A couple of awesome new papers out on superresolution imaging.
The first one is on PALM imaging in live tissues. Despite Eric’s promise that he’s “getting out of PALM imaging”, the Betzig lab continues to pump out papers on the topic. In a May Nature Methods paper, they do the logical extention of their previous work, demonstrating PALM imaging on live tissue. Frame rates of up to 1/25 Hz and spatial resolution of 60nm. Clearly the resolution is not optimal at these speeds, but they are now able to see dynamic processes. Improved photoconvertable fluorescent proteins could dramatically increase the speed. Don’t miss Mats Gustafsson’s informative commentary on the work. We ♥ this Mats, not this Mats.
Next up is a collaboration between Heinrich Leonhardt, John Sedat, and Gustafsson’s groups. Using 3D structured illumination (3D-SIM), they do multicolor 3D superresolution imaging in fixed tissue. SIM works by repeatedly illuminating a sample with gratings of interfering light, rotating the angle of the illumination pattern. The resulting dataset can be used to reconstruct the sample beyond the diffraction limit. Here the authors add an illumination pattern that varies in the z-dimension, allowing them to image with superresolution in all three axes. 3D-SIM seems to be a bit behind competing techniques on the typically achievable spatial resolution. However, it has a distinct advantage over PALM, STORM and STED. It uses standard fluorescent dyes, making it well suited for multicolor acquisition and compatible with the huge library of existing labels.
Finally, Stefan Hell’s group has managed to extend the STED technique into 3D. Using a 4Pi objective configuration (objective on the top and the bottom) with STED, they were able to sculpt their excitation spot into a 45nm sphere. Sweeping this across the sample allowed two-color 3D reconstruction of mitochondria morphology @ 40nm resolution in fixed mammalian cells. The excitation volume can theoretically be continuously tuned to arbitrary size.