The blood-retina barrier normally prevents molecules in the blood from reaching the retina.
Presently, drugs to treat age-related macular degeneration, the leading cause of blindness in the west, have to be injected into the eyeball once a month, which can cause eye infections and even blindness in 1 out of 50 injections.
But now, researchers led by Matthew Campbell have now restored the vision of mice by temporarily weakening the blood-retina barrier to let drug molecules injected into the blood slip through.
Using a technique called RNA interference (RNAi), they could block the production of claudin-5-a protein that normally helps make the blood-retina barrier impermeable.
"It's proof of principle," New Scientist quoted Campbell as saying.
However, RNAi therapy is not yet safe enough for use in people, as it would breach the blood-brain barrier as well as the blood-retina barrier.
The team are already working on a way to block claudin-5 production in the eye only.
This method involves gene therapy using a modified virus that makes the RNA only in the presence of an antibiotic called doxycycline.
With the novel approach, patients would need only one initial injection into their eye to deliver the virus.
Then once a month they would take the antibiotic orally, and two days later the drug to treat blindness would be injected in their blood.
By then the blood-retina barrier would be sufficiently weakened to allow the drug to pass through.
Although the existing drugs to treat age-related macular degeneration would be too big to pass through the barrier, there are other molecules in development that are small enough.
Some of these could be taken orally when the barrier goes down.
The study has been published in Proceedings of the National Academy of Sciences.