Double knockout lands a breast cancer gene.

It takes a one-two punch to knock out many human genes. Since people inherit one set of chromosomes from their father and one from their mother, human cells hold two copies of most genes. If a mutation eliminates the action of one copy, the other alone can often serve the cell adequately. Researchers have recently developed an experimental strategy that allows them to knock out the activity of both copies of randomly chosen genes. In the first test of this approach, the scientists have had the good fortune to identify a previously unknown gene involved in breast cancer. Nearly 50 percent of breast cancer patients tested so far have mutations in one or both copies of the gene, called tsg101.

"It looks like they've latched onto a pretty interesting tumor suppressor gene," says Lawrence C. Brody of the National Center for Human Genome Research in Bethesda, Md. Tumor suppressor genes produce proteins that regulate cell growth; mutations in both copies of such genes are usually required to turn a cell cancerous. Using the new gene discovery method they had developed, Limin Li and Stanley N. Cohen of the Stanford University School of Medicine last year identified the mouse form of tsg101. Working with Stanford colleagues Xu Li and Uta Francke, they have now located the human version of tsg101 on a region of chromosome 11 suspected of containing a tumor suppressor gene involved in breast and other cancers. The Stanford group reports in the Jan. 10 Cell that 7 out of 15 people with advanced breast cancer have mutations in one or both copies of tsg101.

The mutations appear in cancerous tissue but not in the surrounding, normal breast tissue. This observation implies that the mutations were not inherited, though tsg101 may yet be defective in some families plagued by breast cancer. Mutations in BRCA1 and BRCA2, the known inherited breast cancer genes, do not account for all such families. Researchers plan to examine these families immediately for tsg101 mutations.

The DNA sequence of tsg101 suggests that the gene encodes a protein that can bind to other genes. It may also interact with stathmim, a protein implicated in cell growth and differentiation. Li and Cohen's gene discovery method, described in the May 3, 1996 Cell, relies on the ability of retroviruses, such as the AIDS virus, to insert their genetic material into a cell's DNA.

The two researchers add a payload of genetic material to a retrovirus that has been made harmless. In each cell it infects, the virus inserts that payload at a random location on the cell's chromosomes.

The inserted payload not only mutates the gene, a portion of it can also silence the cell's other copy of the gene. When the researchers trigger this portion, the cell reads the sequence of the mutated gene backwards. This creates antisense molecules, which interfere with the protein-coding instructions produced by the normal copy of the gene (SN: 2/16/91, p. 108). To find tsg101, Li and Cohen used their strategy to create mutations in a population of mouse cells. They then isolated any mutant cells that would grow in a thick substance called agar. Since tumor cells grow easily in agar but noncancerous cells do not, they concluded that the selected cells were missing a tumor suppressor gene. In fact, some of the isolated mutant cells produced tumors when injected into mice.

The two scientists finally used another part of the viral payload, an easily detectable genetic tag, to identify and sequence the cells' mutated genes. Li and Cohen's gene discovery strategy may have many uses beyond finding tumor suppressor genes. For example, by isolating mutant cells that resist infection by a specific virus, researchers might unearth cellular genes whose proteins are co-opted by the virus. Indeed, the method used to find tsg101 may become as important as the gene itself, says Brody.