Eye on the Prize

Salamanders and dragonflies are very different animals. One is a four-legged amphibian and the other a six-legged, winged insect. But these evolutionarily distant creatures have figured out how to detect a moving piece of food—and get it quickly into their mouths. Anthony Leonardo, a group leader at Janelia Farm, is studying the neuroscience behind that behavior in both species. Measuring all the parts of this singular action is one way to get at the complexity of the nervous system.

Leonardo likes studying hunting because it doesn’t allow for multitasking. “When a salamander or a dragonfly is foraging and catching things, it has this sort of Zen-like focus where that’s all it’s doing. If you look at the neurons, they may do things that are very complicated, but they’re all directed toward the same purpose.”

The first step in hunting by sight happens in the retina, a tissue at the back of the eye that records light and sends signals to the brain. One of the questions Leonardo wants to answer is, “What is the retina doing while a fly, also known as ‘dinner,’ is moving around in front of it?”

A retina can live for hours outside the body if it’s kept in a mix of salt water and other ingredients. In experiments carried out by Leonardo’s team, a salamander retina rests in a flat glass dish that has a near-invisible pattern of sensors and wires etched onto its surface. The neuron-sized sensors record electrical signals coming out of the retina’s ganglion cells, specialized neurons that collect information from the other cells in the retina and send it to the brain.

In the lab, Leonardo plays a video of a round spot, like a ball, moving on a screen. This is the kind of video he projects onto a retina’s photoreceptor cells. Pointing to one area of the screen, he says, “If a ganglion cell is looking at this part of the screen where my hand is, the neuron goes, nothing … nothing … nothing … and then it goes ‘click!’” when the ball passes in front of it. This shows that a particular cell is interested in a specific part of the scene in front of the animal. By collecting many measurements like that one, Leonardo can start to understand how the retina sees the world.

He can even go one step further, by setting a live insect into a sort of dragonfly lounge chair and then showing it moving images while recording the response in its brain. The team has also made precise measurements of how salamanders and dragonflies move as they catch fruit flies. For example, they learned that the dragonfly doesn’t chase a fruit fly; instead, it figures out the trajectory the fly is following and intercepts it, like a guided missile. Leonardo is working now on perfecting a tiny backpack that would affix to a dragonfly’s belly and record electrical activity in the nerves in its thorax while it’s catching prey. See "Flight of the Dragonfly", HHMI Bulletin, May 2009.

The idea is to take all of this information—how the prey moves, how the animal responds when food is nearby, and what’s going on in the neurons all the while—and put it together to understand how the neurons work together to make dragonflies and salamanders act the way they do. Leonardo says it’s like putting together a puzzle, where the picture you’re trying to assemble is the behavior. “The neurons are the little pieces,” he says. “We want to see if we can fit them together.”

-- Helen Fields
HHMI Bulletin, August 2011

Return to main article >