Comparing Super-Resolution Techniques

Written by Jesse Aaron

---

For many researchers, the prospects afforded by super-resolution (SR) imaging are both vast and exciting. The ability to visualize intact biological specimens at never-before-seen resolutions opens new pathways for elucidating a variety of biological processes, spanning infectious disease mechanisms, carcinogenesis, neurodegeneration, and endless more. But the dizzying array of acronyms reported in the literature, including PALM, STED, SIM, GSDIM, RESOLOFT, iPALM, dSTORM, among many others, can leave researchers unsure as to the optimal technique for their specific application(s). In order to help steer investigators in the right direction, this article will first provide a brief overview of the major classes of SR techniques, with suggested reading for further details. The second part of the article will outline a practical comparison of the two SR systems currently available at the Advanced Imaging Center (AIC).  

Brief Summary of Major Super-Resolution Techniques

In general, super-resolution techniques can be organized into three main classes:  

1. Stimulated Emission Depletion (STED) microscopy

2. Structured Illumination Microscopy (SIM)

3. Single Molecule Localization (SML) microscopy, which includes PALM/iPALM, STORM/dSTORM, GSDIM, RESOLOFT, SOFI, and others.

Stimulated Emission Depletion (STED) Microscopy

STED microscopy was first proposed theoretically in 1994 [1], and experimentally realized approximately several years later by S. Hell and colleagues [2].  The typical STED system operates similarly to a laser-scanning confocal microscope. However, in addition to the conventional excitation, a second illumination spot is super-imposed and scanned in tandem with the first. This second spot is of longer wavelength than the first, and is engineered to take the form of an optical vortex. This appears as a “doughnut” shaped beam, with zero intensity at its center. It acts to cause most of the fluorophores from within the excitation volume to emit at only a single, red-shifted wavelength. This can then be easily removed with a filter, and leaves only the remaining conventional fluorescence from the center of the doughnut to be detected. This remaining signal typically emanates from an area only 30-80nm in diameter, well below the diffraction limit [3]. Importantly, the degree of resolution enhancement will depend strongly on the intensity of the depletion beam, with often very high powers needed. Figure 1 illustrates the basic STED concept.

Figure 1: The basic STED concept. A conventional excitation spot is scanned in tandem with a "doughnut" shaped depletion that prevents normal fluorescence from all but the central 30-80nm area, thus achieving sub-diffraction imaging.

Structured Illumination Microscopy (SIM)

The evolution of SIM occurred more gradually, with the mathematical basis first described over 50 years ago [4]. The first experimental realization in the early 1990s confirmed predictions, but was ultimately limited in application [5].  Following this, strategies were shown that concentrated on improving resolution in the axial direction only [6]. Later incarnations demonstrated resolution enhancement in the lateral direction, and ultimately in all three dimensions [7].  In general, however, SIM relies on illuminating samples with spatially varying light patterns. By rotating these patterns and observing the resulting moiré fringes, information that is previously undetected can be extracted by using Fourier-based analysis. Importantly, however, SIM is fundamentally limited to a 2x improvement in resolution, or approximately 100nm laterally, and 300nm axially, depending on the objective being used.

Figure 2: Structured illumination Microscopy (SIM). (Left) Moiré fringes (dark bands) form between two periodic patterns. (Right) The blue circle (I) indicates the "observable region" of spatial frequencies in a conventional microscope. The three Fourier components of a sinusoidally striped illumination pattern (blue dots) must fall within this circle to be observable within the diffraction limit. (II) Illuminating the sample with structured light extends the observable region in (I) to contain the spatial frequencies within two offset regions (violet). (III) Moving and rotating the structured illumination can extend the frequency space by as much as a factor of 2. Adapted from Gustafsson et al., J. Microscopy 198 82-87 (2000).

Single Molecule Localization (SML) Microscopy

In contrast to STED and SIM instrumentation, which primarily exploit novel illumination principles, SML SR Microscopy makes use of specialized sample labeling/mounting and unique detection strategies. First reported by a flurry of studies in 2006 [8]–[10], this basic strategy has thus far achieved some of the highest effective resolutions available using conventional optics, including 3D SR [11]–[13].

The SML principle relies on two key aspects, and is applicable to most variations such as PALM and STORM:  First, via the use of photoswitchable and/or photoactivatable fluorescent proteins, or dyes in combination with special buffer formulations, samples can be imaged such that only a small, random subset of its fluorophores are visible at any given time.  By tuning the excitation laser power, as well as a UV wavelength “activation” laser, images can be acquired that contain only single, isolated fluorophores. Second, based on the assumption that acquired images contain individual, well-isolated fluorophores, their location can be found with much higher precision than what is possible using normal diffraction-limited microscopy by computing the centroid of the detected point-spread functions. By repeatedly imaging a sample over time, the nanoscale location of nearly all fluorophores can be found.  Figure 3 outlines this general concept. Importantly, the requirement to image samples over thousands of acquisitions can limit its utility for dynamic and/or live cell imaging, although some demonstrations of this have been shown [14]. 

Figure 3: Basic principles of single molecule localization (SML) microscopy. (A) By imaging single fluorophores over time via photoswitching dyes (either endogenous or exogenous), a full reconstruction of nearly all a sample's fluorophores can be achieved. (B) The various SML techniques (such as PALM, STORM, GSDIM, etc.) mainly vary in the way fluorophores are rendered visible and/or invisible over time, and include a number of mechansisms such as photoswitching, photoactivation, and photoconversion. Adapted from Dempsey et al., Nat. Meth. (2011)

A Practical Comparison:  iPALM vs. Multicolor Live Cell SIM

The AIC at Janelia features two super-resolution imaging systems with unique capabilities:  The Multicolor Live Cell SIM [15] and iPALM [13].  The former system makes novel use of a spatial light modulator (SML) to generate SIM patterns with high speed, obviating the need to physically move any microscope hardware component.  This allows rapid image acquisition, and permits temporal resolutions that are unavailable in commercial SIM microscopes.  With the current system, 2-10 SIM Z-stacks can be acquired per second and up to two colors (corresponding to 488nm and 560nm excitation) can be imaged.  Furthermore, sample preparation procedures developed for conventional microscopy can be applied to SIM, with no specialized dyes or imaging buffer formulations required. Multicolor Live Cell SIM samples are housed within a temperature and CO2 controlled incubator, making it well-suited for sensitive and/or longer term live cell experiments.  Lateral (XY) resolutions of 110nm and Z resolution of 360nm can be achieved.

The iPALM system is a novel SML microscopy variant that makes use of an interferometric approach to achieve 3D resolution enhancement.  To date, the iPALM has demonstrated 7-30nm resolution in all three dimensions, giving it the highest overall resolution available at the AIC, and arguably in any imaging facility, using conventional far field optics.  Unlike the Multicolor Live Cell SIM, sample fluorophore brightness will ultimately determine the final resolution, rather than any fundamental instrumentation limit.  Importantly, however, faithful image reconstruction can require on the order of ~104 total raw images, which can span ca. 5-30 minutes of acquisition time.  This puts a major constraint on the ability of the iPALM to capture dynamic events. Thus iPALM typically utilizes fixed samples for optimal results.  Furthermore, as in any SML technique, iPALM requires sample be labeled with fluorphores that can be stochastically switched to/from “dark” and “light” states.  This may include photoswitchable/ photoactivatable proteins, photoswitchable organic dyes, or a combination thereof [16].  A variety of options currently exist across the visible spectrum, and comprehensive lists of available labels can be found in several sources, including HERE.  The iPALM at the AIC can image samples with a variety of wavelengths spanning the optical range, including 488, 561, 647, and 705nm, with a 405nm “activation” laser.

Selecting the optimal super-resolution system at the AIC:  Temporal vs. Spatial Resolution

For researchers interested in utilizing the AIC’s super-resolution capabilities, the most fundamental consideration will typically be the tradeoff between temporal and spatial resolution between the two systems.  For those researchers that are most interested in capturing dynamic events, the Multicolor Live Cell SIM system will likely be the more effective choice.  The iPALM system, however, will be preferred for studies examining detailed molecular organization that require the highest spatial resolution possible.  To help illustrate this point, refer to the figures below that give prototypical application examples that highlight the strengths of each system.

Figure 4: Multicolor Live Cell SIM images of live HeLa cell mitochondria labeled with MitoTracker green. SIM resolution enhancement allows visualization of mitochondrial membranes, which were not apparent in conventional widefield images (not shown). Furthermore, Live Cell SIM allowed following morphological changes over time. In this case, a 38-plane z-stack was acquired in 20.3 seconds, which included 15 pattern rotations in each slice to maximize image quality. Scale bars are 1um. Taken from Shao L et al Nat. Meth. 2011;8:1044–1046

Figure 5: iPALM XY images of integrin αv (top left) and actin (top right) distributions in focal adhesion complexes within U2OS cells. The colormap is coded for z-position (see inset top right). Note that the z-distributions of these proteins (bottom left, bottom right) are clearly discernable using iPALM, despite both proteins occupying only ~100nm of axial distance from the cell membrane. Taken from P. Kanchanawong et al. Nature 468, 580-584 (2010)

In Figure 4, the dynamic behavior of mitochondria in live cells can be imaged with relatively high temporal resolution, even with 38 z-stacks being acquired for each experiment.  Even higher temporal resolution is possible if users are willing to forgo 3D information and utilize the system in 2D TIRF mode.  The resolution enhancement available with the Multicolor Live Cell SIM allows visualization of structural details that are difficult or impossible to see in conventional widefield microscopy, but limited to about half the normal resolution limit.  Figure 5, on the other hand, shows the exquisite resolution available with iPALM imaging.  The molecular distributions of components within the focal adhesion complex are clearly separable in the axial direction, despite the fact that they occupy only 100nm of axial space from the cell membrane.  Such discrimination would be impossible with Multicolor Live Cell SIM, as this represents roughly 1/3rd the resolution limit.  Importantly, however, the iPALM would be a poor choice to visualize these structures in a live, moving cell, as the image reconstruction would contain significant motion artifacts.  Furthermore, the proteins in question were labeled with photoactivatable Eos fluorescent protein variants in order to perform PALM imaging. While photoactivatable labels are similar to other conventional fluorescent proteins in terms of their transfection into cells, added efforts must be made if researchers wish to develop stably expressing cells lines of their choice.  

A further consideration is imaging depth. While both systems can be operated in either TIRF or epi-mode, the advantages of TIRF mode in iPALM are more pronounced. Indeed, while Multicolor Live Cell SIM can effectively image samples to within 5-10 microns of depth, iPALM is most effective at <1um from the coverslip.

In summary, the AIC at Janelia offers two unique super-resolution imaging systems that maximize the advantages of each technique:  the Multicolor SIM system allows unprecedented temporal resolution, while giving the 2-fold resolution enhancement inherent to this method.  The iPALM system allows equally unprecedented spatial resolution, particularly in the axial direction.  In choosing the correct system, potential users should carefully consider the relative strengths of each in light of the particular questions they wish to answer.  

References

1. S. W. Hell and J. Wichmann, “Breaking the diffraction resolution limit by stimulated emission: stimulated-emission-depletion fluorescence microscopy,” Opt. Lett., vol. 19, no. 11, pp. 780–782, Jun. 1994.

2. T. A. Klar, S. Jakobs, M. Dyba, A. Egner, and S. W. Hell, “Fluorescence microscopy with diffraction resolution barrier broken by stimulated emission,” Proc. Natl. Acad. Sci., vol. 97, no. 15, pp. 8206–8210, Jul. 2000.

3. V. Westphal and S. W. Hell, “Nanoscale Resolution in the Focal Plane of an Optical Microscope,” Phys. Rev. Lett., vol. 94, no. 14, p. 143903, Apr. 2005.

4. W. Lukosz and M. Marchand, “Optischen Abbildung Unter Überschreitung der Beugungsbedingten Auflösungsgrenze,” Opt. Acta Int. J. Opt., vol. 10, no. 3, pp. 241–255, Jul. 1963.

5. B. Bailey, D. L. Farkas, D. L. Taylor, and F. Lanni, “Enhancement of axial resolution in fluorescence microscopy by standing-wave excitation,” Nature, vol. 366, no. 6450, pp. 44–48, Nov. 1993.

6. M. A. A. Neil, R. Juskaitis, and T. Wilson, “Method of obtaining optical sectioning by using structured light in a conventional microscope,” Opt. Lett., vol. 22, no. 24, pp. 1905–1907, Dec. 1997.

7. M. G. L. Gustafsson, L. Shao, P. M. Carlton, C. J. R. Wang, I. N. Golubovskaya, W. Z. Cande, D. A. Agard, and J. W. Sedat, “Three-Dimensional Resolution Doubling in Wide-Field Fluorescence Microscopy by Structured Illumination,” Biophys. J., vol. 94, no. 12, pp. 4957–4970, Jun. 2008.

8. M. J. Rust, M. Bates, and X. Zhuang, “Sub-diffraction-limit imaging by stochastic optical reconstruction microscopy (STORM),” Nat Meth, vol. 3, no. 10, pp. 793–796, Oct. 2006.

9. E. Betzig, G. H. Patterson, R. Sougrat, O. W. Lindwasser, S. Olenych, J. S. Bonifacino, M. W. Davidson, J. Lippincott-Schwartz, and H. F. Hess, “Imaging Intracellular Fluorescent Proteins at Nanometer Resolution,” Science, vol. 313, no. 5793, pp. 1642–1645, Sep. 2006.

10. S. T. Hess, T. P. K. Girirajan, and M. D. Mason, “Ultra-High Resolution Imaging by Fluorescence Photoactivation Localization Microscopy,” Biophys. J., vol. 91, no. 11, pp. 4258–4272, Dec. 2006.

11. M. F. Juette, T. J. Gould, M. D. Lessard, M. J. Mlodzianoski, B. S. Nagpure, B. T. Bennett, S. T. Hess, and J. Bewersdorf, “Three-dimensional sub-100 nm resolution fluorescence microscopy of thick samples,” Nat Meth, vol. 5, no. 6, pp. 527–529, Jun. 2008.

12. B. Huang, W. Wang, M. Bates, and X. Zhuang, “Three-Dimensional Super-Resolution Imaging by Stochastic Optical Reconstruction Microscopy,” Science, vol. 319, no. 5864, pp. 810–813, Feb. 2008.

13. G. Shtengel, J. A. Galbraith, C. G. Galbraith, J. Lippincott-Schwartz, J. M. Gillette, S. Manley, R. Sougrat, C. M. Waterman, P. Kanchanawong, M. W. Davidson, R. D. Fetter, and H. F. Hess, “Interferometric fluorescent super-resolution microscopy resolves 3D cellular ultrastructure,” Proc. Natl. Acad. Sci., vol. 106, no. 9, pp. 3125–3130, Mar. 2009.

14. H. Shroff, C. G. Galbraith, J. A. Galbraith, and E. Betzig, “Live-cell photoactivated localization microscopy of nanoscale adhesion dynamics,” Nat Meth, vol. 5, no. 5, pp. 417–423, May 2008.

15. R. Fiolka, L. Shao, E. H. Rego, M. W. Davidson, and M. G. L. Gustafsson, “Time-lapse two-color 3D imaging of live cells with doubled resolution using structured illumination,” Proc. Natl. Acad. Sci., Mar. 2012.

16. G. Patterson, M. Davidson, S. Manley, and J. Lippincott-Schwartz, “Superresolution Imaging using Single-Molecule Localization,” Annu. Rev. Phys. Chem., vol. 61, no. 1, pp. 345–367, Mar. 2010.