(page 2 of 2)

"We're attempting to create mutations that destroy innate immunity and in this way, to identify all of the genes involved in innate immunity," Beutler says, "or at least as many as we can."

The mutagenesis involves attacking genomes with ethylnitrosourea and looking for defective innate immune function phenotypes. Treatment with ethylnitrosourea creates about 10,000 to 30,000 randomly distributed mutations in the germline, mostly A–T or A–G substitutions. About 1.6 percent of these mutations fall within coding regions, and about three quarters of these result in an amino acid change. Thus in every separate mutant genome, there are between 100 and 300 mutations that create a coding change.

By examining several thousand such genomes, one can get very deep coverage of the entire genetic content of a species—hundreds of thousands of mutations are induced in all, with at least one mutation in almost every gene and several mutations in some of the larger genes. Then, Beutler says, one simply looks for phenotypes that are of interest. In his case, these include instances of immunodeficiency.

"Then if we find an individual that is immunocompromised," says Beutler, "we can go back and positionally clone the critical gene that we have hit."

Though this may sound routine, there are many more failures than successes, and the competition can be fierce, especially when the mapping and sequencing steps take years, as they did in the case of Tlr4. Beutler described the long battle in a historical narrative recently published in the Journal of Endotoxin Research. Of the race to clone Lps, he wrote:

"On occasion, we rejoiced, feeling that our goal was within sight, only to have our hopes dashed within days. Ever present was the fear that, upon opening the next issue of Nature, we would read that the gene had been cloned elsewhere, as a gambler fears the cards of those arrayed against him."

Bringing it Home to TSRI

After the successful cloning effort, many new questions related to endotoxin signaling immediately presented themselves, and Beutler continues to study this protein and its relatives, hopeful that a full understanding of the events that occur in signaling may emerge. By making chimeras of human Tlr-4 and mouse Tlr-4 proteins, he and his coworkers hope to pinpoint the part of the protein that actually "touches" the endotoxin molecule, leading to cell activation.

Other follow-up work has already panned out. A standard gene searching program called Basic Local Alignment Search Tool (BLAST) has revealed three novel Toll-like receptors in mammals, which were then independently cloned in 2000 by the Beutler and Ulevitch laboratories. These molecules were dubbed TLRs 7, 8, and 9. TLR10—which may prove to be the "last" TLR—was cloned shortly thereafter in the Ulevitch lab. The binding specificity of most of the TLRs remains unknown. "And," Beutler says, "everybody is interested to learn about the structure of these proteins."

Beutler calls the process of positional cloning 'addictive.' "The methods used in hunting genes are compelling, and I will continue to use them whenever possible," says Beutler. "The power of a mutation to disclose function is tremendous, and provided that one begins with a clear phenotype, the genes that are ultimately identified will most certainly be relevant to the biological question that one has in mind."

There is something in biology referred to as the phenotype gap—the discrepancy between the 30,000 to 40,000 genes we believe are present in the human genome and the mere total of 5,000 distinguishable traits that have been identified through studies of inherited diseases and knockout mutations produced by gene targeting. The inference to be made is that the critical function of most genes remains unknown. This applies in the field of immunology as in many other fields.

The precise question of which genes serve innate immunity is one of several lines of inquiry Beutler brought with him when he came to TSRI last year. He is also interested in discovering why the placenta is not rejected during pregnancy as any "normal" tissue allograft would be. Since half of a fetus's genetic material hails from its father, the fetus is sufficiently non-self to cause an immune reaction. But in most cases, it does not.

Beutler is trying to create a model in which semi-allogeneic pregnancy (in which the fetus is derived from genetically different parents) is not tolerated, but syngeneic pregnancy (in which the fetus is genetically identical to both parents) is tolerated. "We are asking how it happened that placenta could arise in evolution, although when it did, there was already a very good adaptive immune system," Beutler says.

For Beutler, coming to TSRI was a definite homecoming. Raised in southern California and very familiar with San Diego from having attended the University of California at San Diego, he also joins his father, Ernest Beutler, who is chair of the Department of Molecular and Experimental Medicine.

"I love it here," he says. "The scientific environment at Scripps is superb, my colleagues here are wonderfully supportive, and not least, I always wanted to come back to California, having missed it ever since I left 25 years ago."

Beutler got his start in science while working in his father's laboratory as a teenager. "My father worked on red cell enzymes at the time, and my first project involved assaying glutathione peroxidase activity in human erythrocytes," says Beutler. "He was a fine teacher, who gave me an in-depth introduction to science at a young age."

When asked his opinion, Ernest Beutler smiled widely and nodded in approval. "I'm glad he's here," he said.

1 | 2 |


Alignment of TIR sequences from the ten human Toll-like receptors. The TIR is the most conserved portion of members of the Toll-like receptor family. Different amino acid residues are shown in different color in this picture for clarity, with colors grouped according to physico-chemical properties. For example, hydrophobic residues L and V are both colored pink. The sequences are optimally aligned with the inclusion of gaps where required.