Advanced Glycation End Products and Metabolic-Mechanical Coupling in Diabetic Retinopathy

Abstract

Diabetic retinopathy (DR) is a major ocular microvascular complication of diabetes mellitus and one of the leading causes of vision loss among working-age populations worldwide. Hyperglycemia-induced metabolic dysregulation has been regarded as a central mechanism underlying DR; accumulating evidence indicates that biomechanical remodeling of the retinal microenvironment also plays an important role in disease progression. Advanced glycation end products (AGEs) serve as key mediators linking metabolic stress to alterations in mechanical signaling. Through non-enzymatic glycation and covalent crosslinking of long-lived proteins, AGEs promote extracellular matrix stiffening, basement membrane thickening, and retinal vascular rigidity. These mechanical alterations can be sensed by retinal vascular cells through mechanotransduction pathways such as Piezo1, YAP/TAZ, and RhoA/ROCK signaling. Aberrant mechanical signals may further exacerbate oxidative stress, inflammatory responses, blood-retinal barrier disruption, vascular dysfunction, and pathological neovascularization. This review summarizes the role of AGEs in DR from an integrated metabolic-mechanical perspective and discusses therapeutic strategies targeting AGEs, providing new insights into potential interventions aimed at biomechanical abnormalities.

Introduction

DR is a leading cause of irreversible vision loss in working-age adults, with cases projected to rise to 161 million by 2045.¹ Traditionally, research on DR has predominantly followed two paradigms: “metabolic dysregulation” and “vascular pathology”. The former focuses on biochemical disturbances such as hyperglycemia, AGEs, and oxidative stress,² while the latter emphasizes morphological outcomes including microaneurysms, leakage, and non-perfusion.³ The etiology of DR is highly complex. Viewing the retina merely as a passive target of metabolic injury cannot fully explain its progressive nature. Clinically, diabetic retinopathy progresses along a continuum from early non-proliferative retinal changes to more advanced stages characterized by increasing capillary non-perfusion, ischemia, vascular leakage, and, in some patients, proliferative diabetic retinopathy with pathological neovascularization.⁴ The risk of progression is influenced not only by glycemic exposure but also by the duration of diabetes and coexisting systemic factors such as hypertension and renal involvement, including albuminuria.⁵ In addition, social determinants of health, including access to regular screening, continuity of diabetes care, treatment adherence, and socioeconomic barriers, may influence the timely detection and management of retinal disease.⁶ Poor glycemic control is associated with a greater risk of DR development and progression, whereas sustained improvement in glycemic control can substantially reduce this risk.^7,8 These clinical observations highlight the importance of integrating systemic risk-factor control with regular retinal surveillance and early ophthalmic intervention.

Recent studies propose an integrated biomechanical framework in which DR progresses through mutual reinforcement of metabolic and mechanical abnormalities. Hyperglycemia and AGEs stiffen the retinal extracellular matrix (ECM). This stiffness is not merely a pathological outcome but actively disrupts cellular behavior via mechanotransduction, thereby further worsening metabolic dysregulation.⁹ Within this mutual reinforcement, AGEs play a pivotal bridging role connecting metabolic dysregulation and mechanical abnormalities. AGEs bind to their receptors, activating downstream signaling pathways, which trigger oxidative stress and chronic inflammation, leading directly to cellular damage.¹⁰ Additionally, AGEs promote covalent cross-linking of long-lived proteins such as collagen and elastin via non-enzymatic glycation, directly contributing to ECM stiffening and thickening of the vascular basement membrane.¹¹ This effectively converts sustained metabolic toxicity into persistent mechanical signals. Such pathological remodeling of the ECM mechanical microenvironment subsequently activates mechanotransduction pathways—including integrins, mechanosensitive ion channels (eg, Piezo1), and downstream effectors such as YAP/TAZ and RhoA/ROCK—which further disrupt cellular metabolism and drive vascular dysfunction.¹² Advancing DR research requires understanding the key “metabolic-mechanical” interplay driven by AGEs. From a clinical perspective, these mechanosensitive pathways may help explain how persistent metabolic stress and progressive structural remodeling reinforce one another as DR advances, thereby linking molecular abnormalities such as Piezo1, YAP/TAZ, and RhoA/ROCK activation with increasingly severe neurovascular dysfunction.

This review summarizes the mutual reinforcement between metabolic dysregulation and tissue stiffness in DR and elaborates on potential therapeutic strategies directed against AGEs.

The AGE-centered metabolic-mechanical framework proposed in this review is summarized in Figure 1.

Figure 1.

AGE-driven metabolic-mechanical coupling mechanism in DR. Chronic hyperglycemia promotes the accumulation of AGEs, which subsequently activates AGE-RAGE signaling and AGE-mediated ECM cross-linking. These pathological processes exert synergistic effects through biomechanical signal transduction pathways, ultimately leading to blood-retinal barrier disruption, tissue hypoxia, pathological neovascularization, and the progression of DR.

Literature Search Strategy

This narrative review was based primarily on a PubMed literature search covering publications from database inception through January 2026. Search terms included combinations of “diabetic retinopathy”, “advanced glycation end products”, “AGEs”, “extracellular matrix”, “matrix stiffness”, “mechanotransduction”, “Piezo1”, “YAP/TAZ”, “RhoA/ROCK”, “vascular remodeling”, and “retinal fibrosis”. Relevant original studies, reviews, and mechanistic investigations addressing AGE-mediated metabolic and biomechanical alterations in diabetic retinopathy were considered. Additional relevant references were identified from the reference lists of selected articles. Because this article is a narrative review, study selection was guided by relevance to the proposed metabolic-mechanical framework rather than by a formal systematic-review protocol.

Formation and Pathological Significance of AGEs

AGEs are stable compounds generated by non-enzymatic glycation, during which reducing sugars react with amino groups in proteins, lipids, or nucleic acids.¹³ AGEs are formed primarily through the Maillard reaction. In this process, glucose non-enzymatically glycates the epsilon-amino groups of proteins, producing unstable Schiff-base intermediates. These intermediates are subsequently rearranged into more stable Amadori products and ultimately converted into irreversible compounds collectively referred to as AGEs.¹⁴

The pathophysiology of diabetes is multifactorial, and AGEs are widely recognized as key contributors to the development and progression of diabetes and its complications. In patients with DR, persistent hyperglycemia enhances non-enzymatic glycation reactions. Poorly controlled hyperglycemia promotes AGE generation and leads to their progressive accumulation in the retinal microenvironment.² In DR, AGEs represent critical mediators linking hyperglycemic stress, oxidative and inflammatory responses, ECM remodeling, and tissue mechanical abnormalities. With the development of metabolomic analyses of intraocular fluids, AGE-related precursors, reactive carbonyl species, and detoxification pathways have been increasingly identified, enabling a more systematic understanding of the role of AGEs in DR.¹⁵ On this basis, AGEs induce metabolic disturbance in DR and mediate mechanical effects by promoting ECM crosslinking and stiffening. Thus, AGEs may be regarded as a central node through which metabolic abnormalities are converted into structural remodeling and mechanical imbalance, providing an important theoretical framework for elucidating metabolism-mechanics coupling in DR.

Mechanical Effects of AGEs

In DR, AGEs act as convergent toxic mediators that exert direct physical effects by stiffening the ECM. The central mechanism underlying AGE-induced ECM stiffening is molecular crosslinking. Once formed through non-enzymatic glycation, AGEs can establish covalent crosslinks with long-lived ECM components, such as collagen and elastin.^14,16 At the molecular level, such crosslinking directly remodels the network topology of the ECM and restricts the intrinsic mobility and sliding of macromolecules, thereby fundamentally reducing matrix elasticity and increasing mechanical stiffness.^17,18

By covalently crosslinking long-lived matrix proteins, AGEs actively drive ECM stiffening and establish a primary mechanical alteration rather than merely representing a passive consequence of tissue injury. AGE-induced matrix stiffness generates mechanical cues that are sensed by cellular mechanoreceptors, including integrins and mechanosensitive ion channels. These cues are transduced into intracellular pathways, thereby reshaping cellular behavior through mechanotransduction, including metabolic activity, proliferation, and differentiation.¹⁹ Within this framework, the “matrix stiffness-metabolism axis” represents a unidirectional cascade initiated by AGE accumulation, translating metabolic stress into mechanical and cellular dysfunction.²⁰

Although the crosslinking effects of AGEs on collagen have been firmly established through in vitro biochemical assays and ex vivo tissue analyses, directly quantifying their independent contribution to retinal vascular stiffness in vivo remains challenging because of complex interactions with hemodynamic factors.¹⁶ At present, most evidence linking AGE-mediated crosslinking to the progression of DR is derived from animal models and clinically relevant studies.

AGE-mediated ECM stiffening is not limited to structural alterations within the matrix but can further propagate to vascular wall remodeling and microcirculatory dysfunction. In DR, persistent AGE accumulation promotes crosslinking between type I collagen and elastin,²¹ as well as pathological and disorganized deposition, which is associated with AGE-induced resistance to matrix degradation. Consequently, a stiff and disordered microenvironment is formed, leading to capillary rigidity, luminal narrowing, and vascular dysfunction. These changes ultimately impair blood flow and oxygen exchange, resulting in retinal hypoxia.²⁰ Hypoxia induces the upregulation of factors such as VEGF, thereby driving pathological neovascularization. Notably, components of the fibrotic ECM, such as fibronectin, may act synergistically with VEGF, further amplifying the fragility and instability of neovessels. This process closely connects AGE-initiated mechanical alterations with the characteristic vascular pathology of DR.

As upstream regulators of profibrotic signaling pathways, including the TGF-beta/Smad axis, AGE accumulation enhances TGF-beta activation and triggers Smad-dependent transcription. Ji et al²² observed that increased TGF-beta levels in diabetes activate the Smad pathway. This activation upregulates the synthesis of ECM components, such as collagen, while inhibiting matrix metalloproteinase (MMP)-mediated degradation, resulting in abnormal ECM deposition in the retina. This process contributes to retinal vascular stiffening, occlusion, neovascularization, and ultimately visual impairment.

AGEs-Induced Metabolic Dysregulation

The pathological role of AGEs in DR extends beyond structural remodeling of the ECM. Through receptor-dependent signaling pathways, AGEs can amplify local microenvironmental abnormalities into persistent metabolic stress, oxidative-inflammatory responses, and cellular injury. Upon binding to their receptor, the receptor for advanced glycation end products (RAGE), AGEs activate downstream pathways, including NADPH oxidase, mitogen-activated protein kinase (MAPK), and nuclear factor-kappaB (NF-kappaB). These pathways promote reactive oxygen species (ROS) production, inflammatory cytokine release, and apoptosis, thereby inducing oxidative stress and inflammation and aggravating retinal neurovascular injury.²³

NF-kappaB induces the production of inflammatory mediators, accelerates apoptosis, and upregulates RAGE expression, thereby establishing a positive feedback loop that amplifies AGE-RAGE signaling. AGE-RAGE interaction activates NADPH oxidase, increases intracellular ROS production, induces oxidative stress, and causes oxidative cellular damage.²⁴ AGE-mediated RAGE activation further upregulates RAGE expression, creating a feed-forward cycle in which ligand-stimulated RAGE acts as both a propagator and a persistence factor. This mechanism may convert transient hyperglycemia into long-term tissue mechanical memory, allowing DR lesions to continue progressing even after glycemic control has been achieved.^24,25

Reciprocal Interactions Between Mechanical and Metabolic Remodeling

In DR, mechanical remodeling and metabolic abnormalities are not independent pathological processes; rather, they form a mutually reinforcing vicious cycle. Under hyperglycemic conditions, the accumulation of AGEs, oxidative stress, and inflammatory responses can convert metabolic disturbances into tissue-level mechanical abnormalities through ECM deposition, protein crosslinking, and vascular wall stiffening. Conversely, abnormal matrix stiffness, altered hemodynamics, and mechanical stress can be sensed by retinal vascular cells and further exacerbate metabolic reprogramming, inflammatory responses, and microvascular injury through mechanosensitive pathways such as YAP/TAZ, RhoA/ROCK, and Piezo1.

Metabolic Abnormalities Drive Retinal Vascular Remodeling and Mechanical Alterations

In DR, AGE accumulation enhances oxidative stress, activates inflammatory responses, and aberrantly stimulates signaling pathways such as TGF-beta/Smad and NF-kappaB. These processes promote the deposition of ECM components, including collagen, fibronectin, and laminin, and reduce matrix elasticity through AGE-mediated protein crosslinking, leading to basement membrane thickening, ECM stiffening, and decreased vascular wall compliance.^20,22,26 Therefore, metabolic abnormalities are not confined to intracellular energy-metabolism disturbances; they can also be translated into tissue-level mechanical abnormalities, driving the retinal microvasculature from a soft and dynamically adjustable state toward a stiff, poorly compliant pathological phenotype.

As vascular wall stiffening and basement membrane thickening progress, the hemodynamics of the retinal microcirculation also become altered. A thickened and stiffened capillary basement membrane may act as a diffusion barrier, hindering the delivery of oxygen and nutrients as well as the clearance of metabolic waste, thereby further impairing neuronal and photoreceptor function.²⁷ Meanwhile, vascular stiffening increases flow resistance and promotes hypoperfusion, which is reflected by reduced peak systolic velocity and end-diastolic velocity, together with an increased resistance index. Color Doppler and duplex Doppler imaging studies have shown that the progression of DR is accompanied by reduced ocular blood flow velocity, particularly in the central retinal artery, and increased resistance index in retrobulbar vessels. These findings suggest that impaired perfusion, increased distal microvascular resistance, and vascular stiffening may contribute to retinal ischemia and hypoxia during DR progression.^28,29 Accordingly, diabetes-related metabolic disorders may progressively convert “metabolic signals” into “mechanical signals” through ECM remodeling, vascular wall stiffening, and hemodynamic abnormalities.

Mechanical Abnormalities Exacerbate Retinal Metabolic Dysregulation

Altered mechanical cues can be sensed by retinal vascular cells, which respond to abnormal mechanical stimulation. Among these cells, pericytes serve as important mechanosensing nodes. Retinal vascular pericytes are specialized mural cells located on the outer surface of capillaries. Through physical contact and paracrine signaling, they interact with endothelial cells and play essential roles in maintaining vascular stability, regulating blood flow, and supporting the blood-retinal barrier (BRB). Pericytes can sense matrix stiffness, vascular wall tension, and mechanical signals transmitted by endothelial cells through integrins, the cytoskeleton, and mechanosensitive ion channels, and can regulate cellular contractility through pathways such as RhoA/ROCK.^30,31 Under conditions of increased stiffness and abnormal mechanical stress, pericytes may undergo aberrant contraction, apoptosis, or functional impairment, thereby weakening vascular stability and accelerating blood-retinal barrier breakdown and pathological angiogenesis.

Abnormal mechanical signals can also be transmitted to the nucleus through the cytoskeleton, where they regulate mechanosensitive pathways such as the YAP/TAZ signaling and TGF-beta/Smad signaling. These pathways subsequently influence pericyte contractility, ECM remodeling, and interactions between pericytes and endothelial cells.^30,32 YAP/TAZ and RhoA/ROCK pathways not only participate in the regulation of mechanical phenotypes but also exhibit extensive crosstalk with metabolism-related processes, including glycolysis, mitochondrial function, oxidative stress, and inflammatory responses. Studies have shown that YAP/TAZ can promote glycolysis, lipogenesis, glutamine metabolism, endothelial nutrient uptake, and mTORC1 activation.³³ The RhoA/ROCK pathway has also been implicated in high glucose-induced retinal endothelial dysfunction and may promote a glycolysis-like metabolic shift.³⁴ Therefore, altered mechanical signaling may induce metabolic reprogramming in retinal vascular cells through these mechanosensitive pathways.

In addition, mechanosensitive ion channels constitute important mediators through which mechanical signals influence cellular metabolic status. Piezo1 is a canonical mechanosensitive cation channel that participates in sensing membrane tension and changes in matrix stiffness in multiple retinal cell types.^35,36 Under diabetic conditions, AGE-induced ECM stiffening may increase the mechanical stress experienced by retinal cells, leading to Piezo1 activation and Ca²⁺ influx.³⁷ Excessive Ca²⁺ signaling may promote inflammatory cytokine release and reactive proliferation of Müller cells, thereby aggravating neurovascular dysfunction.³⁸ Existing studies have shown that Piezo1 expression is significantly upregulated in diabetic animal models and in Müller cells exposed to high glucose. Inhibition or knockout of Piezo1 can improve retinal dysfunction, reduce inflammatory cytokine release, and attenuate abnormal glial proliferation, suggesting that Piezo1 may serve as an important bridge linking ECM stiffening, mechanical stress, and retinal metabolic-inflammatory responses.³⁵

Furthermore, reduced blood flow and tissue hypoxia caused by mechanical abnormalities can further aggravate local metabolic dysregulation. Hypoperfusion and hypoxia induced by vascular stiffening are not merely downstream consequences; rather, they can further enhance oxidative stress, inflammatory responses, and hypoxia-inducible factor (HIF)-related metabolic adaptation, maintaining endothelial cells, pericytes, and Müller cells in a chronically activated state. A hypoxic and oxidative-stress microenvironment can further promote the activation of TGF-beta/Smad, NF-kappaB, and YAP/TAZ pathways and induce persistent deposition of ECM components such as collagen, fibronectin, and laminin, thereby exacerbating basement membrane thickening and fibrosis-like remodeling.^20,22,26 Therefore, mechanical abnormalities are not only consequences of metabolic dysregulation but can also drive metabolic imbalance, forming a mutually reinforcing state between mechanical alterations and metabolic stress.

This reciprocal feedback between metabolic dysregulation and mechanical remodeling is summarized in Figure 2.

Figure 2.

In DR, metabolic dysregulation and mechanical remodeling interact to form a vicious cycle. Hyperglycemia and AGEs induce ECM collagen deposition, cross-linking, and vascular stiffening. In turn, aberrant mechanical cues further activate mechanotransduction, inflammatory responses, hypoxia-related pathways, and metabolic stress, thereby establishing a self-amplifying pathological cycle.

Targeted Interventions

Chronic hyperglycemia promotes the excessive formation and accumulation of AGEs in the retina, where they exert cytotoxic effects on retinal cells. AGEs induce pericyte apoptosis and endothelial dysfunction, aberrantly activate multiple signaling pathways, and trigger oxidative stress and chronic inflammation. In parallel, they promote ECM stiffening and facilitate pathological neovascularization. Thus, AGEs initiate a progressive pathological cascade in DR, extending from early cellular dysfunction to vascular structural disruption and, ultimately, aberrant neovascular formation. Accordingly, targeting AGEs by inhibiting their formation or promoting the cleavage of AGE-mediated cross-links may represent an upstream therapeutic strategy for DR intervention.

Inhibiting AGEs Formation

Early studies identified aminoguanidine, a nucleophilic hydrazine compound, as an inhibitor of AGE formation. In experimental diabetic retinopathy, aminoguanidine inhibited AGE accumulation and attenuated retinal vascular pathological changes, including endothelial cell proliferation, pericyte loss, acellular capillary formation, and microaneurysm development.³⁹ In contrast, pyridoxamine, a naturally occurring form of vitamin B6, appears to have greater clinical translational potential. It scavenges reactive carbonyl species generated during non-enzymatic glycation and lipid peroxidation, thereby inhibiting the conversion of glycation intermediates into AGEs, while simultaneously attenuating oxidative stress. In this way, pyridoxamine may suppress AGE formation at an early stage of the pathogenic cascade.⁴⁰

Inhibition of the AGEs–RAGE Signaling Pathway

Once AGEs have already been formed, blocking their RAGE-mediated signal transduction becomes an important therapeutic strategy. RAGE is a key receptor mediating AGE-induced cellular inflammatory responses and oxidative stress. Upon binding to RAGE, AGEs can activate downstream signaling pathways, including NF-κB and MAPK, thereby promoting the production of inflammatory cytokines, adhesion molecules, and ROS, contributing to the onset and progression of diabetes and its microvascular complications. Evidence indicates that soluble RAGE (sRAGE) can act as a decoy receptor by binding AGEs and other RAGE ligands, thereby reducing their interaction with membrane-bound RAGE. In addition, anti-RAGE antibodies, RAGE aptamers, and RAGE inhibitors can attenuate AGEs–RAGE axis-mediated inflammation and vascular injury by blocking ligand–receptor interactions or inhibiting downstream signal transduction.⁴¹

Cleavage and Clearance of Preformed AGEs

AGE cross-link breakers are designed to reverse preformed AGEs and their stable cross-linked structures. Studies have shown that the compound N-phenacylthiazolium bromide (PTB) can effectively cleave preformed AGE cross-links in experimental settings. Although direct evidence in diabetic retinopathy remains limited, this strategy provides a potential therapeutic direction for diabetes-related vascular complications by targeting preformed AGE cross-links.⁴²

Targeting Mechanotransduction

As an important mechanosensitive signaling pathway, the RhoA/ROCK pathway plays a critical role in inflammation, neovascularization, blood–retinal barrier disruption, and AGE-related injury in DR. Studies have shown that the ROCK inhibitor fasudil can attenuate high glucose-induced activation of RhoA/ROCK1, reduce the interaction between phosphorylated myosin light chain (P-MLC) and connexin 43 (Cx43), preserve the membrane localization of Cx43, and upregulate the expression of tight junction proteins such as ZO-1, occludin, and claudin-5. These effects collectively alleviate abnormal vascular permeability, inflammatory responses, and apoptosis, thereby protecting BRB function.³⁴ Therefore, the RhoA/ROCK pathway may represent a potential therapeutic target for DR, particularly diabetic macular edema (DME). In addition, modulation of mechanotransduction pathways such as YAP/TAZ and RhoA/ROCK represents an emerging research direction. Among them, targeting YAP may simultaneously suppress ECM deposition and pathological angiogenesis, providing a mechanism-based therapeutic strategy for DR that differs from conventional anti-VEGF therapy.⁴³ However, the evidence supporting fasudil in DR is predominantly experimental, and fasudil is not currently an established clinical therapy for diabetic retinopathy.

Targeted therapeutic strategies against AGEs are summarized in Table 1.

Table 1.

Targeted Interventions for DR Based on AGEs and Mechanotransduction

Intervention Strategy	Specific Target	Mechanism of Action	Therapeutic Effects in DR	References
I. Inhibiting AGEs formation	Aminoguanidine	Nucleophilic hydrazine compound; inhibits the formation of AGEs.	Attenuated retinal vascular pathological changes, including endothelial cell proliferation, pericyte loss, acellular capillary formation, and microaneurysm development.	[39]
	Pyridoxamine (vitamin B6 form)	Scavenges reactive carbonyl species generated during non-enzymatic glycation and lipid peroxidation; inhibits the conversion of glycation intermediates into AGEs; attenuates oxidative stress.	Reduces microenvironmental oxidative stress, inhibits AGE formation, and protects retinal nerve cells.	[40]
II. Inhibition of the AGEs-RAGE signaling pathway	Soluble RAGE (sRAGE)	Acts as a decoy receptor by binding AGEs and other RAGE ligands, thereby reducing their interaction with membrane-bound RAGE.	Reduces AGE-induced cellular inflammatory responses and oxidative stress; attenuates AGEs-RAGE axis-mediated inflammation and vascular injury.	[41]
	Anti-RAGE antibodies; RAGE aptamers	Block ligand-receptor interactions between AGEs and RAGE.	Attenuate AGEs-RAGE axis-mediated inflammation and vascular injury.	[41]
	RAGE inhibitors	Inhibit downstream signal transduction activated by AGEs-RAGE binding.	Reduce the production of inflammatory cytokines, adhesion molecules, and reactive oxygen species (ROS); attenuate AGEs-RAGE axis-mediated inflammation and vascular injury.	[41]
III. Cleavage and clearance of preformed AGEs	N-phenacylthiazolium bromide (PTB)	Cleaves preformed AGE-mediated cross-links, reversing stable cross-linked structures.	Potentially reverses preformed AGEs; provides a potential therapeutic direction for diabetes-related vascular complications.	[42]
IV. Targeting mechanotransduction	Fasudil (ROCK inhibitor)	Attenuates high glucose-induced activation of RhoA/ROCK1; reduces interaction between phosphorylated myosin light chain (p-MLC) and connexin 43 (Cx43); preserves membrane localization of Cx43; upregulates tight junction proteins (ZO-1, occludin, claudin-5).	Alleviates abnormal vascular permeability, inflammatory responses, and apoptosis; protects blood-retinal barrier (BRB) function.	[34]
	YAP/TAZ pathway (targeting YAP)	Modulates mechanotransduction pathways to suppress extracellular matrix (ECM) deposition and pathological angiogenesis.	Simultaneously suppresses ECM deposition and pathological angiogenesis; offers a mechanism-based therapeutic strategy distinct from anti-VEGF therapy.	[43]

Targeting Extracellular Matrix Stiffness and Lysyl Oxidase

In addition to targeting AGE formation and mechanotransduction, modulation of extracellular matrix stiffness may represent another potential therapeutic direction. Lysyl oxidase (LOX) is an extracellular enzyme that promotes covalent cross-linking of collagen and elastin and thereby contributes to matrix stiffening. In the diabetic retinal microenvironment, increased LOX-mediated subendothelial matrix stiffness has been shown to enhance RAGE-dependent retinal endothelial activation, suggesting a mechanistic link between extracellular matrix remodeling and inflammatory vascular dysfunction.⁹

Accordingly, modulation of LOX activity may provide a potential strategy for interrupting the metabolic-mechanical feedback loop by reducing pathological matrix stiffening and its downstream cellular effects. However, this approach remains experimental, and direct evidence supporting LOX-targeted therapy in patients with diabetic retinopathy is currently lacking. Further studies are required to determine the safety, efficacy, and clinical applicability of targeting LOX-mediated matrix remodeling in DR.

Conclusions and Perspectives

This review proposes that the vicious cycle between advanced glycation end products and the metabolic dysfunction and biomechanical dysregulation they induce constitutes a key mechanism driving the onset and progression of DR. This conceptual framework extends beyond the traditional pathological perspective, which tends to focus on isolated biochemical events, by integrating hyperglycemic stress, AGE accumulation, extracellular matrix stiffening, aberrant mechanotransduction, hemodynamic impairment, and pathological angiogenesis into a dynamic, self-amplifying model. This integrative perspective offers a new paradigm for understanding the comprehensive pathophysiology of DR.

Current research in this field remains at an early stage, with an incomplete understanding of the AGE-mediated metabolic-mechanical coupling mechanisms and a lack of in vivo, dynamic validation methods. Future research efforts should prioritize the following directions: Mechanistically, there is a need to elucidate the spatiotemporal interaction networks involving mechanosensitive channels and canonical signaling pathways such as YAP/TAZ and RhoA/ROCK during DR progression. Technologically, the development of in vivo, non-invasive, high-resolution retinal biomechanical imaging tools is crucial to enable quantitative assessment of pathological tissue stiffening, facilitate early diagnosis, and bridge the gap between molecular mechanisms and clinical phenotypes. Translationally, moving beyond conventional strategies that solely target AGE inhibition, innovative therapeutic approaches aimed at actively “softening” the pathological extracellular matrix, such as agents modulating lysyl oxidase activity or cleaving collagen crosslinks, should be explored to fundamentally reverse the dysregulated biomechanical microenvironment in DR. Several limitations should be considered when interpreting the current evidence. Much of the mechanistic evidence linking AGE accumulation, extracellular matrix stiffening, and mechanotransduction in diabetic retinopathy is derived from in vitro studies and animal models, whereas direct longitudinal evidence in patients remains limited. In addition, retinal biomechanical properties are difficult to quantify in vivo, and the independent contribution of AGE-mediated matrix stiffening may be influenced by vascular, inflammatory, metabolic, and hemodynamic factors. Therefore, the proposed metabolic-mechanical framework should currently be regarded as an integrative mechanistic model rather than an established clinical pathway.

From a clinical perspective, this framework reinforces the importance of early identification of patients at high risk of progression, optimization of systemic risk factors, regular retinal screening, and timely ophthalmic intervention. A better understanding of AGE-related matrix remodeling and mechanosensitive signaling may also contribute to the future development of biomechanical biomarkers and therapeutic strategies that complement existing approaches such as glycemic control and anti-VEGF therapy. However, interventions targeting matrix stiffness or mechanotransduction remain investigational and require further validation in human studies before routine clinical application.

Funding Statement

This work was supported by the Open Fund of the School of Optometry and Vision Science, Nankai University, No. NKYKK202525; the Open Project of the Institute of Optometry and Vision Science, Nankai University, Nos. NKSGZ202503 and NKSGY202409; and the Tianjin Key Medical Discipline Construction (TJYXZDXK-3-004A-3).

Ethics

Ethical approval was not required for this review article because it does not involve human participants, identifiable human data, human tissue, animal experiments, or original clinical data. Consent to participate and consent for publication were not applicable.

Author Contributions

Lifang Wang and Xudong Zhao contributed equally to this work and share first authorship. Lifang Wang conceived the topic, performed the literature search, organized the manuscript structure, and drafted the manuscript. Xudong Zhao contributed to literature collection, literature interpretation, and manuscript revision. Yan Wang supervised the study, provided academic guidance, and critically revised the manuscript for important intellectual content. Wei Chen contributed to the conceptual framework, supervised the study, and critically revised the manuscript for important intellectual content.

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

The authors declare that they have no conflicts of interest.