Q&A: A 'Knocked-Out' Mouse May Hold the Key to Better Pharmacology

In the next few years, scientists will complete the monumental task of ''reading'' the entire genetic code of humans. But since scientists don't know what the vast majority of the estimated 80,000 to 100,000 human genes actually do, making sense of the genetic code will take decades.

One promising approach is to disable individual genes. This can't be done in humans, but it can be done in mice. At Lexicon Genetics in Texas, researchers have figured out how to produce thousands of mice, each with one gene ''knocked out,'' or disabled. Dr. Arthur T. Sands, president and CEO of Lexicon, spoke with John Carey, Business Week's Washington-based science correspondent, about the challege and promise of cracking the genetic code.

Q: In the next few years, the entire ''sequence,'' or genetic code, of all human DNA will be read. What will it mean to have the entire genome in hand? Better drugs, for instance? And how long will it take?
A: The Genome Project is extremely exciting and is a revolutionary development. But [when it comes to new drugs and targets for drugs] everyone is looking for the low-hanging fruit. The reality is that most of the fruit may be on the top or the middle. One actually has to get up there. It will pay a rich return, but it is not a trivial undertaking.

Q: You believe that one of the keys to getting there is understanding the functions of the genes. And the best way to find the function of a gene is to knock it out in the mouse -- and see what happens?
A: The knockout mouse has become the gold standard, although no single technology will get us there. We focus on the knockout mouse because we think it is the best starting point. A company creates a genetic knockout first, then launches legions of chemists to produce potential drug candidates that attack the target. It is a truly rational approach to drug development.

Q: What are some examples of how knocking out a gene helps find its function?
A: There are dozens of published examples. One of the most famous is the dopamine inhibitor uptake knockout. The mouse displays addictive behavior, as if addicted to cocaine. Another is the knockout of an adenosine receptor that mimics the effects of caffeine.

Q: Not everyone sees the knockout mouse as the key to finding the functions of all these unknown genes.
A: Some people think of knockouts as modeling human diseases. That's not what they are really for. Rather, you use a knockout to model basic physiological effects. My point is that if you knock out a target [i.e. a gene], and see no effect on the mouse's physiology, then that target should be deprioritized as a target for drug development, since there must be some alternative pathway.

In general, industry believes that there are about 2,000 to 5,000 genes that represent possible targets in the genome. [Using various other methods], you can cull the 100,000 human genes down to 5,000, or maybe 10,000. Then you can ask where are these genes expressed. That can give clues as to what genes are important.

But those two pieces of information are only as far as you can go without turning to the actual biology. That's why you turn to knockouts to tell what a gene does.

Q: It sounds like rapid progress is being made.
A: We're moving headlong into making sense of this vast amount of information. The overall vision is that within 20 to 40 years, we would have all the relevant targets for intervening in human disease. Then we can hone in on the chemical interventions [i.e. drugs] for those targets. The result will be that drugs will get better and better, with greater efficacy and lower side effects.

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

19: We'll Have All the Genetic Pieces. Next, We'll Assemble the Jigsaw Puzzle

ONLINE ORIGINAL: Q&A: A ``Knocked-Out'' Mouse May Hold the Key to Better Pharmacology

Human Genome Project Information

To Know Ourselves: The Genome Project

The Human Genome Project: Links to related sites

E-Mail to Business Week Online

Copyright 1999, Bloomberg L.P.
Terms of Use   Privacy Policy