Photoimmunology of lupus erythematosus and dermatomyositis: a speculative review.

David Norris has proposed a four step model for the pathogenesis of LESSD (1): (1) exposure to UV light induces the release of proinflammaotry epidermal and dermal mediators such as IL-1 and TNF-alpha; (2) these mediators induce changes in epidermal and dermal cells including the induction of adhesion molecules and promotion of the translocation of normally intracellular autoantigen such as Ro/SS-A to the surface of epidermal cells; (3) autoantibody from the circulation binds to autoantigens such as Ro/SS-A that have been translocated to the surface of epidermal keratinocytes and (4) keratinocyte cytotoxicity ensues as the results of lymphoid cells that have been recruited from the circulation recognizing and responding to the Fc domains of autoantibody molecules bound to autoantigen expressed on the surface of keratinocytes (i.e. antibody-dependent cell-mediated cytotoxicity). Although this remains among the most attractive of hypotheses for the explanation of Ro/SS-A antibody-associated forms of LESSD such as SCLE and neonatal LE, it does not address the pathogenesis of other forms of LESSD such as DLE, which are not associated with high-level Ro/SS-A antibody production or other known autoantibody specificities (low-level Ro/SS-A autoantibody production has been noted recently in DLE patients (155)). In addition, this hypothesis implies that the fundamental abnormality in SCLE and neonatal LE is the production of high levels of Ro/SS-A autoantibody; however, equally high levels of Ro/SS-A antibodies having similar molecular specificities are frequently encountered in other conditions such as Sjögren's syndrome in which LESSD is seen only infrequently. Also, a nude mouse model of anti-Ro/SS-A autoantibody in deposition grafted human skin has been developed; however, no inflammation or epidermal injury occurs in these animals (83). Most work has indicated that the action spectrum for Ro/SS-A autoantigen modulation in human epidermal keratinocytes is limited to the UVB spectrum; however, recent studies have suggested that UVA is involved in the elicitation of certain forms of photosensitive cutaneous LE such as SCLE. The hypothesis that CD4+ T cells that are specific for autoantigens in the skin whose expression is altered by UVR exposure might play a role in the expression of LESSD needs to be explored further. Because LE is thought to be a polygenic autoimmune disease, it is possible that polymorphism of genes that govern the skin's response to UVR might be involved in the pathogenesis of photosensitive LESSD. Candidate genes would include: The Ro/SS-A autoantigenic polypeptides and h-YRNA; cytokines, cytokine receptors and adhesion molecules induced on epidermal keratinocytes and dermal endothelial cells by UVR; molecules involved in DNA repair; components of pathways leading to the generation and quenching of oxygen free radicals and components of the UVR-induced apoptosis cascade. Unfortunately, so little is known about DM photosensitivity that it is difficult to even speculate about pathogenetic mechanisms that might be involved other than to extrapolate from the observations and currents of thought relating to photosensitive cutaneous LE. A truly limiting aspect of our understanding in this area has been the absence of working models of the patterns of inflammation seen in LESSD and cutaneous DM. Until such models become available, considering the limitations of human studies, alone it is highly likely that reviews of this subject will continue to be based on much in speculation as observation.