Researchers Successfully Force Evolutionary Leap

The development, reported in the Feb. 20 issue of Science, demonstrated how the bacterium created an entirely new way to make disulfide bonds. These bonds compose a protein’s stiffening struts that helps the protein fold into its proper, functional, three-dimensional shape. Improperly folded proteins are implicated in diseases ranging from mad cow disease to Alzheimer’s disease.

“We were able to make evolution work for us,” says Dr. George Georgiou, professor of chemical engineering and biomedical engineering at The University of Texas at Austin, and a lead author of the paper. “It illustrated the remarkable diversity of biological systems that can result from a small number of mutations.”

In a process similar to natural selection, Dr. Georgiou and graduate student Lluis Masip made random alterations in the DNA of a strain of mutant bacteria developed by Dr. James Bardwell, an associate professor of molecular, cellular and developmental biology at the University of Michigan. Masip forced the bacteria to use the protein thioredoxin, which normally destroys disulfide bonds, to change its role and create the bonds.

The result required a surprisingly small change in thioredoxin’s make-up, says Masip. Only two amino acid changes, or a 2 percent change in the total number of amino acids in thioredoxin, restored the disulfide bond. The new thioredoxin could later form disulfide bonds using its new artificially-produced pathway, using none of its original disulfide bond mechanisms.

Georgiou, known for his work with antibodies that yielded a potential cure for anthrax, discussed this thioredoxin experiment at a scientific meeting in France in spring 2002, where Bardwell asked if they could collaborate to determine the mutant bacteria’s precise cause for success.

With post-doctoral fellow Jean Francois Collet, Bardwell's team found that the two amino acid substitutions in thioredoxin cause a remarkable transformation: they result in the binding of two iron and two sulfur atoms that form a complex that bridges two thioredoxin protein molecules. This iron-sulfur cluster was shown to be necessary for the new enzyme to form disulfides. Iron-sulfur complexes occur in many enzymes, but never before had such a functional iron-sulfur complex been introduced into a protein as a result of laboratory evolution. James Penner-Hahn, a professor of chemistry and chair of the Biophysics Research Division at U-M, showed exactly what kind of iron-sulfur cluster was involved.

Bardwell likens the new pathway for disulfide bond formation to engineering."People often speak of computer-assisted design, where you try things out on a computer screen before you manufacture them,” he said. “We put the bacteria we were working on under a strong genetic selection, like what can happen in evolution, and the bacteria came up with a completely new answer to the problem of how to form disulfide bonds. I think we can now talk about genetic-assisted design.

"The naturally occurring enzymes involved in disulfide bond formation are a biological pathway whose main features are the same from bacteria to man," Bardwell said. "Understanding how disulfide bond formation occurs and figuring out new ways to make it happen could be important to numerous disease states, like Alzheimer's and cystic fibrosis, that result from defective protein folding."

Disulfides are also vital for the activity of most proteins injected into people for medical purposes, such as insulin and TPA, a blood clot dissolver injected into people having heart attacks and strokes.