Exploring beyond the diffraction limit
Despite all the great insights obtained with conventional fluorescence microscopy, it is easy to become frustrated when studying subcellular structures, as the biological entities of interest are often significantly smaller than the diffraction limit. Under conventional far-field fluorescence microscopes, which are incapable of resolving objects closer to one another than about 200 nm, details of interest are lost in a blur. Several methods of far-field superresolution microscopy overcome this fundamental obstacle and allow morphological details to be studied far beyond the diffraction limit. Stimulated Emission Depletion (STED) imaging , as confocal laser scanning microscopy, moves a spot of excitation light over the sample, detects the emitted fluorescence and generates pixel by pixel the image of the observed optical section.
To achieve superresolution, the diffraction-limited excitation spot is overlaid with a donut-shaped point spread function of a second laser: the STED laser. A process called stimulated emission prevents dyes from emitting fluorescence anywhere in the focal volume except for the very center of the donut. The amount by which the effective focus can be shrunk, i.e. the resolution that can be achieved, depends on the intensity of the STED laser as well as on the dye. With the commerically available STED-microscopes, sub 60 nm lateral resolution is routinely achieved with e.g., Atto 647N. Confocal and STED microscopy are a perfect match and can be readily implemented in the same setup. With the available commercial realizations the user can toggle between confocal and STED resolution by a single mouse click.
Achieving 2C confocal superresolution
STED microscopy allows fast and unbiased information to be obtained on the structure and organization of a biological entity of interest. The next step is to put this information into context. Much can already be learned by exploring one structure with a resolution of e.g., 60 nm and imaging several confocal counterstains at the same time. Nevertheless, seeing how two structures are organized in relation to one an-other in subdiffraction-limited detail opens the door to a completely new world of co-localization studies.
Two approaches for achieving two color (2C) STED images have been reported. The first is technically demanding and uses two separate sets of excitation and STED wavelengths for two spectrally separated dyes  The second solution (the "one donut approach") uses a standard fluorophore (Stokes shift 10–30 nm) in combination with a large Stokes shift dye (e.g., >100 nm for Chromeo 494) of partially overlapping emission spectra . This allows the use of only one STED laser. The differences in the excitation spectra are exploit-ed to distinguish the two dyes. To generate a 2C image, two frames are recorded with different excitation laser lines active in each frame. With the right dye combination and a balanced staining, brilliant 2C STED images are collected following this approach, even when the same band is detected for both channels. Of course, the differences in emission spectra can also be exploited to help distinguish the dyes. The advantage of applying the same STED donut for both superresolution channels not only reduces the overall cost and complexity of the system, but also avoids chromatic aberration problems, which other implementations of superresolution microscopy need to compensate.
Providing standard tools for co-localization studies at the nanoscale
Both confocal superresolution systems (pulsed and cw-version) were enabled to perform co-localization studies in the sub 100 nm realm. To this end, a second excitation laser at 531 nm was introduced in addition to the 640 nm line for the pulsed STED system, which realizes confocal nanoscopy with pulsed lasers in the deep red range. Also, a special filter cube for the highly sensitive avalanche photodiode detectors (APDs) was designed for the recommended dye pair Chromeo 494/Atto 647N. This assures optimal dye separation without the need for post processing.
The cw-version uses continuous wave lasers for STED imaging in the visible range. It facilitates confocal nanoscopy with green standard dyes, e.g., Alexa 488, Oregon Green 488, FITC, Chromeo 488, and also autofluorescent proteins (FPs) e.g. eYFP, Citrin and Venus. A newly designed objective – the HCX PL APO 100x 1.40NA Oil STED Orange – allows the cw-version user to freely choose any argon line (458, 476, 488, 496, 514 nm) for excitation together with the 592 nm laser for STED. So a lot of dye and or fluorescent protein combinations have become applicable to 2C confocal nanoscopy.
Excellent 2C images are obtained e.g. from samples stained with BD Horizon V500 and Oregon Green 488 without crosstalk between channels. Fluorophore combinations showing a bigger spectral overlap also yielded good results after dye separation with the appropriate software package. With the multitude of available dyes and FPs, especially in the visible range, more and more fluorophore combinations will prove useful for multicolor confocal nanoscopy.
The optical sectioning capability of confocal microscopy has boosted co-localization studies to the next level. Nevertheless, diffraction-limited resolution often masks important subcellular details in the recorded z-stacks. Fully embedded in state-of-the-art confocal instruments, multicolor STED microcopy allows fast, unbiased and easy access to hidden morphologies within optical sections of cells, tissues or even organisms. The multicolor confocal nanoscopy systems enable life science researchers to explore cellular architecture beyond the diffraction limit.
- Hell SW and Wichmann J: Opt Lett 19 (1994) 780–82.
- Donnert G et al.: Biophys J 92:8 (2007) L67–9.
- Schmidt R et al.: Nat Methods 5:6 (2008) 539–44.