Exendin-4, a GLP-1 receptor agonist peptide originally derived from Gila monster venom, suppressed diabetic retinopathy in both cell and rat models by downregulating the growth factor TGFB2.
TGFB2 identified as key targetExendin-4's retinal protection was abolished when TGFB2 was overexpressed, pinpointing this growth factor as the critical mediator of the peptide's protective effect in diabetic eye disease
What the researchers found
Exendin-4 inhibited the progression of diabetic retinopathy in both a high-glucose-induced human retinal endothelial cell (HREC) model and a streptozotocin (STZ)-induced rat model of diabetic retinopathy.
The peptide's protective effects included reduced cell death, inhibited abnormal blood vessel tube formation, and suppressed inflammatory markers. Mechanistically, Exendin-4 downregulated TGFB2 (transforming growth factor beta-2) expression. When TGFB2 was overexpressed in cells, Exendin-4's protective effects were reversed, establishing TGFB2 as the mediating pathway.
Why it matters
GLP-1 agonists like exenatide (based on Exendin-4) and semaglutide are already widely prescribed for diabetes and obesity. If they also protect against diabetic retinopathy — one of diabetes' most feared complications — patients taking these drugs may get an important bonus benefit. This could influence prescribing decisions toward GLP-1 agonists for diabetic patients at risk of eye disease.
How the study worked
The study used two complementary models: (1) human retinal endothelial cells (HRECs) cultured in high-glucose conditions (in vitro), and (2) STZ-induced diabetic rats (in vivo). Techniques included qRT-PCR for gene expression, CCK-8 for cell viability, TUNEL for cell death detection, western blotting for protein levels, tube formation assays for angiogenesis, and ELISA for cytokine measurements. TGFB2 overexpression experiments confirmed the mechanism.
What this study cannot tell us
The study used STZ-induced diabetes in rats, which models type 1 rather than type 2 diabetes — the form most commonly treated with GLP-1 agonists. The in vitro model used high glucose alone, which doesn't capture the full complexity of diabetic retinopathy. Specific dosing details and long-term effects were not described in the abstract. No human clinical data were presented.
How to read the evidence
This is a preclinical study using cell culture and a rat model of diabetes. While it provides clear mechanistic evidence, no human clinical data exist for Exendin-4's retinal effects specifically, placing this at an early translational stage.
When this study was published
Published in 2025, this is a recent contribution to the growing body of evidence on GLP-1 agonists' protective effects beyond metabolic control.
The bigger picture
This study adds to growing evidence that GLP-1 agonists have benefits far beyond blood sugar control — including neuroprotection, cardiovascular protection, and now retinal protection. As these drugs become some of the most prescribed medications worldwide, understanding their tissue-protective effects could expand their therapeutic indications and inform clinical decision-making for diabetic complications.
Questions still open
- Do patients already taking GLP-1 agonists like exenatide or semaglutide show lower rates of diabetic retinopathy in real-world clinical data?
- Could Exendin-4 or similar peptides be administered directly to the eye (intravitreal injection) for more targeted retinal protection?
- Does the TGFB2-mediated mechanism apply to other diabetic microvascular complications like nephropathy or neuropathy?
Common questions
What is Exendin-4 and how is it related to diabetes drugs like Byetta?
Could GLP-1 drugs prevent blindness in people with diabetes?
Read the original research
Exendin-4, a GLP-1 receptor agonist, suppresses diabetic retinopathy in vivo and in vitro.
Archives of physiology and biochemistry, 131(1), 1-10
Citation
Liu, Jufen; Wang, Huijing; Huang, Cuiting. (2025). Exendin-4, a GLP-1 receptor agonist, suppresses diabetic retinopathy in vivo and in vitro.. Archives of physiology and biochemistry, 131(1), 1-10. https://doi.org/10.1080/13813455.2023.2274279