We combined instant structured illumination microscopy (iSIM) with total internal reflection fluorescence microscopy (TIRFM) in an approach referred to as instant TIRF-SIM, thereby improving the lateral spatial resolution of TIRFM to 115 ± 13 nm without compromising speed, and enabling imaging frame rates up to 100 Hz over hundreds of time points. We applied instant TIRF-SIM to multiple live samples and achieved rapid, high-contrast super-resolution imaging close to the coverslip surface.
Subscribe to Journal
Get full journal access for 1 year
only $4.92 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Tax calculation will be finalised during checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
Poulter, N. S., Pitkeathly, W. T. E., Smith, P. J. & Rappoport, J. Z. in Advanced Fluorescence Microscopy: Methods and Protocols Vol. 1251 (ed Verveer, P. J.) (Springer, New York, 2015).
Fiolka, R., Beck, M. & Stemmer, A. Opt. Lett. 33, 1629–1631 (2008).
Kner, P., Chhun, B. B., Griffis, E. R., Winoto, L. & Gustafsson, M. G. L. Nat. Methods 6, 339–342 (2009).
Chung, E., Kim, D., Cui, Y., Kim, Y. H. & So, P. T. Biophys. J. 93, 1747–1757 (2007).
Gliko, O., Reddy, G. D., Anvari, B., Brownell, W. E. & Saggau, P. J. Biomed. Opt. 11, 064013 (2006).
Li, D. et al. Science 349, aab3500 (2015).
York, A. G. et al. Nat. Methods 10, 1122–1126 (2013).
Roth, S., Sheppard, C. J. R., Wicker, K. & Heintzmann, R. Opt. Nanoscopy https://doi.org/10.1186/2192-2853-2-5 (2013).
De Luca, G. M. et al. Biomed. Opt. Express 4, 2644–2656 (2013).
Curd, A. et al. Methods 88, 37–47 (2015).
Stout, A. L. & Axelrod, D. Appl. Opt. 28, 5237–5242 (1989).
Gould, T. J., Myers, J. R. & Bewersdorf, J. Opt. Express 19, 13351–13357 (2011).
Burnette, D. T. et al. J. Cell Biol. 205, 83–96 (2014).
Sochacki, K. A., Dickey, A. M., Strub, M. P. & Taraska, J. W. Nat. Cell Biol. 19, 352–361 (2017).
Faire, K. et al. J. Cell Sci. 112, 4243–4255 (1999).
Miller, A. L. & Bement, W. M. Nat. Cell Biol. 11, 71–77 (2009).
Apolloni, A., Prior, I. A., Lindsay, M., Parton, R. G. & Hancock, J. F. Mol. Cell. Biol. 20, 2475–2487 (2000).
Grimm, J. B. et al. Nat. Methods 12, 244–250 (2015).
Presley, J. F. et al. Nature 389, 81–85 (1997).
Choy, E. et al. Cell 98, 69–80 (1999).
Takahashi, S. et al. J. Cell Sci. 125, 4049–4057 (2012).
Shen, H. et al. Chem. Rev. 117, 7331–7376 (2017).
Fiolka, R. Opt. Express 24, 29556–29567 (2016).
Olveczky, B. P., Periasamy, N. & Verkman, A. S. Biophys. J. 73, 2836–2847 (1997).
Winter, P. W. et al. Optica 1, 181–191 (2014).
Mattheyses, A. L. & Axelrod, D. J. Biomed. Opt. 11, 014006 (2006).
Fu, Y. et al. Proc. Natl. Acad. Sci. USA 113, 4368–4373 (2016).
Richardson, W. H. J. Opt. Soc. Am. 62, 55–59 (1972).
Lucy, L. B. Astron. J. 79, 745–754 (1974).
Miura, K., Rueden, C., Hiner, M., Schindelin, J. & Rietdorf, J. ImageJ Plugin CorrectBleach V2.0.2. Zenodo. https://doi.org/10.5281/zenodo.30769 (2014).
Li, D. & Betzig, E. Science 352, 527 (2016).
Thévenaz, P., Ruttimann, U. E. & Unser, M. IEEE Trans. Image Process. 7, 27–41 (1998).
Tinevez, J. Y. et al. Methods 115, 80–90 (2017).
Choudhury, A. et al. J. Clin. Invest. 109, 1541–1550 (2002).
Baird, M. A. et al. Mol. Biol. Cell 28, 240–251 (2017).
We thank S. Lee and A. Kumar for useful discussion on instant TIRF-SIM, H. Eden and P. La Riviere for providing feedback on the manuscript, C. Combs (National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, MD, USA) for loaning us the Leica alignment reticle, E. Tyler for help with the illustrations, W. Bement (Department of Integrative Biology, University of Wisconsin–Madison, Madison, WI, USA) for the gift of EMTB–3×EGFP, D. Mullins (Department of Cellular and Molecular Pharmacology, University of California, San Fransisco, San Francisco, CA, USA) for the gift of pEGFP-C1 F-tractin–EGFP, D. Esposito (Frederick National Laboratory for Cancer Research, Frederick, MD, USA) for the gift of the HaloTag chimera of HRas, and L. Lavis (Janelia Research Campus, Howard Hughes Medical Institute, Ashburn, VA, USA) for the gift of Janelia Fluor 549. Support for this work was provided by the Intramural Research Programs of the National Institute of Biomedical Imaging and Bioengineering; the National Heart, Lung, and Blood Institute; and the NSF (grant 1607645 to A.U. and I.R.-S.). J.W.T. and H.S. acknowledge funding from the NIH Director's Challenge Innovation Award Program.
The authors declare no competing interests.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Integrated supplementary information
488 nm and 561 nm lasers are combined and passed through an acousto-optical tunable filter (AOTF) for shuttering, beam expanded and spatially filtered (f = 45 mm and f = 400 mm achromats with 100 μm pinhole placed at the co-focal point between these lenses), and directed to an excitation microlens array. An opaque circular mask placed at the co-focal point between Scan lens 1 and Scan lens 2 (one focal length from the excitation foci produced by the microlens array) serves to block subcritical excitation (blue), producing annular illumination. For clarity, only a single on-axis beamlet produced by the central microlens is shown impinging on the mask. Scan lens 2 and 3 relay the mask to a 2 sided galvanometric mirror, with a Compensator plate reducing astigmatism that would otherwise result as the beam is transmitted through a thick dichroic mirror (TDC). A tube lens and Scan lens 4 further magnify and relay the mask to the back focal plane of a high NA (1.7) objective. Scanning the galvo translates the TIRF excitation pattern at the sample. Fluorescence is collected along the same optical path, descanned from the galvo and reflected from TDC. Emission-side optics are nearly identical to those used in our previous work (see Methods for further description). Note that distances, angles, and ray diagrams are only approximate.
Supplementary Figure 2 Comparative (non-TIRF/TIRF) excitation and emission patterns after micro-optics.
The response of each micro-optic as measured with fluorescence dye solution is shown for non-TIRF (entire objective back aperture illuminated, left column) and TIRF (annular illumination, middle column) excitation. Line scans corresponding to red lines in each image are shown in right column (blue: non-TIRF imaging; red: TIRF imaging). Although contrast using excitation microlenses is much better in TIRF than in non-TIRFl imaging, there is still residual background between excitation foci. This background is removed after adding pinholes; performance after pinholes (second row) and with emission microlenses (third row) is virtually identical whether exciting in TIRF or not. Note that these images were acquired with the galvanometric mirror held stationary, to visualize the emission around individual excitation foci. Scale bar: 2 μm. Experiments were repeated at least 3 times with similar results; representative data are shown.
Supplementary Figure 3 Estimation of evanescent field decay length with silica beads in fluorescent dye.
a) Experimental geometry. Given the known radius of the silica bead R, the apparent radius r recorded from the measurement can be converted to an axial coordinate z using geometry. b) Representative bead image showing dark central region corresponding to excluded dye region, surrounded by fluorescence signal corresponding to dye. Images were inverted and the centroid computed. Based on the centroid, a ring with 1 pixel width and radius of r was drawn, and the intensity of all pixels within the ring averaged. By varying r, the averaged intensity profile versus r, F(r) was obtained. The signal at r = 0 was regarded as background and subtracted from the profile. Then, F(r) was transformed to F(z), by computing z as R – (R2-r2)0.5. Data from n = 7 independent beads were averaged (c), central value represents mean and error bars indicates standard deviation) fitted to a sum of two exponentials as described in Methods, and fit parameters extracted to determine the decay length. Scale bar: 1 µm. Experiments were performed at least 4 times with similar results; representative data are shown.
U2OS cells were fixed, microtubules immunostained with Alexa Fluor 488, and the same field imaged in instant SIM (non-TIRF illumination) (a) and instant TIRF-SIM (b). Arrows in higher magnification views (c, d) highlight features evident in instant SIM imaging but absent or attenuated in instant TIRF-SIM. Images have been delined as descripted in Supplementary Fig. 5–7. Scale bars: 10 μm in a, b; 2 μm in c, d. Experiments were repeated at least 4 times with similar results; representative data are shown.
a) Example of data from Fig. 1a. Top: before delining, bottom: after delining. b) corresponding power spectra of images in a). Top: power spectrum corresponding to the image before delining. Blue arrows highlight locations in Fourier space that correspond to horizontal line artifacts in the image. Bottom: power spectrum corresponding to the image after delining. Notch filters (black rectangular regions of zero intensity) are selectively applied, resulting in suppression of line artifacts. c), d) Additional examples from data presented in Fig. 2c and Fig. 3a, highlighting delining. Insets show higher magnification views of yellow and red rectangular regions. Scale bars: a), c) and d) 5 μm, insets 2 μm, b) 1/200 nm−1. Experiments were repeated at least 4 times with similar results; representative data are shown.
a) Fluorescent dye, imaged in instant TIRF-SIM, flat-fielded and delined according to procedure in Supplementary Fig. 5. Red and blue lines indicate profiles in vertical (perpendicular to line artifact) and horizontal (parallel) directions. b) Variation in intensity along line profiles in a). Although vertical variations are larger than horizontal, they are still only ~2.4% of the mean intensity of the sample. Scale bar: 5 μm. Experiments were repeated 3 times with similar results ; representative data are shown.
Supplementary Figure 7 Intensity fluctuations in delined EGFP–Rab11 data along the vertical direction are no greater than those along the horizontal direction.
a) Data shown in Fig. 3a, with 15 selected tracks (#1–5 highlighted with white arrows, relatively stationary tracks with displacement < 0.3 μm; #6–10, tracks with motion primarily along vertical direction, perpendicular to the line artifact; #11–15, tracks with motion primarily along horizontal direction, parallel to line artifact). b) Higher magnification views of tracks 6–15, highlighted with white arrows. c) Example intensities from tracks 2, 8, 14. Intensities have been normalized to the mean (=1) and then smoothed with the 'smoothdata' function in Matlab, with a Gaussian kernel of window size 20. Both raw (blue) and smoothed (red) traces are shown. d) Examples in c after subtracting smoothed traces, emphasizing high frequency intensity variations. e) As in d), but now showing high frequency intensity variations for all tracks in relatively stationary (top), primarily vertical (middle) and primarily horizontal (bottom) tracks. Also shown are the mean standard deviations <σ> as a percentage of the mean intensity in each panel. Although <σ>s derived from vertical and horizontal tracks are both slightly greater than stationary tracks, they are within 1% of each other. Scale bars: 5 μm in a, 1 μm in b. Experiments were repeated 4 times with similar results; representative data are shown.
Beads were imaged using non-TIRF illumination (a, c and e) and in TIRF (b, d and f). Bead appearance is similar regardless of imaging modality. Scale bars: 200 nm.
a) GFP-Myosin IIA in live U2OS cells. Inset highlights myosin IIA bipolar filaments. b) GFP-FCHO2 in live SK-MEL cells. Inset highlights substructure within individual FCHO2 puncta. Scale bars: a) 5 μm, inset 2 μm; b) 5 μm, insets 1 μm. Experiments were performed at least 3 times with similar results; representative data are shown.
Supplementary Figure 10 Instant TIRF-SIM resolves Ras microdomains better than diffraction-limited TIRF.
a) Image taken of GFP-HRas in U2OS cell in diffraction-limited TIRF, i.e. with pinholes and emission microlenses removed. Insets correspond to higher magnification views of blue and red rectangular regions. b) Image taken in the same cell as in (a) but with instant TIRF-SIM. c) Selected image frames taken of GFP-HRas (green) and pDsRed2-ER (magenta) in U2OS cells from instant TIRF-SIM acquisition. d) The same images as (c) but blurred to simulate diffraction-limited images. Arrows highlight two microclusters resolved in instant TIRF-SIM but not in TIRF. See also Fig. 2i, Supplementary Video 6. Scale bars: a) and b): 5 μm; insets, 1 μm; c) and d): 500 nm. Experiments were performed at least 4 times with similar results; representative data are shown.
a) Synthetic object. b) The object blurred with the raw instant TIRF-SIM PSF, derived from beads (e.g. Supplementary Fig. 8d). This image represents what is captured on the camera. c) The object blurred with the diffraction-limited TIRF PSF (e.g. Supplementary Fig. 8b). This is what the camera would measure with emission micro-optics removed. d) The deconvolved instant TIRF-SIM image. e) The image that results after blurring d) with a PSF with kernel σ =(σ2DIFFRACTION-LIMITED - σ2INSTANT TIRF-SIM)0.5 where σINSTANT TIRF-SIM is derived from the deconvolved instant TIRF-SIM PSF (e.g. Supplementary Fig. 8f). Compare c) to e) – they are near identical as line plots in f) and g) (corresponding to blue and red lines in c, e) confirm.
Histograms of total displacement for all tracks (a) and mean speed of the tracks with displacement greater than 1 μm (b), in addition to visual display of particle tracks with displacement less than (c) and greater than (d) 1 μm overlaid on first frame of image series. Scale bars: 5 μm. See also Fig. 3. Experiments were performed at least 4 times with similar results; representative data are shown.
Supplementary Figure 13 Effect of temporal undersampling and spatial blurring on automated tracking of EGFP–Rab11 particles.
a) First frame of image series corresponding to Fig. 3. b) Effect of sampling at 100 Hz (i.e. native recording rate, top), digitally downsampling to 20 Hz (middle) and digitally downsampling to 10 Hz (bottom). Temporal evolution of each track is indicated with color bar at right, red and orange arrows highlight examples of automatically detected tracks (top) that are artificially shortened (middle) and eventually lost (bottom) due to undersampling. Particle tracking was performed on a different cell imaged in instant TIRF-SIM (c, first frame in 300 frame series) and subsequently pinholes and emission microlenses were removed so that the same sample was imaged (for another 300 frames) in diffraction-limited TIRFM (d). 1477 trajectories were detected in instant TIRF-SIM whereas only 1031 were detected in TIRFM. For clarity, only those tracks that were automatically detected and that cover a displacement > 0.3 μm, are shown in the figure (121 in c, 76 in d). We observed a similar increase in detected particle trajectories in an additional cell (1317 total trajectories in instant TIRF-SIM vs. 1107 in TIRF). Scale bars: 5 μm in a, c and d, 1 μm in b. See also Supplementary Videos 12–13. Experiments were repeated at least 3 times with similar results; representative data are shown.
a) Still image from 100-frame instant TIRF-SIM series, acquired at 37 °C. Plasma membrane and vesicles are stained with CellTracker CM-DiI in a live U2OS cell. The 'hot' colormap in ImageJ was used for image display due to the large dynamic intensity range in the sample. b) Higher magnification view of white rectangular region in a) comparing instant TIRF-SIM (top), 4 successive frames binned in time (middle) to simulate motion blur, and the same region blurred to simulate diffraction-limited acquisition (bottom). Blue arrows highlight vesicle pair resolved in instant TIRF-SIM but not in diffraction-limited image. Green arrow highlights vesicle pair resolved in instant TIRF-SIM but blurred due to motion or diffractive blur. White arrow highlights artifact - an apparent vesicle - due to motion blur. Scale bars: 5 μm in a), 1 μm in b). See also Supplementary Video 15, 16. Experiments were repeated 3 times with similar results; representative data are shown.
EGFP-Rab11 particles in U2OS cells were imaged at 37 °C on a commercial TIRF SIM system (9 diffraction-limited images acquired and processed to reconstruct a single super-resolution image) at a frame rate of 12.9 Hz. a) Single raw frame from diffraction-limited series (time point 62 of 200). b) Corresponding TIRF SIM reconstruction. c, d) Higher magnification images of the yellow dashed rectangular regions in a, b. See also Supplementary Video 17. Motion blur broadens particle extent significantly (compare particles highlighted with red arrows, apparent particle width given in red text), presumably due to particle motion during the 9-frame raw image acquisition period (Supplementary Video 18). Scale bars in a, b 5 μm; in c, d 500 nm. Experiments were repeated at least 3 times with similar results; representative data are shown.
Supplementary Figures 1–15, Supplementary Tables 1 and 2 and Supplementary Note 1
Matlab scripts for processing instant TIRF-SIM data, including 2D deconvolution, flat fielding and image delining. A user introduction and test images are also provided
Time-lapse imaging of Jurkat T cells expressing EMTB–3×EGFP at room temperature. Images were acquired every 50 ms, over 500 time points. See also Fig. 2a-b. Experiments were repeated 4 times; representative data are shown
Time-lapse imaging of U2OS cell expressing EGFP–HRas at 37 °C. Images were acquired every 0.75 s, over 60 time points. Images were binned 2×2 relative to the data for display purposes. See also Fig. 2c. Experiments were repeated 5 times; representative data are shown
Higher-magnification view of subregion in Supplementary Video 2. highlighting the ‘wave-like’ dynamics among Ras microdomains. See also Fig. 2d
Dual-color time-lapse imaging of U2OS cell expressing EGFP–VSVG (green) and HaloTag–Ras (magenta) at 37 °C. Images were acquired every 2.3 s, over 100 time points. Images were binned 2×2 relative to the data for display purposes. See also Fig. 2e-g. Experiments were repeated 4 times; representative data are shown
Dual-color (merge; bottom) time-lapse imaging of U2OS cell expressing EGFP–HRas (green; middle) and pDsRed2–ER (magenta; top) at 37 °C. Images were acquired every 1.2 s, over 100 time points. Images were binned 2×2 relative to the data for display purposes. See also Fig. 2h. Experiments were repeated4 times; representative data are shown
Higher-magnification view of subregion in Supplementary Video 5. highlighting apparent fission and fusion of Ras microclusters and comparing instant TIRF-SIM (left) with simulated diffraction-limited images (right). See also Fig. 2i
Time-lapse imaging of calcium flux within INS-1 cells, as reported by GCamp6S-CAAX at room temperature. Images were acquired every 88 ms, over 500 time points. Note localized calcium activity. Experiments were repeated 3 times; representative data are shown
Time-lapse imaging of Jurkat T cells expressing F-tractin–EGFP at room temperature. Images were acquired every 2 s, over 200 time points. Experiments were repeated 3 times; representative data are shown
Time-lapse imaging of U2OS cell expressing F-tractin–EGFP at 37 °C. Images were acquired every 2.6 s, over 50 time points. Experiments were repeated 3 times; representative data are shown
Dual-color time-lapse imaging of U2OS cell expressing GFP–myosin IIA (green) and mApple–F-tractin (magenta) at 37 °C. Images were acquired every 6 s, over 50 time points. Experiments were repeated 3 times; representative data are shown
Time-lapse imaging of U2OS cell expressing GFP–Rab11 at 37 °C. Images were acquired at 100Hz frame rate, over 600 time points. See also Fig. 3. Experiments were repeated 4 times; representative data are shown
Higher-magnification view of subregion in Supplementary Video 11, highlighting diffusive (blue) versus directed (red) motion of particles. The first 305 frames (from 0 s to 3.04 s) are shown. See also Fig. 3 b-d
Higher-magnification view of subregion in Supplementary Video 11, highlighting tracks derived from many particles. The 300th to 500th frames from the total image series (from 3 s to 5 s) are shown. See also Supplementary Fig. 13a, b
Higher-magnification view of EGFP–Rab11 events in Fig. 3g. The 178th to 278th frames from the image series (from 1.78 s to 2.78 s) are shown, comparing instant TIRF-SIM (left) with simulated diffraction-limited images (right)
Time-lapse imaging of U2OS cell stained with CellTracker CM-DiI at 37 °C. Images were acquired at 36 Hz, over 100 time points. See also Supplementary Fig. 14. Experiments were repeated 3 times; representative data are shown
Higher-magnification view of the region outlined by the white rectangle in Supplementary Video 15. comparing instant TIRF-SIM (left), temporally downsampled instant TIRF-SIM (middle), and simulated diffraction-limited TIRF (right). See also Supplementary Fig. 14
Comparative diffraction-limited TIRF (left) and TIRF-SIM (right) recordings of GFP–Rab11 particles acquired on a commercial TIRF-SIM system. See also Supplementary Fig. 15, corresponding to frame 62 (4746 ms) in this video. Experiments were repeated 3 times; representative data are shown
Motion of GFP–Rab11 particles during raw image capture in commercial TIRF-SIM. Nine raw frames (frame # indicated at top right) used in reconstructing the single super-resolution frame shown in Supplementary Fig. 15. Note movement of particles (e.g. in yellow bounding box) during image acquisition, which leads to motion blur in Supplementary Fig. 15, Supplementary Video 17. Experiments were repeated 3 times; representative data are shown
About this article
Cite this article
Guo, M., Chandris, P., Giannini, J.P. et al. Single-shot super-resolution total internal reflection fluorescence microscopy. Nat Methods 15, 425–428 (2018). https://doi.org/10.1038/s41592-018-0004-4
Novel endoscopic optical diagnostic technologies in medical trial research: recent advancements and future prospects
BioMedical Engineering OnLine (2021)
Imaging the rapid yet transient accumulation of regulatory lipids, lipid kinases, and protein kinases during membrane fusion, at sites of exocytosis of MMP-9 in MCF-7 cells
Lipids in Health and Disease (2020)
Nature Biotechnology (2020)
Video-rate multi-color structured illumination microscopy with simultaneous real-time reconstruction
Nature Communications (2019)
Anchoring cortical granules in the cortex ensures trafficking to the plasma membrane for post-fertilization exocytosis
Nature Communications (2019)