An Introduction to Expansion Microscopy Protocols

Expansion microscopy (ExM) is an imaging technique that embeds samples within a dense, swellable hydrogel and enlarges them isotropically in three dimensions.^1,2 This physical sample enlargement enables ExM to achieve a four- to five-fold linear expansion, reaching the ~60–70 nm resolution range. This significantly enhances the capabilities of conventional diffraction-limited microscopes, whose optical resolving power is capped by the wavelength of visible light at roughly 200 to 300 nm.^2,3 The defining advantage of this methodology is its democratization of nanoscale imaging, leveraging conventional fluorescence microscopes that are already common in most labs. 

Overview of the ExM workflow

ExM is largely defined by its workflow that relies heavily on the chemical preparation of samples. Fixed and permeabilized specimens are first anchored to a polymerizable handle, then embedded in an absorbent polyelectrolyte hydrogel, often acrylamide-based.^2,3 The anchored sample is then homogenized, typically through enzymatic digestion or heat-detergent denaturation, to relieve the protein-protein interactions and mechanical constraints that would otherwise resist uniform swelling. Water then expands the gel, where the strongly hydrophilic, negatively charged network can imbibe roughly 100 times its dry volume.⁴ The anchoring chemistry depends on the type of molecular target. Proteins are typically anchored through amine-reactive reagents such as acryloyl-X, SE (AcX), while RNA is anchored instead through a guanine-reactive reagent such as LabelX. Newer universal reagents, including methacrolein, are increasingly used to anchor proteins, nucleic acids, and lipids simultaneously in a single gelation step.^2,4,5

Once expanded, samples are imaged directly in the water or buffer used to swell them, mounted on a glass-bottom dish or slide and, for extended acquisition, held in place with poly-L-lysine, agarose, or a bead of cyanoacrylate glue to limit drift. Because the expanded gel is mostly water, a water-immersion objective is generally preferred above roughly 20x magnification to avoid any refractive-index mismatch aberrations.^4,6 Widefield epifluorescence and laser-scanning confocal remain the most practical choices for imaging cultured cells and thin sections, while spinning-disk confocal suits larger tissue volumes. Light-sheet microscopy is generally preferred for the largest samples, since its fast optical sectioning offsets the longer acquisition times needed by thicker specimens.^4,6

ExM protocols for specific applications

The original ExM chemistry has since branched into a family of protocols that are now most usefully distinguished by the application they were built to solve. Key differences among these include expansion factor, tissue compatibility, and processing time, which will depend on variables such as sample type and throughput.

Cultured cells and thin tissue

For cultured cells and thin tissue sections, protein-retention ExM (proExM) and its RNA counterpart, ExFISH, can be used as a reasonable starting point. By anchoring proteins directly through AcX in place of custom gel-anchorable fluorophore labels, proExM allows both antibody-conjugated dyes and endogenous fluorescent proteins to survive a standard ~4x expansion using only commercially available reagents.² ExFISH extends the same logic to RNA, anchoring transcripts to the gel through LabelX. Readout is done by post-expansion in situ hybridization, typically single-molecule FISH for cultured cells or hybridization chain reaction amplification for tissue, where unamplified signal is too dim to detect.⁴ Both protocols rely on standard fixation, commercially available reagents, and a benchtop incubator, making this ExM workflow relatively accessible before adapting the chemistry to thicker or more specialized samples.

Tissue blocks

Beyond thin sections, anchoring reagents and digestion enzymes must diffuse fully through the tissue before gelation and homogenization can proceed evenly. Protocols adapted for thick tissue blocks generally slow gelation and extend the digestion time, allowing reagents more time to penetrate before the gel sets.⁴ Whole-mount tissues that are not pre-sectioned can add a further handling challenge, as free-floating samples tend to fold or curl during gelation unless physically mounted. One recent adaptation modified an existing 4x ExM protocol specifically to image the neuromuscular junction in whole-mount human and mouse muscle biopsies without sectioning, extending the digestion step to clear the tissue while preserving fluorescent signal.¹¹ Expansion factors validated on soft tissue such as brain also do not necessarily transfer to mechanically tougher organs, which can distort or require harsher digestion to expand evenly.⁸ As a general rule, the thicker or tougher the tissue, the more a standard protocol needs to be re-optimized.

Fixed tissues

Formalin-fixed, paraffin-embedded (FFPE) tissue, the format in which most clinical and archival specimens exist, poses a distinct challenge because formaldehyde-induced crosslinking resists the digestion that fresh tissue would otherwise tolerate easily. Magnify, a universal anchoring chemistry that links proteins, nucleic acids, and lipids to the gel in a single step, has been validated on FFPE kidney tissue. The protocol includes a harsher, longer homogenization step than those of fresh-tissue methods and has demonstrated substantially better expansion and protein retention on FFPE specimens than earlier, gentler anchoring chemistries.⁵ However, not every high-factor protocol has yet been tested on FFPE tissue, and some newer single-shot approaches cite testing on human clinical tissue more broadly merely as a future direction.⁸ Regardless of anchoring chemistry, the dehydration and embedding steps used to prepare FFPE tissue strip lipids from the specimen, so lipid imaging is generally not recoverable.⁵ Aside from imaging-based readouts, expansion-based proteomics methods have also been demonstrated directly on archival FFPE specimens.¹⁰

Ultrastructure imaging

For applications where near-EM ultrastructural detail matters more than tissue scale, iterative and high-factor protocols have been developed. Ultrastructure expansion microscopy (U-ExM) and its iterative descendant, iU-ExM, were developed largely on cultured cells and isolated organelles, including centrioles, nuclear pore complexes, and Chlamydomonas basal bodies. These have reached a combined 14- to 26-fold expansion sufficient to resolve the eightfold symmetry of the nuclear pore complex on a conventional confocal.⁷ iU-ExM has also been applied to intact mouse retina, embedded whole and sectioned only after gelation, though locating a small region of interest within a much larger expanded volume becomes progressively harder as the expansion factor increases.⁷ Notably, single-shot 20-fold ExM can offer a faster route to comparable resolution without the multiple gelation and re-embedding rounds required by iU-ExM.^7,8

Whole organs and organisms

Currently, true whole-organ and whole-organism imaging by ExM remains in its infancy. Some protocols point to large-working-distance objectives and more permeable gel chemistries as a plausible route toward expanding much larger tissue volumes. However, such protocols have yet to be definitively demonstrated.⁵ Broader reviews of the field note that whole organs and organisms generally require longer processing times and more precise control than smaller specimens, while rigid or calcified tissue such as bone poses a separate challenge, resisting hydrogel infiltration and homogenization outright.¹⁰

High-throughput and expansion omics

High-throughput ExM adapts the protocol to accommodate a multiwell plate format, expanding and imaging many samples in parallel. Instead of the standard slide-sized gel being imaged manually, smaller gels within wells are imaged using automated, high-content confocal imagers.⁹ One study demonstrated this approach's relevance to drug development directly, detecting a dose-dependent phenotype in treated cells that was invisible before expansion, demonstrating ExM as a viable screening readout.⁹

A parallel effort focuses on scaling what each expanded sample can reveal, not simply how many samples get processed. A growing set of protocols, together termed "expansion omics," extends expansion beyond fluorescence imaging to profile the transcriptome (ExFISH, expansion MERFISH, ExSeq, expansion spatial transcriptomics), proteome (FAXP, ExPRESSO), lipidome (GAMSI, LExM, Ex-MSI), and epigenome (SCEPTRE, ExEpi, ChromExM) across a variety of sample types, including suspension cells, fresh-frozen tissue, and FFPE specimens.¹⁰ In practice, these methods remain mostly single-omic, since extracting more than one biomolecule class from the same gel without cross-contamination is still technically demanding. A general sample-preparation workflow for spatial biology has not yet been established.¹⁰

Addressing common obstacles

Several notable issues tend to arise across many ExM protocols, regardless of the specific chemistry used: loss of fluorescent signal, anisotropic (non-uniform) expansion, and a persistent trade-off between homogenization strength and epitope preservation. Signal loss arises when gel polymerization degrades certain fluorophores or when unanchored probes wash out during homogenization and expansion. This risk can be reduced by choosing expansion-validated dyes and fluorescent proteins or by adding a post-expansion amplification step where brightness is limiting.^3,4

Anisotropic expansion, in which some regions of a sample expand more than others, is the more consequential obstacle for quantitative or ultrastructural work. It distorts the spatial measurements ExM is meant to preserve, and it typically stems from uneven fixation, incomplete homogenization, or an unevenly crosslinked hydrogel.^3,6 No single protocol is guaranteed to eliminate this distortion outright, and a user cannot assume that a published expansion factor will transfer unchanged. Thus, it is important to validate the expansion factor and isotropy for each new sample type against a known structure or calibration reference.^5,6

Finally, homogenization strength and antibody compatibility can pull in opposite directions. A harsher digestion expands the sample more evenly but destroys more epitopes, while gentler, heat-based homogenization preserves epitopes at the risk of incomplete or uneven expansion.^2,5 New adopters should treat this trade-off as a protocol-selection decision, addressed early in experimental planning. Consider whether an experiment depends heavily on antibodies validated before gelation or on the flexibility to stain new targets after the expansion. Approached this way, all three obstacles become considerations to plan around from the outset, not failures to troubleshoot after the fact.

Final considerations for new adopters of ExM

New adopters of ExM can benefit from quickly resolving several high-priority considerations. Among these are the resolution required for the biological question, the biomolecule class being targeted, the biological model system, the tolerable signal loss, and the protein density of the structure of interest.³ Resolution needs and biomolecule class are usually the first two to settle, since they determine whether a standard proExM workflow suffices or a higher-factor, iterative protocol is warranted, and which anchoring chemistry applies.^2,3 The remaining variables typically require sample-specific validation, not an answer from a published protocol alone, since fixed archival tissue, dense connective tissue, and soft neural tissue each behave differently under identical chemistry.^3,5,8

Viewing these variables as coupled trade-offs instead of independent parameters, the more useful framing is not so much "which protocol is best," but "which protocol better matches this sample and this question." A choice optimized for one application, such as high-resolution on isolated organelles, often will not work as cleanly with another, such as high-throughput screening of cultured cells.³ Equally relevant are the obstacles that frequently come with adapting that chemistry to a new sample, including signal loss, anisotropic distortion, and the homogenization-epitope trade-off. Fortunately, these have been well-characterized, each with a documented mitigation that a new adopter can plan around.

The growing popularity of ExM is evident with the formation of the International Expansion Microscopy User Group, a community spanning more than 40 countries.⁶ Here, members share protocol failures alongside successes, which shortens the trial-and-error period a new lab would otherwise spend. Reagents that once required custom synthesis are now commercially available, and expansion factors that needed multiple iterative gelation rounds as recently as 2017 are achievable in a single step today, both changes this user community has helped document and share.^6,8 As the underlying chemistry, workflow, and this collaborative community continue to mature, additional protocols suited to new sample types and applications are likely to follow.

References

1. Chen F, Tillberg PW, Boyden ES. Optical imaging. Expansion microscopy. Science. 2015;347(6221):543-548. doi:10.1126/science.1260088

2. Tillberg PW, Chen F, Piatkevich KD, et al. Protein-retention expansion microscopy of cells and tissues labeled using standard fluorescent proteins and antibodies. Nat Biotechnol. 2016;34(9):987-992. doi:10.1038/nbt.3625

3. Zhuang Y, Shi X. Expansion microscopy: A chemical approach for super-resolution microscopy. Curr Opin Struct Biol. 2023;81:102614. doi:10.1016/j.sbi.2023.102614

4. Asano SM, Gao R, Wassie AT, Tillberg PW, Chen F, Boyden ES. Expansion Microscopy: Protocols for Imaging Proteins and RNA in Cells and Tissues. Curr Protoc Cell Biol. 2018;80(1):e56. doi:10.1002/cpcb.56

5. Klimas A, Gallagher BR, Wijesekara P, et al. Magnify is a universal molecular anchoring strategy for expansion microscopy. Nat Biotechnol. 2023;41(6):858-869. doi:10.1038/s41587-022-01546-1

6. Woo N, Brown CM. Review of expansion microscopy combined with advanced imaging modalities. J Microsc. 2026;301(3):335-354. doi:10.1111/jmi.70048

7. Louvel V, Haase R, Mercey O, et al. iU-ExM: nanoscopy of organelles and tissues with iterative ultrastructure expansion microscopy. Nat Commun. 2023;14(1):7893. Published 2023 Nov 30. doi:10.1038/s41467-023-43582-8

8. Wang S, Shin TW, Yoder HB 2nd, et al. Single-shot 20-fold expansion microscopy. Nat Methods. 2024;21(11):2128-2134. doi:10.1038/s41592-024-02454-9

9. Day JH, Della Santina CM, Maretich P, et al. High-throughput expansion microscopy enables scalable super-resolution imaging. Elife. 2024;13:RP96025. Published 2024 Nov 26. doi:10.7554/eLife.96025

10. Dong Z, Xiang W, Jiang W, Guo T. Expansion omics: from expansion microscopy to spatial omics. Mol Syst Biol. 2026;22(2):165-178. doi:10.1038/s44320-025-00171-9

11. Ramadan A, Sheard TMD, Alhindi A, et al. Expansion microscopy reveals nano-scale insights into the human neuromuscular junction. Cell Rep Methods. 2025;5(6):101082. doi:10.1016/j.crmeth.2025.101082