It's not exactly a stop-the-presses scoop that mutations in the genetic code underlie inherited diseases. But, Richard Gatti told BioWorld Today, "when you analyze what those mutations actually do, the focus has been very na ve until recently."

The simplest possibility (aside from a silent mutation that does nothing at all) is that a change in the DNA code leads to an amino acid substitution, which will in turn lead to a protein that is misfolded and so on, not functioning optimally.

But "there is a whole other language" beneath the classic code of DNA triplets making strings of amino acids, said Gatti, a professor of pathology at the University of California, Los Angeles' David Geffen School of Medicine. And one of the things that language describes is how to cut the relevant bits out of the original transcript and paste them together - a process known as splicing.

"Sixty to 70 percent of all proteins are alternatively spliced," he said. In other words, "you can leave out exon 24 or you can keep it in, and that will give you a different protein," Gatti added.

When it goes according to plan, alternative splicing is a comparatively frugal way for proteins to diversify - making a protein that is either membrane-bound or soluble, for example, from the same starting DNA sequence.

But splicing can, and does, go wrong. In fact, in a paper published this week in the early online edition of the Proceedings of the National Academy of Sciences, Gatti and his colleagues stated that "the majority of all inherited diseases result from mutations that cause aberrant splicing." In the paper, they show that mutations leading to cutting errors underlie the genetic disorder ataxia-telangiectasia, and that using a contraption known as antisense morpholino oligonucleotides, or AMOs, to mask such mutations can restore the affected kinase in cells from ataxia-telangiectasia patients.

Ataxia-telangiectasia (A-T) is a DNA repair disease. It results from mutations in a kinase that leave cells unable to properly repair DNA and, consequently, patients with a host of problems, from increased susceptibility to cancer to neurodegeneration. Half of those afflicted do not live past their teen years.

Many different mutations can cause A-T, which, Gatti said, is not all that untypical. "There are a few diseases where one mutation causes most of the disease - cystic fibrosis is a good example," he said. "But in the case of most other genes, there are many different mutations - the BRCA gene is an example. The larger a gene is, the longer it is, the more mutations you'll see."

That variety makes research on cells from A-T patients experimentally tractable. "We have many ways of testing whether the cells function, because the A-T cells have a lot of phenotypes that can be corrected."

Gatti and his team used AMOs to correct mutations in the A-T gene that led to improper splicing. Essentially, AMOs were "put as a patch over the mutation in the pre-RNA," Gatti said. For three different splice mutations, the procedure restored proper splicing and led to kinase levels that were anywhere from 5 percent to 25 percent of normal levels. While that may not sound like much, mutation carriers often are unaffected despite having kinase levels that are only 40 percent to 50 percent of what is normal - so 100 percent restoration probably will not be necessary to achieve clinical success.

The current paper follows another paper published by the same group in 2004 showing a way to correct another type of mutations in the kinase: premature stop codons that lead to truncated proteins, which affect around 30 percent of patients. Splice-site mutations affect about half.

"Our lab is working very hard on bringing the concept of mutation-based therapy to the clinic," Gatti said. The major obstacle is not in the scientific rationale, but in a much more pedestrian area: "We have the same problem as gene therapy - we need a delivery vehicle."

The reason is that, at a cost of roughly $3,000 per gram, morpholinos for A-T patients would be extremely expensive using general delivery. In fact, it would make the $10,000 a month cost of some monoclonal antibodies look like dollar store pricing. Gatti said that according to a rough calculation of his, the cost to treat a single A-T patient with AMOs "worked out to around $5 million," and that's assuming untargeted delivery would work at all.

"It stays in the bloodstream; that's not a problem," Gatti said. "The question is how we're going to get such an expensive compound to concentrate in the organs. Once that's solved, we've shown with this paper that we know exactly how to design [AMOs] for this disease."