Published online 3 April 2005 | Nature | doi:10.1038/news050328-13


Genetic patch treats 'bubble-boy' disease

Targeting sequences may prove key to successful gene medicine.

Earlier treatment for the disease involved encasing the patient in a plastic bubble.Earlier treatment for the disease involved encasing the patient in a plastic bubble.© AP Photo

A technique that can correct the genetic errors behind human disease has recently been demonstrated by US scientists. If the method proves to be reliable, it could be used to make astonishingly specific repairs anywhere in the genome.

Much has been promised by gene therapy, the process that compensates for a faulty gene by adding working copies of it to cells. But the technology has faced numerous setbacks, partly because introducing extra genes can trigger dangerous and unexpected side-effects.

The new technique takes a different approach: it corrects the mutation in the original, disease-causing gene. And the latest study, published online by Nature1, shows that the method can successfully patch up the faulty gene underlying a rare disorder called severe combined immunodeficiency disease (SCID). This condition, which leaves children vulnerable to life-threatening infections, is also known as 'bubble-boy disease' because early treatments involved encasing patients in a protective plastic bubble.

Find and replace

To repair the faulty gene, a team led by Michael Holmes of Sangamo Biosciences in Richmond, California, first engineered a protein that was able to comb through the millions of letters in the human DNA code to find a single 24-letter sequence within the defective gene that causes SCID. They fused this protein to an enzyme that snips through DNA at that site.

The researchers aimed to insert this enzyme into human cells, along with a fresh piece of DNA that matched the defective gene but lacked the mutation. Severing the DNA triggers a repair process called recombination, in which the cell attempts to patch up a cut by swapping the entire region with an undamaged segment, in this case, the unmutated DNA sequence.

In a trial experiment, the team showed that it could alter the gene underlying SCID in around 20% of human immune cells, which are the cells destroyed by the disease. The technique cannot currently be used to correct cells within the body, but the researchers think they could extract stem cells, which give rise to the affected immune cells, correct their mutations, and put them back into the patient's body to restock the immune system.

Researchers at Sangamo are also developing their technique to tackle HIV. Rather than correct a mutation, the idea is to introduce a defect into a gene in immune cells, so that they no longer allow the virus to enter.

Indeed, the study suggests that researchers could design similar enzymes that recognize and cut DNA at any point in the human genome, potentially triggering almost any change or repair. "It's very powerful," says geneticist John Wilson of Baylor College of Medicine in Houston, Texas. "I'm really enthusiastic about this."

Tailor made

Researchers discovered more than 20 years ago that they could entice mammalian cells to swap one segment of DNA for another by recombination. But the process occurs spontaneously in only one in a million cells: far too few to have any impact on disease.

In the past few years, however, scientists have begun to tailor enzymes like the one used by Holmes, which snip DNA at a particular spot and hugely bump up the rate of recombination. They do this by mixing and matching an array of proteins that recognize and bind to specific three-letter sequences of DNA. By sandwiching together eight of these proteins, for example, they can design one that binds to a chosen 24-letter sequence.

Holmes's team is the first to design an enzyme that successfully recognizes a mutated gene underlying human disease, and to show that it prompts efficient gene repair. "It's terrific," says Dana Carroll, who studies these enzymes at the University of Utah in Salt Lake City.


The team hopes that this technique will prove a safer way of treating SCID than traditional gene therapy. A French clinical trial of the latter has been stopped twice because three children developed cancer: probably because the corrective gene, when added to their cells, inserted itself next to a cancer-causing gene.

The gene-correction technique ought to avoid this problem, because the new DNA replaces the old copy at the same site. "The genome is left unscarred," explains Philip Gregory, Sangamo's senior director of research.

Researchers acknowledge that there are still potential pitfalls. It may, for example, prove difficult to engineer enzymes that recognize particular spots in the genome. And there are concerns that the enzyme might snip and damage DNA at an inappropriate place. "We don't want to be cutting up the genome willy-nilly," says Wilson. 

  • References

    1. Urnov F. D., et al. Nature, Advance Online Publication doi:10.1038/nature03556 (2005).