We recently developed a method called expansion microscopy, in which preserved biological specimens are physically magnified by embedding them in a densely crosslinked polyelectrolyte gel, anchoring key labels or biomolecules to the gel, mechanically homogenizing the specimen, and then swelling the gel–specimen composite by ∼4.5× in linear dimension. Here we describe iterative expansion microscopy (iExM), in which a sample is expanded ∼20×. After preliminary expansion a second swellable polymer mesh is formed in the space newly opened up by the first expansion, and the sample is expanded again. iExM expands biological specimens ∼4.5 × 4.5, or ∼20×, and enables ∼25-nm-resolution imaging of cells and tissues on conventional microscopes. We used iExM to visualize synaptic proteins, as well as the detailed architecture of dendritic spines, in mouse brain circuitry.
Subscribe to Journal
Get full journal access for 1 year
only $21.58 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
Chen, F., Tillberg, P.W. & Boyden, E.S. Expansion microscopy. Science 347, 543–548 (2015).
Tillberg, P.W. et al. Protein-retention expansion microscopy of cells and tissues labeled using standard fluorescent proteins and antibodies. Nat. Biotechnol. 34, 987–992 (2016).
Chen, F. et al. Nanoscale imaging of RNA with expansion microscopy. Nat. Methods 13, 679–684 (2016).
O'Connell, P.B.H. & Brady, C.J. Polyacrylamide gels with modified cross-linkages. Anal. Biochem. 76, 63–73 (1976).
Kurenkov, V.F., Hartan, H.-G. & Lobanov, F.I. Alkaline hydrolysis of polyacrylamide. Russ. J. Appl. Chem. 74, 543–554 (2001).
Weber, K., Rathke, P.C. & Osborn, M. Cytoplasmic microtubular images in glutaraldehyde-fixed tissue culture cells by electron microscopy and by immunofluorescence microscopy. Proc. Natl. Acad. Sci. USA 75, 1820–1824 (1978).
Dempsey, G.T., Vaughan, J.C., Chen, K.H., Bates, M. & Zhuang, X. Evaluation of fluorophores for optimal performance in localization-based super-resolution imaging. Nat. Methods 8, 1027–1036 (2011).
Aquino, D. et al. Two-color nanoscopy of three-dimensional volumes by 4Pi detection of stochastically switched fluorophores. Nat. Methods 8, 353–359 (2011).
Chozinski, T.J. et al. Expansion microscopy with conventional antibodies and fluorescent proteins. Nat. Methods 13, 485–488 (2016).
Dani, A., Huang, B., Bergan, J., Dulac, C. & Zhuang, X. Superresolution imaging of chemical synapses in the brain. Neuron 68, 843–856 (2010).
Cai, D., Cohen, K.B., Luo, T., Lichtman, J.W. & Sanes, J.R. Improved tools for the Brainbow toolbox. Nat. Methods 10, 540–547 (2013).
Olivier, N., Keller, D., Gönczy, P. & Manley, S. Resolution doubling in 3D-STORM imaging through improved buffers. PLoS One 8, e69004 (2013).
Wombacher, R. & Cornish, V.W. Chemical tags: applications in live cell fluorescence imaging. J. Biophotonics 4, 391–402 (2011).
Chen, B.-C. et al. Lattice light-sheet microscopy: imaging molecules to embryos at high spatiotemporal resolution. Science 346, 1257998 (2014).
Ke, M.-T., Fujimoto, S. & Imai, T. SeeDB: a simple and morphology-preserving optical clearing agent for neuronal circuit reconstruction. Nat. Neurosci. 16, 1154–1161 (2013).
Choi, H.M.T. et al. Programmable in situ amplification for multiplexed imaging of mRNA expression. Nat. Biotechnol. 28, 1208–1212 (2010).
Cipriano, B.H. et al. Superabsorbent hydrogels that are robust and highly stretchable. Macromolecules 47, 4445–4452 (2014).
Jungmann, R. et al. Multiplexed 3D cellular super-resolution imaging with DNA-PAINT and Exchange-PAINT. Nat. Methods 11, 313–318 (2014).
Lubeck, E., Coskun, A.F., Zhiyentayev, T., Ahmad, M. & Cai, L. Single-cell in situ RNA profiling by sequential hybridization. Nat. Methods 11, 360–361 (2014).
Chen, K.H., Boettiger, A.N., Moffitt, J.R., Wang, S. & Zhuang, X. RNA imaging. Spatially resolved, highly multiplexed RNA profiling in single cells. Science 348, aaa6090 (2015).
Lee, J.H. et al. Highly multiplexed subcellular RNA sequencing. in situ. Science 343, 1360–1363 (2014).
Ku, T. et al. Multiplexed and scalable super-resolution imaging of three-dimensional protein localization in size-adjustable tissues. Nat. Biotechnol. 34, 973–981 (2016).
Klapoetke, N.C. et al. Independent optical excitation of distinct neural populations. Nat. Methods 11, 338–346 (2014).
Chow, B.Y. et al. High-performance genetically targetable optical neural silencing by light-driven proton pumps. Nature 463, 98–102 (2010).
Vaughan, J.C., Dempsey, G.T., Sun, E. & Zhuang, X. Phosphine quenching of cyanine dyes as a versatile tool for fluorescence microscopy. J. Am. Chem. Soc. 135, 1197–1200 (2013).
Legant, W.R. et al. High-density three-dimensional localization microscopy across large volumes. Nat. Methods 13, 359–365 (2016).
Dell, R.B., Holleran, S. & Ramakrishnan, R. Sample size determination. ILAR J. 43, 207–213 (2002).
E.S.B. was funded by the HHMI-Simons Faculty Scholars Program; the NIH Director′s Pioneer Award 1DP1NS087724; the New York Stem Cell Foundation-Robertson Award; the US Army Research Laboratory and the US Army Research Office under contract/grant number W911NF1510548; US-Israel Binational Science Foundation Grant 2014509; the Picower Institute Innovation Fund; IARPA D16PC00008; NIH grants 1R01MH110932, 1R43MH101943, 1R01MH103910, 1R01EY023173, and 2R01DA029639; the IET A.F. Harvey Prize; the Open Philanthropy Project; the Halis Family Foundation; and the MIT Media Lab. J.-B.C. was supported by the Simons Postdoctoral Fellowship. F.C. was supported by the NSF Fellowship and Poitras Fellowship. Y.-G.Y., J.S.K., and H.-J.S. were supported by Samsung Scholarships. P.W.T. and A.T.W. were supported by Hertz Foundation fellowships. Confocal imaging was performed in the W.M. Keck Facility for Biological Imaging at the Whitehead Institute for Biomedical Research J.-B.C. was supported by the Center for Neuroscience Imaging Research. D.C. was funded by NIH grants R21GM114852 and R01MH110932. We acknowledge W. Salmon (MIT) for her assistance with confocal imaging. STORM imaging shown in Figure 2o was performed in the Center for Brain Science at Harvard University. We acknowledge E. Garner, C. Wivagg, and S. Turney (Harvard) for allowing us to use the N-STORM microscope and their assistance with STORM imaging. We acknowledge D. Park (MIT) for assistance with the preparation of cultured neurons. We also acknowledge S. Shim, Y. Sigal, C. Speer, M. Thanawala, D. Kim, M. Sauer, and S. Alon for helpful discussions. We acknowledge University of Michigan, Ann Arbor for providing antibodies against Brainbow fluorescent proteins.
E.S.B., J.-B.C., F.C., and P.W.T. have applied for a patent on iExM (US application 20160305856 A1). E.S.B. is cofounder of Expansion Technologies, a company that aims to provide expansion microscopy kits and services to the community.
Integrated supplementary information
(a) Epifluorescence image of cultured BS-C-1 cells immunostained with an antibody against beta tubulin, and expanded ~16-fold via hp-iExM. (b) Confocal image of BS-C-1 cells with labeled microtubules, after ~16-fold expansion via hp-iExM. The inset in upper right zooms in on the small box at center. (c) Transverse profile of microtubules in the boxed region (dotted lines) of the inset of b after averaging down the long axis of the box and then normalizing to the peak (blue dots), with superimposed fit with a sum of two Gaussians (red lines). The peak-to-peak distance of the two Gaussian functions was 58.3 nm. (d) Population data for 277 microtubule segments from three expanded samples, showing a histogram of the peak-to-peak distances (calculated as in c).
(a) Schematic of simulation strategy (see Supplementary Note 2 for details), along with depiction of the calculated point-spread function of the microscope we used (a spinning disk confocal microscope with a pinhole size of 50 μm equipped with a 40x NA1.15 objective lens). Scale bar, 200 nm. (b) Confocal microscope image of microtubules from cultured BS-C-1 cells after 20-fold expansion via iExM. The microtubule segment in the boxed region was analyzed in c. (c) Transverse profile (red dots) of the microtubule segment in the boxed region of b with superimposed fit with a sum of two Gaussians (red line) and simulated (using the model of Supplementary Fig. 2a) microtubule profile with a Ri of 30.6 nm and Ro of 34.8 nm (green dots) with superimposed fit with a sum of two Gaussians (green line). See section ‘MATLAB simulation of iExM images’ of the Methods for details of the fitting. The green and red numbers indicate the full width at half maximum for individual microtubule sidewalls, both simulated and real (see Supplementary Fig. 6 for more analyses). (d) As in c, but for a different region (not shown in b). (e) Population data for 129 microtubule segments from one culture (numbers indicate mean ± standard deviation), showing a histogram of Ri and Ro. (f) Population data for 129 microtubule segments from one culture (mean ± standard deviation), showing a histogram of Ro-Ri.
The 5’ acrydites of the DNA oligos are distributed in the purple shaded region (corresponding to the cylinder of Supplementary Fig. 2a). We constructed a detailed schematic showing a possible arrangement of antibodies and DNA around microtubules, building from Supplementary Fig. 2. As shown in Supplementary Fig. 3, the radius of the microtubule is 12.5 nm, as previously measured by transmission electron microscopy (TEM)1. The radius of an immunostained microtubule (stained with conventional primary and secondary antibodies) is 30 nm, as measured by previous TEM imaging1. The 5’ acrydites of the DNA are distributed in a cylinder with an inner radius of 26.7 nm and outer radius of 33.5 nm, as derived in Supplementary Fig. 2e. As can be seen in Supplementary Fig. 3, the DNA-conjugated secondary antibody makes the radius of the microtubule 3.5 nm larger than the microtubule labeled with regular antibodies (outer radius of 30 nm).
Supplementary Figure 4 Example of how antibody-bearing microscopy images differ as a function of the geometry of target protein complexes and the size of probes.
See Supplementary Note 3 for related results text. (a) Cross section of a rectangular protein complex of interest, with an end-to-end length of 25 nm, and immunolabeled at the two lateral ends. (b) Cross section of a cylindrical protein complex of interest with a diameter of 25 nm, labeled from all sides. (c) Cross section of the same protein complex as in a, but labeled with a DNA-conjugated secondary antibody. Panel a and c show examples in which primary antibodies (green) bind only to the left and right surfaces of the protein complex. In panel a, fluorophore (yellow star)-labeled secondary antibodies (orange) are used. In panel c, DNA (purple line)-conjugated secondary antibodies (orange) are used. (d) The resulting fluorescence signal profiles (modeled by convolving the patterns of panels a-c with a FWHM of 22.8 nm, the point-spread function (PSF) of iExM excluding the label size (Supplementary Fig. 6b)) approximate how a small object might look when labeled with conventional antibodies (panel a and b) or DNA-conjugated antibodies (panel c) imaged via a super-resolution technique with a resolution of 22.8 nm (such as iExM).
See Supplementary Note 4 for associated results text. (a) Schematic of the DNA-conjugated secondary antibody used in this study. (b) Schematic of a regular secondary antibody bearing fluorophores. (c-e), Three possible designs of DNA-conjugated secondary antibodies with a smaller size. (c) A shorter strand of DNA is used. (d) The acrydite (i.e., gel anchoring group) is positioned at the proximal end of the DNA. (e) A single stranded DNA with an acrydite moiety is conjugated to a secondary antibody.
See Supplementary Note 5 for associated results text. (a) Population data for 129 microtubules from one culture (two such sidewalls were obtained from each microtubule, for a total of 258 individual sidewalls), showing a histogram of the full width at half-maximums (FWHMs) of these single sidewalls (examples of such FWHMs are shown in Supplementary Fig. 2c and 2d by green and red numbers). Green, FWHMs of simulated (according to the model of Supplementary Fig. 2a) microtubule profiles, as shown in green in Supplementary Fig. 2c and 2d. Red, FWHMs of experimental microtubule profiles, as shown in red in Supplementary Fig. 2c and 2d. This is the “overall point spread function (PSF)” of iExM, which includes contributions to the PSF from the labels (primary and DNA oligo-conjugated secondary antibody), gelation and expansion steps, and optical diffraction. (b) A PSF of iExM, generated by deconvolving the experimental microtubule profile of Supplementary Fig. 2c with a simulated microtubule labeled with DNA-conjugated antibodies (simulated as in Supplementary Fig. 3, and not including the effects of optical diffraction), that isolates the gelation, expansion, and optical contributions to the PSF (that is, omitting the label contributions). (c) As in Supplementary Fig. 2c, but simulating additional 5 and 10 nm positional errors.
(a) Confocal microscopy image of microtubules of a 100-μm slice of preserved mouse lung tissue after 22-fold expansion by iExM. Inset zooms in the boxed area in a. (b) Maximum intensity projection of the sample imaged in a. (c) Transverse profile of the microtubules of the inset of a (blue dots), with superimposed fit with a sum of two Gaussians (red line). (d) Population data for 55 microtubule segments from one expanded sample, showing a histogram of the peak-to-peak distances. (e-h) As in a-d, but for a 100-μm thick slice of preserved mouse liver after 19-fold expansion. (h) 95 microtubule segments from one expanded sample were analyzed. (i) Representative image of microtubule bundles observed in mouse brain cells (cortex). The mouse cortex was immunostained with an antibody against tubulin and expanded 18-fold using iExM. z-stack images were acquired and the cross-sectional views shown in the bottom panels were constructed. (j) Confocal microscopy image of a microtubule bundle after 18-fold expansion. (k) Confocal microscopy image of the same microtubule bundle shown in j after shrinking the gel back to 6.5-fold in a salt solution, resulting in a resolution more similar to earlier ExM versions (~4.5-fold). (l) Transverse profiles of microtubules in the boxed regions of j and k (blue: j, orange: k).
Supplementary Figure 8 Schematic of signal amplification based on DNA and locked nucleic acid (LNA).
See Supplementary Note 7 for associated results text. (a) After the first hydrogel (dark blue) is expanded and re-embedded in an uncharged polyacrylamide gel (not shown), a long DNA (we call it ‘linker DNA’, schematized in the bottom left), consisting of the A’ sequence (red) followed by four repeats of a new sequence B’ (light blue), and equipped with a polymerizable group (black dot), is hybridized to the anchored DNA (green strand). (b) The second swellable hydrogel (orange) is formed, incorporating the new linker DNA strand. (c) DNA or LNA with a sequence of B (purple dotted line) with a fluorophore (yellow star) is hybridized to the linker DNA, enabling more fluorophores to be appended to the site of a given biomolecule. (d) The second gel is expanded in 0.2x PBS for DNA and in distilled (DI) water for LNA.
Epifluorescence image of Emx1-Cre mouse cortex labeled with anti-Homer1 (green) and anti-Bassoon (red), after 18-fold expansion via iExM and LNA hybridization-based signal amplification. Lower panels show magnified views of boxed regions in the upper panel.
Supplementary Figure 10 Confocal images of two brain regions stained with anti-Bassoon and anti-GABAARα1/α2.
Brain region: a, globus pallidus; b, cortex. Brains were expanded 16-fold via iExM. See Supplementary Video 4 for 3-D movie of a.
(a-g) Single xy-plane images of the boxed region of Fig. 3iv at different z-heights. (a) z=3.05 μm. (b) z=3.15 μm. (c) z=3.28 μm. (d) z=3.33 μm. (e) z=3.38 μm. (f) z=3.42 μm. (g) z=3.55 μm. (h) Upper left shows a single xy-plane image at a z-height of 3.33 μm (same with d); right shows a single yz-plane image reconstructed from the z-stack image; bottom shows a single xz-plane image reconstructed from the z-stack image; dotted lines of upper left show the single x plane and y plane shown in the right and bottom. Scale bar = 500 nm.
16-fold expanded hippocampal brain circuitry with labeled mTFP (blue), mCherry (green), TagBFP (red). mCherry (green) signals and TagBFP (red) signals were amplified by LNA hybridization-based amplification. mTFP (blue) signals were not amplified. (a) Single confocal xy-plane image; (b-e) single xy-plane images at, and flanking (i.e., at different z-heights above and below the xy-plane of), the boxed region of a. (f) Single yz-plane image reconstructed from the z-stack images shown in b-e. Scale bar = 1 μm. (g) 3-D visualization of the volume shown in b-e. (h) 3-D visualization of the z-stack shown in a. See Supplementary Video 6.
Two confocal z-stack images of 16-fold expanded mouse hippocampal circuitry with labeled mTFP (blue), mCherry (green), TagBFP (red). mCherry (green) signals and TagBFP (red) signals were amplified by LNA hybridization-based amplification. mTFP (blue) signals were not amplified. (a-g) Single xy-plane images at different z-heights. (h) Snapshot of a 3-D visualization of the z-stack shown in a-g; see Supplementary Video 8. (i-o) Single xy-plane images at different z-heights. (p) Snapshot of a 3-D visualization of the z-stack shown in i-o; see Supplementary Video 9.
Supplementary Figure 14 Large-area tiling epifluorescence image of Emx1-Cre mouse cortex expressing membrane-bound fluorescent proteins (Brainbow AAVs).
TagBFP (green), mCherry (red), and EYFP (blue) were immunostained, expanded 16-fold via hp-iExM. TagBFP (green) and mCherry (red) signals were amplified by LNA hybridization-based signal amplification; EYFP (blue) signals were not amplified.
Epifluorescence image of cultured BS-C-1 cells labeled with anti-tubulin, and expanded ~53-fold via three consecutive expansions. Signal was amplified by LNA hybridization-based signal amplification. Associated results in Supplementary Note 8.
Supplementary Figures 1–15, Supplementary Tables 1–14, and Supplementary Notes 1–8 (PDF 3899 kb)
Supplementary Protocol (PDF 408 kb)
iExM simulator. MATLAB program simulating iExM of a microtubule labeled with a primary antibody and DNA-conjugated secondary antibody. (ZIP 155 kb)
3-D visualization of the z-stack confocal image of beta tubulin-stained BS-C-1 cells after 20-fold expansion via iExM shown in Fig. 2g. (MOV 12543 kb)
3-D visualization of the center region of the z-stack shown in Supplementary Video 1 and Fig. 2g. Red numbers of the bounding box of the video show scales in micron. (MOV 3156 kb)
3-D visualization of the z-stack confocal image of Homer1(green)/Bassoon(red) stained medial pallidum after 16-fold expansion via iExM shown in Fig. 3l-o. Red numbers of the bounding box of the video show scales in micron. (MOV 3169 kb)
3-D visualization of the z-stack confocal image of GABAARα1α2/Bassoon stained globus pallidus after 16-fold expansion via iExM shown in Supplementary Fig. 10a. (MOV 16297 kb)
Z-stack confocal image of the molecular layer of the mouse hippocampal dentate gyrus after immunostaining of EYFP (blue) and mCherry (green) and 20-fold expansion via iExM shown in Fig. 4c. This video also shows 3-D visualization and surface rendering of the stack. (MOV 18754 kb)
3-D visualization of the z-stack confocal image of the molecular layer of the mouse hippocampal dentate gyrus after immunostaining of mCherry, TagBFP, and mTFP and 16-fold expansion via iExM shown in Supplementary Fig. 12h. (MOV 17047 kb)
3-D visualization of the z-stack confocal image of the molecular layer of the mouse hippocampal dentate gyrus after immunostaining of mCherry, TagBFP, mTFP, and EYFP and 20-fold expansion via iExM shown in Fig. 4g. (MOV 19376 kb)
3-D visualization of the z-stack confocal image of the molecular layer of the mouse hippocampal dentate gyrus after immunostaining of mCherry, TagBFP, and mTFP and 16-fold expansion via iExM shown in Supplementary Fig. 13h. (MOV 15434 kb)
3-D visualization of the z-stack confocal image of the molecular layer of the mouse hippocampal dentate gyrus after immunostaining of mCherry, TagBFP, and mTFP and 16-fold expansion via iExM shown in Supplementary Fig. 13p. (MOV 15682 kb)
About this article
Cite this article
Chang, JB., Chen, F., Yoon, YG. et al. Iterative expansion microscopy. Nat Methods 14, 593–599 (2017). https://doi.org/10.1038/nmeth.4261
Chemical Communications (2020)
ACS Nano (2020)
Bioconjugate Chemistry (2020)
Systematic Evaluation of Chemically Distinct Tissue Optical Clearing Techniques in Murine Lymph Nodes
The Journal of Immunology (2020)