Nanoscale microscopy goes green

Fluorescence microscopy has become a popular mainstay of subcellular visualization, but its resolving power is limited by the diffraction of light. Several approaches have been taken to break the 'diffraction barrier', including an approach called stimulated emission depletion (STED) microscopy. In this technique, the area of the focal spot of a fluorescence-excitation beam is tightened by precisely overlaying a second doughnut-shaped beam that is tuned to de-excite fluorophores by stimulated emission. By this means, the focal spot area can be reduced below the diffraction barrier of a conventional fluorescence microscope.

Imaging applications of STED microscopy until now have used organic dye staining to visualize cell organelles. Now, the authors have applied STED microscopy to samples labelled with green fluorescent protein (GFP). They achieved a resolution of 70 nm when analysing GFP-tagged virus-like particles and fixed mammalian cells that expressed GFP. A comparison of confocal images and STED images clearly showed the improved resolution in STED images. STED microscopy is compatible with other fluorescent proteins and multicolour labelling, and future technical advances are expected to further improve the resolution. The authors anticipate that nanoscale fluorescent-protein microscopy will “open up a new avenue for answering many key questions in biology.”

Stochastic imaging

Single fluorescent molecules can be detected with sub-diffraction-barrier resolution, but accurately resolving multiple fluorophores that are in close proximity poses a problem. However, a new imaging technique — stochastic optical reconstruction microscopy (STORM) — overcomes this drawback by repeated high-resolution detection of individual fluorophores within a sample. STORM uses fluorophores that can be switched reversibly between a fluorescent and a dark state. A cycle of imaging intentionally turns on only a fraction of these fluorophores at a time, so that each fluorophore can be optically resolved from the rest, which allows the position of the fluorophore to be determined with high accuracy. The key is that a stochastically different subset of fluorophores is turned on in each cycle. The final image of the sample is then reconstructed by combining a set of multiple high-resolution images.

Using STORM, the location of proteins bound to a single DNA plasmid could be mapped with a sub-diffraction-barrier resolution of 20 nm. The method should therefore be a valuable tool for high-resolution fluorescence in situ hybridization and immunofluorescence imaging. Although the concept was tested with a cyanine dye, Cy5, it is also applicable to other photoswitchable fluorophores and fluorescent proteins.