein interaction ; however, the interaction may be weak or transient, and was not observed by other investigators.

This Ro60 binds to a family of small RNA molecules called hYRNA that have important functions in “quality control” and processing of RNAs. The cellular function for YRNAs remains unknown, but Ro60 protein is postulated to play roles in small RNA quality control and the enhancement of cell survival after exposure to ultraviolet irradiation. These are shown schematically in a crystallographic form binding to hYRNA in Fig. 16.2 .

Perhaps the most interesting and pathogenetically important feature of the Ro62 molecule is its ability to withstand apoptotic degradation, maintain its binding to hYRNA, and migrate into the surface apoptotic bleb of dying cells, where it will be phagocytosed by macrophages and subsequently play a role in the perpetuation of the autoimmune process (described in the following).

Another interesting feature of autoantibodies, including anti-Ro/SSA and anti-La/SSB, is their ability to bind near the “active” site of the target protein and to recognize conserved amino acid or structural motifs in patients with different ethnic backgrounds and their different human leukocyte antigen–antigen D–related (HLA-DR) genotypes.

The antibody to La/SSB binds to a distinct 45 kD molecule that is loosely associated with the Ro/hYRNA complex. La/SSB is involved in diverse aspects of RNA metabolism, including binding and protecting 3′ UUU (OH) elements of newly RNA polymerase III-transcribed RNA, processing 5′ and 3′ ends of pretransfer (pre-tRNA) precursors, acting as an RNA chaperone and binding viral RNAs associated with hepatitis C virus.

2

Methods of Detection of Antinuclear Antibodies and Ro/SSA and La/SSB

The “gold standard” of detection of antinuclear antibodies (ANAs) is the IFA, originally using a cell line (Hep2). After many years of propagation in vitro, these Hep2 cell lines were determined to have been cross-contaminated by a myriad of other cell lines such as HeLa. Thus a standardized Hep cell line called Hep2 2000 was suggested and its use as a standardized substrate is strongly advised. This cell line has been transfected with Ro (60 and 52 kD) genes.

The problem in clinical practice is that not all laboratories use the same method, and different methods give different results .

One of the earliest examples of this laboratory “error” was the syndrome of “subacute” lupus characterized by a rash of erythema annulare. The patient would have a “negative” result for ANA, but a positive result for SSA antibody. Because the Ro/SSA is a nuclear component, it is logically impossible for both tests to be correct. The most common explanation was the lability of the Ro/SSA antigen to fixation process during preparation of slides for IFA. When correctly fixed slides were used, the entire syndrome of “subacute lupus” disappeared.

In an effort to cut costs of laboratory evaluation, the IFA method of ANA detection has been replaced in many laboratories by an enzyme-linked immunosorbent assay (ELISA) method. Depending on the quality of the extract used for ELISA, differing results may be obtained. Thus the rheumatologist may obtain a negative ANA result in an ELISA test in the same sample that yields a positive IFA in a different laboratory . Similarly, the finding of a negative ANA by ELISA may accompany a report of positive antibodies to Ro/SSA on the same sample. This is logically inconsistent and indicates that one of the two assays is incorrect.

It is important for the rheumatologist and other physicians to recognize that the ANA by ELISA method in most clinical laboratories is “geared” to detect SLE sera, and thus may miss the patient who has only the Ro/SSA antibody or the anticentromere antibody (ACA).

This interesting discrepancy in results of ANA testing was actually the subject for a New England Journal of Medicine “Clinical Pathology Correlation” case study. The diagnosis of SLE/SS was excluded on the basis of a negative ELISA, and the correct diagnosis was only made when IFA detection was used to detect positive ANA and then the antibody to Ro/SSA antigen.

In a clinical world where diagnosis is often made or excluded on the basis of a laboratory test, the type of test used to support a diagnosis is rarely recorded . The standard positive sera provided to the laboratory for ANA generally are derived from an SLE patient who has very high titers of anti-DNA antibodies. A positive standard sample containing only antibody to Ro/SSA is rarely included. Unless specifically requested, the large clinical laboratory rarely runs an antibody test for Ro/SSA and La/SSB or ACA unless the screening ANA is positive.

The problem of standardization has further been exacerbated in the United States by the proliferation of individual rheumatologists who run the ANA and Ro/SSA tests in their office laboratories, presumably as a method of obtaining increased revenue.

The importance of this simple but important diagnostic “pitfall” in diagnosis of SS deserves repetition:

• If the ANA by ELISA is negative, then an anti-Ro/SSA test is not routinely run, because the “full panel” is only run as a reflex response to the positive ANA.

• This simple laboratory artifact may bias entire disease registries and population studies.

• Perhaps the most important “simple” change in identifying SS patients would be the standardization of ANA by IFA or ELISA tests.

3

Detection of Specific Antibodies to Ro/SSA and La/SSB

Ro60 and Ro52 may be prepared by recombinant methods or by affinity columns. They may be detected either separately or in combination. Similarly, antibodies to La (45 kD), ACA, and rheumatoid factor (RF) are done by different methods. Although large laboratories attempt standardization by exchange of substrate and positive blood samples, there remains continued and significant variation in the results based on the substrate preparation and detection methods.

4

Simple Model for the Role of Ro/SSA and La/SSB in Pathogenesis and as a Possible Target for Therapy

Gordon et al. initially proposed division of the steps for the pathogenesis of Ro/SSA and La/SSB in SS:

• 1.

  the initiation of autoimmunity, probably by an intercurrent viral infection that included damage to epithelial cells;

• 2.

  infiltration of the glands by lymphocytes and the diversification of autoantibodies to apoptosis-associated antigens;

• 3.

  the antibody-mediated tissue injury that perpetuated the cycle of apoptosis and further lymphoid infiltration;

• 4.

  the escape of particular lymphoid B-cell clones to lymphoma.

In the initiation step, the key finding is that intracellular autoantigens, including Ro/SSA and La/SSB, are clustered in membrane blebs (apoptotic blebs) at the surface of apoptotic cells. It has been proposed that apoptotic cells serve as a source of immunogen of intracellular proteins for the production of autoantibodies. Redistribution of Ro/SSA and La/SSB polypeptide into the blebs may produce neoepitopes by mechanisms such as oxidation, proteolytic cleavage, or conformational changes, and these have been termed “apotopes.”

However, the resistance to apoptotic degradation can be only part of the story. It is known that SS is strongly associated with particular HLA-DR alleles. This association may be different among divergent ethnic populations. In Caucasians, extended HLA-DR3/DQ1 haplotype and in Han Chinese or Japanese SS patients, a different HLA-DR haplotype, even though similar Ro/SSA epitopes, are recognized.

The production of autoantibodies to Ro/SSA and La/SSB in SS patients was strongly linked to these same HLA-DR/DQ alleles over 40 years ago. More recently, genome-wide association studies (GWASs) have further extended this result to include other loci associated with SS (and thus autoantibody to Ro/SSA), including genes associated with interferon (IFN) type 1 gene signature and B-cell stimulating factors ( Table 16.1 ).

Taken together, their findings suggest a simplified model that includes the Ro/SSA antigen that can bind to: (1) anti-Ro/SSA antibody, (2) the hYRNA that can bind to Toll-like receptors (TLRs), and (3) particular HLA-DR antigens associated with SS. Details of findings are as follows:

• an environmental agent or genetic tendency for initial damage to the lacrimal or salivary gland; indirect evidence suggesting a potential role for Epstein-Barr virus, a related virus, an endogenous viral fragment, or perhaps production of a microRNA in this initial damage;

• apoptosis of the glandular tissues with resistance of the Ro/SSA antigen from degradation and migration of the Ro/SSA–hYRNA complex into the apoptotic bleb;

• binding of the anti-Ro/SSA antibody to the Ro/SSA–hYRNA complex;

• phagocytosis of the Ro/SSA antigen (bound to hYRNA) by macrophages or antigen-presenting cells (APCs) where its normal fate would be enzymatic degradation;

• entry of the immune complex (antibody to Ro/SSA–Ro/SSA–hYRNA) to be into the lysosome or antigen-binding compartments of the macrophage;

• entry into other compartments of the macrophage by the complex because of the Fc receptors on the lysosomal membrane that bind to the Fc on the anti-Ro/SSA antibody;

• Ro/SSA antigen degraded into a peptide that is able to bind to the specific HLA-DR antigen for presentation to T helper cells that drive B-cell autoantibody production;

• hYRNA (containing both single- and double-stranded RNA regions) in the same lysosomal compartments, able to activate TLR to generate type 1 and γ IFN signatures;

• activated TLRs and activated T helper cells drive B cells to release further cytokines, form immune complexes, and eventually drive lymphocyte “aggressive” behavior as a result of continued antigen-driven processes;

• the role of other factors including hormonal status and local glandular factors in perpetuating this cycle.

In the particular case of antibodies to Ro/SSA and La/SSB, the immune complex of antibody/protein/hYRNA was historically first identified but later shown to play a plausible pathogenetic role ( Figs. 16.3–16.6 ).

The hYRNA has single- and double-stranded regions that may appear similar to APCs as viral pathogens. To provide immediate and strong immune response to these viral challenges, the innate system has developed a family of TLRs that generate potent type 1 IFN and later type 2 IFNγ responses.

To protect delicate areas (such as eye or mouth) from indiscriminate destruction from these innate immune responses, natural evolution has cleverly “hidden” these TLRs inside lysosomal compartments of APCs, where they are not exposed to the direct liberation of RNA debris of dying cells. Thus a rate-limiting step in the inflammatory response is the ability of autoantigens to gain entry into the cytoplasmic component, where they can stimulate and perpetuate autoimmune responses that eventuate in an autoimmune disease.

This “perfect” storm occurs in SS patients with:

• a positive antibody to Ro/SSA;

• Ro/SSA that binds to hYRNA in the apoptic bleb;

• antibody to Ro/SSA forming an immune complex with Ro/SSA–hYRNA;

• internalization of the immune complex into the phagocytic or dendritic cell through its Fc receptor;

• delivery of the immune complex to the intracytoplasmic TLRs;

• stimulation of TLR 7/9 by the single- and double-stranded RNA components to generate a type 1 and type 2 IFN signature;

• further activation of local T cells and B cells that result in cytokine production and continued antibody to Ro/SSA.

  The genes predisposing to this model have been identified in recent GWAS studies (see Table 16.1 ).

The steps in this pathogenesis are shown in Fig. 16.1 .

5