The Human Receptor for Advanced Glycation End Products, Created By: Alison Bellew
The Human Receptor for Advanced Glycation End products (RAGE, 3O3U) is a cell surface signaling multi-ligand receptor protein for various ligands of the S100, high mobility group protein box-1 (HMGB1), amyloid β families and many other ligands. It was first described as a receptor for Advanced Glycation End products (AGE) (1,2). RAGE, primarily, is involved in various ligand binding and the subsequent signaling events such as inducing a pro-inflammatory response (1). Its ability to sustain cell activation sites where the ligands accumulate, leads to an exaggerated response of the host, the human, which can lead to diabetes, inflammatory responses, cancer and neurodegenerative disorders (1,2). RAGE, upon ligand recognition, stimulates unique and chronic inflammatory responses. Therefore, by understanding its mechanism for ligand recognition and atomic level structural details, RAGE may prove valuable in developing receptor antagonists, and helping those with common disorders such as diabetes (1). Additionally, understanding RAGE could lead to further interpretations of how one receptor has the ability to recognize multiple ligands (1). The molecular weight of RAGE is 63360.21 Da and its isoelectric point is 7.58 (3).
Ultimately, in proteins, structure determines function. Structurally, RAGE is a cell surface receptor that is composed of three immunoglobulin (Ig) domains, followed by a single transmembrane spanning helix and a short C-terminal cytoplasmic tail (1,2). RAGE was crystallized successfully as a maltrose-binding protein (MBP), although it retains its physiological structure, and was refined to a 1.5Å resolution. The structure was determined by x-ray crystallography. The model of RAGE has 550 amino acid residues, a maltotriose, 985 H2O molecules, and a sulfate ion. Of the three Ig domains, domain 1 (residues 23-118) and domain 2 (residues 121-231) are analyzed crystallized fragments. Relative to each other, the domains are twisted about 70° from their respective axis, resulting in a 140° angle (1). The structure is a rigid unit with the hydrophobic units buried inside the interface (1).
Although there may be more residues involved, experiments have proven explicitly that the hydrophobic region’s interactions from Asn-54 to Gln-67 are an entropically favorable process that drives the biding of RAGE to at least one of its ligands, S100B (1). Here, it is important to note why the S100 family remains relevant in today’s research in both the biochemical and medical fields. The S100 family represents the largest subgroup of EF-hand, Ca2+-binding proteins (4). These S100 proteins regulate processes such as cell growth, transcription and motility (4). When binding to Ca2+, the S100 proteins undergo a conformational change, which is important for its ability to bind to RAGE, because the change exposes the protein-protein interaction site (4). There are many members of the S100 family that bind to RAGE, but all to different structural domain regions which will most likely lead to different intracellular signals (2).
Concentrating on S100B, a particularly important member of the S100 family, the first thing that must happen in order for it to bind to RAGE is to undergo a conformational change in the presence of Ca2+ to expose its hydrophobic helices (2). Although the exact active sight is not known, experiments have shown that on Domain 1, the C’D loop bears a resemblance to the known consensus sequence where S100B binds. Due to the flexibility of RAGE, it may contain the ability to conform into the necessary conformation to bind S100B (1,2). S100B is an important protein that is expressed in the human brain and promotes neuron survival (4). The S100B ligand protein when a tetramer, as opposed to a dimer, has been shown to show stronger activation of cell growth as well as promoting cell survival (4). Studies have shown that the tetrameric S100B also has a greater binding affinity with RAGE than the dimeric S100B (4). S100B is important in the field of science and medicine because higher levels of the S100B protein are found after traumatic brain injuries, as well as in neurodegenerative disorder such as Alzheimer disease and Down’s Syndrome (4). Since the effects of S100B in the body are mediated by the receptor RAGE, understanding RAGE could potentially help doctors understand more about these debilitating diseases and disorders (4).
Some other important members of the S100 family are S100A8/9 and S100A12, with the latter found in humans (2). They increase their binding potential to RAGE once RAGE has been enriched by carboxylated glycans. This increase can sometimes be as great as 100 fold (2). Adding the carboxylated glycan onto RAGE is a post-translational modification and possibly involves an amide linkage to the N-glycans (2). These N-glycans, which increase binding potential, show restricted expression in human monocytes and endothelial cells but are expressed in various tumor cells (2). S100A12 in endothelial cells activates NF-κB and the expression of proinflammatory genes. In order for S100A12 to bind to RAGE either zinc or calcium must be present (2).
The secondary structure of Domain 1 involves front and back sheets with β-strands B and F connected by a disulfide linkage (1). The back β-strands, B, D, and E are relatively short leading to longer interstrand loops strongly structured by intramolecular hydrogen bonds (1). Domain 1 has an extensive hydrophobic region which is highly conserved (1,2). It also has a positively charged region due to various Arg and Lys residues that wrap around, diagonally, across the front β-strands, A, G, F and C (1). By contrast, domain 2 has few distinctive surface properties, except a small positively charged patch continuing from domain 1 consisting of only Lys-123, Arg-216, Arg-218 and Arg-221 next to Arg-29 and Arg-114 of domain 1 (1). These positively charged region and patches on Domain 1 and 2 are important for RAGE to recognize the negative patches on AGE and similar anionic ligands, which facilitates binding; however, the exact residue of the active site is unknown (1). Only the region that participates in binding is known, and it has been shown experimentally that without this site binding will not occur (1). The binding affinity for AGE is increased by the glycation of AGE itself (5). AGE is important medically because it is associated with sites of inflammation in renal failure, and under conditions of hyperglycemia and oxidant stress (5). Evidence has shown that AGE might aid in the development of diabetes, as well as aging, uremia, Alzheimer disease and various inflammatory disorders (5). Once AGE is bound to RAGE, a receptor-dependent signaling is activated in the cell leading to inflammation (5).
Additionally, the positively charged region on RAGE forms a site where oligonucleotides, such as a 19-bp DNA duplex, can bind. In general, a sulfate-coordinated region represents the presence of a biding site for the phosphate backbone of oligonucleotides (1). In RAGE, the sulfate ion’s position is confirmed by understanding that the electron density between the BC and FG loops, found during refinement of the crystal structure of RAGE, is due to the sulfate ion (1). The sulfate ion is highly stabilized by the ordered water molecules. In the FG loop, the sulfate ions interact with two specific residues, Asn-103 and Arg-104 and, on the BC loop, the basic residues Lys-37, Lys-39, Lys-43 and Lys-44 which form a positively charged region (1). There is a possibility that the phosphate backbone of DNA coordinates with the sulfate ion region along the basic patch on the BC loop (1).
The PSI-BLAST computer program is used to find proteins with similar protein structure, represented by an E-score value of less than 0.05. When RAGE was searched a similar protein, Thermotoga maritima maltotriose (2FNC), was found with an E value of 4E-166 (6).
The Dali Server is used to find proteins with similar tertiary structure, which is represented by a Z-score above 2. Performing the Dali search for Domain 2 of RAGE shows that the closest protein structurally is the second domain of Lutheran blood glycoprotein (2PET), with a Z-score of 15.0 (7). Also, Lutheran blood glycoprotein compared to RAGE has around 25% conserved primary structure (1). In comparing the secondary structures of the proteins, RAGE has 21 alpha-helices while Lutheran blood glycoprotein has only four alpha-helices, whereas the β-sheets are more conserved. In comparing RAGE and Lutheran blood glycoprotein, both contain immunoglobulin-like domains, however RAGE functions as a receptor while Lutheran blood glycoprotein functions as a ligand (8). Lutheran blood glycoprotein plays a significant part in sickle cell disease (8).