By David N. Leff

"Take two aspirin tablets and call me in the morning."

That's the classic tongue-in-cheek scenario for modern doctor-patient interfacing. Central to the savor of this set-piece is the all-purpose, risk-free rep of aspirin and its look-alike non-steroidal pain-killing drugs.

Aspirin — acetylsalicylic acid — has been with us since the German chemical firm of Friedrich Bayer developed and launched the drug in 1899. But its analgesic properties were well known for centuries, if not millennia, ever since — according to legend — Hippocrates (460-377 B.C.) prescribed boiling the bark of a willow tree as an aspirin-precursor potion to relieve aches, pains, fevers and inflammations.

Now, millions of Americans pop a daily pill of low-dose "baby aspirin" — 81 milligrams — to ward off heart disease. The rationale for this cardiac prophylaxis is that aspirin binds irreversibly to blood platelets, thus inhibiting the formation of clots that dam coronary arteries.

That's the plus side. On the minus side, habitually-used regular high-dose aspirin is a rot-gut molecule that damages the gastrointestinal tract's inner lining, to cause bleeding, ulcers and kidney damage — but prevents colon cancer. At the root of these side-effects are two yin-and-yang enzymes, cyclooxygenase-1 and cyclooxygenase-2 (COX-1 and COX-2). Their genes reside on human chromosomes 9 and 1, respectively.

COX-1 thins the blood, which prevents the aggregation of clot-triggering platelets. But COX-1 inhibition also irritates the stomach; COX-2 takes care of aspirin's analgesic and anti-inflammatory business.

Biochemist Lawrence Marnett, at Vanderbilt University School of Medicine, in Nashville, Tenn., has been working for a number of years on the interaction of aspirin with COX-1. "And we also had some ideas," he told BioWorld Today, "that might help us design a molecule that would preferentially react with COX-2."

Both enzymes, he added, "are very similar in overall structure, but there are some critical differences in the region of the protein where they bind a substrate molecule, arachidonic acid. This is in the same position where COX inhibitors bind. And drug designers have been exploiting those differences, although retrospectively.

"People have found compounds," Marnett went on, "and then subsequently elucidated why they are selective. We're kind of in the same boat. We've found compounds that are not absolutely selective for COX-2, but which do reverse the selectivity from aspirin. Now we're in the process of trying to understand the reason for the selectivity of COX-2 over COX-1."

In that process, Marnett and his colleagues have developed and tested a prototype drug on which they report in the current issue of Science, dated May 21, 1998. Their paper bears the title: "Aspirin-like molecules that covalently inactivate cyclooxygenase-2."

Next-Generation Drug Looks At $14-Billion Market

This analog is elbowing its way into an anti-inflammatory world market estimated at $14 billion dollars a year. Besides the non-steroidal analgesics, of which aspirin is one, now already in that market, and on TV commercials, pharmaceutical companies have perhaps a dozen novel contenders that inhibit COX-2 and its side effects.

Front-runners among these are Merck & Co. Inc., in Whitehouse Station, N.J., and G.D. Searle & Co., of Skokie, Ill., a division of Monsanto Co., of St. Louis. "Their compounds," Marnett pointed out, "work by binding very tightly, but not covalently, to the enzyme. So they may eventually fall off. Ours binds covalently; that is, it's the first irreversible COX-2 inhibitor. We got that effect by modifying a serine residue in exactly the same way that aspirin does."

After developing their most potent product, dubbed APHS, and testing it in various test-tube settings, the Vanderbilt co-authors moved recently to in vivo trials in rats. Under the skin of six animals, they created small (20-milliliter) pockets of sterile air. They filled these cavities with a pro-inflammatory substance, carrageenan, which induces COX-2 expression and prostaglandin (PG) biosynthesis. (Aspirin's key action is to block this PG aggregation.)

Three hours later, they administered their APHS aspirin-like compound, then measured inhibition of PG. It was complete, thus testifying to selective blockade of COX-2 in vivo.

APHS proved 60 times as reactive against COX-2 as store-bought aspirin, and 100 times as selective for the enzyme's inhibition, Science reported.

"However," Marnett allowed, "our APHS compound doesn't inhibit COX-1 very well, as aspirin does. And COX-1's inhibition is involved in preventing cardiovascular disease. So we've got a lot more to do before we can really tell how efficacious these compounds might be.

"What we need to do now is begin pharmacokinetic and efficiency testing — a range of things more associated with the drug development phase. We're a small group," he observed. "We really have only a limited number of synthesis people working in this. A big pharmaceutical company, at this point, would put a lot of synthetic chemists on refining the analog's structure."

Marnett continued, "So I think the most efficient way for us to get a good candidate drug quickly is to establish some sort of a working relationship with a pharmaceutical company that can bring much deeper resources to this than we can."

One of the co-authors on his Science paper is pharmacologist Karen Seibert, of Searle. "She did the in vivo experiments," Marnett recounted, "but as a friend; no relationship to her company."

Inquiries Coming From Big Pharma

He's been fielding inquiries from such large companies, and expects closure in the near future. Vanderbilt has applied for a patent on this work, covering "selected inhibitors of prostaglandin endoperoxide synthetase-2."

There's another kicker in Marnett's research, as reported: APHS showed in vitro signs of being able to ward off colon cancer.

He and his co-authors treated two types of growing tumor cells with their compound, one that expresses COX-2 in quantity and one that does not. APHS inhibited growth of the first tumor type, not of the second. These results, Marnett concluded, "confirm that COX-2 is important for the growth of colon cancer cells that express the enzyme." *