Researchers used electron microscopy to reconstruct a set of neurons in the fly brain (in color, above) that provides odor information to a brain region involved in memory and learning.

Z. Zheng et al./Cell 2018

In a ‘tour de force,’ researchers image an entire fly brain in minute detail

For the first time, scientists have imaged the entire brain of the fruit fly Drosophila melanogaster in enough detail to detect the individual junctions, or synapses, between every neuron. The resulting database of images could help researchers map the neural circuits that underlie every sniff, buzz, and aerial maneuver in a fly’s behavior.

“This data set—and the opportunities it creates—are … arguably one of the most important things to have happened in neurobiology recently,” says Rachel Wilson, a neurobiologist at Harvard University who was not involved in the new work. “Anyone in the world who is interested can download the data set and determine whether any two neurons … talk to each other.”

The 100,000-neuron fruit fly brain is elementary compared with the roughly 100 billion neurons in our own skulls. But the fly is still “much more than this little speck that you swat away from your wine glass over dinner,” says Davi Bock, a neuroscientist at the Howard Hughes Medical Institute’s Janelia Research Campus in Ashburn, Virginia. Some systems in the fly brain—like those responsible for detecting and remembering smells—likely share “common principles” with those in humans, he says.

To make out features of individual synapses, where a signal from one neuron travels to another, Bock and colleagues used an electron microscope, which can resolve much finer detail than a traditional light microscope. They soaked a fly’s brain in a solution containing heavy metals, which bind to the membranes of neurons and to proteins at the synapses. That made the brain look like a wad of noodles, each dark on the outside but white on the inside, Bock explains. Then, a diamond knife cut the brain into about 7000 slices, each of which was struck with a beam of electrons from the microscope to create an image.

The process required a camera that could capture 100 frames per second, a robotic system to scoot each brain slice into place with nanometer precision, and software to stitch together the resulting 21 million pictures. The result is a reconstruction that lets researchers zoom in on the features of an individual synapse. 

“This paper is the absolute definition of a tour de force in terms of the technical accomplishment,” says neurobiologist Cornelia Bargmann of The Rockefeller University in New York City. She studies the nervous system of the nematode Caenorhabditis elegans; a wiring diagram, or connectome, of its 302 neurons was published in 1986. To get a similar diagram for the fly brain, researchers will have to use the new images to trace each neuron to every other neuron that it listens and talks to across the brain.

So far, Bock and his team have done that for a small subset of neurons in a part of the brain involved in learning and remembering smells called the mushroom body. That initial project, described today in Cell, gives new details about the fly’s well-studied olfactory system. For example, neurons that relay odor information to cells in the mushroom body form unexpectedly tight bundles, which Bock’s team is now studying for clues about how flies sample odors from their environments.

If teams across the world manage to make a complete wiring diagram of the fly’s brain, they’ll then need to combine that information with other technologies that record brain activity from living flies. The strength of the connections between neurons changes in different contexts and over time, Bargmann notes. “I’ve worked on an organism with a connectome for 30 years, and we’re still discovering how that nervous system works.”

But the technical capabilities described in the new paper suggest it will soon be possible to map the connectome of a creature that’s yet another evolutionary step closer to humans. “Given that they’ve got the fly working, the zebrafish is just about the same order of complexity,” Bargmann says. “I think we could get to vertebrates in not too long.”