Science Editor

Splicing made headlines earlier this week when researchers from the National Institutes of Health reported that telomere-induced changes to the splicing machinery appear to be a major mechanism of cellular aging. (See BioWorld Today, Jun 14, 2011.)

Now, a separate study has concluded that splicing may have a larger role in hereditary diseases than is currently recognized as well.

"We are proposing that 22 percent of human mutations in the Human Gene Mutation Database have some kind of splicing defect," William Fairbrother told BioWorld Today. Fairbrother is an assistant professor at Brown University and the senior author of the paper describing the findings, which appeared in the Jun 13, 2011, online edition of the Proceedings of the National Academy of Sciences.

Adding those 22 percent to the diseases where alternate splicing has already been identified as the cause, nearly one in three hereditary diseases may have malfunctioning splicing at their root.

Splicing, which cuts out and pastes together the coding parts, or exons, of a gene after transcription and pastes them together into the final messenger RNA that will be used to make the protein, is commonplace: "It happens about 10 times, on average, in a human gene," Fairbrother said. And sometimes, alternative splicing is part of the genome's way of making slightly different versions of the same protein. But at other times, alternative splicing can wreak havoc on a protein's function. (See BioWorld Today, July 20, 2009.)

Splicing may be basic, but that doesn't mean that figuring it out was simple. "For a long time, it was not clear how the cell knew" which parts to cut, Fairbrother said. The RNA sequence contains "a little bit of information," on where to splice the unprocessed RNA transcript, he said, but not nearly enough. In an average intron of perhaps 100,000 base pairs length, "could have hundreds of sites that look like perfectly good splice sites, but are never chosen."

The discovery of enhancers and silencers – helper sequences located within the exons themselves – solved part of the mystery of how the spliceosome recognizes its work sites. They are "very dependent on position," Fairbrother said – "they had to be at a certain place" relative to the splice site itself. Some splicing proteins can either increase or decrease splicing, depending on whether the RNA sequence they bind to is in the exon or the intron of an RNA sequence.

In their new paper, Fairbrother and his team took an alternate approach to identifying splicing signals. They began from the idea that "we know where all the splice sites are," and went from there to analyze all the hexamers, or six-nucleotide sequences, surrounding them.

In a nutshell, the approach amounts to counting how often a given six-letter sequence appears around the known splice sites, and comparing it to how often that sequence would be expected. Fairbrother and his team soon found certain sequences "will either tend to be avoided around splice sites, or be more frequent."

After identifying potential splice signaling sequences, Fairbrother and his team tested those sequences further, to see how different their distribution was from what would be expected. They found that sequences with a larger difference between the expected and observed distributions were more likely to affect splicing.

The authors next went back to the bench and tested mutations associated with diseases including albinism and colorectal cancer that their program had predicted would affect splicing. They found that four out of the six mutations they tested did indeed affect splicing.

If splicing is in fact behind a large number of hereditary diseases, that could be good news in terms of treating such diseases – both because it might enable targeting of a few splicing signals, rather than a plethora of mutated proteins, and because the targeting itself could be easier than alternative approaches such as gene therapy.

"A processing defect," Fairbrother said, "may be able to be detected and fixed much more easily and safely than a protein coding defect."

Companies such as Isis Pharmaceuticals Inc. and AVI Biopharma Inc. are in the clinic with RNA-based therapeutics that attempt to influence splicing.