The logic of the review begins with chemistry. In type 2 diabetes, sustained hyperglycemia and dyslipidemia push cells into a state of chronic bioenergetic and oxidative stress. Excess glucose flux accelerates glycolysis, and once mitochondrial oxidative capacity is exceeded, intermediates spill into subsidiary routes — the polyol pathway, the hexosamine pathway, and diacylglycerol–protein kinase C signaling — while methylglyoxal accumulates as a reactive glycolytic byproduct. The polyol pathway itself consumes NADPH as aldose reductase converts glucose to sorbitol, weakening glutathione-dependent antioxidant defenses, and the resulting shift in the NADH/NAD+ ratio can promote mitochondrial reactive oxygen species production. Elevated free fatty acids add further injury through mitochondrial dysfunction, incomplete beta-oxidation, endoplasmic reticulum stress, and more ROS.

The consequence is carbonyl stress. Reactive carbonyl species such as methylglyoxal, glyoxal, and 3-deoxyglucosone modify proteins, lipids, and nucleic acids far faster than glucose itself, driving the nonenzymatic formation of advanced glycation end-products. Under healthy conditions the glyoxalase system, particularly glyoxalase 1, keeps methylglyoxal in check, but in diabetes the substrate burden and oxidative stress overwhelm this detoxification pathway. The AGEs that accumulate are strikingly heterogeneous: adducts such as Nε-carboxymethyl-lysine, Nε-carboxyethyl-lysine, the methylglyoxal-derived hydroimidazolone MG-H1, pyrraline, and argpyrimidine, alongside crosslinking species like pentosidine, glucosepane, and methylglyoxal-lysine dimer that stiffen extracellular matrix proteins and resist proteolytic turnover. Crucially, the authors note, the relative abundance of each species depends not only on glycemia but on tissue context, redox state, substrate availability, and glyoxalase activity — and it remains unclear which individual AGE species, if any, is the principal pathogenic driver in vivo.

What is clear is how these modified molecules are sensed. RAGE is a pattern-recognition receptor of the immunoglobulin superfamily, built from an extracellular V-type ligand-binding domain, two C-type domains, a transmembrane segment, and a short cytoplasmic tail. The full-length membrane-bound form, mRAGE, is the signaling-competent version, and it is particularly abundant on inflammatory cells such as neutrophils and macrophages. Its ligand repertoire extends well beyond AGEs to danger-associated molecular patterns including HMGB1, multiple S100 family proteins, and amyloid-beta, and it can even participate in recognizing pathogen-associated molecular patterns such as lipopolysaccharide and extracellular nucleic acids. This breadth places RAGE at the interface of metabolic stress, sterile inflammation, and innate immune activation.

Two structural features give mRAGE its signaling power. First, oligomerization is a prerequisite rather than a passive byproduct: mRAGE forms homodimers at the cell surface, and disrupting this self-association markedly suppresses ligand-induced signaling, while multivalent ligands such as S100B enlarge or stabilize RAGE–ligand oligomeric complexes. Second, the cytoplasmic tail, though lacking intrinsic catalytic activity, binds diaphanous-related formin 1, or DIAPH1, which couples extracellular ligand recognition to intracellular signal propagation. Following ligand-induced clustering, DIAPH1 recruitment activates the small GTPases Rac1 and Cdc42, driving actin polymerization, membrane protrusion, and cell migration, while parallel redox-dependent signals activate MAP kinases, TAK1, the IKK complex, and NF-κB. The result is transcription of cytokines, chemokines, adhesion molecules — and RAGE itself, completing a self-amplifying feed-forward loop in which oxidative stress, glycoxidation, inflammation, and ligand accumulation perpetuate one another.

Against this pathogenic machinery stands a family of molecular decoys. Soluble RAGE isoforms lack the transmembrane and cytoplasmic domains needed for signaling and instead sequester extracellular ligands before they can reach mRAGE. One major form, sRAGE, is generated predominantly by proteolytic shedding of mRAGE from the cell surface — accounting for more than half of circulating sRAGE — through metalloproteinases including ADAM10 and MMP9, a process that may be enhanced under inflammatory conditions. The other, endogenous secretory RAGE or esRAGE, arises from alternative splicing and is secreted directly; because it retains a protease-sensitive site, both routes can converge on the same final soluble structure. Far from being a lung curiosity, esRAGE is widely distributed across human tissues, including neurons, vascular endothelium, pancreatic beta cells, macrophages, and arterial walls.

The clinical picture of these decoys, however, is more complicated than a simple protective story. In the ARIC cohort, low sRAGE predicted incident diabetes, coronary heart disease, and all-cause mortality in people with preserved kidney function, and long-living individuals show increased esRAGE, hinting at a role in healthy aging. Yet the CARDS and ADVANCE trials reported that higher sRAGE or esRAGE levels were associated with coronary events, renal outcomes, and mortality in patients with type 2 diabetes. The Kanazawa group’s own work offers a resolution: impaired renal function correlates positively with circulating esRAGE in diabetes, suggesting that the kidneys are a major clearance route. Elevated soluble RAGE in high-risk diabetic patients may therefore reflect failed clearance and vascular or renal injury rather than increased protective decoy activity — a caution for anyone hoping to use these isoforms as biomarkers without accounting for kidney function.

Then comes the twist. RAGE also mediates the physiological transport of oxytocin, the hypothalamic neuropeptide governing social bonding, emotional regulation, and maternal behavior, across the blood–brain barrier. Remarkably, oxytocin’s engagement of RAGE is mechanistically distinct from that of inflammatory ligands: it does not compete with other major RAGE ligands, does not trigger canonical intracellular mRAGE signaling, and does not interfere with signaling induced by other ligands. Oxytocin appears to use mRAGE as a transport-associated binding partner rather than a pro-inflammatory receptor, with recent evidence showing delivery to the medial prefrontal cortex after peripheral or intranasal administration. Soluble isoforms, meanwhile, do not act as effective decoys for oxytocin: in reconstituted blood–brain barrier models, luminal esRAGE did not inhibit transport, and esRAGE overexpression in vivo did not reduce oxytocin entry into cerebrospinal fluid. Complement component 4a has also emerged as a circulating oxytocin-binding carrier that modulates free hormone availability. The clinical relevance is underscored by evidence that circulating oxytocin is reduced in obesity and both major forms of diabetes, and that postmortem analysis of patients with type 2 diabetes reveals a selective loss of oxytocin-immunoreactive neurons in the hypothalamic paraventricular nucleus.

This duality directly informs therapy, which the review organizes into four layers. Upstream ligand reduction aims to limit AGE formation through metabolic control and enhanced carbonyl detoxification. Extracellular blockade targets the ligand–mRAGE interaction, with compounds ranging from the prototypic antagonist FPS-ZM1 to azeliragon (TTP488), which reached late-stage trials in Alzheimer’s disease: lower doses showed potential benefit in a phase II study, but a higher-dose arm was halted over safety concerns including confusion, falls, and cognitive decline, and the phase III STEADFAST trial missed its co-primary endpoints. Azeliragon has since been repositioned for glioblastoma, where early dose-finding data suggest feasibility but unproven efficacy, with a grade 3 maculopapular rash among reported adverse events. Drug repurposing has identified papaverine and low-molecular-weight heparin as ligand–RAGE modulators, while newer chemistries such as SBLR876 and benzo[b]thiophene-2-carboxamide derivatives expand the preclinical landscape. Intracellular interface blockade is exemplified by RAGE229 and RAGE406R, small molecules that disrupt the mRAGE–DIAPH1 interaction and reduced diabetic complications in mice. A fourth strategy, gene silencing, is embodied by ARO-RAGE, an inhaled RNA interference therapeutic that markedly reduced soluble RAGE in serum and bronchoalveolar lavage fluid in early human testing.

The authors’ conclusion is a call for selectivity rather than brute-force suppression. Because RAGE carries out a physiological transport job in the brain, complete or prolonged blockade may be the wrong goal; instead, strategies that transiently shift the balance from signaling-competent mRAGE toward soluble decoy forms — for example by promoting controlled mRAGE shedding — could dampen pathogenic ligand-driven signaling while preserving oxytocin transport. The review also flags unresolved questions: which AGE species matter most in vivo, how isoform balance is regulated across tissues, and how mRAGE signaling intersects with neuroendocrine circuits. If those answers come, RAGE may finally be druggable on the disease side of its double life — a context-dependent regulator of metabolic, inflammatory, and neuroendocrine crosstalk rather than a single switch to be switched off.

Subject of Research: The role of the receptor for advanced glycation end-products (RAGE) and its soluble decoy isoforms in diabetic complications and oxytocin transport

Article Title: Decoding RAGE in diabetes: a pathogenic signaling hub modulated by decoy receptors in diabetic complications

Article References: Decoding RAGE in diabetes: a pathogenic signaling hub modulated by decoy receptors in diabetic complications. (n.d.). https://doi.org/10.3389/fcdhc.2026.1854214

Image Credits: AI Generated

DOI: 10.3389/fcdhc.2026.1854214

Keywords: RAGE, advanced glycation end-products, diabetic complications, type 2 diabetes, NF-κB, inflammation, methylglyoxal, soluble RAGE, DIAPH1, oxytocin, blood–brain barrier, drug development