By David N. Leff

Back when families baked their own bread, and some home-brewed their own beer or wine, baker's yeast was an important item on the grocery shopping list.

With the advent of genetics, and now biotechnology, Saccharomyces cerevisiae has a whole new career — as a stand-in for the human genome.

That single-cell eukaryotic fungus possesses only 6,000 genes or so, compared with Homo sapiens' estimated 70,000. All 12,057,500 base pairs of DNA in the yeast's genome were sequenced about a year ago. (See BioWorld Today, April 25, 1996, p. 1.) The human genome runs to 3 billion.

The big news is that a sizable fraction of S. cerevisiae's genes are virtually interchangeable with those of H. sapiens'.

A paper in the current issue of Science (March 21, 1997) reports efforts by a biotech start-up company to discover anticancer drugs, based on this yeast-human homology, as applied to the Ras protein. Its title: "Modulation of Ras and a-factor function by carboxyl-terminal proteolysis."

"Ras is a protein that functions in the cell as a molecular switch," explained molecular biologist Matthew Ashby, director of biology at Acacia BioSciences Inc., in Richmond, Calif., and one of the Science article's three co-authors.

"Ras transmits signals to the cell from outside," he told BioWorld Today, "either to divide, differentiate, or remain quiescent. It's an on-off switch, which can become locked into an 'on' state by a whole number of mutations. And those Ras mutations also turn up in a number of very important tumors."

For Ras to function in this oncogenic way, Ashby said, "three of its terminal amino acids must get clipped off."

This post-translational step, he pointed out, "potentially could be a pharmacological target to inhibit Ras function, and thereby Ras-based tumors."

Seven-Year Quest Finds Tumorigenic Yeast Protease

When Ashby and his mentor, geneticist Jasper Rine, at the University of California, Berkeley, started this line of research seven years ago the amino-acid-cleaving protease remained to be found.

"Finally," he said, "through some genetic trickery, we were able to isolate the defective mutants, and then clone what turned out to be two — not one — protease genes. In yeast, we determined, Ras processing uses only one of the two, which we named Rce1. It appears that humans have homologues for both proteases. So we think, though it's not proven, that the human homologue of Rce1 will be the Ras protease."

Working in yeast, the group reported, "if you knock out the gene for Rce1, the enzyme that does Ras, the cells appear completely indifferent to its loss.

"On the other hand, if you take a yeast cell that has a Ras that's activated [stuck in the 'on' position], then it has very characteristic changes, such as aberrant growth — cancer. There's a number of mutations in Ras that lock it into this 'on' state.

"And in yeast and human cells, those genes are functionally interchangeable.

"If you delete Ras from yeast cells," Ashby went on, "they're dead. They must have Ras in order to survive. And the same goes for a human cell."

When the team knocked out the Rce1 protease in a Ras-normal cell, "we couldn't tell that there was anything wrong with the cell. However, knocking it out in a cell that has an activated Ras, this aberrant form is largely returned to normal."

Ashby pointed out that "one potential problem for designing a therapeutic is that completely inhibiting Ras function would be lethal to all cells. But if you don't kill the cell, you cause a lot of normal cells to be affected badly — manifested in the clinic as side effects and toxicity."

For example, he continued, "If you give a person with pancreatic cancer a drug, 99.99 percent of their cells are normal; you wouldn't want the drug to hurt those. So in an ideal setting, it would selectively affect cells that have activated Ras, and not cells that are normal."

He and his co-authors "have made an inhibitor that seems to act the same in yeast, in amphibians and in small mammals."

Now they are "very interested in isolating the human gene. We want to really assess the viability of the protease as a target for anticancer therapy."

Acacia Plans Instant Gene-Profiling Service

Ashby and Rine are scientific co-founders of Acacia, together with Nobelist Walter Gilbert, of Harvard University, and chemist C. Dale Poulter, of the University of Utah, Salt Lake City.

The University of California has licensed to Acacia rights to U.S. patent No. 5,569,588, "Methods for Drug Screening," issued Oct. 29, 1996. Its inventors are Ashby and Rine.

Acacia calls its trademarked method "Genome Reporter Matrix."

"We don't do drug screening; we do drug profiling," Ashby pointed out. "Screening is where you're looking at one target. We're characterizing the complete biological activity of any one candidate compound. We will be knocking out the 6,000 yeast genes, one at a time. Then see how the absence of that one gene affects the expression of all the other 5,999 genes."

He described a hypothetical example of this service to drug discoverers:

"Say they give us 10,000 chemicals. We can profile them, and tell the client: 'You have 18 lead compounds that inhibit cholesterol biosynthesis, three that inhibit this Rce1 protease that we're working on, and 20 that inhibit various map kinases.'"

Ashby observed, "It seems that about every week, there's another human disease gene identified. A quarter or more of them have homologues in yeast, so you can imagine that if tomorrow somebody tells us they've just identified a human gene for pick-your-favorite-disease, and it's announced in the media.

"Then we just go to our data base the same day, and figure out that we already have 27 lead compounds for that gene. Other companies would have to figure out, A, What does that gene product do? B, Assay. C, high-throughput. Then screen other libraries. That can take years. We could do it within a minute."

Much of Rine's research for the past dozen years at the University of California has been funded in part by the National Institute of General Medical Sciences. Its program director for cell biology and biophysics, Jean Chin, told BioWorld Today anent the current Science report:

"They've been looking for this protease for a long time, before finding the ultimate step that identified what it was that did what it did. This is a real breakthrough toward finding a good target drug to inhibit Ras activation." *