Label-retention expansion microscopy

Abstract

Expansion microscopy (ExM) improves the resolution of fluorescence microscopy by physically expanding the sample embedded in a hydrogel 1-4 . Since its invention, ExM has been successfully applied to a wide range of cell, tissue and animal samples 2-9 . Still, fluorescence signal loss during polymerization and digestion limits molecular-scale imaging using ExM. Here we report the development of label-retention ExM (LR-ExM) with a set of trifunctional anchors that not only prevent signal loss but also enable high-efficiency protein labeling using enzymatic tags. We have demonstrated multicolor LR-ExM for a variety of subcellular structures. Combining LR-ExM with super-resolution Stochastic Optical Reconstruction Microscopy (STORM), we have achieved 5 nm resolution in the visualization of polyhedral lattice of clathrin-coated pits in situ.By physically expanding the sample before image acquisition, ExM has enabled the use of a conventional confocal microscope to achieve ~ 70 nm lateral spatial resolution 2-4, 6, 9, 10 . Recent efforts have further enhanced the resolution of ExM either by increasing the volume expansion ratio 11,12 or by combining ExM with super-resolution microscopy such as Structured Illumination Microscopy (SIM) 5,8,13 and Stimulated Emission Depletion (STED) microscopy 7,14,15 . In all these cases, the homogenization of the specimen through either protease digestion 10 or protein denaturation 2, 3 is essential to achieve isotropic expansion without structural distortion. To retain the spatial information of the target structures, the biomolecules of interest (e.g. protein 2-4 , RNA 1 ) and/or labels (e.g. dye-labeled DNA 10 , dye-labeled antibodies 2, 4 or fluorescent proteins 4 ) are anchored to the hydrogel matrix prior to digestion or denaturation. Nevertheless, digestion and denaturation cause incompletely anchored proteins or protein fragments to be washed out, the polymerization reaction to make the hydrogel produces free radicals that readily destroy fluorophores, and both factors can damage fluorescent proteins. Consequently, more than 50% of the target molecules can lose labeling after expansion 4 , which is a major issue of current ExM methods 16,17 . This label loss is exacerbated when aiming for higher spatial resolution. First, ensuring isotropic expansion at nanometer scale requires more thorough digestion or denaturation, hence more wash-out. Second, superresolution microscopy often requires certain dyes that do not survive hydrogel polymerization reaction. For example, Alexa Fluor (AF) 647, one of the best fluorophores for STORM, suffers > 90% loss of the fluorescent molecules after polymerization and

Super-Resolution Label-free Volumetric Vibrational Imaging

“…Indeed, our quantification in CH3 channel showed that the protein loss could reach 79% ( Fig. 1f), consistent with a recent fluorescence analysis 18 . With such a high protein content loss and an approximate 64-fold signal dilution due to 4-fold isotropic sample expansion, SRS signals are therefore diminished.…”

Molecular resolution imaging by post-labeling expansion single-molecule localization microscopy (Ex-SMLM)

“…8 and 9) 34,35 our findings are highly relevant. Very recently 36,37 , trifunctional linkers have been introduced that are inert to polymerization, digestion and denaturation, and enable direct covalent linking of target molecules and functional groups to the hydrogel. Therefore, trifunctional linkers can retain a high number of labels and fluorescence molecules available for postexpansion imaging.…”

A Picture Worth a Thousand Molecules—Integrative Technologies for Mapping Subcellular Molecular Organization and Plasticity in Developing Circuits

“…For larger volume reconstructions, ExM has been combined with lattice-light-sheet microscopy for high-throughput imaging of neural circuits including entire fly brains (Gao et al, 2019a; Figure 5J). To further increase image resolution, Ex-SIM (Cahoon et al, 2017;Halpern et al, 2017), Ex-STED (Gao et al, 2018), and Ex-STORM (Cang et al, 2016;Shi et al, 2019) have each combined physical expansion with optical superresolution imaging to achieve multiplicative increases in spatial resolution. By combining the advantages of single-molecule photoswitching and spatially-patterned excitation, MINFLUX achieves the highest resolution reported in far-field fluorescence imaging by using a central excitation minimum to localize individual molecules based on their minimal fluorescence emission (Balzarotti et al, 2017).…”