New way to see through brains reveals minute cell structures

A new chemical cocktail that renders brains transparent allows a powerful microscope to resolve minute details within the brain¹.

Among the tricks researchers are using to peer into whole animal brains, the new approach — called SeeDB2 (for See Deep Brain, version 2) — may be the best yet for resolving synapses, or neuronal connections, which may be altered in autism.

To create a picture of the inside of a mouse brain, researchers have two choices: Slice it up and piece together images of the slices, or render it transparent and illuminate the interior with light from a fluorescent microscope. In the past five years, researchers have been working fervently to refine the latter approach.

For example, CLARITY — a method that wowed scientists with its potential for spectacular images when it debuted more than three years ago — allows researchers to reconstruct long-range connections between many types of neurons.

That method is ill-suited to visualizing small details, however, says Takeshi Imai, team leader at the Riken Center for Developmental Biology in Japan, who developed SeeDB2.

CLARITY renders the brain transparent by extracting its fat molecules, which scatter light. But this harsh process temporarily causes the brain to swell and jostles the fine structures, such as synapses, inside. Other clearing methods, such as Scale or CUBIC, similarly lead to temporary changes in brain size. Some, including Scale, rely on soaking the brain in a solution that changes the fat molecules to make them transparent. But this process takes at least two weeks.

In SeeDB, a previous version of the new method, researchers soak brains in a water-soluble sugar solution that gently generates transparent brains within roughly three days². The method does not change the size of a brain or disturb its synapses.

Deep dive:

To resolve specific proteins within the mouse brain, researchers tag them with fluorescent molecules. Unlike some other methods, SeeDB minimizes the fading of the fluorescence, allowing researchers see small areas that emit little light.

Still, SeeDB is not ideal for high-resolution microscopy because the resulting brain has a different refractive index — a measure of how much light bends when moving through a material — than the oil in which the microscope’s lens sits. Because of this, the microscope’s light scatters as it passes through the tissue, limiting how far it can enter the brain.

In the new study, Imai and his colleagues altered the recipe so that the refractive index of the treated brain matches that of the oil. They tested various substances, looking for those that clear the tissue and don’t cause it to swell, or the fluorescent molecules to fade. They finally landed on iohexol, a contrast agent used for X-ray imaging, and detergents that help the iohexol seep deeper into the brain.

The new method, SeeDB2, yields significantly more detailed images of fluorescent neurons in a chunk of mouse brain than do SeeDB, CLARITY, CUBIC or Scale (see inset above).

The researchers used SeeDB2 to distinguish synapses that dampen neural activity from those that excite it, by tagging a protein specific to inhibitory synapses with a fluorescent molecule. The unusual distribution of this protein in the brain demonstrates how altering an excitatory pathway can also perturb an inhibitory one.

They also used the method to build a three-dimensional picture of an entire fly brain (see video below), in which neurons are tagged with a fluorescent protein.

The new method is a relatively easy way for researchers to trace the shape and number of synapses, something that is important for studying conditions such as autism, says Imai. The details of the protocol are freely available.