Chimeraplasty Potential

As research advances, hopes rise, but efficiency and safety are still concerns

By Paul Smaglik

If single-gene disorders are akin to minor misspellings in the human genome, then it stands to reason that the biological equivalent of a word processor's search-and-replace function could correct them. Some researchers are hoping that chimeraplasty can be that tool. The biological software--a chimeric oligonucleotide constructed from both DNA and RNA--was invented and first tested in vitro in 1994. But before it can be "shipped" to the clinic, its developers and others must optimize it and debug it.

Gene therapy researchers, in particular, are enthusiastic about the tool, because correcting defects in existing genes rather than flooding a patient's body with duplicate, correct copies seems safer. The promise--and the elegance--of the chimeraplasty approach lured R. Michael Blaese away from his National Institutes of Health post and into a position with Kimeragen Inc., Newtown, Pa., firm that plans to develop and market the technology invented and patented by Eric Kmiec.^1,2 Kmiec is a Kimeragen founder and stockholder, who recently set up and became director of the Laboratory of Gene Therapy at the University of Delaware. "What's really neat about this oligonucleotide is that it's specifically designed to correct the misspelling of the gene that you are interested in," comments Blaese. The molecule, designed to be homologous to the target gene, contains the normal spelling of the genetic typo. The oligo then base-pairs with the endogenous genomic sequence. But because it contains a correctly spelled gene and the target gene contains a misspelling, the resulting mismatch triggers the cell's repair machinery, researchers think. But they need to understand the actual mechanism better, agree many, including Kyonggeun Yoon, a Thomas Jefferson University researcher who worked with Kmiec to refine the technique. She agrees that, with further refinement, the technique could be safer than viral-based gene therapy. "Immune reaction is the primary concern [with viral-based gene therapy], and so is nonspecific integration." The first concern is moot, because chimeraplasty uses a nonverbal approach; liposomes--with different cargo--are frequently used in medicine. Chimeraplasty also avoids the random integration problem because of its specific targeting mechanisms--unless, of course, other, unknown chemical pathways are also involved in repair. Yoon and a Jefferson colleague used the gene-repair technique to change color-producing cells in white albino mice to black. This in vitro experiment corrected a point mutation in the enzyme tyrosinase, involved in making melanin and in pigmentation. Within six days after the chimeraplasty procedure, tyrosinase activity restarted, resulting in melanin production.³ Still, the gene-repair technique is far from perfected, Yoon notes.

Animal Models

Kimeragen researchers and colleagues used this technology in vivo to first disable, then repair clotting ability in rats.⁴ In a more recent study, a Kimeragen-sponsored group treated the Gunn rat for jaundice.⁵ Those rats were bred to be defective for the enzyme that clears the toxin bilirubin from their livers. The rat models a rare disease, more common in Amish children, called Crigler-Najjar (C-N). Blaese and Kimeragen plan to conduct the first human clinical trial with chimeraplasty next year in some of those children, if they get U.S. Food and Drug Administration approval. Kmiec remains "cautiously optimistic" about using the oligo in clinical trials. "It's five years out of the box," he says. "It's pretty early, but it looks great." He would like to first better understand how the oligo initiates repair, perhaps by characterizing all related signal pathways. He's fairly certain the mut-s analogue HMSH2 is involved, but he suspects others. "We just haven't nailed down all the genes yet," Kmiec notes. "We now have some evidence that other pathways are involved, some of which may act to suppress the precision of the correction". Cliff Steer, a University of Minnesota researcher conducting the Kimeragen animal studies and working to set up the C-N trial, agrees that other pathways likely come into play. But mechanistic knowledge based on bacterial and eukaryotic experiments makes him confident that technique works by "actual removal of the single base and replacement of the single base." He thinks that the single biggest issue, as with conventional gene therapy, remains delivery. "If you can deliver this material to the cell, you're going to get it to work," Steer notes. "One of the reasons that there's been such controversy about this technology is that many different laboratories have tried it and have failed in significantly delivering this material to the cells. If you don't get a boatload of this material into the cells, the kinetics just doesn't favor the reaction." He estimates that getting 100 copies into the nucleus won't do the trick, but 1,000 or 10,000 might. For the liver, at least, Steer thinks he has a solution. "We're using a very specific targeting mechanism for the hepatocyte." To reach other organs or tissue where specific receptors can't be favorably exploited, Steer suspects that directly injecting "boatloads" into the organ may also work. "Then you don't need targeting."

Li-wen Lai, research associate professor at the University of Arizona at Tucson, in collaboration with nephrologist Howard Lien, has used chimeraplasty to correct another metabolic disorder in mice, carbonic anhydrase type III deficiency. However, the results differed widely, depending on the type of nonviral carrier she used. "There's quite a bit of variation of conversion--from 1 percent to 40 percent." However, pairing the most effective carrier with targeting may be more effective, she says, adding that Steer has done a "terrific job" applying organ-specific ligands. "I think that will probably improve our conversion rate." Tom Rando, assistant professor of neurology at Stanford University, agrees that low efficiency rates have thus far kept some gene therapy researchers away from the technique. "We weren't dissuaded by people's earlier technical problems." Rando and colleagues decided to apply chimeraplasty to a tissue type other than liver: muscle. They treated mice for muscular dystrophy (MD), with limited success: "Our efficiency is very low." That animal model experiment, he's quick to point out, doesn't necess arily mean that a treatment for MD in humans is on the horizon. "Most [MD] in humans does not involve point mutations." Chimeraplasty can theoretically correct mutations in up to three base pairs in a row, but most researchers are sticking with fixing one. However, Rando's experiment does suggest that tissue other than liver can be targeted. Still, treating large areas of muscles rather than the local area targeted in this proof-of-principle study could be very difficult; muscle, unlike liver, does not have a series of blood vessels leading directly into a specific region.

Targeting Plants

Two recent papers demonstrate that the technique pioneered in animals can also be applied to plants--with some caveats. Chris Baszczynsky, research coordinator with Pioneer Hi-Bred International of Johnston, Iowa, tried the technique to modify maize. "It wasn't directly transferable," he remarks. "There are a lot of differences between animal and plant cells." Because plant cells have tougher membranes than animal cells, the group used a particle gun--a physical method to force DNA into a cell's nucleus--instead of encasing the chimera in a liposome vector. They did see some change after blasting the oligos into the cells, but not as much as they need to make that technique an efficient way to modify plant cells to guarantee commercially desirable traits in crops.⁶

Even the best efforts resulted in a lower percentage of changed cells than in the animal liver literature. "We don't feel it's at the level of efficiency we'd like to see," Baszczynsky notes. In plants, the technique has limitations beyond delivery methods and efficiency levels. It can only be applied to plants with correctable phenotypes and will be limited to correcting more recessive traits, because at this point, modifying two alleles is out of the question, for practical purposes. Nonetheless, Baszczynsky sees promise in the approach. "We think it's doable in plants. We just need to get the efficiency up." Until that issue is addressed, Baszczynsky thinks the chimeraplasty could be applied as a basic research tool rather than as an applied genomic modifier. For example, introducing targeted mutations could help parse out specific gene functions in a complicated pathway. "We think it's a way to provide us a lot of information on what genes do that we don't know yet." He also agrees that understanding the basic mechanisms underlying chimeraplasty "could open up some other ways of using the approach." Gregory D. May, an associate scientist with the Noble Foundation of Ardmore, Okla., along with Kmiec, is doing just that. May, like Baszczynsky, applied the technique to plants, with limited results.⁷ But unlike Baszczynsky, May suspects the answer lies beyond delivery. "We feel that there are a sufficient number of molecules delivered to plant cells in our current methods," he comments. Understanding the basic ,echanisms behind the system should provide a way to boost the system's efficiency. "What's the rate-limiting step? Is it a particular set of proteins or cofactors that are involved in the mechanism?" They will also look closely at the pairing mechanisms, the mismatch-repair component, and the target-selection process within cells. "We've developed a cell-free extract system that will allow us to essentially biochemically reconstitute the key conversion mechanism in a test tube." He hopes that the high-throughput system will allow them to quickly analyze all parameters. May suspects that a good test of whether chimeraplasty really works the way some proponents think will be to try it on plants that have compromised DNA repair machinery. "Are these plant mutant lines capable of directing a conversion? If not, that gives us an insight into what may be involved in the process," he comments. That insight could lead to refining some useful software that, with work, could run on plant, animal, and human machines.

References

1. K. Yoon et al., "Targeted gene correction of episomal DNA in mammalian cells mediated by an RNA/DNA chimeric oligonucleotide," Proceedings of the National Academy of Sciences, 93:2071?6, 1996.

2. A. Cole-Strauss et al., "Correction of the mutation responsible for sickle cell anemia by an RNA-DNA oligonucleotide," Science, 273:1386-9, 1996.

3. V. Alexeev, K. Yoon, "Stable and inheritable changes in genotype and phenotype of albino melanocytes induced by an RNA-DNA oligonucleotide," Nature Biotechnology, 16:1343-6, 1998.

4. B.T. Kren et al., "In vivo site-directed mutagenesis of the factor IX gene by chimeric RNA/DNA oligonucleotides," Nature Medicine, 4:285-90, 1998.

5. B.T. Kren et al., "Correction of the UDP-glucuronosyltransferase gene defect in the Gunn rat model of Crigler-Najjar syndrome type I with a chimeric oligonucleotide," Proceedings of the National Academy of Sciences, 96:10349-54, Aug. 31, 1999.

6. R. Zhu et al., "Targeted manipulation of maize genes in vivo using chimeric RNA/DNA oligonucleotides," Proceedings of the National Academy of Sciences, 96:8768-73, July 20, 1999.

7. P.R. Beetham et al., "A tool for functional plant genomics: chimeric RNA/DNA oligonucleotides cause in vivo gene-specific mutations," Proceedings of the National Academy of Sciences, 96:8774-88, July 20, 1999.

The Scientist 14[1]:13, Jan. 10, 2000