BioWorld International Correspondent

LONDON The “junk” DNA in the human genome so-called because no one knows what its function is may not be so useless after all. Researchers have now described a novel type of genetic mutation, which leads to disease even though it does not lie in the part of the DNA that normally contributes information to make a protein.

The gene concerned is known as ataxia-telangiectasia mutated, or ATM. People who inherit ATM have the disease ataxia-telangiectasia. One of the abnormalities associated with the disease is a predisposition to developing cancer.

Francisco Baralle, director of the International Centre for Genetic Engineering and Biotechnology in Trieste, Italy, together with collaborators at the Hannover Medical School in Hannover, Germany, report their results in a letter to Nature Genetics (the March 11, 2002, advance online publication) titled “A new type of mutation causes a splicing defect in ATM.”

Baralle told BioWorld International, “This study began when we found a patient with ataxia-telangiectasia for whom the only change in ATM was in the middle of an intron in other words, in functional terms, it was in the middle of nowhere.” It is, he said, the first time that anyone has described a mutation of this kind.

The human genome is composed of DNA sequences called exons, which code for proteins; sequences that separate exons within the gene, called introns; and large intergenic sequences. Most of the latter two types of sequences lack an ascribed function and hence have been thought of as “junk” DNA. But the Nature Genetics paper showed that there is more function in junk DNA than anyone would ever have thought, Baralle said. “In this case, the lack of four bases in the middle of what everyone thought was junk DNA is causing a disease.”

Normally, prior to protein manufacture, the cell’s enzymes first splice the RNA, removing the introns and joining up the exons to provide the message required for making the protein. If researchers were to see a small mutation in one of the introns, Baralle said, they would probably assume that it was a polymorphism a small variation in the non-coding sequences, of a type that occurs quite frequently between normal individuals and is of little consequence.

This was certainly not the case in the patient Baralle and his colleagues were studying. Experiments carried out by the team showed that the mutation abolished a recognition site for a molecule called U1 snRNP. The name stands for U1 small nuclear ribonucleoprotein. The new sequence itself was termed the intron-splicing processing element, or ISPE.

Baralle explained: “Normally, U1 snRNP participates in the recognition of one of the boundaries of the exon called the 5’ splice site, defining the exons from the introns. In this case, however, U1 snRNP was found interacting with a sequence thousands of nucleotides away from any known exon of the gene. The function of this interaction, which has been unknown up till now, was revealed by the mutation in the patient, which abolished the interaction with U1. The lack of binding of U1 to this intronic region, the ISPE, resulted in the surrounding sequence being treated in the mutant like a cryptic exon that is, included in the messenger RNA and hence producing a non-functional, truncated protein.”

The group then tried moving the ISPE sequence to another exon. They introduced the sequence into exon 9 of the cystic fibrosis transmembrane regulator. The cells made the resulting messenger RNA without including exon 9.

In another set of experiments, they were able to “rescue” the function of the patient’s cells in vitro by introducing a modified U1 snRNP molecule that was able to recognize the mutated site.

Baralle said that the ISPE could play a role in a hitherto unknown RNA processing step. “The mutation that we have observed gives us the opportunity to look at something that has been missed up till now. We speculate that the ISPE takes part in normal RNA processing and that it may define an intermediate step in RNA processing,” he said.

Finally, an important lesson from this study, Baralle concluded, is that the team would never have made its discovery purely by computer scanning of the genome. “If you don’t have a functional assay for changes in the genome, then you may decide that something is just a polymorphism, simply because you don’t know what its normal function is,” he said.

Baralle pointed out that U1 snRNA may have therapeutic potential because it can correct genetic defects in vivo. “We have shown that when a modified U1 snRNP binds to an exon,” Baralle said, “it makes that exon disappear. This type of molecule has, for example, the potential to be used to knock out an oncogene that is causing cancer.”