By David N. Leff

"What makes her so different - the girl with the three blue eyes?"

This old-time musical comedy hit could well serve as the theme song for efforts to tell the deep-down molecular difference between healthy and malignant cells. "One approach to understanding the molecular basis of cancer," observed molecular oncologist Diane Pennica, "is to identify differences in gene expression between cancer cells and normal cells."

Pennica is a senior scientist at Genentech Inc., of South San Francisco. She recently carried out a project comparing the expression of Wnt-1 regulated genes in colon cancer tumors with matching healthy colon tissues from the same patient.

Pathologists at the University of Pittsburgh and the University of Leeds, in the U.K., contributed the tumor specimens and control tissues from 25 patients who had had adenocarcinomatous sections of their bowel removed surgically.

The Genentech researchers were looking in particular for versions of a gene called WISP, Pennica told BioWorld Today. It stands for "Wnt-induced secreted protein." The Wnt gene family, she explained, "sends signals downstream to target genes involved in cell growth. These, such as beta-catenin and APC, [adenomatous polyposis coli] are mutated in many cancers. APC mutates when people get multiple polyps, which can go on to colon cancer, a hereditary disease."

Some 60 to 80 percent of all colorectal cancers involve mutations in the APC tumor suppressor gene. And these lead to overexpression of beta-catenin, by disrupting the Wnt signaling pathway.

"So these mutations," Pennica continued, "prevent beta-catenin from being degraded. Instead, you get high levels of it in the cells' cytoplasm. When that happens, the beta-catenin gets transported to the cell's nucleus, where it binds to transcription factors such as TCF - T-cell factor. Then, this hyperactive interaction of beta-catenin with TCF activates the genes involved in growth control. So you have uncontrolled growth where you don't want it. In other words, cancer. But, if these genes are not mutated, beta-catenin gets degraded normally, and Wnt target genes are turned off."

Mutated Genes Help Tumors Grow

Wnt is a busy molecule. "It's involved in a number of activities," Pennica observed. "For example, cell proliferation, differentiation, growth control, a role in brain development."

She and her group at Genentech recently discovered one other key role for Wnt: It gives birth to a novel gene family called WISP.

"We were looking for downstream genes," she recounted, "that were up-regulated in Wnt-1 transformed cells. And after a differential subtraction between normal mouse mammary epithelial cells and those same cells that were stably transformed with Wnt-1, we found a subset of genes that were up-regulated in those Wnt-transformed cells. One of the genes we named WISP-1, and when we discovered others we dubbed them WISP-2 and WISP-3, because, interestingly enough, they were largely homologous. The three reside on the long arms of human chromosomes 8, 20 and 6 respectively. The cDNA sequence of human WISP-1 is 2,830 base pairs long and encodes proteins containing 367 amino acids. These proteins define a subfamily of the connective tissue growth factor family."

Reverting to the 25 colon cancer patients, Pennica recalled that WISP-1's genomic DNA was amplified in the 19 tumors tested, and its RNA overexpressed from four to more than 25-fold in 16 of them (84 percent), compared with healthy adjacent mucosal tissue. WISP-3 turned in a similar expression performance.

"In contrast," Pennica said, "while WISP-2's DNA was also amplified, its RNA expression was reduced from two-fold to more than 30-fold in 15 of the 19 tumors examined (79 percent)."

Hunt Is On For Tumor-Marker Antibody Bait

"What we found was that these genes are aberrantly regulated in human colon tumors, in that WISP-1 and WISP-3 were more highly expressed in tumors than in normal mucosa from the same patient," Pennica said. "These results suggest that the WISP genes' altered expression may confer selective growth advantage on the tumor."

Pennica is first author of a paper in the current Proceedings of the National Academy of Sciences (PNAS), dated Dec. 8, 1998. Its title is "WISP genes are members of the connective tissue growth factor family that are up-regulated in Wnt-1-transformed cells and aberrantly expressed in human colon tumors." Its senior co-authors are molecular geneticist David Bottstein of Stanford University, in Palo Alto, Calif., and molecular biologist Arnold Levine, new president of The Rockefeller University, in New York.

As for eventual clinical applications, Pennica said researchers "are trying to find cell-surface molecules - markers that could be targets for potential antibodies. So if you found the cell-surface receptors for these molecules, you could target an antibody to either stimulate or inhibit WISP's activities. These could be markers for colon tumors or other tumors. Now we have to look at breast, lung, melanoma and other potential cancers, to see if they're regulated differently as well." n