Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Brief Communication
  • Published:

3D high- and super-resolution imaging using single-objective SPIM

This article has been updated

Abstract

Single-objective selective-plane illumination microscopy (soSPIM) is achieved with micromirrored cavities combined with a laser beam–steering unit installed on a standard inverted microscope. The illumination and detection are done through the same objective. soSPIM can be used with standard sample preparations and features high background rejection and efficient photon collection, allowing for 3D single-molecule-based super-resolution imaging of whole cells or cell aggregates. Using larger mirrors enabled us to broaden the capabilities of our system to image Drosophila embryos.

This is a preview of subscription content, access via your institution

Access options

Rent or buy this article

Prices vary by article type

from$1.95

to$39.95

Prices may be subject to local taxes which are calculated during checkout

Figure 1: Principle and 3D high-resolution capabilities of the soSPIM method.
Figure 2: 3D super-resolution capabilities of the soSPIM method.

Similar content being viewed by others

Change history

  • 11 June 2015

    In the version of this article initially published online, panels a, b and d in Figure 2 were mislabeled. The correct labels should be soSPIM-PALM, soSPIM-STORM and soSPIM–3D dSTORM, respectively. The errors have been corrected for the print, PDF and HTML versions of this article.

References

  1. Tokunaga, M., Imamoto, N. & Sakata-Sogawa, K. Nat. Methods 5, 159–161 (2008).

    Article  CAS  PubMed  Google Scholar 

  2. Huang, B. et al. Science 319, 810–813 (2008).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  3. Pavani, S.R. et al. Proc. Natl. Acad. Sci. USA 106, 2995–2999 (2009).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  4. Juette, M.F. et al. Nat. Methods 5, 527–529 (2008).

    Article  CAS  PubMed  Google Scholar 

  5. Shtengel, G. et al. Proc. Natl. Acad. Sci. USA 106, 3125–3130 (2009).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  6. Cisse, I.I. et al. Science 341, 664–667 (2013).

    Article  CAS  PubMed  Google Scholar 

  7. Cella Zanacchi, F. et al. Nat. Methods 8, 1047–1049 (2011).

    Article  PubMed  Google Scholar 

  8. Planchon, T.A. et al. Nat. Methods 8, 417–423 (2011).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  9. Wu, Y. et al. Proc. Natl. Acad. Sci. USA 108, 17708–17713 (2011).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  10. Gebhardt, J.C. et al. Nat. Methods 10, 421–426 (2013).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  11. Chen, B.C. et al. Science 346, 1257998– (2014).

    Article  PubMed  PubMed Central  Google Scholar 

  12. Hu, Y. et al. Opt. Nanoscopy 2, 7 (2013).

  13. Engl, W. et al. Nat. Cell Biol. 16, 587–594 (2014).

    Article  CAS  PubMed  Google Scholar 

  14. Betzig, E. et al. Science 313, 1642–1645 (2006).

    Article  CAS  PubMed  Google Scholar 

  15. Hess, S.T., Girirajan, T.P. & Mason, M.D. Biophys. J. 91, 4258–4272 (2006).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  16. Thompson, R.E., Larson, D.R. & Webb, W.W. Biophys. J. 82, 2775–2783 (2002).

    CAS  PubMed  PubMed Central  Google Scholar 

  17. Rust, M.J., Bates, M. & Zhuang, X. Nat. Methods 3, 793–795 (2006).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  18. Heilemann, M. et al. Angew. Chem. Int. Edn. Engl. 47, 6172–6176 (2008).

    Article  CAS  Google Scholar 

  19. Fölling, J. et al. Nat. Methods 5, 943–945 (2008).

    Article  PubMed  Google Scholar 

  20. Manley, S. et al. Nat. Methods 5, 155–157 (2008).

    Article  CAS  PubMed  Google Scholar 

  21. Giannone, G. et al. Biophys. J. 99, 1303–1310 (2010).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  22. Visser, T.D., Oud, J.L. & Brakenhoff, G.J. Optik (Stuttg.) 90, 17–19 (1992).

    Google Scholar 

  23. Strandman, C., Rosengren, L., Elderstig, H.G.A. & Backlund, Y. J. Microelectromech. Syst. 4, 213–219 (1995).

    Article  CAS  Google Scholar 

  24. Barycka, I. & Zubel, I. Sens. Actuators A Phys. 48, 229–238 (1995).

    Article  CAS  Google Scholar 

  25. Resnik, D., Vrtacnik, D., Aljancic, U., Mozek, M. & Amon, S. J. Micromech. Microeng. 15, 1174–1183 (2005).

    Article  CAS  Google Scholar 

  26. Kechkar, A. et al. PLoS One 8, e62918 (2013).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  27. Izeddin, I. et al. Opt. Express 20, 2081–2095 (2012).

    Article  CAS  PubMed  Google Scholar 

  28. Holtzer, L., Meckel, T. & Schmidt, T. Appl. Phys. Lett. 90, 053902 (2007).

    Article  Google Scholar 

  29. Martin, A.C. et al. J. Cell Biol. 188, 735–749 (2010).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

This work was supported by joint funding from the Ministère de l'Enseignement Supérieur et de la Recherche (ANR-10-INBS-04 FranceBioImaging), the Regional Council of Aquitaine and the Singapore National Research Foundation (NRF). V.V., G.G. and R.G. acknowledge financial support from the Mechanobiology Institute through an NRF grant. This research was conducted within the scope of the International Associated Lab LIA CAFS. The authors acknowledge A. Liew for help in the initial tests; P. Legros and C. Butler for help and feedback; S. Wolf and A. Wong of the Mechanobiology Institute for help editing the manuscript; C. Armouda and T. Saunders from the Mechanobiology Institute for help with Drosophila embryo imaging and for feedback; C. Viasnoff and G. Viasnoff for help with drawing conception; and K. Monier from ENS Lyon, W.J. Nelson from Stanford University and L. Hufnagel from EMBL for their kind gifts of stable fluorescent cell lines and a stable fluorescent embryo line. The authors thank Nikon France and Errol for loaning equipment.

Author information

Authors and Affiliations

Authors

Contributions

R.G. performed the experiments and designed the optical setup. G.G. and R.G. designed and manufactured the micromirror chips. R.G. and A.A. imaged embryos. V.S. and V.V. came up with the original idea. J.-B.S., V.S. and V.V. conceived the experiments, supervised the work and contributed equally to the development of the project. R.G., G.G., V.V., V.S. and J.-B.S. contributed to the manuscript.

Corresponding author

Correspondence to Virgile Viasnoff.

Ethics declarations

Competing interests

The authors are named on UK patent application GB1315248.3 and French patent FR1358226 related to this work.

Integrated supplementary information

Supplementary Figure 1 soSPIM principle and representation of the soSPIM beam-steering unit.

Schematic representation of the soSPIM principle showing a sample holder comprising 45° micromirrored cavities deposited on a glass coverslip and an excitation beam–steering system mounted on a conventional inverted microscope. Beam steering of multiple laser lines (see Online Methods) is achieved by reflection on an x-axis galvanometric (XG) mirror conjugated by relay lenses (RL) to a y-axis galvanometric mirror (YG). A focus tunable lens (TL) allows the localization of the light sheet on the sample after its reflection onto the mirror. The system is mounted on the epifluorescence port of a standard inverted microscope. A cylindrical lens can be inserted into the optical path to create a static light sheet. XG, YG and TL are all conjugated to the back focal plane (BFP) of the objective so that the excitation beam is always emitted parallel to the objective optical axis. Finally, a variable telescope enables the adjustment of the beam diameter at the BFP to optimize the light-sheet thickness. Upper right: scanning electron microscopy (SEM) image of the section of a silicon wafer displaying a 45° surface created by anisotropic etching and a microwell created by dry etching (see Online Methods) and SEM images of a silicon wafer displaying 45° surfaces and an array of 22 × 22 µm2 microwells.

Supplementary Figure 2 Light-sheet characterization.

(a) Top, light sheet imaged in the xy plane of a fluorescent solution obtained by scanning a Gaussian beam along the mirror axis. Bottom, single Gaussian beam at different defocusing strengths illustrating the ability to localize the thinnest portion of the light sheet at a variable distance from the reflecting mirror. The red rectangles represent the position of the mirror where the excitation beam is reflected. (b) Top, spatial characterization of the Gaussian excitation laser beam reflected on the 45° mirror (red rectangle). Bottom, Gaussian beam cross-section (red dots) and Gaussian fits (black lines) at the three positions (i, ii and iii) represented in the top panel along the light propagation. The cross-sections show perfect Gaussian profiles, as expected. (c) Beam profile along the light propagation (red dots) measured from the width of the Gaussian fitting represented in b all along the propagation axis of the Gaussian beam. The black line represents the fit with the equation of propagation of a Gaussian beam (Online Methods). The dotted lines represent the position of the beam cross-section represented in the lower panels of b. (d) Light-sheet field of view (two times the Rayleigh length) and thickness (FWHM) for different objectives (100×/1.4-NA, 60×/1.2-NA, 40×/0.75-NA, 20×/0.5-NA, 10×/0.3-NA) and beam diameters at the back focal plane of the objective. The black line represents the theoretical relation between the field of view and the thickness of a light sheet created by the focalization of a Gaussian beam at 561 nm (Online Methods).

Source data

Supplementary Figure 3 Deviation compensation of the light-sheet position.

(a) The top panel shows the excitation beam visualized through a fluorescent solution at two different depths (∆z apart from each other) within the sample as represented in the lower panel without defocusing compensation (∆f = 0) for the axial movement of the objective. (b) Light-sheet profiles at different depths within the sample without compensation for the axial movement of the objective. (c) The top panel shows the excitation beam visualized through a fluorescent solution at two different depths (∆z apart from each other) within the sample as represented in the lower panel with defocusing compensation (∆f α ∆z) for the axial movement of the objective. (d) Light-sheet profile at different depths within the sample after compensating for the axial movement of the objective. Scale bars correspond to the length of the light sheet (22.4 µm) and are positioned at the level where the light sheet is thinnest.

Source data

Supplementary Figure 4 Illustration of the devices displaying 45° mirrors used for soSPIM imaging.

(a) Top, SEM image of the section of a silicon wafer displaying a 45° surface created by anisotropic etching and a microwell created by dry etching (Supplementary Fig. 5a and Online Methods). The microwell depth was designed to be 5 µm lower than the 45° surface depth to allow 3D imaging of cell from the basal to the apical domain. Bottom, final micro-fabricated device displaying 45° mirroring surfaces and microwells mounted in a bottom-free petri dish for easy cell culture and imaging. (b) Top, a millimeter-sized 45° mirror made in UV-curable polymer molded onto a microprism and metallized with a gold layer (Supplementary Fig. 5b and Online Methods). Bottom, mounted system for embryo imaging representing an index-matched capillary inserted below a millimeter-sized 45° mirror mounted on a coverslip. (c) SEM images of a silicon wafer displaying 45° surfaces and 60 × 300 µm2 (left) or 22 × 22 µm2 (right) microwells.

Supplementary Figure 5 Micro-fabrication process for the device comprising 45° mirroring surfaces and potentially microwells.

(a) Micro-fabrication process for an index-matched UV-curable polymer chip displaying microwells flanked by 45° metallized mirroring surfaces. Anisotropic wet etching and dry etching processes were sequentially used to create 45° surfaces and microwells, respectively, within a silicon wafer (i,ii). A PDMS replica of this wafer was created (iii) and used to reproduce the wafer shape on a coverslip in a UV-curable polymer via a capillary filling process (iv,v). After the PDMS replica was peeled off (vi), the polymer surface was metallized with a gold layer by thermal evaporation in vacuum (vii). The 45° mirrors were then protected by an additional layer of polymer using a capillary filling process (viii), and the unprotected metal was removed by wet etching (ix). (b) Fabrication process for millimeter-sized 45° mirrors in polymer. A right-angle glass prism flanked by two PDMS sheets was replicated in UV-curable polymer (i–iii) The polymer replica of the right-angle prism imprint was peeled off (iv) and metallized by thermal evaporation under vacuum (v). The right-angle prism imprint was finally sealed onto a coverslip ready for imaging. Lower panels: schematic representation of the multiview rotation stage and picture of the multiview rotation stage displaying the rotation motor, the capillary containing the embryos and the 45° millimeter-sized mirror sealed onto a coverslip. (c) Atomic force microscopy (AFM) characterization of the surface roughness of a UV-curable and metallized 45° mirror created as described in a and the Online Methods. Left, AFM 3D profile of a 45° mirror made in UV-curable polymer and coated with gold. Right, line profile along the dashed line in the left-hand panel. The roughness root mean square of the 45° surface was determined to be 13.27 nm.

Source data

Supplementary Figure 6 3D volume imaging within microwells.

(a) Top, schematic representation of the microwell architecture enabling image capture of the entire height of cells and cell aggregates within a well. By tuning the duration of deep reactive-ion etching of the silicon wafer, we were able to design the well depth so as to be 5 µm lower than the 45° mirror. This enabled imaging of the bottom of the wells in the soSPIM configuration without having the excitation beam clipped by the base of the 45° mirror. Bottom, soSPIM optical sections of the basal region (displaying stress fibers) to the apical poles of spread hepatocyte cell aggregates labeled with phalloidin-Alexa-647 Fluor 488 within a 40 × 40 µm2 well. The sectioning capability of the soSPIM microscope allowed clear identification of the actin stress fibers and structures of the basal region of the spread hepatocyte cell. (b) Microwells aligned along the mirror allow simultaneous imaging of multiple cells. Two-color soSPIM optical sections of three S180 cells expressing the membrane protein E-cadherin–GFP (upper panel) and the cortical actin protein F-Tractin–RFP (middle panel), each positioned in a different 24 × 24 μm² microwell. Each color was acquired sequentially. This capability leads to an increase in the imaging throughput. The scanning light sheet illuminates only the zones defined by the red bars, and the readout speed of the camera is equivalent to that of single-well imaging. The lower panel shows a phase contrast image of the three wells. Scale bars, 5 µm.

Supplementary Figure 7 soSPIM imaging of a large cell aggregate.

(a) Two-color soSPIM optical sections at different depths of an S180 cell aggregate cultured for 48 h in a 40 × 40 µm2 microwell coated with fibronectin (green, E-cadherin–GFP; red, Lamin-B1–Alexa Fluor 568). (b) Image of a whole cell aggregate of MDCK cells expressing the cytoskeleton protein actin-GFP acquired with a 1.9-µm-thick and 31.4-µm-long light sheet. (c) Image of the cell aggregate in b reconstructed from three images acquired at x0(I1), x0(I2) and x0(I3) positions to create an image with optimized sectioning, higher contrast and reduced shadowing effects without moving the sample. This image was obtained by combining the three images acquired with a 1.6-µm-thick and 25.7-µm-long light sheet as represented in d. (d) Schematic representation of a large cell aggregate within a 60 × 300 µm2 microwell and the positions of a thin light sheet shorter than the whole cell aggregate. The shorter thin light sheet allowed us to image the whole aggregate while optimizing the sectioning performance. The light sheet was defocused at three different positions (x0(I1), x0(I2) and x0(I3)), with three defocus strengths (Ii) to image the whole cell aggregate. (e) Stitching of the three images acquired at three different x0 positions of the light sheet were used to create an image of the whole cell aggregate as represented in c. Upper panels: the masks defined by the equation described in the Online Methods were used to select the image parts acquired with the thinnest part of the light sheet. Lower panels: corresponding images of the cell aggregate after multiplication with the masks represented in the upper panels.

Supplementary Figure 8 Characterization of 3D super-resolution performance.

(a) Example of one of the 8,000 frames acquired during a PALM sequence acquisition of U2-OS cells expressing fibrillarin-Dendra2, showing two molecules located 7 µm above the coverslip. The low background is due to the high sectioning capability of the soSPIM approach. Inset: zoom on a molecule detected in the white box. (b) Upper left: the time trace recorded in the white box in a over 300 frames, illustrating the typical single-molecule signature. Upper right: the single-molecule intensity profile (red line) of the molecule represented in the inset in a along the white box, with its Gaussian fit (black line) giving a FWHM of 2.51 pixels (402 nm). Lower panel: a histogram of the collected number of photons per localization during the PALM acquisition, giving a median localization precision of 38.2 nm. (c) Leftmost panel, image of Alexa Fluor 647 dye embedded in the polymer of a device comprising 45° mirroring surfaces and 24 × 24 µm2 microwells. This dye was used as a beacon to estimate the 3D localization-precision performance of the soSPIM system. The image was acquired with a 1.6-µm-thick and 25.7-µm-long light sheet and an astigmatism lens in front of the camera for 3D localization. Second panel from left: image of three Alexa Fluor 647 single dye spots localized at three different z-positions illustrating the PSF shaping due to the astigmatism lens used for 3D localization. Second panel from right: sum of the xy localizations of the Alexa Fluor 647 dye shown in the second panel from the left for a 500-frame acquisition. Rightmost panel: sum of the yz localization of the Alexa Fluor 647 dye represented in the left panel for a 500-frame acquisition. (d) Left, histogram of the xy and z astigmatism-based localization performance of the soSPIM super-resolution method on 35 detected Alexa Fluor 647 dyes embedded in the chip polymer and localized at least 128 times each. We measured xy localization precision of 23.9 nm in xy and 73.6 nm in z. Right, histogram of the xy and z astigmatism-based localization performance of the soSPIM super-resolution method on 23 detected FLIP 565 dyes embedded in the chip polymer and localized at least 72 times each. We measured xy localization precision of 21.5 nm in xy and 46.3 nm in z.

Source data

Supplementary Figure 9 3D super-resolution reconstruction and soSPIM versus wide-field single-molecule detection.

(a) Super-resolution lateral (xy) and orthogonal (xz and yz) reconstructions of the nucleus membrane protein Lamin-B1 labeled with the photoactivatable dye FLIP 565 in suspended S180 cells within a 24 × 24 µm2 microwell. Orthogonal reconstructions were performed from 14 planes (Fig. 3) acquired every 540 nm at between 6 and 13 µm above the coverslip. Scale bar, 5 µm. (b) Example of one of the 4,000 frames acquired during a STORM sequence acquisition on the nucleus membrane protein Lamin-B1 labeled with Alexa Fluor 647 dyes in a suspended S180 cell within a 24 × 24 µm2 microwell. The image on the left was acquired with a 1.5-µm-thick light sheet using soSPIM. The image on the right was acquired using a wide-field illumination performed with a collimated laser beam centered on the microwell. The graph shows the line intensity profiles of two single-molecule events acquired with soSPIM illumination (red line) and wide-field illumination (blue line). Scale bar, 1 µm.

Source data

Supplementary Figure 10 Time-lapse imaging of the early stage of development of a Drosophila embryo.

(a) Time sequence of soSPIM optical sections 21.2 µm deep within a Drosophila embryo expressing the nuclear protein histone-mCherry imaged with a 20×/0.5-NA objective and a 4.3-μm-thick light sheet. A 35-µm Z-stack with a 1.35-µm z-step was acquired every 150 s for 220 min. We can observe the nucleus migrating toward the embryo’s outer surface (t = 0 to 52.5 min), further division stages of the nucleus at the outer surface of the embryo (t = 52.5 to 97.5 min) and the beginning of the gastrulation stage of the embryos (t = 190 min) demonstrating the viability of the embryos, despite the few apoptosis cellular events (t = 97.5 min). (b) Upper panel: maximum intensity projection along the z-axis (Z-MIP) at t = 40 min, illustrating the early migration of the nucleus toward the outer surface of the embryo. Lower panel: maximum intensity projection along the y-axis (Y-MIP) at t = 40 min, with segmentation of the nuclei within the embryo (depth is color-coded). Nuclei were detected up to at least 35 µm inside the embryo, illustrating the high penetration depth of the soSPIM method. (c) Three soSPIM optical sections of the embryo at t = 87.5 min, illustrating the development of the embryo as well as the penetration depth. (d) soSPIM optical section at t = 207.5 min and 30-µm depth, illustrating the embryo viability after 207 min of imaging, as the embryo is about to enter gastrulation. Scale bars, 50 µm.

Supplementary Figure 11 Multiview imaging of Drosophila embryo.

(a) soSPIM optical sections of a Drosophila embryo expressing the nuclear protein histone-mCherry imaged with a 20×/0.5-NA objective and a 4.3-μm-thick light sheet for three different rotation angles of the embryo along the 45° mirror. (b) Schematic representation of the soSPIM multiview system. The embryo is inserted inside a UV-curable index-matched capillary polymer that rotates along a millimeter-sized mirror allowing multiview imaging of the embryo (Supplementary Fig. 5b and Online Methods). Scale bar, 50 µm.

Supplementary Figure 12 soSPIM calibration.

(a) Plot showing the measurements of the axial z-position of the light sheet depending on the light-sheet displacement along the mirror x-axis for the 60× water-immersion objective (blue dots) and the 40× (red dot), 20× (red dots) and 10× (yellow dots) objectives as described in the Online Methods. The black lines represent the linear fits used as calibration for soSPIM 3D-volume imaging. (b) Plot showing the position of the thinnest part of the light sheet (x0) along the x-axis depending of the defocus strength applied to the focus-tunable lens (red dots). The black line represents the linear fit used to calibrate the defocus system of the soSPIM microscope. (c) Plot showing the ratio between the x-axis drift of the light-sheet position (dx0) per z-axis objective displacement steps (dz) depending on the defocus strength applied to the focus-tunable lens (dI) per z-axis objective displacement steps (dz) (red dots) for the 60× water-immersion objective. The black line represents the linear fit. To fully compensate for the x-drift of the light sheet for 3D imaging the ratio dx0/dz must be set to 0, i.e., dI/dz = 0.013 mA/µm.

Source data

Supplementary information

Supplementary Text and Figures

Supplementary Figures 1–12 and Supplementary Table 1 (PDF 1792 kb)

Time-lapse imaging of the early stages of development of a Drosophila embryo.

Time-lapse movie of soSPIM optical sections 21.2 μm deep within a Drosophila embryo expressing the nuclear protein histone-mCherry imaged with a 20x/0.5-NA objective and a 4.3-μm-thick light sheet. A 35-μm z-stack with a 1.35-μm z-step was acquired every 150 s for 220 min. We observed the nucleus migrating toward the embryo's outer surface, further division stages of the nucleus at the outer surface of the embryo, and the beginning of the gastrulation stage of the embryos, demonstrating the viability of the embryos, despite the few cellular apoptosis events observed. Scale bar, 50 μm. (AVI 6234 kb)

Multiview rotation stage.

Imaging chamber for soSPIM multiview imaging of embryos. The multiview imaging chamber is composed of a coverslip and a glass capillary extended by an index-matched UV-curable polymer–based capillary containing embryos embedded in low-melting agar gel (Online Methods and Supplementary Fig. 5b). The movie illustrates the well-controlled rotation of the capillary (left, 90° steps; right, 360° steps) driven by a computer-controlled step-rotation motor. For imaging, a 45° millimeter-sized mirror was sealed onto the coverslip over the index-matched capillary containing the embryos. (AVI 18874 kb)

Source data

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Galland, R., Grenci, G., Aravind, A. et al. 3D high- and super-resolution imaging using single-objective SPIM. Nat Methods 12, 641–644 (2015). https://doi.org/10.1038/nmeth.3402

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/nmeth.3402

This article is cited by

Search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing