What if you could power your house with sewage? Or run your pacemaker with blood sugar rather than a traditional battery? Scientists hope that microbial fuel cells -- devices that use bacteria to generate electricity -- could one day make this vision a reality.

Scientists hope to design new strains of bacteria that produce energy more efficiently.

While typical fuel cells use hydrogen as fuel, separating out electrons to create electricity, bacteria can use a wide variety of nutrients as fuel. Some species, such as Shewanella oneidensis and Rhodoferax ferrireducens, turn these nutrients directly into electrons. Indeed, scientists have already created experimental microbial fuel cells that can run off glucose and sewage. Although these microscopic organisms are remarkably efficient at producing energy, they don't make enough of it for practical applications.

Tim Gardner, a bioengineer at Boston University (and member of the 2004 TR35), has developed a new technique for understanding the networks of genes that regulate the chemical reactions taking place in bacterial cells. The resulting "map" will be an advance for the field of synthetic biology: the quest to design and build biological systems that can perform specific functions. Gardner's team aims to harness the genetic control system to engineer bacteria that can produce energy more efficiently.

As a test run of their technique, Gardner and coworkers analyzed the regulatory network of Escherichia coli, a common bacteria often used in research studies. The researchers identified more than 200 gene regulators that could be used in synthetic biology circuits. And they are now applying the technology to Shewanella bacteria.

Technology Review interviewed Tim Gardner about his plans during the Synthetic Biology 2.0 conference, held this week at the University of California, Berkeley.

Technology Review: What is the potential for microbial fuel cells?

Tim Gardner: Microbial fuel cells could really happen, but we have a ways to go in improving the power output. Right now, the output is so low that the technology is unlikely to be able to generate power for homes and cars. But there are some applications for which fuel cells might be appropriate. Some devices don't need much power or could benefit from the ability to use unusual fuel sources -- a medical implant that gets power from the blood, for example, and never needs to get charged. Or robots in the field that could grab a plant and convert it to power.

TR: How will you improve on nature's design of bacteria?

TG: We want to rationally design a cell by manipulating existing machinery. A lot of the early work in synthetic biology was to try to build complete devices from scratch. But we realized we were fundamentally limited using a wholly synthetic approach -- we were trying to build what evolution had built over millions of years. So we said, let's try to tweak what evolution already built.