The flash of insight, says lead investigator Adam Cohen, a biophysicist at Harvard University, was accidental. "None of us had any knowledge or inclination to study bacteria," he says; his group was interested in neurons. Many researchers are trying to engineer neurons to fire when light-sensitive proteins in their membranes detect a flash of light. This ability would be useful for studying neuron firing and might even help restore vision in people with damaged retinas. But researchers have lacked a way to measure how much voltage neurons produce when light activates them. To solve the problem, Cohen's group altered a light-sensitive bacterial protein called green-absorbing proteorhodopsin (GPR) so that it could detect membrane voltage and give off a flash of light when the membrane depolarizes. By measuring the amount of light produced, they could find out how much voltage the neurons produced. 

But before they put the modified GPR into neurons, the researchers engineered E. coli, the workhorse of molecular biology, to carry the gene. On a whim, postdoc Joel Kralj placed some bacteria on a slide and looked at them under a microscope. He was shocked to see them blinking on and off like Christmas lights, suggesting that they lost their charge altogether rather than using the stable charge difference to pump ions across the membrane. Even though the bacteria had all grown together in the same flask and were supposedly identical, each cell flashed at a different rate. Some blinks lasted for a fraction of a second, others for several seconds. Why the bugs flash at different rates is unclear, the researchers say in their paper published online today in Science, but when the researchers counted the flashes and graphed them, the pattern looked similar to the electrical activity of a neuron when it fires.

John Spudich, a biochemist at the University of Texas Health Science Center in Houston, calls the discovery "very exciting." The work suggests that the phenomenon is "something important in cell physiology." Neurons, powered by energy-producing organelles called mitochondria, can afford to depolarize and fire when they want to communicate with one another, but in a bacterium, the electrical difference is the "fundamental energy currency of a bacterial cell. That's not something it's going to turn off lightly, only if it needs to," he says.

But why would it need to, even for a second? The researchers aren't sure, but they believe that depolarizing their membranes may allow the bacteria to kick out charged molecules they've accumulated, such as toxins or antibiotics. In fact, Cohen says, the finding could be a potential antibiotic resistance mechanism. When the researchers put a dye molecule that, like many antibiotics, was positively charged into the bacteria, they kicked it out. Previous researchers had observed this phenomenon but couldn't explain it.

Cohen and his group plan to investigate the phenomenon in many kinds of cells and even in mitochondria and light-capturing chloroplasts, which evolved from bacteria. Cohen says his group is also interested in returning to the original question of how to get mammalian cells to read out their voltage.