By David N. Leff

Most wanted among the criminal cancer oncogenes is ras.

This murderous mutated molecule is found in flagrante delicto at the scene of 30 percent to 40 percent of all human malignancies - in half of the colorectal, a quarter of the lung and 90 percent of pancreatic cancers.

The italicized acronym ras stands for "rat sarcoma," which is the rodent tumor where molecular biologist Edward Scolnick and his colleagues at the National Cancer Institute (NCI) cornered the mutant oncogene in 1978.

Non-mutated - benign - ras encodes a protein that controls normal proliferative responses in the body, and indeed acts as a tumor suppressor. So when a wound strikes that body, Ras protein calls up growth factors, which persuade cells that heal to proliferate and repair the lesion. What then persuades Ras to stop, to call off its pro-growth, cell-division activity when the body no longer needs it? That's a good question, the answer to which is being sought by anticancer drug discoverers at a score of pharmaceutical and biotechnology companies, plus academia and the NCI.

Normal regulatory mechanisms shut down Ras, unless its gene has been damaged, usually by a single point mutation. Then its mutant product stays stuck in the "on" position, continuing to make cells multiply indefinitely. At that point, the only way cancer clinicians have had to stop those tumors from growing and spreading is to cut them out with surgery, burn them off with radiation, or poison them with chemotherapy. As for innocent bystander cells - good luck.

Only in the present decade have oncologists fixed on the recently found oncogenes, such as ras, p53, p16 and a growing roster of others, as a way someday to move from the current kill-em-all shotgun onslaught against tumors to a more rifle-like effort at halting and reversing malignancies by interdicting their oncogenes

Ras proteins rendered oncogenic by their mutated ras gene "are specifically modified by an enzyme called farnesyltransferase [FT]," explained tumor biologist George Prendergast, at the Wistar Institute in Philadelphia. "Finding a small molecule that inhibits FT," he added, "is a Grail target that all the companies are looking for."

So is he, together with his lab colleagues at Wistar.

"We're interested in ways to inhibit the activity of Ras in these tumor cells," Prendergast told BioWorld Today, "as a way to attack cancer." FT gets in its dirty work initially by shunting aside isoprenoid, a molecule in the cell that's an intermediate in synthesizing cholesterol, for one thing.

On The Trail Of Protein X

"Enzymes called geranylgeranyl transferase [GGT]," Prendergast went on, "also shunt isoprenoid intermediates off onto proteins." All of these molecules are like concentric circular targets converging on the bull's eye of a therapeutic FT-inhibiting drug. But Prendergast's latest work suggests that there may be an alternative bull's eye, which he calls "protein X."

In 1993, Merck Research Laboratories in West Point, Pa. (where Prendergast was working at the time), and Genentech Inc., of South San Francisco, both announced, he recalled, "FT inhibitors that were quite potent in blocking Ras-mediated malignant cell transformation, and could reverse malignancy preclinically in some types of tumor cells - both in vitro and in mice. This raised a lot of excitement."

Prendergast is senior author of a paper in the March issue of the journal Molecular and Cellular Biology. Its title: "Cell growth inhibition by farnesyltransferase inhibitors is mediated by gain of geranylgeranylated RhoB."

Rho is his protein X.

"Ras is a small GTPase [guanosine triphosphate-cleaving enzyme]," Prendergast explained, "common in animal cells. The suggestion that emerged from subsequent preclinical studies was that while FT inhibitors [FTIs] can block farnesylation of Ras, whether that event is crucial for the antitumor properties of the inhibitory drugs is unclear.

"The idea was that maybe some other type of farnesylated protein, other than Ras - protein X - was the key target for alteration by FTIs; that perhaps members of the Rho protein family were relevant targets," he continued. "Rho, which stands for 'Ras homology protein,' is another kind of FT-like GTPase, which affects a normal cell's cytoskeleton and shape, but not its growth or viability. And it changes a malignant cell's shape.

"One way to think about this," Prendergast suggested, "is that for many years we've diagnosed cancer by looking at shapes of cells." That is, tumor pathologists peer through their microscopes at sample tissues from patients. Cells that stuck to the bottom of a petri dish in a single contiguous layer were adjudged anchorage-dependent, hence non-malignant. Those that piled up in a random heap demonstrated malignancy.

Location Changes Function

"What happens to this Rho protein," Prendergast pointed out, "is that it becomes unfarnyselated in cells, but then is immediately picked up by the enzyme responsible for geranylgeranylation. Rho retains its ability to associate with cell membranes, but its location in the cell changes, and we believe its function changes as a result.

The Wistar scientists had expected FTIs to interfere somehow with Ras activity. Instead, they found that FTIs affect Rho proteins, which regulate cell shape, motility and survival. As Rho levels are high in many metastatic tumors, they wondered whether the connection was causative or merely correlative.

"What we did in this paper, which favors causation," Prendergast recounted, "was to engineer a recombinant Rho protein that's only geranylgeranylated. Targeted expression of that protein was sufficient to mimic many of the inhibitory drug phenotypes. What the FT inhibitor has to do now is produce that protein.

"This is important because it tells us about drug mechanism - something the FDA wants to know. Clinicians want to know what to measure and what to look at in the cancer patients they treat in their trials. And the scientists want to know, because they'd like to find other compounds that would take advantage of the same anti-transforming principle as FTIs, which are surprisingly tumor-selective.

"Now the question is," Prendergast observed, "can we find a drug that will mimic Rho's activity? For example, can we make a piece of Rho?"

In pursuit of such questions, the Wistar scientist is moving his seven-person lab this June to Dupont Pharmaceutical's research laboratories in Wilmington, Del., where he will head the division of cell adhesion and apoptosis.