Graves' disease (GD) is a complex autoimmune disorder characterized by hyperthyroidism and diffuse goitre. 25-50% of the patients with GD develop eye symptoms, Graves' ophthalmopathy (GO). The overall aim of this thesis was to explore the molecular etiology of GD and GO.
By studying gene expression in Studies I, III, and IV, we demonstrated that adipocyte-related immediate early genes (IEGs) including CYR61, COX2, BTG2, and ZFP36 were overexpressed in active GO and to a lesser degree in chronic GO, but not in chronic lymphedema. CYR61-responsive genes were also upregulated in active GO. In Study V, associations were found between SNPs in CYR61, BTG2, and ZFP36 and GD and/or GO. These findings support a role for the IEGs in the pathogenesis of GO.
Inflammation and adipogenesis decreased with decreased disease activity in GO, as demonstrated in Studies I and III by the expression pattern of COX2 and SCD in active and chronic GO. In Study III, we showed that the anti-inflammatory drug diclofenac could inhibit adipogenesis in differentiating 3T3-L1 preadipocytes, decreasing the number of mature adipocytes by approximately 50%.These results suggest that diclofenac may be a therapeutic alternative for GO.
In Study II, we investigated a potential role of the recently discovered hormone thyrostimulin, in GD and GO. We were not able to detect the expression of the beta subunit of thyrostimulin in human healthy and diseased orbital or thyroid tissues. These findings argue against the role of thyrostimulin in human thyroid physiology or disease.
In Study IV, we compared gene expression in chronic arm lymphedema and chronic GO. The expression pattern in lymphedema involved upregulation of genes with roles in wound healing, formation of extracellular matrix and fibrosis, but not upregulation of genes with functions in early adipogenesis. This finding was in contrast to GO, where genes related to adipogenesis, but not fibrosis, were upregulated. To summarize, more differences than similarities exist between chronic arm lymphedema and chronic GO.
In Study V, we investigated whether genetic variation in genes with high expression in GO is associated with GD and/or GO in a Swedish and Polish case/control material. The results of the Swedish study showed associations of SNPs in BTG2, CYR61, ZFP36, and SCD with GD and/or GO; however, rs12136280 in BTG2 and rs3753793 in CYR61 were not associated with GD or GO in the Polish material and the meta-analysis was not significant. The shown associations have to be replicated before they can be considered to be true.