PHILADELPHIA – In 1990, scientists discovered a genetic mutation that turns an important light-sensing protein in the retina into a toxin that steadily destroys the ability to see in low light.

Similar mutations have been identified since then, but there is still no way to treat or prevent the loss of night vision.

Now, University of Pennsylvania researchers have used gene therapy to do just that in a canine version of the inherited retinal disorder.

The success in six dogs paves the way for clinical trials in humans, perhaps in just a few years, and comes as gene therapy raises hopes for turning debilitating hereditary eye diseases into treatable conditions.

The trailblazing blindness drug, Luxturna — developed by Spark Therapeutics and scientists at Penn and Children's Hospital of Philadelphia — was approved last December by the U.S. Food and Drug Administration to restore some sight in a different inherited form of blindness.

The new research was conducted by Penn's School of Veterinary Medicine and the Perelman School of Medicine in collaboration with the University of Florida. "This is a huge step forward," said Stephen Rose, chief scientific officer of Foundation Fighting Blindness, one of the funders of the research.

The light-sensing protein that turns destructive is called rhodopsin. Normally, this molecule — located in the rod cells of the retina — converts dim light into electric signals that the retina sends to the brain, which interprets them as images. But in mutated form, rhodopsin ruins the rod cells.

Rhodopsin gene mutations account for about 20 percent of all cases of retinitis pigmentosa, a group of inherited retinal diseases that rob people of night and peripheral vision before progressing to blindness.

Over the past 20 years, rhodopsin mutations have basically defied gene therapy, which involves using an inactivated virus, called a vector, to ferry corrective genetic material into cells. One challenge has been the dominance of rhodopsin mutations; a person who inherits just one copy of the defective gene develops the disease. What's more, about 150 mutations have been identified.

The Penn team overcame that by disabling both the abnormal and normal copies of the rhodopsin gene — so the specific mutation didn't matter — while simultaneously providing a healthy copy to make the vital protein.

"For 38 years, we've known one of the most common causes of retinal degeneration, and yet we've had no success in treating it," said Artur V. Cideciyan, an ophthalmologist at Penn's Scheie Eye Institute and co-lead author of the paper. "With these results, we can start working toward treating these mutations … in humans."