RAGE Biology and Inflammation

Abstract

The receptor for advanced glycation end products (RAGE) is a cell surface signal transduction receptor that amplifies inflammation, principally through the activation of nuclear factor (NF)-κB. A circulating soluble form of RAGE (sRAGE) acts as a “decoy” receptor to alleviate the pro-inflammatory activity mediated by membrane-bound RAGE (mRAGE)-ligand interaction. N-linked glycosylation of RAGE plays an important role in the regulation of ligand binding. Two sites for N-linked glycosylation, at Asn25 and Asn81, are implicated, one of which is potentially influenced by a naturally occurring polymorphism that substitutes Gly82 with Ser (G82S). The G82S RAGE polymorphic variant is associated with reduced plasma sRAGE levels and enhanced ligand-binding. As a consequence, there is increased pro-inflammatory activity associated with the G82S variant. The overall aims of this study were to establish the effects of the G82S polymorphism on RAGE glycosylation and the consequences for sRAGE production and mRAGE-ligand binding. To compare the glycosylation patterns of the G82S variant of RAGE with that for wild-type (WT) receptor, WT-mRAGE or G82S-mRAGE were produced by transfecting human embryonic kidney 293 (HEK293) cells and the glycosylation patterns of expressed proteins were compared. Enzymatic deglycosylation showed that WT-mRAGE and the G82S-mRAGE are glycosylated to the same extent. Furthermore, various mutant forms of mRAGE (N25Q, N81Q, N25Q+G82S, and N25Q+N81Q) were produced to further investigate the complexity of the RAGE glycosylation patterns. Cell surface biotinylation and flow cytometry analysis was used to establish the sub-cellular distribution of WT and all of the mutant forms of mRAGE. All forms reached the cell surface when expressed by HEK293 cells and African Green Monkey SV40-transfected kidney fibroblast cells (COS-7). Mutagenesis and mass spectrometry analysis showed that Asn25 is always “fully” glycosylated in both WT-mRAGE and G82S-mRAGE. However, Asn81 may or may not be glycosylated in WT-mRAGE, whereas in G82S-mRAGE, Asn81 is always “fully” glycosylated. Combined these data indicate that the G82S polymorphism promotes N-linked glycosylation of Asn81. Subsequently, the effects of the enhanced Asn81 glycosylation were investigated. The production of sRAGE via mRAGE ectodomain shedding and the binding of ligand to WT and the variously glycosylated, mutant forms of mRAGE were investigated. Human embryonic kidney 293 cells were transiently transfected with plasmids containing WT or various mutant forms of mRAGE cDNA. To assess matrix metalloproteinase (MMP)-induced mRAGE shedding, surface-expressed mRAGE was biotinylated and transfected cells were treated with potential “shedding inducers” - 4-aminophenylmercuric acetate (APMA) or phorbol 12-myristate 13-acetate (PMA). Levels and component species of mRAGE remaining on the cell surface and of sRAGE released into the cell culture medium were compared by western blot analysis. The data show that reduced amounts of sRAGE were released from transfected cells expressing G82S-mRAGE, when compared to WT receptor. It appears that enhanced N-linked glycosylation associated with G82S-mRAGE reduces the efficiency with which the mRAGE ectodomain is shed from the cell surface. Increased G82S-mRAGE remained on the cell surface. The combination of reduced sRAGE production and retention of cell surface mRAGE reveals one mechanism whereby the G82S polymorphic variant is associated with increased pro-inflammatory activity. Finally, binding of WT and mutant forms of mRAGE by S100B and high mobility group B (HMGB)-1 ligands were compared. Rather than consider the ligand binding directly, the consequences for NF-κB activation were assessed by NF-κB p65 nuclear translocation and western blot analysis. The data show that with WT-mRAGE, S100B requires Asn25 glycosylation for binding while glycosylation of both Asn25 and Asn81 is required for the mRAGE-HMGB-1 interaction. In contrast, N-linked glycosylation of Asn81 was sufficient to mediate S100B and HMGB-1 binding to G82S-mRAGE. The results suggest that the enhanced Asn81 glycosylation seen in G82S-mRAGE introduces an additional point for mRAGE-ligand interaction. This may contribute to the increased ligand binding displayed by the G82S variant of RAGE. In support of this an adapted surface molecule structure model for RAGE shows Asn81 is located in close proximity to the ligand-binding region of RAGE. Therefore the glycosylation of Asn81 may have an impact on RAGE structure. Given that the glycosylation does appear to influence RAGE-ligand interaction, Asn81 is ideally positioned to perform a sentinel role, controlling RAGE receptor function. The results in this thesis highlight different N-linked glycosylation patterns for WT-mRAGE and G82S-mRAGE, and provide detail of how N-linked glycosylation regulates aspects of RAGE biology. Enhanced N-linked glycosylation, promoted by the G82S polymorphism contributes to reduced sRAGE production and mRAGE-ligand interaction. The data go some way to explaining why there is increased inflammation associated with the G82S RAGE polymorphic variant.