Breasts may secrete tumor suppressor.

Since the 1994 identification of the BRCA1 gene and its link to breast and ovarian cancer, researchers have wondered how the gene normally prevents malignant cell growth. Now, investigators from the University of Washington in Seattle and Vanderbilt Unive rsity School of Medicine in Nashville have shown that adding a working BRCA1 to some cancer cells inhibits their proliferation and suppresses tumor growth.

This finding supports the idea that understanding BRCA1 may lead to new cancer therapies. Adding to that hope is a more controversial result from the same researchers: They contend that BRCA1, the protein encoded by BRCA1, is secreted by breast cells and belongs to an obscure family of proteins called granins. If these observations are correct, BRCA1's role may differ from that of other known tumor-suppressing genes, most of whose proteins act within the cell to control proliferation.

The surprising claim that BRCA1 is secreted raises the possibility that the protein stops cancerous cell growth by binding to molecules on the surfaces of breast and ovarian cells. If so, investigators might design cancer-fighting drugs that mimic BRCA1's ability to latch onto those surface molecules.

"If I were a pharmaceutical firm and heard about this, I would be really excited," says Vanderbilt's Roy A. Jensen, an author of the BRCA1 reports, which appear in the March Nature Genetics.

"It opens the door to what would appear a quite simple therapeutic measure," adds Simon A. Gayther of the Cancer Research Campaign in Cambridge, England.

In one series of experiments, led by Vanderbilt's Jeffrey T. Holt, researchers added functional copies of BRCA1 to a variety of cancer cells growing in test tubes. The added genes slowed the growth of breast and ovarian cancer cells but had no effect on c ells from other cancers or normal muscle.

Then the investigators added BRCA1 to other cancerous breast cells and injected the cells into mice. Ordinarily, cancer cells of this type form large tumors that kill mice within 2 weeks, but such cancer cells with the added gene either developed slowly i nto small tumors or didn't generate tumors at all.

The investigators also inserted copies of BRCA1 into breast cancer tumors growing in the abdomens of five mice. When they autopsied the mice, two of them no longer had discernible tumors.

"For people who already have tumors, BRCA1 may be therapeutically useful. [In mice,] you can not only prevent tumors but make them regress," says Barbara Weber of the University of Pennsylvania Medical Center in Philadelphia.

In other experiments, led by Jensen, investigators found evidence that the BRCA1 protein is secreted. For example, they were able to deplete breast cells of BRCA1 by adding a compound that stimulates secretion.

Researchers then noted that BRCA1 contains an unusual string of 10 amino acids, known as the granin motif. In addition to secretion and the motif, granins share other characteristics with BRCA1, such as an unusual resistance to heat. "[Granins] get secret ed and chopped into little bioactive peptides that do all sorts of stuff," says Jensen.

The granin model for BRCA1 appears to offer further evidence that the location of a mutation in the BRCA1 gene determines the risk of ovarian cancer, a hotly debated suggestion (SN: 12/21/95, p. 374). The investigators found that families with mutations i n BRCA1 almost never had ovarian cancer if the mutation in the gene was to one particular side of the stretch of DNA encoding the granin motif.

The secreted nature of BRCA1 could also help explain the observation that pregnancy before age 20 mysteriously cuts a woman's risk of breast cancer in half. Studies by several research teams recently have revealed that BRCA1 production in mice rises drama tically during pregnancy and lactation.

Taken together, says Jensen, the findings suggest that there is a protective effect if women secrete more growth-inhibiting BRCA1 while their breasts are still maturing.

Other researchers, including a group at the University of Texas Health Science Center at San Antonio, have asserted that BRCA1 normally resides in the nucleus (SN: 11/18/95, p. 334). "We stand by our results. We're even more confident based on some recent data," says C. Kent Osborne of the San Antonio team.

"There are different groups getting different results, and it's not yet clear what the explanation for the discrepancy is," concludes Weber.