Gene therapy strategy repairs RNA, not DNA.

In a significant test of a novel strategy for gene therapy, investigators have inserted genes for unusual enzymes called ribozymes into mammalian cells and repaired the faulty protein-making instructions sent out by a mutant gene. The ribozymes, which derive from single-celled pond organisms, had previously performed genetic repair jobs in test tubes and inside bacteria. Researchers were uncertain, however, whether these corrective enzymes could function within the radically different chemical milieu of mammalian cells. "It's fantastic that it's working," says Nava Sarver of the National Institute of Allergy and Infectious Diseases in Bethesda, Md., which funds several efforts to treat AIDS with ribozymes.

Unlike most enzymes, which are proteins, ribozymes are made of RNA, a single-stranded, DNA-like molecule consisting of sequences of chemical components called nucleotides. The ribozymes used in the gene therapy experiments have the unusual ability to splice part of their nucleotide sequence onto other RNA molecules, explains Bruce A. Sullenger of Duke University Medical Center in Durham, N.C., who heads the team that performed the recent gene therapy experiments.

That splicing talent has therapeutic potential because RNA is a vital cog in the cell's protein-building machinery. When a cell makes a gene's protein, it first copies the instructions stored in the gene's DNA into an RNA strand. This messenger RNA, or mRNA, then travels to sites in the cell where it directs the assembly of amino acids into proteins.

Two years ago, Sullenger and Thomas R. Cech of the University of Colorado in Boulder showed that in bacteria, ribozymes can repair the mRNA of a deliberately shortened gene. The ribozymes, engineered to include the missing portion of mRNA, match part of their sequence to a complementary sequence on the faulty mRNA and add on the absent nucleotide sequence.

Almost the same results have now been achieved in mouse cells, Sullenger and his colleagues reported last week at the RNA Society meeting in Madison, Wis., and in the June Nature Medicine. Even though they clearly observed the correction of mRNA in the mammalian cells, the researchers have not yet established that the repaired mRNA produces a functioning protein, as it did in their bacterial studies.

In addition to repairing the mRNAs for which they were intended, the ribozymes frequently spliced their cargo onto the mRNAs of other genes. That "sloppiness" can be overcome, asserts Sullenger. "We know so much about these ribozymes there's some obvious things to try to increase their specificity." Though many genetic diseases are caused simply by a dearth of a gene's normal protein, ribozyme-based gene therapy would be especially useful for diseases in which a genetic mutation results in the production of a damaging protein. RNA-correcting ribozymes would both eliminate these deleterious proteins and restore synthesis of the correct protein. "You swap out the bad information and put in the good information," says Sullenger.

Correcting genetic diseases at the RNA level instead of the DNA level offers another potential advantage-timing. A disease produced by a mutant gene will not necessarily be corrected by adding a copy of a normal gene into a patient's cells; investigators cannot yet ensure that the inserted gene will turn on and off when it should. In contrast, using ribozymes to repair mRNAs made by a mutant gene preserves the normal timing of the gene's activity, says Sullenger.