Diabetic retinopathy is often described as a slow burn: years of elevated blood sugar gradually weaken the retina's delicate blood vessels until vision begins to fail. Yet for many patients, the earliest stages are silent. By the time symptoms appear, the disease can be harder to reverse.
Now, researchers led by scientists at University College London (UCL) have identified a protein that appears to act much earlier than the classic hallmarks clinicians look for. The protein, called LRG1, was found to trigger the first functional damage in diabetic retinopathy by constricting tiny retinal blood vessels and reducing oxygen delivery. In mouse experiments, shutting down LRG1 protected the retina from these early changes.
The work points to a different way of thinking about diabetic eye disease: not only as a problem of leaky, abnormal new vessels that emerge later, but as a problem that may begin with subtle vessel narrowing and oxygen deprivation. If that early step can be blocked, the trajectory of the disease could change.
What diabetic retinopathy is-and why early damage matters
The retina is one of the most metabolically demanding tissues in the body. It needs a steady supply of oxygen and nutrients to convert light into signals the brain can interpret. That supply depends on a dense network of microscopic blood vessels that must continuously adjust their diameter to match changing demand.
In diabetes, chronic high blood sugar stresses blood vessels throughout the body. In the retina, that stress can lead to a cascade of problems: impaired blood flow, inflammation, breakdown of the blood-retina barrier, microaneurysms, bleeding, and in advanced stages, the growth of fragile new vessels that can scar and detach the retina.
Today's most widely used treatments target later-stage disease, particularly abnormal vessel growth and leakage. Those therapies can preserve vision, but they are typically deployed after damage is already underway. A mechanism that explains the earliest shift-before obvious structural injury-could open the door to prevention rather than rescue.
LRG1 and the first squeeze on retinal blood flow
The UCL-led team reports that LRG1 appears to initiate early diabetic retinopathy by causing constriction in the smallest retinal blood vessels. When those microvessels narrow, less blood can pass through, and oxygen delivery drops. The retina, already hungry for oxygen, becomes vulnerable to hypoxia-driven stress responses.
That detail is important because it reframes the earliest stage of disease as a functional problem-blood flow regulation going wrong-rather than only a structural one. Vessel constriction can occur before the more visible signs of retinopathy show up in routine eye exams.
In the reported mouse work, blocking or shutting down LRG1 prevented these early changes. While mouse models are not a guarantee of human outcomes, they are often used to map cause-and-effect in ways that are difficult to prove in people. The implication is that LRG1 is not merely associated with retinal damage but may be driving it.
How a single protein can shift a whole tissue's oxygen balance
Proteins like LRG1 can act as signaling molecules, influencing how cells behave and how tissues respond to stress. In the retina, the diameter of microvessels is controlled by a coordinated system involving endothelial cells lining the vessels, supporting cells around them, and chemical signals that tell vessels when to dilate or constrict.
If LRG1 pushes that system toward constriction, the effect is amplified by the retina's tight margins. A small change in vessel diameter can translate into a meaningful change in blood flow. Reduced flow means reduced oxygen, and low oxygen can trigger a chain reaction: metabolic strain, inflammatory signaling, and eventually the pathological vessel growth seen in advanced diabetic retinopathy.
That sequence helps explain why an "early trigger" matters. If the first step is a squeeze on microcirculation, then later-stage features may be downstream consequences of a retina that has been operating under oxygen debt for too long.
Why current treatments don't fully solve the early-stage problem
Standard management of diabetic retinopathy includes controlling blood sugar and blood pressure, regular screening, and-when disease progresses-procedures and drugs aimed at stopping leakage and abnormal vessel growth. These approaches can be effective, but they are not designed to correct the earliest microvascular dysfunction.
Many patients also face practical barriers. Treatments for advanced disease can require repeated clinic visits and ongoing monitoring. Screening programs catch many cases, but not all, and early disease can still progress between exams.
A therapy that targets an early molecular driver could complement existing strategies. It would not replace the need for metabolic control, but it could provide a more direct way to protect retinal circulation from diabetes-related stress.
What an LRG1-targeting approach could look like
The new findings suggest that inhibiting LRG1 might preserve retinal blood flow and oxygenation in diabetes. The mouse results indicate that shutting down the protein can prevent early damage, which raises a practical question: how would such inhibition be delivered in people?
Eye diseases are often treated locally, including by injections into the eye, because local delivery can concentrate the drug where it's needed while limiting systemic exposure. That said, the best delivery method depends on the biology of the target, how long inhibition needs to last, and safety considerations.
Any translation from mice to humans would require careful testing to confirm that LRG1 plays the same initiating role in human diabetic retinopathy, to determine the right therapeutic window, and to ensure that blocking the protein does not interfere with normal vascular function elsewhere.
Implications for screening and risk prediction
If LRG1 truly sits near the start of the disease pathway, it could influence more than drug development. It could also affect how clinicians identify risk and monitor progression.
Today, diabetic retinopathy is largely staged by visible retinal changes. But functional changes in blood flow and oxygenation can precede those visible signs. A biomarker tied to early microvascular constriction could, in principle, help flag patients whose retinas are under stress even when imaging looks relatively normal.
That could lead to more personalized follow-up schedules and earlier intervention. It could also help researchers design clinical trials that enroll patients at the stage when prevention is still plausible, rather than after damage has accumulated.
A broader shift: treating microvascular disease at the source
Diabetic retinopathy is one of several diabetes complications rooted in microvascular dysfunction. Kidneys, nerves, and other tissues also depend on small vessels that can be disrupted by chronic metabolic stress. A protein that drives early vessel constriction in the retina raises the possibility that similar mechanisms might exist in other organs, even if the specific molecules differ.
For the eye field, the immediate value is conceptual clarity. The discovery supports a model in which early diabetic retinopathy is not only about vessels becoming leaky or proliferative, but about vessels becoming too narrow, too soon.
That model aligns with a prevention-first mindset: keep the retina oxygenated, keep the microcirculation responsive, and the later destructive stages may be less likely to take hold.
What to watch next
The next steps will likely focus on validating the role of LRG1 in human disease and determining whether LRG1 inhibition can be achieved safely and durably. Researchers will also need to map where LRG1 sits in the broader network of diabetic signaling in the retina-what turns it on, which cells produce it, and how it interacts with other pathways involved in vascular tone and inflammation.
For patients, the findings do not change current advice: regular eye screening and good diabetes management remain the most reliable tools for protecting vision. But for the industry and research community, the work adds a promising target to the early-stage landscape-one aimed at the first functional insult rather than the final structural damage.
If that approach holds up in further studies, it could expand the therapeutic playbook for diabetic retinopathy from damage control to true early intervention.
