Antibody scrabble: two genetic players proposed.

Antibody Scrabble: Two Genetic Players Proposed

Just as the millions of words in the English language arise from rearrangements of only 26 letters, the body's billions of different antibodies come from an alphabet of a new hundred genes. Two years ago, the discovery that certain white blood cells recombine these "letters" using a genetic cut-and-paste technique garnered a Nobel prize (SN: 10/17/87, p.244). But exactly how the cells do this remains unclear.

Researchers now report they have identified two genes apparently involved in the puzzling permutation. At the Whitehead Institute of Biomedical Research in Cambridge, Mass., David G. Schatz, Marjorie A. Oettinger and David Baltimore cloned a gene that activates recombination when injected into muscle cells, which normally don't tamper with DNA. And a group headed by Tasuku Honjo of Japan's Kyoto University found a gene that encodes a protein similar to DNA-cutting enzymes in viruses, bacteria and yeast. Those enzymes belong to a class called the recombinases, which recombine specific portions of DNA.

The Whitehead and Kyoto teams report their findings in the Dec. 22 CELL and the Dec. 21/28 NATURE, respectively.

Although the Whitehead researchers do not yet know the exact function of their new find -- which they have named RAG-1, for recombination activating gene -- they suspect it might encode a recombinase that reshuffles antibody genes. Humans, mice, chickens and frogs share similar RAG-1 genes, they report.

"We don't actually know if RAG-1 is the [gene for] recombinase, or if it's some sort of genetic switch that turns on other genes," Oettinger says. However, the gene does behave as they would expect a recombinase gene to act, she adds. For example, it is active in young B- and T- cells -- the white blood cells that recombine DNA to produce antibodies -- but not in mature ones.

The group used "a mixture of perseverance and a creative assembly of existing techniques" in making its discovery, comments molecular biologist Michael R. Lieber of Stanford University. First, Schatz and Baltimore developed a way to put genes into young muscle cells called fibroblasts. To determine whether an inserted gene prompted recombination, they added on a short DNA segment containing a scrambled gene that would enable a fibroblast to resist a cell-killing drug -- but only if the fibroblast first restored the gene's proper DNA sequence.

Whereas the Whitehead group started with a gene and looked for clues to its function, the Kyoto researchers started with a function and looked for a gene. They purified a protein that binds to a "signal sequence" of DNA -- a segment that tells the cell where antibody recombination should occur -- and then used information about that protein to find two genes, which they dubbed RBP-1 and RBP-2. Parts of these genes code for an amino acid sequence resembling that of the recombinases found in viruses, bacteria and yeast.

The results from the two labs raise as many questions as they answer, says molecular biologist George D. Yancopoulos, who coauthored an editorial on the U.S. and Japanese findings in the Dec. 21/28 NATURE. "Even after the cloning of these genes, no one knows anything about the actual mechanism" of antibody recombination, he says. RAG-1 could represent the main switch that controls the recombination, or it might encode just one of many enzymes that are controlled by some other gene. The only known function of the RBP genes is to encode a protein that binds to DNA, but that protein's role in recombination is likewise a mystery. "It could be that the Baltimore [group's] gene is a gene that turns on many other genes, one of which is the Honjo [group's] gene," Yancopoulos says. "There are loads of possibilities."