Researchers from the University of Pennsylvania report in the March 1 issue of Sciencexpress, the early online edition of Science, that they have managed to model an important feature of many diseases, the instability of trinucleotide repeats.

Trinucleotide repeat expansions are more or less DNA's equivalent of getting stuck in a rut. Repeats of three base pairs, they are a common feature of DNA. However, expansion of such repeats accounts for more than 30 human diseases.

In the polyglutamine class of those diseases, a CAG repeat occurs within the protein coding region of the gene. Because CAG is the DNA sequence that codes for glutamine, a DNA sequence with the trinucleotide repeat will code for a protein that contains a string of glutamines. Such repeats do not necessarily cause disease - as long as they are not too long.

That's where the problem with the instability comes in. "In normal alleles, the length of the repeat will be different in different people, but it will all be relatively short," senior author Nancy Bonini, a professor of biology at Penn, told BioWorld Today. But in the disease situation, the repeat becomes abnormally long, and then "the repeat is dynamically changing from generation to generation."

In practice, that means a parent with a fairly high, but still in the normal range, number of repeats is at risk for having children with a still greater number of repeats and outright disease. Proteins whose glutamine stretches are too long are toxic, and can cause a number of diseases including Huntington's disease and spinocerebellar ataxia.

The research was done using flies as model animals, Bonini said, because they are a complex organism, but have a short lifespan (making it possible to study disease features over many generations) and genetics that are easy to manipulate.

Instability, though, is one feature that has proved surprisingly hard to model in the fly. "It hasn't been easy coming," said Bonini. She and postdoctoral research Joonil Jung were successful by using flies outfitted with a human exon that is found in spinal cerebellar ataxia patients that have a high number of CAG repeats, and then expressing the gene in germline cells, of the egg and sperm.

The model still could use some tweaking. The flies show instability only in about 1 percent to 3 percent of cases, which is much lower than humans. But on the plus side, Bonini noted, "we are seeing very large jumps and very large contractions - in the range of 15-20 repeats," which is similar to what can happen in humans.

Jung and Bonini used their model to study possible molecular details of polyglutamine diseases that may affect repeat instability, focusing on Creb-binding protein or CBP.

Cells just don't take polyglutamine disease lying down. The DNA repeats that are at the heart of polyglutamine diseases "make unusual structures that get recognized by DNA repair machinery," Bonini explained.

But the polyglutamine proteins also affect that DNA repair machinery - among other things, by reducing the activity of CBP. And this appears to make instability self-reinforcing. Reduced CBP activity leads to "even greater instability of the repeat," Bonini said.

CBP works, among other things, by modifying histones, and Jung and Bonini found that treating the flies with an HDAC inhibitor reduced repeat instability. This set of experiments highlighted another advantage of fruit flies as model animals, particularly for neurodegenerative disorders. "They don't have much of a blood-brain barrier," Bonini said, which makes it easy to feed them compounds one wants to test.

Of course, critics might scoff that fruit flies don't have that much of a brain. Fruit flies are sometimes used to study higher functions such as learning, but Bonini readily concedes they have their limits for modeling complex diseases.

"We don't expect all features to be there," she said. But flies "are a useful model for studying molecular mechanisms in a simple, manipulable, yet still complex animal."