Article | Published:

Video-rate nanoscopy using sCMOS camera–specific single-molecule localization algorithms

Nature Methods volume 10, pages 653658 (2013) | Download Citation

Abstract

Newly developed scientific complementary metal-oxide semiconductor (sCMOS) cameras have the potential to dramatically accelerate data acquisition, enlarge the field of view and increase the effective quantum efficiency in single-molecule switching nanoscopy. However, sCMOS-intrinsic pixel-dependent readout noise substantially lowers the localization precision and introduces localization artifacts. We present algorithms that overcome these limitations and that provide unbiased, precise localization of single molecules at the theoretical limit. Using these in combination with a multi-emitter fitting algorithm, we demonstrate single-molecule localization super-resolution imaging at rates of up to 32 reconstructed images per second in fixed and living cells.

Access optionsAccess options

Rent or Buy article

Get time limited or full article access on ReadCube.

from$8.99

All prices are NET prices.

Accessions

References

  1. 1.

    Microscopy and its focal switch. Nat. Methods 6, 24–32 (2009).

  2. 2.

    , & Optical nanoscopy: from acquisition to analysis. Annu. Rev. Biomed. Eng. 14, 231–254 (2012).

  3. 3.

    , & Live-cell super-resolution imaging with synthetic fluorophores. Annu. Rev. Phys. Chem. 63, 519–540 (2012).

  4. 4.

    , , & Optimized localization analysis for single-molecule tracking and super-resolution microscopy. Nat. Methods 7, 377–381 (2010).

  5. 5.

    , , & Ultrahigh accuracy imaging modality for super-localization microscopy. Nat. Methods 10, 335–338 (2013).

  6. 6.

    et al. Dynamic clustered distribution of hemagglutinin resolved at 40 nm in living cell membranes discriminates between raft theories. Proc. Natl. Acad. Sci. USA 104, 17370–17375 (2007).

  7. 7.

    , , & Fast, three-dimensional super-resolution imaging of live cells. Nat. Methods 8, 499–505 (2011).

  8. 8.

    et al. Super-resolution fluorescence imaging of organelles in live cells with photoswitchable membrane probes. Proc. Natl. Acad. Sci. USA 109, 13978–13983 (2012).

  9. 9.

    , & Evaluation of sCMOS cameras for detection and localization of single Cy5 molecules. Opt. Express 20, 7338–7349 (2012).

  10. 10.

    , & Localization-based super-resolution microscopy with an sCMOS camera. Part II: experimental methodology for comparing sCMOS with EMCCD cameras. Opt. Express 20, 17741–17759 (2012).

  11. 11.

    , & Localization accuracy in single-molecule microscopy. Biophys. J. 86, 1185–1200 (2004).

  12. 12.

    , , & Fast, single-molecule localization that achieves theoretically minimum uncertainty. Nat. Methods 7, 373–375 (2010).

  13. 13.

    , , & Simultaneous multiple-emitter fitting for single molecule super-resolution imaging. Biomed. Opt. Express 2, 1377–1393 (2011).

  14. 14.

    , , & Faster STORM using compressed sensing. Nat. Methods 9, 721–723 (2012).

  15. 15.

    et al. Rational design of true monomeric and bright photoactivatable fluorescent proteins. Nat. Methods 9, 727–729 (2012).

  16. 16.

    , , & Structural organization of the actin cytoskeleton at sites of clathrin-mediated endocytosis. Curr. Biol. 21, 1167–1175 (2011).

  17. 17.

    et al. Automatic quantification of microtubule dynamics enables RNAi-screening of new mitotic spindle regulators. Cytoskeleton (Hoboken) 68, 266–278 (2011).

  18. 18.

    , , & Three-dimensional super-resolution imaging of the midplane protein FtsZ in live Caulobacter crescentus cells using astigmatism. ChemPhysChem 13, 1007–1012 (2012).

  19. 19.

    & Mechanisms of endocytosis. Annu. Rev. Biochem. 78, 857–902 (2009).

  20. 20.

    & Three-dimensional tracking of single fluorescent particles with submillisecond temporal resolution. Nano Lett. 10, 4657–4663 (2010).

  21. 21.

    et al. A near-infrared fluorophore for live-cell super-resolution microscopy of cellular proteins. Nat. Chem. 5, 132–139 (2013).

  22. 22.

    & Live-cell dSTORM of cellular DNA based on direct DNA labeling. ChemBioChem 13, 298–301 (2012).

  23. 23.

    , & Ultrabright photoactivatable fluorophores created by reductive caging. Nat. Methods 9, 1181–1184 (2012).

  24. 24.

    , , , & Enzymatic activation of nitro-aryl fluorogens in live bacterial cells for enzymatic turnover-activated localization microscopy. Chem. Sci. 4, 220–225 (2013).

  25. 25.

    et al. High-density mapping of single-molecule trajectories with photoactivated localization microscopy. Nat. Methods 5, 155–157 (2008).

  26. 26.

    et al. Three-dimensional sub–100 nm resolution fluorescence microscopy of thick samples. Nat. Methods 5, 527–529 (2008).

  27. 27.

    et al. Two-color nanoscopy of three-dimensional volumes by 4Pi detection of stochastically switched fluorophores. Nat. Methods 8, 353–359 (2011).

  28. 28.

    et al. Multicolor fluorescence nanoscopy in fixed and living cells by exciting conventional fluorophores with a single wavelength. Biophys. J. 99, 2686–2694 (2010).

  29. 29.

    et al. Superresolution imaging of multiple fluorescent proteins with highly overlapping emission spectra in living cells. Biophys. J. 101, 1522–1528 (2011).

  30. 30.

    The large-sample distribution of the likelihood ratio for testing composite hypotheses. Ann. Math. Stat. 9, 60–62 (1938).

  31. 31.

    & Optimal acquisition scheme for subwavelength localization microscopy of bleachable fluorophores. A. Opt. Lett. 36, 289–291 (2011).

  32. 32.

    , , & The role of photon statistics in fluorescence anisotropy imaging. IEEE Trans. Image Process. 14, 1237–1245 (2005).

Download references

Acknowledgements

We thank J. Munro, P. Pellett, L. Schroeder, F. Bottanelli and M. Gudheti for helpful discussions about the buffer and sample preparation, J. Spatz for support, and P. de Camilli, O. Idevall-Hagren, T. Gould, E. Allgeyer and E. Kromann for helpful comments on the manuscript. We thank P. Xu (Chinese Academy of Sciences) for providing the mEos3.2 plasmid for initial experiments and G. Patterson (US National Institutes of Health) for the human clathrin light chain plasmid. This work was supported by grants from the Wellcome Trust (095927/A/11/Z), US National Institutes of Health (R01 CA098727 to W.M.) and Raymond and Beverly Sackler Institute for Biological, Physical and Engineering Sciences.

Author information

Author notes

    • Tobias M P Hartwich
    •  & Felix E Rivera-Molina

    These authors contributed equally to this work.

Affiliations

  1. Department of Cell Biology, Yale University School of Medicine, New Haven, Connecticut, USA.

    • Fang Huang
    • , Tobias M P Hartwich
    • , Felix E Rivera-Molina
    • , Whitney C Duim
    • , Jordan R Myers
    • , Derek Toomre
    •  & Joerg Bewersdorf
  2. Department of Biophysical Chemistry, University of Heidelberg, Heidelberg, Germany.

    • Tobias M P Hartwich
  3. Department of New Materials and Biosystems, Max Planck Institute for Intelligent Systems, Stuttgart, Germany.

    • Tobias M P Hartwich
  4. Department of Biomedical Engineering, Yale University, New Haven, Connecticut, USA.

    • Yu Lin
    •  & Joerg Bewersdorf
  5. Integrated Graduate Program in Physical and Engineering Biology, Yale University, New Haven, Connecticut, USA.

    • Yu Lin
    •  & Joerg Bewersdorf
  6. Yale College, Yale University, New Haven, Connecticut, USA.

    • Jane J Long
  7. Department of Microbial Pathogenesis, Yale University School of Medicine, New Haven, Connecticut, USA.

    • Pradeep D Uchil
    •  & Walther Mothes
  8. National High Magnetic Field Laboratory and Department of Biological Science, Florida State University, Tallahassee, Florida, USA.

    • Michelle A Baird
    •  & Michael W Davidson

Authors

  1. Search for Fang Huang in:

  2. Search for Tobias M P Hartwich in:

  3. Search for Felix E Rivera-Molina in:

  4. Search for Yu Lin in:

  5. Search for Whitney C Duim in:

  6. Search for Jane J Long in:

  7. Search for Pradeep D Uchil in:

  8. Search for Jordan R Myers in:

  9. Search for Michelle A Baird in:

  10. Search for Walther Mothes in:

  11. Search for Michael W Davidson in:

  12. Search for Derek Toomre in:

  13. Search for Joerg Bewersdorf in:

Contributions

F.H. and J.B. conceived the project. F.H., T.M.P.H., Y.L. and J.B. built the setup and designed the bead experiments. All authors designed the biological imaging experiments. F.H., T.M.P.H., Y.L., J.J.L., P.D.U. and J.R.M. performed the fixed-cell experiments. F.H., F.E.R.-M., W.C.D. and J.J.L. performed the live-cell experiments. M.A.B. and M.W.D. generated the mEos3.2 and tdEos plasmids. F.H. wrote the software and performed the simulations and analysis. All authors wrote the manuscript.

Competing interests

F.H. and J.B. are co-inventors on a patent application related in part to the material presented here. J.B. is consultant, equity holder and member of the scientific advisory board of Vutara, Inc., which makes super-resolution microscopes.

Corresponding author

Correspondence to Joerg Bewersdorf.

Supplementary information

PDF files

  1. 1.

    Supplementary Text and Figures

    Supplementary Figures 1–13, Supplementary Table 1 and Supplementary Note

Zip files

  1. 1.

    Supplementary Data

    ZIP archive of obtained localization estimates and uncompressed super-resolution images for Figure 2a. (i) Uncompressed super-resolution image stretched for visualization purpose. (ii) Uncompressed 2D histogram image. (iii) List of localization estimates containing x, y position estimates and their averaged localization uncertainty (square root of their mean variance from CRLBsCMOS). Units are in pixels (103 nm).

  2. 2.

    Supplementary Software

    Example of the developed algorithms implemented in Matlab and CUDA.

Videos

  1. 1.

    Super-resolution video of mEOS3.2-labeled CCPs in a live HeLa cell

    Raw data were recorded as described in the Online Methods. Acquired images were analyzed using single-emitter fitting (Online Methods). 1,200 frames were combined to reconstruct each super-resolution image corresponding to a 2-s time window. Localization estimates in each image were binned into 20-nm pixels for display. To generate the video, we combined all 40 super-resolution images into a three-dimensional (3D) data stack and smoothed with a 3D Gaussian kernel with σx,y = 20 nm and σt = 2 s to aid visualization. The resulting stack was converted into a video playing back at 15 frames per second. Scale bar, 5 μm.

  2. 2.

    Super-resolution video of mEOS3.2-labeled CCPs in a second live HeLa cell

    This video corresponds to the data set shown in Figure 3a–c. Raw data were recorded as described in the Online Methods. Acquired images were analyzed using single-emitter fitting (Online Methods). 1,200 frames were combined to reconstruct each super-resolution image corresponding to a 2-s time window. Localization estimates in each image were binned into 20-nm pixels for display. To generate the video, we combined all 22 super-resolution images into a 3D data stack and smoothed with a 3D Gaussian kernel with σx,y = 20 nm and σt = 2 s to aid visualization. The resulting stack was converted into a video playing back at 15 frames per second. Scale bar, 5 μm.

  3. 3.

    Super-resolution video for a small cutout of a larger data set of mEOS3.2-labeled CCPs in a live HeLa cell (1)

    Raw data were recorded as described in the Online Methods. Acquired images were analyzed using single-emitter fitting (Online Methods). 1,200 frames were combined to reconstruct each super-resolution image corresponding to a 2-s time window. Localization estimates in each image were binned into 10-nm pixels for display. To generate the video, we combined all 28 super-resolution images into a 3D data stack and smoothed with a 3D Gaussian kernel with σx,y = 10 nm and σt = 2 s to aid visualization. The resulting stack was converted into a video playing back at 15 frames per second. Scale bar, 500 nm.

  4. 4.

    Super-resolution video for a small cutout of a larger data set of mEOS3.2-labeled CCPs in a live HeLa cell (2)

    This video corresponds to the data set shown in Figure 3a–c. Raw data were recorded as described in the Online Methods. Acquired images were analyzed using single-emitter fitting (Online Methods). 1,200 frames were combined to reconstruct each super-resolution image corresponding to a 2-s time window. Localization estimates in each image were binned into 10-nm pixels for display. To generate the video, we combined all 19 super-resolution images into a 3D data stack and smoothed with a 3D Gaussian kernel with σx,y = 10 nm and σt = 2 s to aid visualization. The resulting stack was converted into a video playing back at 15 frames per second. Scale bar, 500 nm.

  5. 5.

    Super-resolution video for a small cutout of a larger data set of mEOS3.2-labeled CCPs in a live HeLa cell (3)

    This video corresponds to the data set shown in Figure 3a–c. Raw data were recorded as described in the Online Methods. Acquired images were analyzed using single-emitter fitting (Online Methods). 1,200 frames were combined to reconstruct each super-resolution image corresponding to a 2-s time window. Localization estimates in each image were binned into 10-nm pixels for display. To generate the video, we combined all 29 super-resolution images into a 3D data stack and smoothed with a 3D Gaussian kernel with σx,y = 10 nm and σt = 2 s to aid visualization. The resulting stack was converted into a video playing back at 15 frames per second. Scale bar, 500 nm.

  6. 6.

    Super-resolution video of tdEos-labeled human PDHA1 in COS-7 cells

    Raw data were recorded as described in the Online Methods. Acquired images were analyzed using our multi-emitter fitting algorithm. 200 frames were combined to reconstruct each super-resolution image corresponding to a 0.5-s time window. Localization estimates in each image were binned into 20-nm pixels for display. To generate the video, we combined all 144 super-resolution images into a 3D data stack and smoothed with a 3D Gaussian kernel with σx,y = 20 nm and σt = 2 s to aid visualization. The resulting stack was converted into a video playing back at 25 frames per second. Scale bar , 5 μm.

  7. 7.

    Super-resolution video of mEOS3.2-labeled EB3 in live HeLa cells

    Raw data were recorded as described in the Online Methods. This video corresponds to one of the data sets shown in Supplementary Figure 11. Acquired images were analyzed using our multi-emitter fitting algorithm. 600 frames were combined to reconstruct each super-resolution image corresponding to a 1-s time window. Localization estimates in each image were binned into 20-nm pixels for display. To generate the video, we combined all 21 super-resolution images into a 3D data stack and smoothed with a 3D Gaussian kernel with σx,y = 20 nm and σt = 2 s to aid visualization. The resulting stack was converted into a video playing back at 15 frames per second. Scale bar , 5 μm.

  8. 8.

    Super-resolution video of tdEos-labeled peroxisome membrane protein in live COS-7 cells

    This video corresponds to the dat aset shown in Figure 3d–f. Raw data were recorded as described in Online Methods. Acquired images were analyzed using our single-emitter fitting algorithm. 300 frames were combined to reconstruct each super-resolution image corresponding to a 0.5-s time window. Localization estimates in each image were binned into 20-nm pixels for display. To generate the video, we smoothed all 165 super-resolution images with a 2D Gaussian kernel with σx,y = 20 nm to aid visualization and then combined them into a 3D data stack. The resulting stack was converted into a video playing back at 15 frames per second. Scale bar, 5 μm.

  9. 9.

    Super-resolution video of Alexa Fluor 647–labeled transferrin receptor cluster dynamics as shown in Figure 4a

    Each frame corresponds to a 31-ms reconstructed super-resolution image and is played back at four frames per second.

  10. 10.

    Super-resolution video of Alexa Fluor 647–labeled transferrin receptor cluster dynamics as shown in Figure 4b

    Each frame corresponds to a 31-ms reconstructed super-resolution image and is played back at four frames per second.

About this article

Publication history

Received

Accepted

Published

DOI

https://doi.org/10.1038/nmeth.2488

Further reading