Nanoparticles that shrink over time deliver eye drugs to the retina without injections


Jan 20, 2026

Eye drop nanoparticles that shrink over time overcome the eye’s natural defenses, delivering protein drugs to the retina and matching the effectiveness of direct injection in mice.

(Nanowerk Spotlight) Every year, millions of people worldwide submit to needle injections directly into their eyes. The procedure, called intravitreal injection, delivers powerful drugs to treat conditions like diabetic retinopathy and age-related macular degeneration. Despite its effectiveness, the invasive nature of this approach carries genuine risks: infection, retinal detachment, and optic nerve damage among them. Patients often require repeated injections over months or years, each visit bringing anxiety and potential complications. The obvious alternative would be simple eye drops. Yet delivering large protein drugs to the back of the eye through topical application poses significant difficulty. The eye evolved sophisticated defenses against foreign substances. Tears constantly wash away applied medications. The cornea forms a nearly impenetrable obstacle. Even if a drug molecule somehow reaches the interior, the blood-retinal barrier stands as a final line of defense, protecting the delicate retinal tissue from circulating substances. Scientists have tried various strategies to overcome these challenges. In 2014, researchers used liposomes coated with a phospholipid-binding protein to ferry bevacizumab toward the retina. Others modified chitosan nanoparticles with fluorocarbon compounds to pry open tight junctions between corneal cells. Despite scattered successes in laboratory settings, none of these approaches achieved the consistent, clinically meaningful drug concentrations needed for real-world treatment. The fundamental problem remains paradoxical: large particles resist being washed away by tears but cannot penetrate tissue; small particles penetrate well but disappear before they can work. This size dilemma has stymied progress toward non-invasive delivery of anti-VEGF (vascular endothelial growth factor) protein drugs, which remain the gold standard for treating retinal vascular diseases affecting over 100 million people globally. A study published in Advanced Functional Materials (“Size‐Evolving Protein Drug Nanoparticles: A Promising Non‐Invasive Strategy for Retinal Neovascularization by Overcoming Multiple Ocular Barriers”) offers a potential solution. Researchers at Wenzhou Medical University and the Wenzhou Institute of the University of Chinese Academy of Sciences developed nanoparticles that change size over time, exploiting the advantages of both large and small carriers in sequence. Size-EvolvingProteinDrugNanoparticles Schematic overview of the experiment. (a) The construction of progressively smaller, carrier-free nanoparticles, (CG2R9 & Beva)@Zn, was achieved through the co-assembly of CG2R9 peptide and Beva in the presence of zinc ions, facilitated by hydrophobic, coordination, and electrostatic interactions. (b) The (CG2R9 & Beva)@Zn nanoparticles increased ocular residence time and subsequently penetrated various ocular barriers by breaking down into smaller particles. These smaller particles traversed the corneal pathway and the conjunctiva–sclera–choroid–retinal pathway, thereby significantly enhancing the bioavailability of Beva at the retina. (c) The (CG2R9 & Beva)@Zn nanoparticles improved inhibition of retinal neovascularization (RNV) through non-invasive delivery, with comparable therapeutic outcomes to those of vitreous injection, offering a potential new treatment option for RNV. (Image: Reproduced with permission from Wiley-VCH Verlag) The team created particles through a one-step assembly process combining three components: a cell-penetrating peptide called CG₂R₉, composed of cysteine, glycine, and nine arginine amino acids, the anti-VEGF drug bevacizumab (commonly known by its brand name Avastin and already FDA-approved for cancer treatment), and zinc ions. When mixed under controlled conditions, these ingredients spontaneously form nanoparticles approximately 214 nm in diameter. The critical innovation lies in what happens next. When exposed to physiological conditions similar to the eye’s environment, these particles progressively shrink to about 44 nm over 12 hours. This size evolution allows the particles to first resist rapid clearance from the ocular surface while large, then penetrate tissue as they become smaller. The CG₂R₉ peptide drives this transformation through its arginine and cysteine components. Arginine disrupts particle assembly through electrostatic competition, while cysteine interferes with the zinc coordination that holds the structure together. Working in concert, these amino acids enable controlled, gradual disassembly. The nanoparticle formulation achieved impressive drug loading characteristics. Encapsulation efficiency reached 91.50%, meaning almost all of the expensive bevacizumab ended up incorporated into the particles rather than wasted. Loading capacity hit 80.70%, indicating that the drug itself constitutes most of the particle’s mass rather than inert carrier material. Most conventional nanocarrier systems must sacrifice one metric for the other. Critically, the assembly process preserved bevacizumab’s biological activity. Circular dichroism spectroscopy confirmed that the protein’s secondary structure remained intact. Functional testing showed the released drug retained 98.76% of its original activity. This matters because proteins are fragile molecules that easily lose function when handled roughly. The researchers tested extensively in both cell cultures and animal models. In cell studies, the nanoparticles potently inhibited endothelial cell migration and blocked the formation of tube-like structures that mimic blood vessel development. When tested on choroidal tissue explants from mice, the treatment reduced sprouting vascular area by 90% compared to untreated controls. Ocular distribution studies in mice revealed that topically applied nanoparticles reached the retina at measurable concentrations. Drug levels in the cornea peaked around 3 hours after application, while retinal concentrations reached their maximum of 253.3 ng/g at 6 hours. This timing aligns with the size evolution profile, suggesting that the particles need time to shrink sufficiently for deep penetration. Surface retention experiments in rats demonstrated another advantage of the size-evolving approach. The nanoparticles maintained 39.46% of their initial presence on the ocular surface after 3 hours, compared to just 7.25% for unformulated bevacizumab. This fivefold improvement in retention provides more time for the drug to penetrate while avoiding the rapid washout that plagues conventional eye drops. Therapeutic benefits emerged clearly in a mouse model of oxygen-induced retinopathy, which mimics aspects of human retinal neovascular disease. Twice-daily eye drop treatment for five consecutive days produced notable improvements. The non-perfusion zone and pathological blood vessel growth both decreased by approximately 52% compared to saline-treated animals. These results approached those achieved by direct intravitreal injection, the current clinical standard. Thirty days of continuous administration in rats revealed no signs of toxicity. Slit-lamp examination showed clear corneas without inflammation. Corneal endothelial cells maintained their normal hexagonal shape and density. Electroretinography, a test measuring the retina’s electrical response to light, indicated preserved retinal function with no significant differences between treated and untreated eyes. Several limitations remain before this approach could reach patients. Species differences between mouse and human eyes may affect how well these results translate clinically. The potential for zinc accumulation in ocular tissues during long-term use warrants further investigation. The researchers also note that while their data consistently support a size-evolution mechanism, directly visualizing particle size changes during corneal traversal proved technically challenging due to low concentrations and tissue interference. The work represents a meaningful step toward non-invasive treatment of retinal vascular diseases. If the approach proves successful in larger animal models and eventually human trials, it could spare countless patients from repeated needle injections while maintaining therapeutic effectiveness.


Michael Berger
By
– Michael is author of four books by the Royal Society of Chemistry:
Nano-Society: Pushing the Boundaries of Technology (2009),
Nanotechnology: The Future is Tiny (2016),
Nanoengineering: The Skills and Tools Making Technology Invisible (2019), and
Waste not! How Nanotechnologies Can Increase Efficiencies Throughout Society (2025)
Copyright ©




Nanowerk LLC

For authors and communications departmentsclick to open

Lay summary


Prefilled posts