By toggling antibody binding, researchers see into dense tissues

Understanding how single cells express proteins and other gene products has revolutionized biology, but understanding those single cells in their tissue context has remained technically challenging. Researchers at the Massachusetts Institute of Technology (MIT) have developed an antibody labeling method they say will enable single-cell protein imaging in intact tissues (Nat. Biotechnol. 2025, DOI: 10.1038/s41587-024-02533-4).

Finding specific proteins in dense tissue using an antibody label is difficult because antibodies are slow to diffuse through tissue but quick to react with their targets. Like a person who arrives at a party intending to mingle but gets absorbed in conversation just inside the front door, antibodies tend to concentrate at the margins of tissues and never encounter targets deeper inside. To overcome uneven labeling, researchers often section organs into slices just a few microns wide, sacrificing some spatial context.

Researchers in the laboratory of MIT chemical engineer Kwanghun Chung are announcing a new way to make labeling antibodies perfuse fixed tissues more evenly by muting those antibodies’ reactivity while they diffuse through fixed tissue and then gradually restoring reactivity. Chung helped to invent the tissue-clearing method CLARITY more than a decade ago and has continued to develop antibody-based fluorescence imaging methods. The new method depends on a highly basic starting environment and a detergent, deoxycholic acid, to inhibit antibody-target binding and uses a rotating electrical field to draw the negatively charged proteins through a porous tissue sample more quickly than they would move by diffusion. The deoxycholic acid is gradually removed by dialysis, and the pH is reduced by an oxidation reaction in the buffer, which restores the antibodies’ ability to bind their targets. The researchers demonstrate that the method achieves single-cell neuron labeling in an intact rat brain and that they are able to study multiple protein targets at a time in brains, other organs, and whole embryos. To label a larger tissue sample, such as a whole human brain, they would need to find a way to diffuse the excess heat from electrotransport.