May 1, 2003 — By exploiting cells’ natural ability to repair genetic damage, University of Utah scientists developed a new method that dramatically improves the efficiency of the “gene-targeting” technology that has revolutionized biology and medicine. The method employs enzyme “scissors” to replace genes or knock them out of action.
“Gene targeting is a method for making specific mutations in a gene of choice,” says Dana Carroll, chair and professor of biochemistry at the University of Utah School of Medicine. “It has been used very effectively in studying the normal functions of genes in organisms from bacteria to mice. This has included putting the equivalent of human disease mutations into experimental animals so we can learn about the development and potential treatment of inherited disease.”
Carroll says that in the May 2 issue of the journal Science, he and his colleagues report “we have developed a new approach that improves the efficiency of gene targeting, and we have applied it to a specific gene in fruit flies.”
The method, which makes the “targeted” gene more susceptible to being replaced by a “donor” gene, “could make possible gene targeting in experimental organisms where it is now not possible, and could simplify gene targeting in mice and make it less expensive,” he says.
“Combined with the many existing and emerging genome sequences [genetic blueprints of organisms], this should speed the analysis of gene function by allowing the use of a wide variety or organisms to understand genes” and what each of them does, Carroll says.
If the method ultimately proves workable in people, it might be used to cure human genetic diseases by replacing defective genes with working copies, Carroll says, noting that current gene therapy experiments in humans add working genes to cells without actually replacing the defective versions.
“My guess is we will know in much less than 10 years whether we can get it to work in mice in an application similar to what would be used in people,” he says. “Whether you would actually do this in people would depend on more than whether it works. There are ethical considerations to permanently changing genes in people. Such procedures should not be undertaken without very thorough public discussion.”
Mario Capecchi, co-chair of the University of Utah Department of Human Genetics and a developer of gene targeting in mice, praised Carroll’s research.
“Any improvement in the efficiency [of gene targeting] will make it more likely that in the future it will be used to treat genetic diseases,” Capecchi says. “Efficiency is going to be a very important parameter with respect to feasibility.”
Gene targeting – old and new
Gene targeting works when a target gene is removed from a strand of DNA and replaced by a donor gene – a process called “homologous recombination.”
The mouse method used by Capecchi – who has won numerous awards for his role in developing gene targeting – involves removing embryonic stem cells from mice, and growing them in culture, where a donor gene is introduced to replace a target gene.
“The efficiency of this reaction is quite low, about one in a million,” meaning only one in a million stem cells accept the donor gene, Carroll says.
Capecchi’s method overcomes that by linking the donor gene with other genes that make cells resistant to an antibiotic drug if they accept the donor gene and susceptible to another antibiotic if they accept the donor gene in the wrong spot. The cells are then grown in culture with both antibiotics. The only cells that survive are those one-in-a-million in which the donor gene has replaced the target gene. Those cells then are inserted into mouse embryos, which in turn are implanted into female mice, allowing Capecchi to take several additional steps to breed mice with the desired genetic change.
Carroll says his new method in fruit flies resulted in a donor gene replacing the targeted gene at a rate of one in 50, and avoids the multi-step process Capecchi must use to find the one-in-1-million genes with the desired change. The technique also will make it easier to make more subtle changes in a target gene rather than completely disabling it. And the method may make it possible to replace genes in plants and in animals for which existing methods don’t work, including the nematode worm, zebrafish, frogs and mammals other than mice. Carroll plans to try his method in mice, and says it already is 10 times more efficient than the existing method of replacing genes in fruit flies – a method developed a few years ago by University of Utah biologist Kent Golic.
“If we could make it so we could achieve efficiencies between 1 percent and 10 percent in mouse embryos, we could eliminate the need for the embryonic stem cell technology altogether” by allowing donor genes to be injected directly into mouse embryos, Carroll says. “It would save lots of time and lots of money.”
Carroll’s method involves cutting the target gene so it is more likely to be replaced by the donor gene. Other researchers showed previously that when a gene is broken by radiation, chemicals or enzymes, a natural recombination process is activated to repair the damage.
Carroll and colleagues exploited that repair process using enzymes called zinc finger nucleases, which were developed by Srinivasan Chandrasegaran at Johns Hopkins University. The enzymes are pieces of protein that include “zinc fingers” – finger-shaped molecules that contain zinc and are designed to attach to a target gene – and a “DNA cleavage domain” – basically a pair of molecular “scissors.” The enzyme “scissors” find and cut the target gene.
“The cell sees that as damage, and initiates a process to repair that damage,” prompting the donor gene to replace the targeted gene, Carroll says.
The new study
To demonstrate the method, Carroll and colleagues designed zinc finger enzyme “scissors” to find and cut a fruit fly gene named yellow and replace it with a donor gene that was a deactivated yellow gene. The donor gene, along with genes that carried the code to produce enzyme “scissors,” were physically injected into tiny fruit fly larvae using a microscopic needle. Those larvae then grew into adult flies, which were mated to produce flies that contained both the donor gene and genes that produced the zinc finger enzymes. Those enzyme “scissors” were activated when the fruit flies were exposed to warm temperatures, a process called heat shock that flies normally use to activate protective genes.
“The yellow gene controls pigmentation on the surface of the adult fruit fly,” Carroll says. “Normally, the fly has a specific pattern of light and dark segments. When the yellow gene is mutant or inactive, the dark segments become yellow.”
That is exactly what happened, showing that the method successfully replaced the target gene – a working yellow gene – with a donor yellow gene that was inactive.
Carroll’s method incorporated Golic’s gene-targeting technique. Golic found that when a donor gene is excised from a fruit fly chromosome and then straightened out or “linearized,” it later is more likely to replace the target gene than if the donor gene remains coiled in a circle, the shape it takes after being excised.
The combination of Golic’s method with Carroll’s enzyme “scissors” approach produced the desired mutation (disabling the yellow gene) in 1.5 percent of all male fruit flies and 0.5 percent of all females. Golic’s method alone produces the desired mutation only in 0.2 percent of females and in no males.
Stimulating gene targeting and replacement by using zinc finger “scissors” to break the target gene “should work in all organisms,” Carroll says. The challenge will be to design zinc finger enzyme “scissors” that can break a wide variety of genes, and to find better ways of delivering the scissors and the donor gene to target genes in different organisms.
Carroll conducted the new study with Marina Bibikova, a former University of Utah research assistant professor who now works at Illumina, Inc., a San Diego biotechnology company; Kelly Beumer, a postdoctoral fellow in biochemistry at Utah; and Jonathan K. Trautman, a laboratory technician.