Better animal models for genetic defects.

Better animal models for genetic defects

If researchers want animal models with one of 2,000 known genetic diseases affecting humans, they can expose laboratory animals to chemicals or radiation and have a million-to-one chance of getting the desired mutation in any given animal.

However, a new technique developed by University of Utah biologists may improve the odds to even money, perhaps making it easier to understand why mutated genes cause such diseases as cystic fibrosis and muscular dystrophy. Researchers want to know, for instance, whether mutated genes underproduce or overproduce certain substances, such as enzymes.

Eventually, the technique, which makes good and bad genes interchangeable, may allow researchers to reduce the occurrence of genetic diseases in humans. They also may be able to place mutated human genes in mice to see which of the 50,000 genetic defects in humans they cause, according to Mario R. Capecchi, whose report appears in the Nov. 6 CELL.

"With this method, we can change the gene the way we want it,' he told SCIENCE NEWS.

Using a variation of gene therapy that researchers first used in 1980 to produce a black-and-white-haired mouse by injecting a black-hair gene into an albino mouse embryo, Capecchi and postdoctoral fellow Kirk R. Thomas mutated the human hypoxanthine phosphoribosyl transferase (HPRT) gene and successfully injected it into mouse stem cells. Stem cells are embryo-derived cells that have not yet decided what they want to be.

The next step will be to inject the altered cells into mouse embryos, which would then express the mutated gene. The cell insertion step, Thomas says, is difficult, but it has been done in several other laboratories.

In humans, the mutated HPRT gene causes Lesch-Nyhan syndrome, characterized by mental retardation and self-mutilation, including finger biting, eye gouging and head banging. The normal HPRT gene produces an enzyme that converts a nucleic acid, guanine, into precursors for RNA and DNA. It is not known why reductions in the enzyme cause the syndrome.

To accomplish the change, Capecchi and Thomas went against a common perception among scientists: When DNA strands are injected into stem cells, they will randomly exchange information with other genes, but only 1 of 1 million interchanges will be correct. The researchers showed that they could increase the number to 1 of 1,000 by using larger strands and that they could increase the odds to even money by setting up a selection system that allows only the cells undergoing the preferred recombination to live.

The two researchers placed modified guanine into a dish of stem cells. Those cells undergoing the unwanted recombination died because the HPRT gene manufactured an enzyme that tried to convert the modified guanine into RNA and DNA precursors. The few remaining cells survived because they made the precursors from smaller building blocks. It is not known why a similar process is not sufficient in humans with mutated HPRT genes.

While this procedure will allow interchangeability between good and bad genes in mouse stem cells for experimental purposes, the same techniques may be used in human bone marrow, which would make the change only in that individual. Exchanging good genes for bad in human stem cells, however, poses technical and ethical problems, Thomas says.