Electron microscopy welcomes a versatile genetically encoded protein tag.
Fluorescence microscopy experienced a 'green' revolution when GFP sent a surge of excitement for live imaging through the life sciences. Meanwhile, electron microscopy (EM) has been waiting patiently for the right probe to come along. EM offers brilliant high-resolution images of ultrastructure, but labeling molecules with specific antibodies is finicky business: the same permeabilization that grants access to targets also degrades fine structures in the cell.
Researchers led by Alice Ting at the Massachusetts Institute of Technology are pinning their hopes on a new generation of genetically encoded probes that avoid this trade-off. Their approach was to ask how the classic label horseradish peroxidase, which fails in highly reducing and calcium-poor environments such as the cytosol, could be made to work anywhere in the cell. “Initial experiments convinced us that this engineering would be extremely difficult, so we instead searched for heme peroxidases that are naturally active in the cytosol,” says lead author Jeffrey Martell.
Horseradish peroxidase uses peroxide to convert 3,3'-diaminobenzidine into a product that can provide contrast for EM. “With guidance from Tom Poulos, one of our coauthors and a leading expert on heme peroxidases, we decided to test ascorbate peroxidase,” says Martell. The peroxidase, abbreviated APX, hails from the pea plant. At first glance, it would seem to make a lousy probe: it exists as a dimer, making it more likely to perturb protein function in the cell, and its native substrate, ascorbate, is not very similar to diaminobenzidine.
The team compensated with several engineering improvements. To make an active monomeric form, they introduced mutations along the dimer interface with the help of the known crystal structure, prior mutagenesis results and sequence alignment to monomeric APX from corn. To compensate for weak heme binding, they modified the active site to resemble that from horseradish peroxidase, yielding faster kinetics and improved activity.
The final product is a tough little tag (about the size of GFP) dubbed APEX, which resists harsh fixation conditions and works in diverse compartments of the cell. Among other examples, APEX-tagged histones and intermediate filaments highlighted chromatin and cytoskeletal features at high resolution, and functional tagged versions of a mitochondrial ion transporter confirmed that it faces the interior of the organelle from its perch in the inner membrane.
APEX joins ranks with the genetically encoded reporter miniSOG, which converts diaminobenzidine for EM contrast by generating singlet oxygen in the presence of light. Although miniSOG is smaller than APEX (12 compared with 28 kilodaltons), the need for light limits how deeply thick tissue sections can be imaged. APEX has yet to be tested on tissue, but diaminobenzidine and peroxide diffusion are expected to allow deep imaging.
Both reporters can also be used for correlative light and electron microscopy. Fusing a fluorescent protein to the EM reporter (or using the mild inherent fluorescence of miniSOG) allows fluorescent features to be correlated and labeled sites to be located quickly by light microscopy before high-resolution electron scanning. The tools also carry the caveats of any genetically encoded reporter—they may alter the function of the protein that they report on—but unlike the partially genetically encoded ReAsH system, they do not suffer from nonspecific labeling.
The green-pea revolution may have just begun for EM microscopy. “We hope to develop improved versions of APEX with improved stability, heme binding, and reactivity toward diaminobenzidine,” says Martell.
Martell, J.D. et al. Engineered ascorbate peroxidase as a genetically encoded reporter for electron microscopy. Nat. Biotechnol. 30, 1143–1148 (2012).
About this article