By David N. Leff

Baker's yeast may be the lowest organism on the eukaroytic totem pole (humans beingtops), but that doesn't meant Saccharomyces cerevisiae cells don't have a sex life.

"Yeasts engage in their own form of sex," observed molecular biologist Alexander Kamb."There are two mating types, A and alpha, and they communicate via pheromonehormones.

"When mating," Kamb continued, "one cell type secretes alpha factor. The other cell type— the opposite sex, so to speak — responds to that signal by cell-cycle arrest.By thus ceasing to grow, the yeast is basically preparing for mating. This involves the twocells getting together, fusing and mixing their DNA."

Kamb is vice president of research at Ventana Genetics Inc., in Salt Lake City. Yeast'ssexual union, as he described it, offered him and his company a model proving ground fortesting their new approach to drug discovery.

"That particular growth-control pathway," he explained to BioWorld Today, "is aparadigm for cell-to-cell interactions, cellular signaling and response. It is probably thebest-characterized, non-metabolic biochemical pathway known in eukaryotes.

"The central piece of this approach," Kamb said, "is our perturbagen technology. Wewanted to test in a well-understood system how well it worked. So what we sought to dowas to identify perturbagens that would abrogate this growth control pathway."

Kamb is senior author of a paper reporting that test in today's Proceedings of theNational Academy of Sciences (PNAS), dated June 23, 1998. Its titleis "Transdominant genetic analysis of a growth control pathway."

Kamb coined the word "perturbagen," he recounted, "by analogy with mutagens. But amutagen alters the DNA, whereas a perturbagen perturbs interactions between proteins, orbetween enzymes and their substrates.

"One of the big problems faced by the pharmaceutical industry," he went on, "is to finddruggable targets that can be inhibited by chemical entities, and in that process, alter cellphysiology in some interesting way.

"Perturbagens provide a tool to find those druggable targets — the cell's Achillesheel. They're basically a way of screening inside human cells for those proteins that arevulnerable to traditional kinds of pharmaceutical interventions. But doing it in a very high-efficiency way."

Curing Cancer In Mating Yeasts

Kamb described his screening experiment as "like trying to cure cancer in yeast," adding,"and I think what we found we're pretty excited about."

What he and his co-authors found "was, in a relatively simple experiment with a simplescreen, multiple perturbagens, which we could map to specific targets in the yeastgrowth-control pathway. In fact," he recounted, "we actually discovered a couple ofcomponents, members of the pathway, that were previously unknown — which isalso surprising, given how well-studied that pathway is."

Ventana is now "well into studying perturbagens in human cells," Kamb continued,"melanoma specifically, and viral infections with antivirals to herpes and adenovirus."Here's how he and his team are going about it:

"Basically, the approach is the same as in our yeast experiment," he said. "That is, wegenerate perturbagen libraries in much the same way we did for yeast, to introduce theminto human cells in culture. And then we use those perturbagens as probes to identify thetarget in the cell, to which they bind in exerting their physiologic effect, such as growtharrest in cancer."

Kamb's perturbagen approach "differs in a number of ways from other drug discoveryscreening methods," he explained.

"First of all, the perturbagen libraries can be thought of as biological combinatorialchemical libraries. But with a difference, in that one can perform genetics with them. It'seasy to recover rare events and amplify them.

"In addition, you can deliver them into cells using gene-transfer procedures that are well-established. So you don't have the problem of a cell-membrane barrier. You can also dothe experiments en masse, with large populations of cells. In the case of a small-molecule screen, you typically have to screen each compound individually in a well. Herewe can screen many millions of perturbagens, but en masse, in a culture dish orflask of tissue culture cells."

On The Trail Of 'Druggable' Small Molecules

"If you get a physiological effect in a small molecule," Kamb pointed out, "it's typicallyvery difficult to figure out the mechanism, because you can't attach labels and recover thecompound as easily as you can with a protein fusion. But you can use all the molecularbiology tricks to get you from a perturbagen to the target in the cell."

Kamb makes the point that "we are in this era of what some call post-genomics, whereeverybody takes as a given that you can get your hands on genes fairly easily.

"So the question is how to sift through all the numerous genes out there, to figure out themedically relevant ones. Our idea is that in perturbagens we have a technology thatbasically lets us sift through all the potential targets in a cell for those that really aredruggable, that can be inhibited in a transdominant fashion a la small molecule.

"What transdominant means in this context," he explained, "is that perturbagens act in thesame way a small molecule does. That is, it's produced, and then it acts on other moleculesin the cell. As an inhibitor it can float around in the cell and bind to proteins that arethemselves floating around. We produce these things that act on proteins, and therefore itdoesn't matter if there's one allele or two alleles actually producing the protein.

"Perturbagens are really designed to solve that problem, to pinpoint real validated targetsin human cells, without having to understand in great detail how all the bits and pieces of acell are put together and work together.

"We're trying explicity to avoid reverse engineering of a cell," Kamb concluded, "figuringthat that's a very difficult thing, which will take probably decades of intensive work. Sowe're trying to avoid deciphering all the secrets of the cell — how it does itsbusiness. But rather, trying directly to find the medically important proteins in a cell, thatcan be targets for pharmaceutical intervention." *