In gene therapy promoters, too, less is more.

That is the conclusion reached by Lynda Ostedgaard, associate research scientist at the University of Iowa Medical College in Iowa City and her colleagues at the College of Medicine and the National Institute of Dental and Craniofacial Research in Bethesda, Md.

Their research is published under the title "A shortened adeno-associated virus expression cassette for CFTR gene transfer to cystic fibrosis airway epithelia" in the Feb. 19, 2005, issue of the Proceedings of the National Academy of Sciences, now available online.

Cystic fibrosis is a grim reminder that identifying the genetic basis of even a monogenic disease does not mean that either therapy becomes trivial or a cure waits around the corner. The defective gene causing cystic fibrosis was identified in 1989, but though the discovery improved both diagnostics and treatment of CF, there still is no cure and the average life expectancy of cystic fibrosis patients is a paltry 30 years.

Different mutations of the same gene can cause cystic fibrosis, but the net defect is simple and always the same: a nonfunctional chloride channel. While other organs are affected by that defect, as well, it usually is the lungs that give cystic fibrosis sufferers the most trouble. Because the movement of chloride ions in airway epithelial cells is blocked by the defective channel, the mucus coating the lungs cannot be kept moist by normal osmotic processes. As a result, the patient becomes prone to bacterial infections.

Because cystic fibrosis is a monogenetic disorder, and because transfecting just one cell type - the airway epithelial cells - might be able to help alleviate a large chunk of symptoms, gene therapy has long been considered a promising line of attack against cystic fibrosis.

Adeno-associated virus 5, or AAV5, has several advantages that make it an attractive vector for gene therapy. It does not cause any known human diseases, but there is a receptor for it on epithelial cells, meaning it can be transported reasonably easily into those cells without wreaking havoc once it has arrived there. Also, AAV5 can transfect non-dividing cells, and messages expressed via an AAV5 vector tend to last longer than those expressed by other gene therapy vectors.

However, AAV5 also has some disadvantages, one being that it cannot be used to package DNA of more than about 5,000 base pairs in total length. Given that the chloride channel gene has 4,450 bases of coding sequence, things get crowded quickly in the AAV5 vector.

"The full-length [gene] is too long to be used with any of the promoters that have been used to date," Ostedgaard told BioWorld Today.

Ostedgaard and her colleagues in the laboratory of Michael Welsh already published work on shortening the gene itself, which showed that a clipped but still functional gene can be created by deleting certain portions of the transporter gene. In the research now published in PNAS, they turned their attention toward shortening the expression cassette support staff.

"We started with the idea that there were certain regions we could get rid of" in the supporting cast, as well, Ostedgaard said. "We needed something that worked well and something that was short."

The scientists looked at three parts of the expression cassette. The main focus of the research was on shortening the promoter itself, which is borrowed from another virus. The scientists also shortened inverted terminal repeats, inverted sequences, and the poly-A tail, all of which are not, technically speaking, parts of either the gene itself or the promoter, but necessary for gene expression to occur. They also tested whether an intron would enhance expression of the transgene.

By tinkering with inverted terminal repeats, inverted sequences, poly-A tails and introns, Ostedgaard's group managed to shorten the length of those expression cassette elements to 480 base pairs. They then constructed two vectors by combining the optimized elements, which differed only in the length of the cytomegalovirus promoter, which was 173 base pairs long in the larger expression cassette and 113 base pairs long in the shorter one, for a total length of 4,940 and 4,880 base pairs, respectively. Both vectors were used to transfect cultured fibroblasts from cystic fibrosis airway epithelia.

Two to four weeks later, the scientists measured chloride currents in those cells and found them improved. In fact, because there is strong variation in chloride currents in normal epithelial cells, Ostedgaard pointed out that 28 days after transfection, which was the latest time point tested, chloride currents in the transfected cells were "in the same ballpark" as the controls.

She noted that while her group wanted to express the chloride channel transgene, the expression cassette could be used for other genes, as well.