The worm Caenorhabditis elegans has a protein called NHR-49. It used to be thought of as a mere overseer of the body's fats. Now it turns out this protein also runs the sense of smell, like a conductor who suddenly starts directing not just the warehouse but the music too.
Scientists switched off this protein in the worms and placed the smell of diacetyl into a Petri dish, since that's what food smells like to a worm. Normal worms crawled toward the smell, while the mutants did not. Yet they responded normally to other tasty smells. So the protein is needed for just one antenna: the AWA nerve cell, which detects diacetyl.
The protein works in nerve cells or muscles. Switch it back on there, and the behavior returns. What's more, mutants have less of the glowing protein odr-10 in their AWA neurons, and that protein is the very antenna that catches the smell of diacetyl. NHR-49, it seems, helps this antenna do its job.
The same protein is needed for habituation. Worms were poked 30 times in a row, with an interval of 10 seconds. Healthy worms backed away less and less with each poke, while the mutants never got used to it. So NHR-49 helps the worm's nervous system remember that a repeated poke is harmless.
This discovery changes how we think about metabolic proteins influencing the brain. NHR-49 was once believed to act only through fats, like a conductor who just changes the lighting in the hall. Now it's clear it can walk up to a particular musician and tune the instrument directly.
Despite its tiny nervous system of 302 neurons, C. elegans distinguishes odors using specialized sensory neurons: AWA handles attractive aromas, AWC handles other attractants, and ASH handles repulsion. NHR-49 turned out to be a "conductor" only for AWA-dependent responses.
Not published on arXiv. Here are its page, the PDF and the formalization: you can check it yourself, not only read our account.