← Latest papers
📄 bioengineering

Biochemical selection on conserved glycans yield a cross-species AAV capsid for efficient panretinal intravitreal delivery

By mapping conserved retinal glycans across mammals to establish a receptor-guided, animal-free selection strategy called CReST, researchers engineered the AAV2.IvtB capsid, which enables efficient, panretinal gene delivery via intravitreal injection by attenuating heparan sulfate binding while engaging terminal N-acetylglucosamine.

Original authors: Wang, J.-H., Cui, M., Zhao, W., Giguere, D., Yang, Y., Su, Q., He, R., Xie, J., Xie, Q., Ke, X., Seddon, J. M., Punzo, C., Gao, G.

Published 2026-11-19
📖 4 min read☕ Coffee break read

Original authors: Wang, J.-H., Cui, M., Zhao, W., Giguere, D., Yang, Y., Su, Q., He, R., Xie, J., Xie, Q., Ke, X., Seddon, J. M., Punzo, C., Gao, G.

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

Imagine trying to deliver a life-saving message to a specific room deep inside a vast, fortified building. The message is carried by a tiny, harmless virus, a biological courier designed to slip into cells and deliver a corrected gene. For decades, doctors have used this method to treat inherited eye diseases, but the delivery method has been a major hurdle. To reach the light-sensing cells at the very back of the eye, the virus currently has to be injected underneath the retina. This is a delicate, invasive surgery that carries risks and can only treat a small, localized patch of the eye at a time. A much simpler, safer approach would be to inject the virus into the main fluid-filled cavity of the eye, the vitreous, allowing it to float naturally to the back and treat the entire retina at once. However, nature has built a formidable barrier for this route. The inner surface of the retina is coated with a sticky, sugary layer that acts like a trap, catching most viral vectors before they can reach their destination.

For years, scientists have tried to engineer viruses that can bypass this trap, often by testing thousands of variations in mice. But a virus that works perfectly in a tiny mouse eye often fails when the same approach is tried in a larger eye, like a human's, because the biological barriers are different. The challenge has been finding a viral shell that can navigate the complex, sugary landscape of the eye without getting stuck, while still knowing exactly where to go once it gets there.

In a new study, researchers took a different approach. Instead of guessing which virus might work and then testing it in animals, they first mapped the terrain. They examined the retinas of seven different mammals, from mice to humans, looking for the specific sugar molecules that line the inner surface of the eye and the light-sensing cells at the back. They found a consistent pattern across all species: the inner barrier is rich in a sticky sugar called heparan sulfate, which acts as a trap, while the target cells at the back are decorated with a different sugar, terminal N-acetylglucosamine, which acts as a welcome sign. The researchers realized that to succeed, a virus needed to be engineered with two specific traits: it must be slippery enough to avoid getting stuck on the inner barrier, but sticky enough to grab onto the welcome sign at the back.

Using this map, the team developed a new method to design a virus entirely in a laboratory, without needing to test thousands of candidates in animals first. They created a library of millions of slightly different viral shells and subjected them to a two-step selection process. First, they filtered out any viruses that got stuck on the sticky inner barrier. Then, they selected only the survivors that could successfully grab onto the welcome sign sugar found on the light-sensing cells. This process, which they called Conserved Receptor-Selection Technology, yielded a new viral variant named AAV2.IvtB.

When the researchers tested this new virus, the results were striking. In mice, the new virus delivered genetic material to the light-sensing cells six and a half times more effectively than the current best-performing virus used in research. Crucially, this success translated to larger eyes. In pigs, which have eyes much closer in size and structure to humans, the new virus still outperformed the standard options, delivering genetic material to the light-sensing cells two and a half times more efficiently. The virus did not just reach the cells; it worked better once it arrived. In models of retinal degeneration, where the light-sensing cells die off, the new virus helped preserve the structure of the retina significantly better than previous methods. In another model involving abnormal blood vessel growth, it suppressed the damage more effectively, showing that it could deliver both gene replacements and therapeutic proteins.

The study also revealed that this new virus triggers a much weaker immune response in pigs compared to older versions, which is a vital safety feature for human treatments. The researchers confirmed that the virus works by specifically binding to the welcome sign sugar on the target cells, proving that their design principle was correct. By focusing on the conserved, unchanging features of the eye's biology rather than the quirks of a single animal model, the team created a vector that is better suited to cross the species barrier. This work suggests a new path forward for gene therapy: instead of hunting for a solution in the dark, scientists can now design their tools based on the universal map of the human body, potentially making treatments for blindness safer, simpler, and more effective for everyone.

Drowning in papers in your field?

Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.

Try Digest →