By David N. Leff

Who would have thought of little old aspirin as a tumor suppressor — right up there with such esoteric proteins as p53?

What's more, who would have thought the everyday, over-the-counter medicine prevents some types of cancer by a mechanism that looks a little like gene therapy.

Aspirin's main prophylactic target seems to be the second most deadly malignancy in Western countries, namely colorectal cancer. But just how that ubiquitous drug gets in these licks has been a mystery.

"Aspirin is a well-known prophylactic for cancer," observed molecular geneticist Richard Fishel. "The new twist is that it suppresses the accumulation of gene mutations that cause hereditary nonpolyposis colorectal cancer [HNPCC]." Fishel is a professor of microbiology and immunology at Thomas Jefferson University, in Philadelphia.

Between 15 and 30 percent of colorectal cancers apparently have a hereditary origin, and half of them involve germ-line mutation of the human mismatch repair genes.

"They were not the very first genes to be associated with DNA repair in cancer," Fishel told BioWorld Today. "Xeroderma pigmentosa was. But HNPCC's genes were the first to be associated with a very common form of cancer. And more importantly," he pointed out, "those individuals, unlike xeroderma patients, were heterozygous carriers. That is, they had inherited one good copy of the gene from one parent, and one bad one from the other. In the xeroderma story both parental genes were knocked out.

"So, here's a situation where you essentially have half of a smoking gun," he continued. "Some time later in life you have a greater chance of knocking out the second half in some single cell in the body. And once that happens, the mutation rate increases, and that cell line goes on to develop disease — at an earlier age than commoner forms of colorectal cancer."

Fishel recalled "the original definition for those early-onset cancers: 'Three family members in two generations; one before the age of 50.' So that was really the first indication that the mutator hypothesis might be correct."

That seminal hypothesis goes back to 1991, he observed, "when a clinical researcher named Larry Loeb at the University of Washington, in Seattle, suggested that tumor cells contained way too many mutations to be accounted for by spontaneous mutagenesis events. Loeb proposed that, even if exposed to high levels of every carcinogen under the sun, the tumor cells had far too many mutations to account for them.

"He suggested that there are general DNA repair genes in all organisms that suppress spontaneous mutations," Fishel went on. "And that if one or several of those genes were altered, you could accumulate mutations at a much faster rate, which would account for the multiple mutations you see in tumors."

Meet The Mutator

Fishel is senior author of a paper in today's Proceedings of the National Academy of Sciences (PNAS), dated Sept. 15, 1998. It bears the title: "Aspirin suppresses the mutator phenotype associated with hereditary nonpolyposis colorectal cancer by genetic selection."

He and his co-authors undertook that research to answer the question, "Was there any way we could affect this increase in mutation rate?"

What made their project feasible, he pointed out, "was our development of a rapid methodology by which we could actually ask that question — a patent-pending technology called microclone analysis.

"In order to look at mutation rates," he explained, "you have to examine individual clones, and ask whether there are different mutations in them. Essentially, we took cells and split them in culture to very low density onto plates — of which the bottoms had microscope slides in them. Those cell lines were derived from patients with HNPCC disease. When the cells began to seed themselves onto these slides, and make very small microcolonies, we harvested them. On a single slide you can get multiple different colonies that arose from single cells."

The co-authors isolated DNA from individual microclones, and then by PCR analysis detected microsatellite sequences in specific regions of the cells' chromosomes.

"We could ask in those individual clones whether the microsatellites were stable or unstable," Fishel recounted. "We could examine individual cells quite rapidly. It turned out that the old technology for such analysis would take something like three months of repeatedly splitting cells and growing them up. With our microclone assay, we could do it in three days."

That was the turning point in the team's aspirin inquiry.

"Once you have that technology in hand," Fishel continued, "you can take every compound you can think of off the shelf, and throw it at these cell lines." He pointed out, though, that the cells are "different from those patients' cells in that they have both genes knocked out, so they're homozygous; they have the mutator phenotype."

How Aspirin Terminates Malignant Mutants

The team tried a number of compounds, Fishel said, "but the ones that worked remarkably well were the two non-steroidal anti-inflammatory drugs we threw at these cells. One is aspirin, the other, sulindac, which is related to indomethacin.

"When we treated those cells with these drugs, we found that the mutator phenotype was dramatically reduced," he said.

More to the point, the drugs "divide the cells into two populations: a population that is going to go into apoptosis, and never give rise to these tumor cells, and a population that is non-apoptotic, and gives rise to healthy cells," Fishel said.

"Next," he went on, "we would like to know whether this works in humans.

"Even before they found out our results," he said, "the Europeans had scheduled an aspirin trial for hereditary cancer patients. This PNAS paper puts their study on a firm basic-science foundation.

"We in the U.S.," he continued, "myself and oncologist Henry Lynch, of Creighton University, in Omaha — one of the founders of HNPCC — have proposed a program with the National Cancer Institute, which would examine the effects of aspirin and other non-steroidal anti-inflammatories in cancer prevention in these hereditary diseases. And we are getting together with the Europeans to develop a much larger international collaborative trial to determine if this actually does work in humans.

"But aspirin and sulindac have their downsides," Fishel cautioned. "They cause intestinal bleeding and liver toxicities, respectively. So," he concluded, "one should never take these or similar drugs long-term without consulting a physician." *