Codexis's early results show that its modified enzymes can survive at temperatures above 85 °C for half an hour. This is high enough for the enzyme to survive in smokestacks, but not at the temperatures needed for freeing the carbon dioxide for storage (130 °C). Lalonde says the company has seen large improvements since these initial results were disclosed, but the company hasn't released the new figures yet.

The company has successfully engineered enzymes for drug development in the past. It has won two "green technology" awards from the U.S. Environmental Protection Agency for developing enzymes for making two drugs--atorvastatin, the active ingredient in the cholesterol-lowering drug Lipitor, and sitagliptin, the active ingredient in the diabetes drug Januvia. The enzymes simplified drug synthesis and reduced waste.

Codexis uses a proprietary version of directed evolution. In its simplest form, directed evolution involves making random changes to existing genes. These mutations alter one amino acid in the enzyme at a time. The genes that work best are then selected and changed to further increase performance. Codexis's researchers have developed a faster version of the process that involves swapping relatively large segments of the gene sequence--making multiple changes to amino acids each time. They've also developed computational techniques that allow them to determine what parts of the gene are most likely to lead to improvements in performance if they are modified. The changes make the process more efficient, and lead to big changes in performance in a relatively short amount of time.

"Codexis's technology has certainly proven itself quite powerful," says Stefan Lutz, associate professor of biomolecular chemistry at Emory University. He cautions that it may be more difficult to work with carbon dioxide than with pharmaceuticals. "If they do succeed, it would be a huge deal," he says.