Viruses contribute to the development of Sjögren's syndrome

Sjögren's syndrome (SS) is a chronic autoimmune epithelitis [1] manifested predominantly by a disruption of epithelial cells making up the acini and salivary ducts of exocrine glands. The ensuing lymphoplasmocytic infiltration carries on with the induction of epithelial cell apoptosis. This disease may occur alone as primary SS (pSS), or be associated with any other connective tissue disease as secondary SS (sSS). Based on their overwhelming predominance, T cells used to occupy central stage of this pathophysiological network and considerable research has focused on disturbances in this cell compartment; however, there have been recent advances in the parameters that modulate the fate of autoreactive B cells [2]. These lymphocytes display a flurry of self-reactivities, including antifodrine and antisalivary duct antibodies, rheumatoid factor especially of the IgA isotype, and antinuclear antibodies, most notably anti-Ro/SSA and anti-La/SSB antibodies.

Ro/SSA presents as one of four small RNA particles associated with a 52-kD and a 60-kD protein. Autoantibodies to the latter peptide are found in SS and in systemic lupus erythematosus (SLE). Although the antigenicity of this 60-kD peptide has been ascribed to its tertiary structure, antibodies that bind to synthetic (and therefore linear) peptides constitute one part of the population of antibodies that recognize the native particle. It is relevant to the pathogenicity of the two diseases that at least three studies [3–5] have revealed distinct differences in the 60-kD epitopes targeted by anti-Ro/SSA antibody-containing sera from patients with pSS, compared with anti-Ro/SSA antibody-containing sera from those with SLE.

The observation that infection can precipitate autoimmune aberrations dates back more than a century. For example, a central role for an infectious agent in the development of pSS has, more than once, been advocated, and indeed a number of viruses have been suspected to launch such a disease. These include herpes virus 6, cytomegalovirus, Epstein-Barr virus, hepatitis C virus, human T lymphotropic virus type 1, human immunodeficiency viruses, human intracisternal A-type retroviral particle, and human retrovirus 5. Nonetheless, the last 20 years have witnessed numerous initially promising causative agents that proved disappointing.

Several lines of epidemiological, serological and experimental evidence have since suggested that enterovirus infections, especially those due to group B. Coxsackieviruses B4, are implicated in insulin-dependent diabetes mellitus [6], and coxsackieviruses B-1 in myocarditis [7]. Last year, by differential display technology using samples from one patient with the primary form of SS, Moutsopoulos’ group [8] discovered a 94 base pair gene which showed full homology with the P2A gene of the coxsackievirus B4. This finding was verified in a large group of pSS patients, but, interestingly enough, was shown to be absent in sSS samples. The disease may thus be associated with coxsackievirus A13 and B4 infection of the salivary epithelial cells. The observation was then confirmed at the protein level, since immunohistochemistry for the enteroviral capsid protein VP1 yielded positive staining in epithelial cells and lymphocyte infiltrates in all the pSS samples tested, but again such staining was never found in salivary gland samples from patients with sSS. Some B lymphocytes were found hosting the viral protein within the local infiltrate in salivary gland biopsies. Furthermore, coxsackievirus sequences were detected in cultured epithelial cells eluted from the salivary glands of pSS, but not sSS [9]. These data indicate that, rather than an acute ongoing infection, the coxsakievirus infection of epithelial cells is persistent. Furthermore, similar observations were achieved in samples from patients suffering from pSS, and originating from different countries (Prof H.M. Moutsopoulos et al. unpublished observations).

A coxsackievirus infection has thus been put forth as an environmental trigger for the development of various autoimmune conditions. This process would occur either by inducing bystander inflammation in the target organ, or by providing a molecular mimic of the candidate autoantigen or mobilizing endogenous antigens [10]. These two processes are not mutually exclusive; however, molecular mimicry prevails since the favoured hypothesis for the propensity of virus infection to precipitate an autoimmune reaction is that autoantibodies are generated by cross-reaction between foreign and self determinants. In this respect, it is interesting that autoantigenicity of the Ro/SSA particle is related to a nucleocapsid protein of vesicular stomatitis virus [3], and that the viral P2C gene product of the B4 strain resembles the pancreatic islet glutamate decarboxylase [6]. A similar observation is reported by Stathopoulou et al. [11] in the current issue of Clinical and Experimental Immunology, on the basis of sequence homology. These investigators showed elegantly that one-third of sera from patients with pSS recognized a synthetic peptide spanning 17 amino acid residues within protein 2B of coxsackievirus A21, as well as a 216–232 amino acid sequence of the 60-kD protein of Ro/SSA particles. Interestingly, the latter peptide has previously been identified by the same group, from among 22 synthetic peptides, as a major epitope for anti-Ro/SSA antibody [5].

This scenario raises a number of important issues which converge to incriminate coxsackieviruses in the aetiology of pSS. This interpretation is consistent with the view [12] that clonal proliferation of B lymphocytes within the target tissue is antigen-driven. At this moment, what we require are new approaches for subjecting the virus candidates to experimental validation. Prospective clinical trials, to determine whether removal of coxsackievirus, would benefit pSS patients also need to be established. These studies warrant completion inasmuch as molecular mimicry has important implications for vaccination, and would permit novel therapeutic approaches aimed at modulating B cell function to be designed in the near future.