By David N. Leff

Picture a shortstop who feints a throw that fakes a runner out of stealing a base.

This decoying ploy has something in common with a just-discovered decoy receptor that keeps a cell from getting put out by programmed cell death -- apoptosis.

Assisted cell suicide engages more death-dealing molecules than there are players on a baseball team.

"The diversity is in the regulation," said molecular biologist Avi Ashkenazi, adding, "It seems that the machinery that actually carries out the execution of a cell is the same, or similar. But this is a common theme in biology," he pointed out. "Something that works well gets duplicated in many ways, which are variations on the same theme. Each is used in a particular biological context, say, within the immune or neural or endocrine system, and so on.

"Apoptosis is a very important process for a given cell," Ashkenazi told BioWorld Today. "You can make only one decision to die. So you need very effective ways of making that decision, and implementing it -- or not."

Molecular biologists are just beginning to identify the proteins and receptors that team up to trigger apoptosis, when its cell-signaling coach calls for that final play.

The newest name of their game is TRAIL -- TNF-related apoptosis-inducing ligand. TNF, of course, stands for tumor necrosis factor, which in turn stands for trouble.

At first, tumor oncologists bet heavily on TNF as being a cytokine that would kill tumor cells, while leaving healthy cells alone. But they soon found that TNF is too toxic to play such anticancer games in animal models, let alone humans.

So too was another player, FasLigand, originally scouted as a professional tumor-killing star. The Kevorkian-like Fas receptor, a key molecular manager of the apoptotic "death domain," belongs to the multitudinous family of TNF receptors. (See BioWorld Today, Dec. 19, 1996, p. 1.)

TRAIL, A Cytokine With A Difference

In 1995, Ashkenazi, at Genentech Inc., in South San Francisco, and immunologist Ray Goodwin, at Seattle-based Immunex Corp., simultaneously and separately cloned TRAIL (a.k.a. Apo2L). This new cytokine is related both to TNF and the FasLigand, which activates the Fas receptor.

"TRAIL," Ashkenazi told BioWorld Today, "was very much like FasLigand in its ability to rapidly trigger cell death in a wide variety of tumor cell lines. But unlike FasLigand, which is very toxic to animals -- it will kill a mouse in 24 hours -- TRAIL does not have that toxicity."

This raised one interesting question: "Why is TRAIL so cytotoxic to tumor cells, but not to normal cells? The explanation," Ashkenazi suggested, "seems to lie in the receptors that mediate the signal carried from the cytokine into the cell."

Now Ashkenazi at Genentech, Goodwin at Immunex, and apoptosis researcher Vishva Dixit at the University of Michigan, Ann Arbor, have come up with not one but three receptors for the TRAIL protein. Two of them, aptly dubbed DR-4 and DR-5 (for Death Receptor), communicate the cytokine's apoptotic signal to the cell's interior for execution.

But surprisingly, the third receptor acts conversely; it nullifies the death sentence.

"This is a decoy receptor," Ashkenazi explained, "that sits on the tumor cell surface but, unlike DR-4 and DR-5, has no cytoplasmic tail by which to relay the message into the cell's interior. However," he continued, "it binds to TRAIL and protects the cell from induction of forced suicide by that cytokine. It does so," he surmises, "by sequestering it away from the other two receptors."

Ashkenazi continued: "TRAIL is absent from virtually all of the tumor cell lines that we looked at. That is, normal, healthy cells that carry the decoy receptor on their surface appear to be resistant to it, whereas cells that don't -- i.e., many tumor-cell types -- are sensitive to the decoy."

This, the Genentech scientist sees as "significant."

For one thing, he observed, "it's the first example of a cell-autonomous mechanism of protection against cell death. That means an individual cell can protect itself against TRAIL by expressing this decoy receptor. Practically speaking," he went on, "we are interested in TRAIL because it triggers apoptosis, and because, for a number of reasons, it has potential as an anticancer agent.

"First, it seems to have a cytotoxic selectivity toward tumor cells vs. normal cells. Second, it activates apoptosis independently of the well-known p53 tumor suppressor."

Putative Clinical Approach Bypasses p53 Mutation

He explained: "Many chemotherapeutic drugs, as well as radiation therapy, also trigger apoptosis in tumor cells, but those approaches require intact p53 function. And it's mutations in the p53 gene -- which half of all tumor cells carry -- that render the tumors resistant to a lot of these treatment approaches."

He also pointed out that "if we combine TRAIL with some of these chemotherapeutic agents, we see a synergism in the killing of the tumor cells."

Ashkenazi is senior author of a paper in the current issue of Science, dated Aug. 8, titled: "Control of TRAIL-induced apoptosis by a family of signaling and decoy receptors." Back to back with this report is a nearly identical article by Dixit: "An antagonist decoy receptor and a death-domain-containing receptor for TRAIL."

Immunex scientists have a similar manuscript in press at the Journal of Experimental Medicine.

In recent weeks, Dixit has moved from Michigan to Genentech, where he now heads the department of molecular oncology, in which Ashkenazi is a senior scientist. Both submitted their papers to Science in June 1997. "We were aware of each other's work relatively late in the process," Ashkenazi observed.

"One interesting aspect of these two papers," he added, "is that we each discovered the same new molecules primarily by searching different databases for ESTs -- expressed sequence tags. Dixit's lab used the database of Human Genome Sciences, in Rockville, Md. We used the LifeSeq database of Incyte Pharmaceuticals Inc., in Palo Alto, Calif., plus our own internal database of secreted proteins."

Ashkenazi concluded: "One thing that these Science papers convey is the usefulness of those databases in discovering new molecules related to things that we already know, but keep surprising us with new functions." *