Albumin Makes Living Mouse Brains Transparent - and the Effect Is Reversible

Kyushu University

Late one night in a Kyushu University lab, Assistant Professor Shigenori Inagaki returned to an idea so basic it seemed almost naive: proteins are polymers. He grabbed a bottle of bovine serum albumin - a common, inexpensive blood protein used in laboratories worldwide - and tested whether it could make brain tissue transparent without killing the cells.

He tested it three or four times before he believed the result. Alone in the lab, he let out a shout. Of all the nearly 100 compounds the team had screened over years of work, a protein that exists naturally in blood turned out to have the lowest osmotic pressure at precisely the refractive index needed to clear living tissue.

The reagent, named SeeDB-Live, is published March 12 in Nature Methods.

The physics of seeing through brains

Brain tissue is opaque for the same reason a bag of glass marbles looks cloudy in air but nearly transparent in oil. Light scatters when it passes between materials with different refractive indices - a measure of how much a material bends light. Inside the brain, lipids and other cellular components create countless tiny mismatches, scattering light and hiding structures deeper than a few hundred micrometers from microscopes.

The solution is conceptually simple: bathe the tissue in a liquid whose refractive index matches the cells, and light passes through uniformly. The team's systematic experiments determined that living cells become most transparent when the extracellular solution reaches a refractive index of 1.36 to 1.37.

The challenge was finding a molecule that could reach that refractive index without damaging the cells. Previous attempts used sugars, but these required concentrations so high they dehydrated cells through osmotic pressure. The team needed something large - large molecules require fewer particles to shift the refractive index, keeping osmotic pressure low.

From blood to brain clarity

Albumin, it turns out, checks every box. It is a large protein (roughly 66 kilodaltons), highly soluble, non-toxic to cells, and achieves the target refractive index at low osmotic pressure. By adding albumin to culture medium at the right concentration, the team created a solution that renders mouse brain slices transparent within one hour of immersion.

When combined with calcium indicators - fluorescent molecules that light up when neurons fire - the transparent tissue revealed neuronal activity deep inside brain slices that would normally be invisible. Applied to living mouse brains through a surgical window, SeeDB-Live made fluorescence signals from deep neurons three times brighter than without the reagent.

This opens clear views of layer 5 of the cerebral cortex, where richly branched neurons help process information and translate neural activity into action. Before SeeDB-Live, obtaining crisp images at this depth with standard two-photon microscopy was extremely difficult.

Reversible by design

Perhaps the most practically significant feature is that the effect washes out. The brain's own extracellular fluid clears the albumin within hours, returning tissue to its original opacity. Because the method causes no permanent changes, the same mouse can be imaged repeatedly to track brain activity over days or weeks.

Senior author Takeshi Imai, Professor at Kyushu University's Faculty of Medical Sciences, noted that albumin is abundant in blood and highly soluble, making it well-suited for clearing. He described the discovery as accidental in origin but almost natural in hindsight, a tribute to what evolution has shaped over millions of years.

A decade of being told it was impossible

The achievement has personal significance for Imai. He developed SeeDB for fixed (dead) tissue in 2013 and an improved version, SeeDB2, in 2016. After each publication, he was asked whether live tissue clearing was possible. By his count, the question came about a hundred times, and each time he answered that it was impossible.

Ten years later, the answer changed. The key was shifting from synthetic polymers to a biological molecule that evolution had already optimized for compatibility with living tissue.

Surgical windows and other constraints

SeeDB-Live is not without limitations. Accessing the brain in living mice still requires a surgical cranial window - an invasive procedure that can cause stress and inflammation, potentially affecting the tissue being studied. The reagent works well for brain tissue specifically, but biological barriers limit its delivery to other organs.

The technique has been demonstrated only in mice. Whether it translates to larger brains or different species is untested. The depth improvement, while substantial, does not make the entire brain transparent - it extends the imaging range by a meaningful but finite amount.

And the reliance on a surgical window means this is fundamentally a research tool for animal studies, not a clinical technique for human imaging. The path from clearing mouse brain tissue to any human application would be long and uncertain.

Inagaki acknowledged that the team has not yet fully realized the reagent's potential, noting that future efforts will focus on less invasive delivery methods to improve penetration for deeper imaging.

What seeing deeper could mean

Complex brain functions like memory and decision-making arise from the coordinated activity of neurons distributed across multiple layers. Existing imaging methods can observe superficial layers clearly but lose resolution with depth. By extending the range of live imaging, SeeDB-Live could help neuroscientists study how deep cortical neurons contribute to integrative brain functions - the kind of multi-layer coordination that defines cognition.

The reagent may also prove useful for evaluating three-dimensional brain organoids and tissue models used in drug discovery, where seeing deep into a structure without sectioning it is a persistent challenge.