Over the Brainbow

Two years after the colorful project made a splash, most researchers are still relying on older techniques to map neural linkages.

Courtesy of Jean Livet

Neurons in young brains form a riot of interconnections that fan out in all directions, with multiple nerve cells often converging on a single target cell. As brains mature, some of these overlapping connections are pruned. Being able to visualize this pruning process for individual cells could help scientists answer fundamental questions about how information is propagated through the neural network. Yet understanding how different connections are strengthened or eliminated has proved difficult, because neighboring neurons stained with a single dye blur together under a microscope.

But in 2007, a team led by Jeff Lichtman, a neurobiologist at Harvard University, developed a technique called Brainbow to label neurons in mouse brains with a palette of over 100 different colors, described in this...

Turning on cre expression caused the protein to splice out different parts of the Brainbow DNA segment and express either the red, blue, or green fluorescent protein. Depending on which lox site cre snips (a stochastic process), each segment expresses one of the three hues at a time.  The team placed 8–16 copies of these Brainbow genes into mice. "The number of copies is like dabs of paint; how many drops [of each color] are added" determines the final color expressed by the neuron, Lichtman says.  With multiple copies of the DNA segments, the mouse brain can be painted with over 100 different hues, which enables researchers looking under a light microscope to distinguish between near-neighbor neurons.

Lichtman's "approach allows us to look at connectivity of individual neurons, and that's something we simply didn't envision would be possible," says Indiana University neuroscientist Olaf Sporns. "The images were so stunning, so beautiful, and they have so much biological information," says Steven Paddock, a developmental microscopist at the University of Wisconsin in Madison, who is studying development in butterflies and fruit flies.

Since researchers released the mouse Brainbow, other teams have begun to use it to color-code cells in other model organisms, including zebrafish and Drosophila. But despite the beauty of Brainbow, some researchers say the technique is still struggling to find a niche and has yet to answer critical questions about neural connectivity.

Branching Out

Alex Schier, a developmental biologist and neurobiologist at Harvard University, has been applying the Brainbow technique to zebrafish. Because the fish are transparent, Schier can watch the play of colors in live animals as they develop. One of the main draws of the Brainbow technique is the potential for mapping long-range connections between neurons, says Schier. "Imagine you have a bowl of spaghetti—there is no way to follow each [piece of] spaghetti from end to end. But if each [piece of] spaghetti had a different color, you could. That's what Brainbow does for the nervous system," he says.

But some groups are skeptical of the technique's reach. "I think that the Brainbow mice have been significantly oversold in the press in terms of brain-wide connectivity projects," says physicist and neuroscience researcher Partha Mitra of Cold Spring Harbor Laboratories in New York. Brainbow techniques can only paint neurons in one of a hundred different colors, but there are thousands and thousands of neurons to map in any given slice of brain. Using the technique, researchers have to "stitch" small images together, Mitra says. But since more than one neuronal connection could be painted with the exact same shade of red, computer algorithms would have to guess which pieces fit together, leaving room for error. For long distances across the entire brain, that could lead to a faulty wiring diagram, he says.

In March of this year, Mitra and 37 other scientists published a paper calling for a systematic effort to map neural circuitry in the mouse. Instead of using technicolor approaches like Brainbow, the group calls for developing high throughput systems for processing more old-fashioned images, in which a single-color tracer dye is injected into the nervous system.1 The technique is more established, and less time-consuming than Brainbow. Tracer injection techniques cannot distinguish individual neurons and provides only about 200-micrometer resolution, but Mitra argues that higher resolution isn't necessary to understand the wiring in the brain.

Neurons up close

Two years after the initial Brainbow project, most researchers mapping neural connectivity are using other approaches. At the Howard Hughes Medical Institute's Janelia Farm Research Center in Ashburn, Va., researchers like Dmitri Chklovskii are taking a zoomed-in approach to the brain connectome. His group is using electron microscopy to characterize the circuitry in the Drosophila brain. Using 50-nm-thick slices of fruit fly brain tissue, they can capture every neuron, axon, and synapse in three dimensions. The method, however, can only be used for small volumes, so the group is piecing together thousands of images using computer vision, he says.

Karel Svoboda, a biophysicist who is also at Janelia Farm, is trying to directly map connections between neurons by measuring electrical activity at synapses, the junction between different neurons. He doesn't use the Brainbow technique because he wants to directly track the electrical signals, not just the anatomical structures that carry the signals. He's optimistic about Brainbow's future in research: Though it hasn't been widely used yet, it will find its niche eventually, he says. "Right now it is a bunch of nice pictures, but I think there's a lot of promise there," Svoboda says.

Data derived from the Science Watch/Hot Papers database and the Web of Science (Thomson ISI) show that Hot Papers are cited 50 to 100 times more often than the average paper of the same type and age.
J. Livet et al. "Transgenic strategies for combinatorial expression of fluorescent proteins in the nervous system," Nature, 450:56–63, 2007. (Cited in 51 papers)

1. J. Bohland et al. "A proposal for a coordinated effort for the determination of brainwide neuroanatomical connectivity in model organisms at a mesoscopic scale. PLoS Comput Biol, 2009;5(3):e1000334.

Interested in reading more?

Magaizne Cover

Become a Member of

Receive full access to digital editions of The Scientist, as well as TS Digest, feature stories, more than 35 years of archives, and much more!
Already a member?