Contact

Duke Health News

919-660-1306

DURHAM, N.C. -- Even before the first generation of human gene therapy trials yielded spotty results, a new generation of researchers had begun rethinking how best to correct faulty genes.

Duke University Medical Center's Bruce Sullenger believes successful gene therapy may lie not in correcting faulty DNA, the storehouse of genetic information, but in correcting the RNA, the messenger of that genetic information. He has already begun exploring use of the new therapy approach for sickle cell anemia and even to "sabotage" the AIDS virus.

Sullenger outlined his strategy and results to date at the 33rd New Horizons in Science Briefing, held this year at Duke University. The briefing is organized by the Council for the Advancement of Science Writing, a non-profit group of journalists aimed at enhancing the quality of science writing. Besides Duke, this year's briefing is supported by the Burroughs Wellcome Fund, with additional support from Sigma Xi, the Scientific Research Society, and by Baxter International Inc.

"Our strategy is to employ the cell's existing editing machinery to correct faulty messenger RNA by performing a type of molecular reconstructive surgery." Sullenger said. He said his research, sponsored by Duke's department of surgery and the National Institutes of Health, has already shown that the strategy works in bacteria, and preliminary experiments with mammalian cells are under way.

The key to Sullenger's strategy is a molecule called a ribozyme, a type of RNA that can find a specific sequence of RNA code, cut out a section of the RNA and splice in another section. Ribozymes, discovered in 1982 by Thomas Cech of the University of Colorado and Sidney Altman of Yale, are now known to be key players in editing the flow of genetic information from DNA to protein. Before ribozymes' discovery, which earned Cech and Altman a shared Nobel Prize in chemistry in 1989, scientists believed only proteins could perform such cutting and splicing activities.

"The fundamental role of RNA molecules in cells is to help manage the use of genetic information," Sullenger said. "Therefore, RNA seems the most logical place to try to correct faulty genetic information." To do this, Sullenger has harnessed the power and specificity of ribozymes to correct defective genetic messages in living cells. In initial experiments, he designed a ribozyme that could find a defective genetic message inside a bacterial cell and replace faulty genetic instructions with corrected genetic information. The corrected genetic message was then translated into an enzyme whose activity could be measured in the living cells. His current studies are focusing on demonstrating that the same strategy could work in mammalian cells. Preliminary results from these experiments are encouraging, Sullenger said. He said there are problems with the first attempts at replacing defective genes, which enlisted viruses to carry "good" copies of genes into cells and integrate them into the human genome. Viruses aren't designed to carry large pieces of DNA, so genes have to be inserted without the normal regulatory segments that control production of the protein product. What's more, he said, researchers can't control where in the genome the virus puts back the correct gene. In addition, many genetic diseases are caused by expression of a defective gene, so simply adding back a normal gene to cells doesn't decrease the amount of the "bad" gene product.

"Simply delivering a gene who's expression won't be regulated correctly to cells via a viral vector is a shotgun approach to gene therapy," Sullenger said. "It's like noticing the instructions for building a radio are missing a key step to tighten bolt B after step 3. If you randomly add back the missing directions, you lose the original context. It may be okay to tighten bolt B earlier or later in the process, but chances are that the order of the steps is just as important in the process. That's what we're facing when we add back genes with gene therapy. The RNA repair approach may allow one to maintain the original regulatory context of the gene."