This setup is commonly used to study addiction: when a drug such as cocaine or amphetamine is consistently administered in the reward room, a mouse learns to associate that room with the reward and later chooses to spend most of its time there.

"What we found, very strikingly, was that this worked," says Deisseroth. One of the optoXRs, built from a receptor protein that normally responds to adrenaline and noradrenaline, produced results much like those seen with drug-based rewards. When loosed after training, mice with this optoXR strongly preferred to spend time in the reward room, where they had received light pulses activating their nucleus accumbens. The results of the study were published this week in Nature.

Many of the same proteins that Deisseroth's team activated with light can be targeted by drugs. But light has a number of advantages over drugs, says Michael Häusser, a professor of neuroscience at University College London, who was not involved in the research. While drugs take time to work and linger after administration, light allows for exquisite control over timing. And while drugs don't pick and choose which cells to affect, optoXRs can be genetically engineered to be expressed in only a specific type of cell.

The hybrid proteins represent a new molecular toolbox with applications beyond drug addiction and the brain's reward system. "There are all kinds of really cool games you can play with these new molecular tools to look at aspects of signaling pathways and how they interact," says Häusser. For example, they open the door to studying more mysterious receptor proteins, which can't be activated pharmaceutically and whose function is not well understood. "For some of them," says Häusser, "our toolbox for manipulating them is limited. So this gives us a fantastic new handle on these receptor classes, and allows us to manipulate them in a really powerful way."