Many species naturally make small amounts of hydrocarbons. Now researchers at the startup LS9, based in South San Francisco, CA, have described the genes and enzymes responsible for this production of alkanes, the major components of fuels such as diesel. The findings, reported in the current issue of the journal Science, have allowed the researchers to engineer E. coli bacteria that can secrete alkane hydrocarbons capable of being burned in diesel engines.

LS9 had previously reported using bacteria to produce hydrocarbon fuel, but this is the first time the researchers have revealed how they did it. “This is the first characterization of these enzymes. Virtually nothing was known about what enzymes were responsible, and how do they do it,” says Frances Arnold, a professor of chemical engineering, bioengineering, and biochemistry at Caltech. Arnold was not involved in the LS9 work. The discovery “opens up a whole new set of possibilities,” she says. “These reactions are very interesting. Nature has made a few versions of them. Now, in the laboratory, we can make many more versions, so your imagination can run wild.” Any commercial applications Arnold and others discover, however, will likely require a licensing agreement with LS9, which has filed for a patent for its discovery.

The LS9 researchers discovered the genes involved by comparing the genomes of 10 strains of cyanobacteria (also called blue-green algae) that naturally produce alkanes with a very similar strain that produces no alkanes. They identified 20 genes that the alkane-producing strains had but that the non-alkane-producing strain lacked. From there, the researchers narrowed down the possibilities until they identified the genes and enzymes necessary for alkane production. They confirmed their discovery by incorporating the genes into E. coli and measuring the alkanes that the bacteria subsequently made. The bacteria secrete the alkanes, which can then by easily collected and used as a fuel.

Organisms make alkanes via a complex process that produces fatty acids from carbon dioxide or sugars. The fatty acids are then converted by the organisms to an aldehyde that includes a carbon atom bonded to an oxygen atom (together they create what’s called a carbonyl group). The enzyme aldehyde decarbonylase helps remove this group to form a chain of hydrogen and carbon atoms–the hydrocarbon. The natural process produces a collection of hydrocarbons of various lengths that are comparable to the hydrocarbon molecules in diesel.

Several research groups at universities and companies have been searching for ways to make renewable fuels that are similar enough to petroleum-based gasoline, diesel, and jet fuel to be used in existing vehicles. Such fuels would be more versatile than ethanol, which can’t be used in high concentrations in ordinary engines. The LS9 discovery is a boost to this effort.

But work remains before the genes can make commercial quantities of fuel at prices that can compete with fossil fuels. “This is a long way from describing a commercially viable process for making alkanes,” Arnold says. “Fuel has to be dirt cheap. It’s not clear that we’re ever going to make it cheap and easy by this route.” One fundamental challenge is scaling up the process. “It all comes down to whether you can send enough carbon through that pathway to get to industrial levels,” she says.

LS9 has been genetically engineering E. coli to optimize the process by which the bacteria converts sugar into fuel.For example, the E. coli naturally feeds on some of the fatty acids it produces, rather than using them as a feedstock for producing alkanes. LS9 is altering the bacteria so that they don’t eat the fatty acids, which helps increase fuel yields, says Andreas Schirmer, LS9’s associate director of metabolic engineering.

The company’s first fuel to market will probably not be a hydrocarbon. Four years ago, the company began developing a fuel based on fatty esters that it says could serve as a replacement for diesel fuel, and is closer to market than the hydrocarbon fuel it started developing two years ago when it first identified the alkane genes and enzymes.