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.

Optimizing imaging speed and excitation intensity for single-molecule localization microscopy

Abstract

High laser powers are common practice in single-molecule localization microscopy to speed up data acquisition. Here we systematically quantified how excitation intensity influences localization precision and labeling density, the two main factors determining data quality. We found a strong trade-off between imaging speed and quality and present optimized imaging protocols for high-throughput, multicolor and three-dimensional single-molecule localization microscopy with greatly improved resolution and effective labeling efficiency.

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

Relevant articles

Open Access articles citing this article.

Access options

Buy article

Get time limited or full article access on ReadCube.

$32.00

All prices are NET prices.

Fig. 1: Effect of laser intensity and imaging speed on image quality.
Fig. 2: Optimization of various imaging applications.

Data availability

All processed data (lists of localizations) and for each condition at least one example file of raw data (camera frames of blinking fluorophores) are deposited on BioStudies (https://www.ebi.ac.uk/biostudies/S-BSST425) under access number S-BSST425. Source data are provided with this paper.

Code availability

The software for the data acquisition and analysis used in this paper is available at github.com/jdeschamps/htSMLM and github.com/jries/SMAP, respectively.

References

  1. Betzig, E. et al. Imaging intracellular fluorescent proteins at nanometer resolution. Science 313, 1642–1645 (2006).

    Article  Google Scholar 

  2. Rust, M. J., Bates, M. & Zhuang, X. Sub-diffraction-limit imaging by stochastic optical reconstruction microscopy (STORM). Nat. Methods 3, 793–796 (2006).

    Article  CAS  Google Scholar 

  3. Heilemann, M. et al. Subdiffraction-resolution fluorescence imaging with conventional fluorescent probes. Angew. Chem. Int. Ed. 47, 6172–6176 (2008).

    Article  CAS  Google Scholar 

  4. Huang, F. et al. Video-rate nanoscopy using sCMOS camera–specific single-molecule localization algorithms. Nat. Methods 10, 653–658 (2013).

    Article  CAS  Google Scholar 

  5. Barentine, A. E. S. et al. 3D multicolor nanoscopy at 10,000 cells a day. Preprint at bioRxiv https://doi.org/10.1101/606954 (2019).

  6. Lin, Y. et al. Quantifying and optimizing single-molecule switching nanoscopy at high speeds. PLoS ONE 10, e0128135 (2015).

    Article  CAS  Google Scholar 

  7. Mortensen, K. I., Churchman, L. S., Spudich, J. A. & Flyvbjerg, H. Optimized localization analysis for single-molecule tracking and super-resolution microscopy. Nat. Methods 7, 377–381 (2010).

    Article  CAS  Google Scholar 

  8. Jones, S. A., Shim, S.-H., He, J. & Zhuang, X. Fast, three-dimensional super-resolution imaging of live cells. Nat. Methods 8, 499–505 (2011).

    Article  CAS  Google Scholar 

  9. Pennacchietti, F., Gould, T. J. & Hess, S. T. The role of probe photophysics in localization-based superresolution microscopy. Biophys. J. 113, 2037–2054 (2017).

    Article  CAS  Google Scholar 

  10. Shroff, H., Galbraith, C. G., Galbraith, J. A. & Betzig, E. Live-cell photoactivated localization microscopy of nanoscale adhesion dynamics. Nat. Methods 5, 417–423 (2008).

    Article  CAS  Google Scholar 

  11. 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).

    Article  CAS  Google Scholar 

  12. Legant, W. R. et al. High-density three-dimensional localization microscopy across large volumes. Nat. Methods 13, 359–365 (2016).

    Article  CAS  Google Scholar 

  13. Thevathasan, J. V. et al. Nuclear pores as versatile reference standards for quantitative superresolution microscopy. Nat. Methods 16, 1045–1053 (2019).

    Article  CAS  Google Scholar 

  14. Nieuwenhuizen, R. P. J. et al. Measuring image resolution in optical nanoscopy. Nat. Methods 10, 557–562 (2013).

    Article  CAS  Google Scholar 

  15. van de Linde, S. et al. Direct stochastic optical reconstruction microscopy with standard fluorescent probes. Nat. Protoc. 6, 991–1009 (2011).

    Article  CAS  Google Scholar 

  16. Dempsey, G. T. et al. Photoswitching mechanism of cyanine dyes. J. Am. Chem. Soc. 131, 18192–18193 (2009).

    Article  CAS  Google Scholar 

  17. van de Linde, S. et al. Photoinduced formation of reversible dye radicals and their impact on super-resolution imaging. Photochem. Photobiol. Sci. 10, 499–506 (2011).

    Article  CAS  Google Scholar 

  18. Zheng, Q. et al. Ultra-stable organic fluorophores for single-molecule research. Chem. Soc. Rev. 43, 1044–1056 (2014).

    Article  CAS  Google Scholar 

  19. Donnert, G., Eggeling, C. & Hell, S. W. Major signal increase in fluorescence microscopy through dark-state relaxation. Nat. Methods 4, 81–86 (2007).

    Article  CAS  Google Scholar 

  20. Huang, B., Jones, S. A., Brandenburg, B. & Zhuang, X. Whole-cell 3D STORM reveals interactions between cellular structures with nanometer-scale resolution. Nat. Methods 5, 1047–1052 (2008).

    Article  CAS  Google Scholar 

  21. Kim, J. et al. Oblique-plane single-molecule localization microscopy for tissues and small intact animals. Nat. Methods 16, 853–857 (2019).

    Article  CAS  Google Scholar 

  22. Gustavsson, A.-K., Petrov, P. N., Lee, M. Y., Shechtman, Y. & Moerner, W. E. 3D single-molecule super-resolution microscopy with a tilted light sheet. Nat. Commun. 9, 123 (2018).

    Article  CAS  Google Scholar 

  23. Bossi, M. et al. Multicolor far-field fluorescence nanoscopy through isolated detection of distinct molecular species. Nano Lett. 8, 2463–2468 (2008).

    Article  CAS  Google Scholar 

  24. Zhang, Y. et al. Nanoscale subcellular architecture revealed by multicolor three-dimensional salvaged fluorescence imaging. Nat. Methods 17, 225–231 (2020).

    Article  CAS  Google Scholar 

  25. Balzarotti, F. et al. Nanometer resolution imaging and tracking of fluorescent molecules with minimal photon fluxes. Science 355, 606–612 (2017).

    Article  CAS  Google Scholar 

  26. Schlichthaerle, T. et al. Direct visualization of single nuclear pore complex proteins using genetically-encoded probes for DNA-PAINT. Angew. Chem. Int. Ed. 58, 13004–13008 (2019).

    Article  CAS  Google Scholar 

  27. Nahidiazar, L., Agronskaia, A. V., Broertjes, J., van den Broek, B. & Jalink, K. Optimizing imaging conditions for demanding multi-color super resolution localization microscopy. PLoS ONE 11, e0158884 (2016).

    Article  CAS  Google Scholar 

  28. Wäldchen, S., Lehmann, J., Klein, T., van de Linde, S. & Sauer, M. Light-induced cell damage in live-cell super-resolution microscopy. Sci. Rep. 5, 15348 (2015).

    Article  CAS  Google Scholar 

  29. Reymond, L., Huser, T., Ruprecht, V. & Wieser, S. Modulation-enhanced localization microscopy (meLM). J. Phys. Photon. 2, 041001 (2020).

    Article  Google Scholar 

  30. Li, Y. et al. Real-time 3D single-molecule localization using experimental point spread functions. Nat. Methods 15, 367–369 (2018).

    Article  CAS  Google Scholar 

Download references

Acknowledgements

We thank A. Rahmani and V. Dunsing for assistance in the data acquisition and processing. We thank Y. Lin and J. Bewersdorf for sharing their raw data and discussing the reanalysis. This work was supported by the European Research Council (grant no. ERC CoG-724489 to J.R.), the National Institutes of Health Common Fund 4D Nucleome Program (grant no. U01 EB021223 to J.R.), the Human Frontier Science Program (grant no. RGY0065/2017 to J.R.), the Engelhorn Foundation (Postdoctoral Fellowship to R.D.) and the European Molecular Biology Laboratory.

Author information

Authors and Affiliations

Authors

Contributions

R.D., M.K. and J.R. conceived the project. R.D. built the main optical setup. M.K., A.S. and U.M. prepared the samples. R.D., M.K. and A.S. acquired and analyzed the data. J.D. wrote the acquisition software. J.R. wrote the analysis software and performed the simulations. R.D., M.K., A.S. and J.R. wrote the manuscript with input from all authors.

Corresponding author

Correspondence to Jonas Ries.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Peer review information Rita Strack was the primary editor on this article and managed its editorial process and peer review in collaboration with the rest of the editorial team.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data

Extended Data Fig. 1 Effect of exposure time for constant excitation intensity in simulations.

Realistic simulations of camera frames (see Supplementary Note 2) for a constant excitation intensity when varying the exposure time over four orders of magnitude. Our standard analysis workflow was used to merge localizations in consecutive frames, stemming from the same fluorophore, and to determine the photons per localization, localization precision, on-time and background per localization per pixel. These simulations show an optimal exposure time that is shorter than the mean on-time, which leads to the highest localization precision. Around this value, the photons per localization and the localization precision display only minimal dependence on the exposure time. For much shorter exposure times, however, merging fails due to missed localizations and large localization errors, reducing photons per localization and deteriorating the localization precision. For exposure times longer than the mean on-time, the background increases linearly with the exposure time, whereas the photons per localization stay constant, also leading to a deteriorated localization precision. Simulation parameters: Mean on-time 30 ms, mean photon emission rate 66 667 Hz, camera read-noise 1.8 e-.

Extended Data Fig. 2 Model for blinking and bleaching of dSTORM dyes.

Model 1 (red curves, inspired by van der Linde et al, Photochem Photobiol Sci, 2011 and Lin et al., PLOS ONE 2015) assumes off-switching from the triplet state, whereas Model 2 (blue curves, inspired by Dempsey et al., Nature Methods, 2011) assumes intensity dependent off-switching from the bright state. B: bright state, T: triplet state, D: dark state, X: bleached state. The graphs show Monte Carlo (MC) simulations (solid line) and numerical integration of partial differential equations (PDE, dashed lines) in comparison to data (dots) for the number of photons per localization, the effective labeling efficiency, the localizations per fluorophore and the on-time. The simulations are described in detail in Supplementary Note 2. For MC simulations, we simulated 100,000 fluorophores for each intensity value, we calculated photons per frame and merged and filtered fluorophores similar to our experimental data analysis. Rate constants from literature Lin et al., PLOS ONE, 2015: kbt=10cm2s1kW1, ktb=70s−1, ktd=15s−1. Rate constants that were chosen to result in similarity of simulations with the experiment: kbd=2cm2s−1kW−1, ktx=0.1cm2s−1kW−1, kdb=0.001cm2s−1kW−1, kdx=0.0002cm2s−1kW−1. Exposure time texp I=0.176skW/cm2, off-switching time toffswitch I=30skW/cm2 with excitation intensity I.

Source data

Extended Data Fig. 3 Raw data.

Four consecutive raw camera frames for the different dyes used in this manuscript, comparing different exposure times and excitation intensities. See Supplementary Table 1 for sample size and replicates for the individual conditions. Size of each image 100 x 100 pixels, scaling 0-900 photons (dyes), 0-300 photons (mMaple). Scale bars 1 µm.

Extended Data Fig. 4 Whole cell imaging with AF647 and CF660C.

Whole cell 3D images of microtubules in U2OS cells immunostained with AF647 (left) or CF660C (right). Individual slices at increasing, subsequently recorded z-positions are shown. The z-distance between individual slices was 300 nm and 50,000 frames were recorded for each slice. The image reconstruction shows a representative result of nc = 2 individual cells from the same preparation (N = 1). Excitation intensity 24 kW/cm2, effective exposure time 7.5 ms.

Extended Data Fig. 5 3D imaging of a mitotic cell with CF660C.

Volume 3D imaging of microtubules stained with CF660C recorded at 24 kW/cm2 excitation intensity and 7.5 ms effective exposure time covers a volume of about 40 µm × 40 µm × 6 µm and, hence, enables SMLM imaging of whole mitotic cells. Top-view projection and orthogonal projections over the indicated regions. See Supplementary Video 3 for a 3D reconstruction. The image reconstruction shows a representative result of nc = 2 individual cells from two different preparations (N = 2).

Extended Data Fig. 6 Three-color ratiometric 3D imaging.

Three-color ratiometric 3D imaging of the nuclear pore complex. Nup96-SNAP is tagged with BG-AF647, ELYS is visualized by indirect immunostaining with CF660C and the central channel is stained with WGA-CF680. a, Lateral view of the lower nuclear envelope. b, Magnified region as indicated in a. c, Side view on region indicated in b. See Supplementary Video 4 for a 3D reconstruction corresponding to c. d, Crosstalk between different channels. Bars show the fraction of the true label being assigned to each channel. The image reconstruction shows a representative result of nc = 3 individual cells from the same preparation (N = 1). The crosstalk was determined from nc = 6 individual measurements from the same preparation (N =1).

Source data

Extended Data Fig. 7 Crosstalk in two-color ratiometric 3D imaging.

Crosstalk in slow ratiometric imaging using Dy634 and AF647. Bars show the fraction of the true label being assigned to each channel. The crosstalk was determined from nc = 6 individual measurements from the same preparation (N =1).

Source data

Extended Data Fig. 8 GLOX+BME buffer long-term stability.

Characterization of long-term stability of GLOX+BME imaging buffer. a, pH value of glucose oxidase/catalase buffer supplemented with 143 mM BME either in direct contact to ambient air or stored in a sealed chamber, measured directly after preparation and after 24 h. Results from six independent experiments each. Error bars denote the accuracy of the pH indicator strips. b, Effective labeling efficiency, c, photons per localization, and, d, localizations per fluorophore for nc = 4 (before 24 h) and nc = 5 (after 24h) individual cells from the same preparation (N = 1) using a sealed chamber. The sample was stored on the microscope at room temperature. Error bars in b-d denote mean ± standard deviation. e, Representative image of Nup96-SNAP BG-AF647 before 24 h and, f, after 24 h storage in the imaging buffer. dSTORM data was recorded at 6.4 kW/cm2. See Supplementary Video 7 for a time-lapse of the buffer response in an open and sealed chamber side by side. The image reconstructions show a representative result of nc = 6 individual cells from one preparation (N = 1).

Source data

Extended Data Fig. 9 Discrepancy to previous work by Lin et al.

Lin et al. reported a nonmonotonic dependence of the photons per localization as a function of the excitation intensity (blue curve, original data read from the publication) for dSTORM imaging of immunostained microtubules. To investigate the discrepancy with our data (gray curve), that is monotonically decreasing photons per localization as a function of the excitation intensity, we partially (nc = 1 for each intensity, N = 1) refitted and reanalyzed the original data kindly provided by Yu Lin and Joerg Bewersdorf. Using our fitter, our localization filter settings and our analysis pipeline, we find higher photons per localization but the same general trend (magenta curve). However, Lin et al. did not keep the product of excitation intensity and single frame exposure time constant. For our data it is constant with tex I=0.18kW/cm2s. Instead, Lin et al. used exposure times as short as 20 ms even for the lowest intensities for example resulting in tex I=0.02kW/cm2s. As a consequence, blink events easily persist over tens of frames, which makes grouping of localizations error prone. To virtually adjust the product of excitation intensity and single frame exposure time for the data of Lin et al., we summed up single frames of the raw data before fitting (9 frames each for 1 kW/cm2, 5 frames each for 1.9 kW/cm2, 2 frames each for 3.9 kW/cm2) or selected the conditions closest to our value chosen for the product (7.8 kW/cm2 to 97 kW/cm2). Additionally, Lin et al. used a less effective setup of emission filters which we estimate to transmit 79 % less fluorescence than our excitation filters. Combining both the refit data of the virtually restored product of excitation intensity and exposure time as well as scaling up the photons by 79 % (green curve) is well in agreement with our results. For each intensity in our data, the mean value ± standard deviation is shown over the data points representing individual cells. For our data, see Supplementary Table 1 for sample size and replicates.

Source data

Extended Data Fig. 10 Effect of filtering on photon counts.

The localization filters applied for organic dyes in this study influence the localization statistics. To visualize the influence, we applied different filter settings to the localizations and re-analyzed the data with respect to the photon counts. Though the absolute numbers change, the general trend of decreasing photon counts with increasing excitation intensity is maintained. These graphs are for the sample Nup96-SNAP-AF647 in 143 mM BME (Supplementary Fig. 2). See Supplementary Table 1 for sample size and replicates. Precision: localization precision filter, PSF: filter on fitted size of the PSF, llrel: filter on the normalized log-likelihood ratio. Error bars denote mean ± standard deviation.

Supplementary information

Supplementary Information

Supplementary Table 1, Figs. 1–12 and Notes 1–3.

Reporting Summary

Supplementary Video 1

Whole-cell 3D, yz reconstruction yz slice translation through the cell shown in Fig. 2a. Scale bar, 10 µm.

Supplementary Video 2

Whole-cell 3D, 3D reconstruction 3D reconstruction of a part of the cell shown in Fig. 2a. Scale bar, 10 µm.

Supplementary Video 3

Whole-cell 3D, mitotic cell 3D reconstruction of the mitotic cell and surrounding cells as shown in Extended Data Fig. 5. Scale bar, 10 µm.

Supplementary Video 4

Three-color 3D data on the NPC 3D reconstruction of the three-color ratiometric image of the NPCs shown in Extended Data Fig. 6. Colors: red, Nup96-SNAP-AF647; green, ELYS immunostained with CF660C and magenta, WGA-CF680. Scale bar, 100 nm.

Supplementary Video 5

Raw camera images, organic dyes Raw data from dSTORM imaging of Dy634, AF647, CF660C and CF680 at different excitation intensities, corresponding to Extended Data Fig. 3. The movies are scaled from 0–900 photons per pixel and per frame. The photon counts reported in this manuscript correspond to merged localizations; that is, are the photons summed over all pixels in the PSF and summed over all consecutive frames in which a fluorophore is in its bright state.

Supplementary Video 6

High-precision NPC 3D reconstruction of an individual nuclear pore complex from a low-intensity, slow dSTORM acquisition of Nup96-SNAP BG-AF647 shown in Fig. 2c,d.

Supplementary Video 7

Blinking buffer long-term stability Time-lapse video of imaging chambers filled with GLOX buffer, either unsealed (left) or sealed with parafilm (right). The buffer was supplemented with 0.005% w/v phenol red which changes its color from red to yellow when the pH changes from pH 8 to pH 6. Corresponding to Extended Data Fig. 8.

Supplementary Video 8

Raw camera images, photoconvertible protein Raw data from PALM imaging of Nup96-mMaple corresponding to the reconstruction shown in Fig. 2e.

Supplementary Dataset 1

Summary of all data: spreadsheet containing data of individual analyzed cells.

Source data

Source Data Fig. 1

Results of the image analysis pipeline for samples relevant to the corresponding figure.

Source Data Extended Data Fig. 2

Results of the image analysis pipeline for samples relevant to the corresponding figure.

Source Data Extended Data Fig. 6

Analysis results to calculate the crosstalk between AF647, CF660C and CF680.

Source Data Extended Data Fig. 7

Analysis results to calculate the crosstalk between AF647 and Dy634.

Source Data Extended Data Fig. 8

Results of the analysis pipeline for samples imaged for buffer-acidification testing and corresponding pH measurements.

Source Data Extended Data Fig. 9

Comparative imaging data of microtubule imaged in this study and in a study by Lin and colleagues.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Diekmann, R., Kahnwald, M., Schoenit, A. et al. Optimizing imaging speed and excitation intensity for single-molecule localization microscopy. Nat Methods 17, 909–912 (2020). https://doi.org/10.1038/s41592-020-0918-5

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s41592-020-0918-5

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