Super-resolution Microscopy with Single Molecules in Biology and Beyond-Essentials, Current Trends, and Future Challenges

Abstract

Single-molecule super-resolution microscopy has developed from a specialized technique into one of the most versatile and powerful imaging methods of the nanoscale over the past two decades. In this perspective, we provide a brief overview of the historical development of the field, the fundamental concepts, the methodology required to obtain maximum quantitative information, and the current state of the art. Then, we will discuss emerging perspectives and areas where innovation and further improvement are needed. Despite the tremendous progress, the full potential of single-molecule super-resolution microscopy is yet to be realized, which will be enabled by the research ahead of us.

Figure 1

Key ideas of single-molecule active control microscopy (SMACM). (A) The camera image of a point emitter or, in good approximation, of a single molecule (point-spread-function, PSF) showing pixelation from a camera detector. (B) Determination of the emitter location by fitting with a model function (here, a 2D Gaussian). (C) Result of the localization procedure. The uncertainty in the position estimate is much smaller than the original width of the PSF. (D) Switching of a single molecule between an emissive state (orange) and a dark state (gray). (E) A simulated structure consisting of concentric rings with different diameter (gray lines) and fluorescently labeled with the orange emitters. The widest ring can just barely be resolved whereas the three smaller rings are not resolved. (F) Active control of emission, yielding a sparse subset of emitting molecules in each single camera frame, and sequential acquisition of many emitter subsets. (G) Super-resolved reconstruction.

Figure 2

Three key experimental components required of all SMACM experiments. (A) Sample preparation and labeling. (B) Optical setup, from fluorophore excitation to detection. (C) Data analysis, i.e. identification and localization of single-molecule signals for image reconstruction.

Figure 3

Selected examples of 3D super-resolution images using single-molecule active control microscopy (SMACM) in cells. (A) Super-resolution image of actin in a COS7 cell revealing the dual-layer organization, where labeling has been done by an Affimer reagent. A site-specifically attached DNA strand has been added to the protein via a Cys mutation and then in the fixed cell, reaction with maleimide-DBCO is performed, and an azide-terminated target DNA strand has been covalently added. Then fluorescently labeled DNA-capture strands are added, and the active control mechanism is binding and unbinding of singles (DNA-PAINT). (B) 3D super-resolution images of synaptonemal complexes in whole mouse spermatocytes, imaged by immunolabeling synaptonemal complex protein 3 (SYCP3) with a blinking fluorescent label. This image required 21 optical sections through six cycles, covering nearly 9 μm in the cell nucleus, and shows the twisting band of the paired lateral elements; the resulting organization of the chromosomes is clear. (C) 3D super-resolution images of tiny membrane tubules on the surface of a cancer cell, where the sialic acids have been covalently labeled with a blinking dye. Here the cell has been imaged with the TILT3D microscope, and individual hollow tubules are shown in the insets.

Figure 4

Importance of sampling the imaged structure. (A) Test patterns consisting of five (left columns) or 20 (right column) stripes, sampled with low (top row) or high numbers (bottom row) of localizations. (B) Corresponding 2D power spectra, radially averaged. PSD = Power spectral density.

Figure 5

High localization precision, small labeling footprint, and high labeling efficiency are equally important for high-quality reconstructions. Shown are simulated patterns of 4 × 4 infinitely small spots with a spacing of 25 nm. Each spot contains 10 binding sites. For each combination of localizations precision and labeling footprint, 0.2, 0.5, and 0.8 average labeling efficiencies are simulated (standard deviation, SD, 0.1 between spots). Each labeled binding site yields on average 10 localizations (SD: 5 localizations). The localizations are binned into 2D histograms with 2 nm bin width, and the edge length for each reconstruction is 125 nm. The color bar ranges from blue through green to yellow. Blue always corresponds to 0 localizations, and yellow to 2 localizations for poor reconstructions and up to 12 localizations for good reconstructions, respectively.

Figure 6

Cooperation between advances in the three key components of SMACM, allowing for improvements in current applications and development of novel applications.