By David N. Leff

Take a look at the circuit-breaker panel in your home or lab. Its rows of automatic switches protect the electrical wiring from meltdown caused by an appliance power overload or a short circuit. In more ways than one, that panel acts like the genes in the human genome. These encode the myriad proteins that build, constitute and operate every living organism - whether mammal, bird, insect or microbe.

"Protein activity in cells is regulated through the catalytic activity of enzymes," observed organic chemist Kevan Shokat at the University of California, San Francisco. "The most common way that it's regulated," he explained, "is by protein phosphorylation - that is, adding phosphates to their molecules. This ubiquitous process is carried out by enzymes called protein kinases. To give you a scale, it seems that 30 percent of all proteins in the human body are phosphorylated - and 1 percent are protein kinases [PKs]."

"I think of these phosphorylation events," Shokat went on, "as microscopic switches. Adding a phosphate group can turn a protein on; removing it can turn it off. So phosphorylation is the most common switch, really, in all biology. And those circuit-breaking switches are turned on and off on a millisecond time scale.

"But there are so many protein kinases," he continued, "upwards of 900 per genome - that it's difficult to isolate and study one at a time. So for 30 years people have primarily used genetics to study protein kinases. One of the more common ways to investigate a protein is just to inactivate it - knock out the gene for that kinase. Then typically, when people look at those genetically altered organisms, they observe one of three things:

"First and most common, that there's nothing they can see that's different. That's because one of the other 900 kinases took its place. It's the biggest protein superfamily, so there's always a surrogate waiting around to move in and take over.

"A second frequent event is that the embryonic KO organism can't grow because you've inactivated a protein kinase that is regulating a switch that's important in development.

"And a third possibility: You thought you'd inactivated a kinase because you're interested in the immune system. And what happened is that the transgenic mouse is born missing a leg. The gene deletion has affected some other cellular pathways that you didn't know about."

Chemical Catalysis Beats Gene Knockout

Shokat pointed out, "The genetic KO process we were observing happened much faster than the tool we were using to perturb it. So we made the obvious leap and said that genetics is not ideal for knocking out PK function. If we had small-molecule inhibitors, that would be great, because we could add them to the organism, and see the effect within the next 5 seconds. But the large size of the PK family is the problem again. Nobody has been able to identify perfectly specific inhibitors of individual kinases. They always inhibit two or three or 10 at once. If we could develop a PK inhibitor that inhibited two or three kinases out of the entire 900, that would be worth a big publication.

"We decided even that wasn't satisfactory. We wanted something perfectly specific. So we developed this combination system, where we used genetics to sensitize one protein kinase, to be inhibited chemically by a small molecule that doesn't touch anything else.

"And the genetics is basically a protein engineering approach where we changed the active-site pocket of one kinase molecule so it's different from all others. Then we used that difference as a key design element to construct small-molecule PK inhibitors.

"They're great for research purposes," Shokat pointed out. "However, it's even more significant when you think about making drugs because protein kinases are about to explode as inhibitor-drug targets, they being the central switches. So the idea would be we rewire a diseased cell by inhibiting a key kinase, and get it to go back to a non-diseased state.

"No drug company," Shokat pointed out, "can make PK inhibitors quickly enough to test whether they're really going to work in terms of curing the disease, and not causing a side effect. So that's where our system - which we can immediately make specific and use - has a real advantage."

Shokat is senior author of a paper in the current issue of Nature, dated Sept. 21, 2000, titled, "A chemical switch for inhibitor-sensitive alleles of any protein kinase."

The co-author's KO compound against a targeted PK works by binding in the active site of the kinase, at the location where ATP (adenosine triphosphate) binds.

"There's one amino acid residue in that ATP-binding site," Shokat noted, "which is located in the back of the pocket that is conserved in all protein kinases. It's always a bulky amino-acid side chain, and we mutated that to the smallest amino acid side chain - glycine."

Shokat originally conceived his concept in yeast cells. Then he and his team tested the PK-curbing compound in mammalian cell lines, and most recently in vivo in mice. "In the mice," he recounted, "we mutated a brain kinase, then spiked their drinking water with our orally available inhibitor. It altered the animals' behavior in terms of their ability to do learning and memory tasks. So we've made a link between a particular kinase in the brain and particular types of cognitive functions, such as conversion of short-term memories to long-term memories - memory consolidation.

"We chose that process," he added, "because it happens within five days during mouse development. No genetics can operate with that type of time window. Humans," Shokat said, "have the same memory-consolidation process, and mice are considered a pretty good mimic of that. They have the same sort of physiology in the brain; their hippocampus is organized the same way."

Second Opinion: 'Beneficial Breakthrough'

Princeton University, where Shokat did the decisive PK inhibitor research, has licensed the technology to Cellular Genomics Inc., of New Haven, Conn. "They're starting to manufacture the compound, and use it in their own experiments," he concluded.

At Glaxo-Wellcome Inc. in Research Triangle Park, N.C., chemist Stephen Frye, who heads that firm's medicinal chemistry department, commented: "In Kevan's technology, the drug design he has developed is beneficial, really a breakthrough, because engineering the protein's active site and the inhibitors, you can more readily attribute actions of the compound to on-target effects. Inhibiting the kinase of interest, I think, is very significant for drug discovery."

n