Skip to main content

Thank you for visiting 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.

Ultrafast, temporally stochastic STED nanoscopy of millisecond dynamics


Electro-optical scanning (>1,000 frames/s) with pixel dwell times on the order of the lifetime of the fluorescent molecular state renders stimulated emission depletion (STED) nanoscopy temporally stochastic. Photon detection from a molecule occurs stochastically in one of several scanning frames, and the spatial origin of the photon is known with subdiffraction precision. Images are built up by binning consecutive frames, making the time resolution freely adjustable. We demonstrated nanoscopy of vesicle motions in living Drosophila larvae and the cellular uptake of viral particles with 5- to 10-ms temporal resolution.

Access options

Rent or Buy article

Get time limited or full article access on ReadCube.


All prices are NET prices.

Figure 1: Ultrafast STED scanning and temporally stochastic image assembly.
Figure 2: STED imaging of millisecond-scale dynamic processes.


  1. 1

    Hell, S.W. Science 316, 1153–1158 (2007).

    CAS  Article  Google Scholar 

  2. 2

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

    CAS  Article  Google Scholar 

  3. 3

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

    CAS  Article  Google Scholar 

  4. 4

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

    CAS  Article  Google Scholar 

  5. 5

    Sharonov, A. & Hochstrasser, R.M. Proc. Natl. Acad. Sci. USA 103, 18911–18916 (2006).

    CAS  Article  Google Scholar 

  6. 6

    Egner, A. et al. Biophys. J. 93, 3285–3290 (2007).

    CAS  Article  Google Scholar 

  7. 7

    Biteen, J.S. et al. Nat. Methods 5, 947–949 (2008).

    CAS  Article  Google Scholar 

  8. 8

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

    Article  Google Scholar 

  9. 9

    Hell, S.W. & Wichmann, J. Opt. Lett. 19, 780–782 (1994).

    CAS  Article  Google Scholar 

  10. 10

    Tsien, R.Y. & Backsai, B.J. in Handbook of Biological Confocal Microscopy 2nd edn. (ed. Pawley, J.) 459–478 (Plenum Press, New York, 1996).

    Google Scholar 

  11. 11

    Westphal, V. et al. Science 320, 246–249 (2008).

    CAS  Article  Google Scholar 

  12. 12

    Huang, F. et al. Nat. Methods 10, 653–658 (2013).

    CAS  Article  Google Scholar 

  13. 13

    Lauterbach, M.A., Ullal, C.K., Westphal, V. & Hell, S.W. Langmuir 26, 14400–14404 (2010).

    CAS  Article  Google Scholar 

  14. 14

    Wu, X., Toro, L., Stefani, E. & Wu, Y. J. Microsc. 257, 31–38 (2015).

    Article  Google Scholar 

  15. 15

    Rietdorf, J. & Stelzer, E.H.K. in Handbook of Biological Confocal Microscopy 3rd edn. (ed. Pawley, J.) Ch. 3, 43–58 (Springer, New York, 2006).

    Book  Google Scholar 

  16. 16

    Chen, X., Leischner, U., Rochefort, N.L., Nelken, I. & Konnerth, A. Nature 475, 501–505 (2011).

    CAS  Article  Google Scholar 

  17. 17

    Kremer, Y. et al. Opt. Express 16, 10066–10076 (2008).

    CAS  Article  Google Scholar 

  18. 18

    Salomé, R. et al. J. Neurosci. Methods 154, 161–174 (2006).

    Article  Google Scholar 

  19. 19

    Engelhardt, J., Özcelik, J. & Hell, S.W. Elektrooptische Anordnung zum ultraschnellen Abtasten eines Objektes. German patent DE 202010004547 U1 (2011) and European patent application WO2012171999 A1 (2012).

  20. 20

    Diaspro, A., Chirico, G., Usai, C., Ramoino, P. & Dobrucki, J. in Handbook of Biological Confocal Microscopy 3rd edn. (ed. Pawley, J.) Ch. 39, 690–700 (Springer, New York, 2006).

    Book  Google Scholar 

  21. 21

    Eggeling, C., Widengren, J., Rigler, R. & Seidel, C.A.M. Anal. Chem. 70, 2651–2659 (1998).

    CAS  Article  Google Scholar 

  22. 22

    Tsien, R.Y., Ernst, L. & Waggoner, A. in Handbook of Biological Confocal Microscopy 3rd edn. (ed. Pawley, J.) Ch. 16, 338–352 (Springer, New York, 2006).

    Book  Google Scholar 

  23. 23

    Donnert, G., Eggeling, C. & Hell, S.W. Nat. Methods 4, 81–86 (2007).

    CAS  Article  Google Scholar 

  24. 24

    Bingen, P., Reuss, M., Engelhardt, J. & Hell, S.W. Opt. Express 19, 23716–23726 (2011).

    CAS  Article  Google Scholar 

  25. 25

    Reuss, M., Engelhardt, J. & Hell, S.W. Opt. Express 18, 1049–1058 (2010).

    CAS  Article  Google Scholar 

  26. 26

    Vogelsang, J. et al. Angew. Chem. Int. Ed. 47, 5465–5469 (2008).

    CAS  Article  Google Scholar 

Download references


We thank D. Richardson (MPI-BPC, now at Harvard University) and C. Gregor (MPI-BPC) for providing the TAT-EGFP samples, as well as H. Ta for advice on single-emitter STED measurements. We thank T. Hope (University of Illinois) for the plasmid pEGFP.Vpr. This work was supported by a Leibniz Award of the Deutsche Forschungsgemeinschaft (to S.W.H.) and by the German Federal Ministry of Education and Research (BMBF) within the project STEDlight (FKZ: 13N11173).

Author information




J.S. and J.E. developed the EOD scanning system and set up the microscope. J.S. and J.Z. performed the experiments and analyzed the data. M.M., S.J. Sigrist, J.C. and H.-G.K. provided samples. J.M. performed single-antibody imaging. J.S., J.E., S.J. Sahl and S.W.H. wrote the manuscript. The latter also conceived and guided the project and is responsible for its main points. All authors discussed the data and commented on the manuscript.

Corresponding author

Correspondence to Stefan W Hell.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Integrated supplementary information

Supplementary Figure 1 Increased photon yield for several fluorophores by ultrafast scanning compared to the conventional slow-scanning mode.

More than 40 comparative measurement series were performed in order to assess to which extent our fast scanner reduced bleaching. Samples with different labeling were studied with the different available laser configurations. For a single measurement, areas of the same size and of originally similar brightness within the same sample were scanned fast and slow respectively. The total image acquisition duration was kept the same for both cases to ensure the same light dose over the area. The pixel dwell time was 6.25 ns for the ultrafast scanning and 20 µs for the slow-scanning modality. Fluorophore molecules gradually bleached both in ultrafast- and slow-scanned areas, but the photon yield per applied light dose in ultrafast-scanned areas was clearly higher. (a-c) Examples of three measurement series on samples expressing yellow fluorescent protein (YFP): (a) labeled mitochondriae in fixed cells, subjected to continuous-wave (CW) excitation. (b) living neurons (cytosolic expression) in brain slices, subjected to pulsed excitation. (c) fixed neurons (cytosolic expression), subjected to pulsed excitation and CW-STED. The left axis shows the mean value of detected photons in the scanned areas after each consecutive scan. After each measurement series, the mean number of photons collected in the fast scanned area was divided by the mean number of photons in the equivalent slow-scanned area. High peak power of pulsed lasers presumably caused a significant amount of bleaching directly from the excited state before the molecules could cross to the triplet state. Thus, allowing triplet relaxation had a smaller effect. The obtained values are shown as green triangles (right axis of the graphs). These signal increase factors over many different analyzed samples are summarized in (d) with the photon yield in slow scanning given as 1.

Source data

Supplementary Figure 2 Resolution in ultrafast scanning STED nanoscopy.

Confocal and STED images of (a) yellow green fluorescent beads with a diameter of 24 nm, (b) MreB-rsEGFP in eisosomes in living E. coli, and (c) single-strand DNA labeled with the dye YOYO acquired by electro-optic temporally stochastic STED recordings. The line profiles indicate the resolution performance in the ultrafast scanning setup.

Supplementary Figure 3 Temporally stochastic STED imaging of single antibodies conjugated to Atto 647N.

(a) Examples of raw data and (b) data smoothed with a Gaussian filter (width of 2 pixels). Scale bars: 300 nm. (c) Histogram of numbers of detected signal photons per antibody for n=62 separate single-antibody scans. (d) Histogram of full-widths at half-maximum (FWHM) inferred by fitting symmetric Gaussian functions (with variable additive offset) to the smoothed data, for the same 62 single antibodies. (e) Number of signal photons vs. FWHM. The data show a mean resolution of ~58 nm for a mean of ~38 photons per antibody detected, at a background of ~0.06 photons per pixel (or ~0.6 photons per FHWM area). These data were recorded with fluorescence excitation at 640 nm (8 μW in back focal plane, 20 MHz) and STED at 775 nm (37.5 mW in back focal plane, 20 MHz) using a Leica 100× 1.46 NA oil immersion objective. (Pixel size: 17 nm, 30 000 frames of 5 ns dwell time, i.e. 150 μs effective pixel dwell time).

Supplementary information

Supplementary Text and Figures

Supplementary Figures 1–3 (PDF 786 kb)

Image build-up with ultra-fast scanning

This video shows how images of static samples are acquired. The sample with EGFP-labeled HI-Viruses is scanned with a line frequency of 250 kHz. The photons collected in each frame are accumulated online. One can see both the single frames and the cumulative image during data acquisition in STED as well as in confocal mode. When a sufficient number of photons are collected or when the number of photons per frame decreases, for example due to bleaching, the acquisition can be stopped manually. The video is displayed in real time and is repeated three times. (AVI 11537 kb)

Imaging of diffusing beads

Moving yellow green beads with a diameter of 100 nm are imaged with a frame rate of 793 fps. Many beads were diffusing freely in the water/glycerol film between the cover slip and the cover slide. Some beads were attached to the cover slip. The video in the upper-left panel shows the raw data during STED acquisition. A Gaussian filter with a full width at half maximum of 90 nm was applied to the data shown on the upper-right. The bottom-left panel depicts the raw data of the diffraction-limited confocal imaging of the same area. These data were smoothed with a Gaussian filter of 180 nm width and shown on the bottom-right. The video is slowed down 10 times for display. The original acquisition time was 1.66 sec. (AVI 19927 kb)

Imaging of vesicle dynamics in living Drosophila larvae

Video acquisition of vesicle trafficking in living Drosophila larvae with an effective frame rate of 125 fps through online binning (original scanning rate corresponds to a frame rate of 500 fps). The dynamics of neuropeptidergic vesicles were investigated, which carry EGFP-labeled proANF neuropeptide. Raw data is shown. (AVI 2555 kb)

Imaging of vesicle dynamics in living Drosophila larvae (laterally smoothed)

The raw data of Supplementary Video 3 is laterally smoothed with a Gaussian of 100 nm width and temporally smoothed over two frames. (AVI 3026 kb)

Imaging of HI-Virus uptake in living cells

Video acquisition of dynamics of EGFP-labeled HI-Viruses taken up by living HeLa cells. The effective frame rate of the acquisition is 125 fps. The data shown are laterally smoothed with a Gaussian of 100 nm and temporally smoothed over five frames. (AVI 1958 kb)

Source data

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Schneider, J., Zahn, J., Maglione, M. et al. Ultrafast, temporally stochastic STED nanoscopy of millisecond dynamics. Nat Methods 12, 827–830 (2015).

Download citation

Further reading


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