ARTICLE SERIES: Imaging Cell Science at a Glance 1607

Super-resolution visualization of ensemble perturbations. It is and >450–700 nm in the z direction. This limit, now possible to visualize the individual also called the point-spread function (PSF), is microscopy at a glance molecules as they dynamically interact. Super- the fixed size of the spread of a single point of resolution microscopy offers exciting light that is diffracted through a microscope; it is Catherine G. Galbraith1,* and opportunities for biologists to ask entirely new also a measure of the minimum-size point James A. Galbraith2,* levels of questions regarding the inner workings source or object that can be resolved by a 1National Institute of Health, 1NICHD and 2NINDS, of the cell. microscope. Objects that are smaller than the Bethesda, MD 20892, USA The impact of super-resolution microscopy is PSF appear to be the same size as the PSF in *Authors for correspondence ([email protected]; rapidly expanding as commercial super- the microscope, and objects that are closer [email protected]) resolution microscopes become available. together than the width of the PSF cannot be Journal of Cell Science 124, 1607-1611 However, super-resolution microscopes are not distinguished as separate. A commonly used © 2011. Published by The Company of Biologists Ltd doi:10.1242/jcs.080085 based on a single technology, and the differences measure of the PSF width is the Rayleigh (R) between the individual technologies can criterion: R0.61l/NA, where NA is the Advances in microscopy and cell biology are influence how suited each approach is to address numerical aperture. Any microscopy technique intimately intertwined, with new visualization a specific cell biological question. Here, we that overcomes the resolution limit of possibilities often leading to dramatic leaps in highlight the main technologies and conventional light microscopy by at least a our understanding of how cells function. The demonstrate how the differences between them factor of two is considered to be a super- recent unprecedented technical innovation of can affect biological measurements. resolution technique. super-resolution microscopy has changed the Super-resolution techniques break the limits of optical resolution from ~250 nm to Defining super resolution diffraction limit by temporally or spatially ~10 nm. Biologists are no longer limited to The resolution limit of conventional light modulating the excitation or activation light. For inferring molecular interactions from the microscopy is ~250 nm in the x and y direction, example, structured illumination microscopy Journal of Cell Science

(See poster insert) 1608 Journal of Cell Science 124 (10)

(SIM) illuminates the entire field with a striped the PSF to reduce its effective diameter. They toactivation localization microscopy (PALM) pattern of light (Gustafsson, 2000). When this surround a laser-scanning focal excitation PSF (Betzig et al., 2006) and fluorescent PALM excitation pattern mixes with the spatial pattern with an annulus of longer-wavelength light that (fPALM) (Hess et al., 2006) use genetically of the sample they produce an interference has an intensity high enough to saturate all expressed photoactivatable probes to achieve pattern (called a moiré fringe) that is much fluorophores in the annulus to the ground state the on–off states of the fluorophore. An coarser than either pattern alone and is (STED) (Hell and Wichmann, 1994) or the alternative approach, stochastic optical detectable by the microscope. The excitation meta-stable dark state (GSD) (Bretschneider et reconstruction microscopy (STORM) (Rust et pattern is translated and rotated to generate a al., 2007) to suppress their fluorescent emission. al., 2006), uses pairs of cyanine (Cy) dyes, series of images with different moiré fringes. As Increasing the intensity of light in the annulus typically coupled to antibodies up to 15 nm in the illumination pattern is known, it can be expands the zone of saturation to create an length, to act as reporter and activator pairs in mathematically removed from the moiré to gain increasingly smaller excitation PSF that is order to cycle multiple times between the dark access to the normally irresolvable higher smaller in diameter than the diffraction limit of and light states. In direct STORM (dSTORM) resolution information in the sample. SIM 200 nm. The smaller PSF is then scanned over (Heilemann et al., 2008), several stand-alone increases resolution to ~100 nm in the x-y the sample to generate the enhanced-resolution synthetic dyes, such as Alexa-Fluor dyes, can direction and ~400 nm axially (Schermelleh et image. The best two-dimensional (2D) also be used in a blinking mode to attain super- al., 2008). SIM is limited to this factor-of-two resolution in a biological context for STED or resolution images. Despite the technical improvement because the periodicity of the GSD that has been achieved so far is 20–30 nm differences between these techniques, they all illumination pattern is created by diffraction- full width at half maximum (FWHM) across the localize the position of the molecule of interest limited optics and is, therefore, limited by the PSF (Westphal and Hell, 2005). The commercial to ~20 nm. PSF of conventional microscopy (Gustafsson, implementation of STED currently has a 2000). resolution of ~70 nm laterally with no increase The resolution of super resolution Other super-resolution imaging techniques in axial resolution above diffraction. Although The resolution of any image, conventional or modulate the excitation light to exploit the the excitation volume is below the diffraction super resolution, depends on the number of ability to saturate the emission of fluorophores limit, these techniques only make ensemble points that can be resolved on the structures in order to break the diffraction barrier by a measurements because they do not distinguish of interest. According to the Nyquist–Shannon greater amount. Saturation can be achieved by between individual molecules within an criteria (Shannon, 1949), a structural feature can using intense illumination to produce a excitation volume. only be resolved when the distance between two photophysical transition of the fluorophore to There are also good prospects for improving labels is less than half the feature size. a transient dark state that can lead to either a the resolution of SIM. The non-linear Therefore, at least two points need to be permanently dark state (bleaching) or the relationship between excitation and emission resolved within the minimum spatial feature size emission of light on a microsecond or can be combined with the illumination pattern that is to be imaged. Localizing individual millisecond time scale, which is much slower used in SIM. This technique – saturated SIM molecules with high precision does not create a than the nanosecond time scale of fluorescence. (SSIM) (Gustafsson, 2005) – can be thought of super-resolution image when there are not Alternatively, super-resolution techniques can as the inverse of STED, in which sharp dark enough labeled molecules within the PSF to

Journal of Cell Science use light to induce photochemical reactions regions are created instead of sharp bright identify the spatial and temporal features of the in photoswitchable or photoactivatable regions. The resolution of SSIM scales with the structure. Thus, when a single-molecule-based fluorophores, and either transition them level of saturation, and 50 nm lateral resolution super-resolution image is reported as having a between on and off states or change their color. has been demonstrated using fluorescent beads resolution of 20 nm, the number must refer to As long as these transitions can be limited to a (Gustafsson, 2005). The general term for super- structural resolution and not simply indicate that subset of fluorophores that are spatially resolution techniques that make use of the image is only displaying molecules whose separated by the distance of the microscope PSF, the reversible non-linear switching of the centroids could be identified with 20 nm the molecules can be located with precision fluorophore state is reversible saturable optical certainty. approaching 5 nm. Super-resolution techniques fluorescence transitions (RESOLFT) (Hell, can be separated into two categories depending 2003). Speed of acquisition on whether these effects are exploited at the Many super-resolution techniques obtain ensemble level or at the single-molecule level. Single-molecule techniques increased resolution at the cost of the speed of Single-molecule-based techniques overcome image acquisition, simply because they use Ensemble techniques the diffraction limit by using light to turn on only conventional microscope optics and hardware. Ensemble-based techniques increase resolution a sparse subset of the fluorescent molecules of It takes longer to scan the smaller PSF of the by shaping the excitation light, and the interest. Even though the visualized molecules STED excitation beam across a specimen than it resolution of the images obtained by these that are turned on appear to have the size of a would take to scan the larger PSF of a techniques is determined by the size of the PSF in a conventional microscope, if they are conventional spot-scanning microscope. The super-resolution PSF. By contrast, single- separated from each other by at least 200 nm, temporal resolution of STED microscopy has molecule techniques rely on localization then the concept of single-molecule localization been reported as 35 mseconds per image with a precision, which is the uncertainty in the can be used to determine their centroids with 1.8 m ´ 2.5 m field of view, at a 2D identification of the center of a molecule’s PSF nanometer-level precision (Gelles et al., 1988; resolution of 62 nm (Westphal et al., 2008). and depends on photon output (Thompson et al., Yildiz et al., 2003). This process is repeated over However, with a larger field of view, imaging 2002). Importantly, localization precision is not many cycles, with new molecules turning on and speed slows down dramatically. Acquisition the resolution of the image. Stimulated emission other molecules turning off in a stochastic times of 10 seconds per image over a 2.5 m ´ depletion (STED) microscopy and ground-state manner. Super-resolution images are created by 10 m field of view for a 50 nm resolution have depletion (GSD) microscopy optically modify assembling all of the localized points. Both pho- been reported for the very bright fluorescence- Journal of Cell Science 124 (10) 1609

filled dendritic spines in living cells (Nagerl et reversibly or irreversibly switchable between a excitation and the STED beam (Wildanger et al., al., 2008). Similarly, if image acquisition light and a dark state, or they need to change 2008). Multi-color STORM requires either requires moving a grid and collecting as many as from one wavelength to another wavelength. pairing the same activator Cy dye to one of three 15 images to generate one SIM super-resolution The probes should be as bright as possible and spectrally different reporter Cy dyes or pairing image, then super-resolution images will be should have a high contrast ratio between the spectrally distinct activators to the same obtained at a slower rate than conventional two states. Of course, different techniques have reporter. In the first scenario, distinct emission images, i.e. at ~30 seconds per image. However, different criteria for what makes a good probe. spectra are the multicolor read out and, in the a recent implementation of SIM using a For STED, dyes need to be easily driven into second scenario, distinct activation spectra are ferroelectric liquid crystal on a silicon spatial stimulated emission, have no excitation at the used to temporally separate the constant light modulator to produce the patterns has depletion wavelength, and have photostability emission (Bates et al., 2007). Finally, multi- proven to be much faster. Image fields of 32 m to withstand high intensities at both the color PALM schemes originally imaged ´ 32 m and 8 m ´ 8 m have been depletion and the excitation wavelengths. ATTO multiple colors sequentially, by photoconverting successfully imaged at 3.7 Hz and 11 Hz, dyes were originally used for STED; however, one fluorophore from a lower-wavelength to a respectively (Kner et al., 2009). Therefore, it is an extensive list of additional dyes, their higher-wavelength color and then reversibly important to consider whether dynamic resolution, parameters used for depletion and switching a lower-wavelength fluorophore biological structures change on a time scale that excitation, and references can be found online at (Shroff et al., 2007). More recently, two-color is slower than the time required to collect an the website of the Department of imaging has been done by using spectrally image when choosing a super-resolution NanoBiophotonics at the Max Planck Institute separated dark-to-light probes (Subach et al., method. for Biophysical Chemistry, Göttingen, Germany 2010). However, there is still a need for The trade-off between speed and resolution is (http://www.mpibpc.mpg.de/groups/hell/STED additional probe development, especially typically more pronounced with single- _Dyes.html). for green fluorophores, where low photon molecule techniques. Two molecules cannot be All single-molecule super-resolution outputs and the inability to identify transfected turned on within the same PSF at any given time, techniques rely on localization precision, which cells prior to activation have slowed down which limits the speed of the molecular read- is approximated as s/ͱ (N), where s is the simultaneous dual-color super-resolution out. Although the concept of turning on a sparse standard deviation of the PSF, and N is imaging efforts. The continued development of subset of labeled molecules has enabled high- the number of photons detected (Thompson et technologies such as dSTORM, which rely on speed single-molecule tracking with high al., 2002). Therefore, the approximately tenfold manipulating the photophysics of dyes used in molecular density (Hess et al., 2007; Manley et higher number of photons detected with an conventional light microscopy, will increase the al., 2008), the spatial constraint on molecular inorganic dye compared with that when using a number of color choices. proximity has made live-cell super-resolution photoactivatable fluorophore suggest that imaging challenging. However, large fields inorganic dyes provide a far greater localization Other dimensions (28 m ´ 28 m) with a single-molecule precision – approaching 5 nm FWHM (Shtengel Super-resolution microscopy began by localization precision of 20 nm can be obtained et al., 2009). However, in biological improving resolution in the x and y dimensions, every 25–60 seconds, but the images have a applications, inorganic dyes are often coupled but biological structures are 3D. Although the

Journal of Cell Science spatial resolution of ~60 nm (Shroff et al., 2008). to antibodies whose size (>10 nm) is two- to simplest way to generate 3D super-resolution Spatial resolution is normally less than threefold larger than the 3–4 nm of genetically images is to combine the serial sectioning of localization precision, not only because of the expressed fluorescent proteins. The length of the tissue with standard lateral super-resolution number of fluorophores localized along a given antibody adds uncertainty to the location of techniques (Punge et al., 2008), many super- feature, but also because the structures in living the centroid and decreases the higher precision resolution techniques have now been extended cells translocate, gain molecules and lose that is inherent to inorganic dyes. However, new to 3D. In 3D STORM a cylindrical lens is simply molecules during image acquisition. Hence, labeling strategies are continuously emerging added in the light path to create astigmatic live-cell single-molecule super-resolution with the use of Halo-, SNAP- and CLIP-tag imaging so that the ellipticity of the PSF requires that sampling is fast enough – both protein labeling systems that enable specific, becomes a sensitive measure of its distance from temporally and spatially – to avoid blurring and covalent attachment of virtually any molecule the focal plane, and yields a resolution of 20– to quantify dynamics. It also requires that to a protein of interest and have already been 30 nm laterally and 50–60 nm axially (Huang et control experiments that estimate the number of used in PALM (Lee et al., 2010), STED (Hein et al., 2008). Isotropic STED (isoSTED) is a more molecules in a structure are used as a guide for al., 2010) and STORM (Dellagiacoma et al., complicated hardware implementation of STED parsing the correct number of single-molecule 2010). These new technologies promise the that uses opposing objectives to create a hollow frames into single-molecule super-resolution possibility of continually increasing localization sphere of light, which replaces the more- time series images. precision. cylindrical doughnut-shaped STED, yielding As super resolution frequently relies on 30 nm isotropic resolution (Schmidt et al., Super-resolution probes photophysics, the task of selecting multiple 2009). To date, the greatest axial improvement Like all fluorescent techniques, the key to the labels requires more consideration than simply has been achieved by using opposing objectives super-resolution techniques lies in the choice of choosing spectral separation. For example, to combine axial interferometry with 2D PALM probes. SIM is the closest to traditional multi-color SIM is best achieved with two dyes in a technique termed iPALM (Shtengel et al., fluorescence microscopy, requiring no special that have similar excitation spectra but a large 2009). iPALM demonstrates 20 nm lateral probes but, as multiple intermediate images are enough shift between their emission maxima to resolution and 10 nm axial resolution (Shtengel collected, photobleaching must be considered. spectrally separate the dyes. Similarly, to avoid et al., 2009). Previously, structures of this size The other super-resolution techniques require using pairs of either excitation or depletion could only be accurately visualized by electron fluorescent probes whose state can be lasers, a pulsed-supercontinuum laser was microscopy (EM), but now the EM resolution controlled. The probes need to be either recently used with two-color STED for both the can be combined with the molecular specificity 1610 Journal of Cell Science 124 (10)

challenging, and single-molecule techniques now need to localize millions of molecules to have adequate spatial resolution in a single Resolution 250 nm image. Each of these considerations represents DIFFRACTION LIMITED an imaging or a biological trade-off that must be evaluated for each experiment (Fig. 1). Finally, it will be crucial to develop ways to put super- resolution images into context: conventional light or electron microscopy can provide an important background for interpreting the new Standard Photocontrollable level of ultrastructure that can be visualized fluorophore fluorophore (Watanabe et al., 2011). Super-resolution technologies will continue to evolve, and the exciting commercialization of STED, SIM, Reversible PALM, STORM and dSTORM microscopy photoswitcher Photoactivatable holds great promise for cell biology. SIM SUPER We apologize to the authors of papers we could not 120 nm RESOLUTION include due to space limitations. Partial support was Nikon, OMX, Zeiss provided by the intramural program of NINDS, NIH. STED Deposited in PMC for release after 12 months. 80 nm Leica An individual poster panel is available as a JPEG file online at Ensemble http://jcs.biologists.org/cgi/content/full/124/10/1607/DC1 measurements This article is part of an Article Series on Imaging available online at http://jcs.biologists.org/cgi/content/full/124/10/1607/DC2

Genetically Cy dye Alexa-Flour expressed pairs dyes References (f)PALM STORM dSTORM Bates, M., Huang, B., Dempsey, G. T. and Zhuang, X. (2007). Multicolor super-resolution imaging with photo- 25 nm Nikon, Zeiss switchable fluorescent probes. Science 317, 1749-1753. Single-molecule Betzig, E., Patterson, G. H., Sougrat, R., Lindwasser, O. W., Olenych, S., Bonifacino, J. S., Davidson, M. W., measurements Lippincott-Schwartz, J. and Hess, H. F. (2006). Imaging Speed intracellular fluorescent proteins at nanometer resolution. Science 313, 1642-1645. Dependent on Bretschneider, S., Eggeling, C. and Hell, S. W. (2007). desired field size Breaking the diffraction barrier in fluorescence microscopy and resolution by optical shelving. Phys. Rev. Lett. 98, 218103. Dellagiacoma, C., Lukinavicius, G., Bocchio, N., Banala,

Journal of Cell Science Fig. 1. Pathways to consider when choosing a super-resolution method. Super-resolution microscopy either visualizes S., Geissbuhler, S., Marki, I., Johnsson, K. and Lasser, ensembles or individual molecules. Within these categories the different technologies require different fluorescent probes. T. (2010). Targeted photoswitchable probe for nanoscopy of In general there is an inverse relationship between image acquisition speed and image resolution. The key to obtaining a biological structures. Chembiochem. 11, 1361-1363. super-resolution image is to understand the trade-offs between probes, speed and resolution. Each of the methods shown Ding, J. B., Takasaki, K. T. and Sabatini, B. L. (2009). is currently available from one or more commercial vendors, and more companies will undoubtedly enter the field. Supraresolution imaging in brain slices using stimulated- emission depletion two-photon laser scanning microscopy. Neuron 63, 429-437. Gelles, J., Schnapp, B. J. and Sheetz, M. P. (1988). and high copy number of genetically expressed achieved with this tool. Dynamic molecular Tracking kinesin-driven movements with nanometre-scale probes to truly visualize molecules within cells. structures can now be visualized at their correct precision. Nature 331, 450-453. Although these techniques are currently size, and resolution similar to electron Gustafsson, M. G. (2000). Surpassing the lateral resolution limit by a factor of two using structured illumination limited in the distance from the cover glass that tomography can be obtained with a molecular microscopy. J. Microsc. 198, 82-87. they can penetrate into the specimen, there have labeling density that satisfies the Gustafsson, M. G. (2005). Nonlinear structured- been several attempts to image deeper into Nyquist–Shannon criteria. However, there are illumination microscopy: wide-field fluorescence imaging with theoretically unlimited resolution. Proc. Natl. Acad. thicker specimens using two-photon excitation still challenges to be faced in embracing these Sci. USA 102, 13081-13086. in combination with PALM (Vaziri et al., 2008; new technologies. Image resolution often comes Heilemann, M., van de Linde, S., Schuttpelz, M., Kasper, R., Seefeldt, B., Mukherjee, A., Tinnefeld, P. and Sauer, York et al., 2011) and STED (Ding et al., 2009). at the price of acquisition speed. Living cells can M. (2008). Subdiffraction-resolution fluorescence imaging As biological structures are dynamic ensembles, potentially move molecules faster than super- with conventional fluorescent probes. Angew. Chem. Int. Ed. often located deep within tissues, the next resolution microscopes can capture images. Engl. 47, 6172-6176. Hein, B., Willig, K. I., Wurm, C. A., Westphal, V., challenges in 3D super-resolution microscopy Localization precision is frequently much Jakobs, S. and Hell, S. W. (2010). Stimulated emission are to adapt the ideas and technologies in order smaller than spatial image resolution. depletion nanoscopy of living cells using SNAP-tag fusion to penetrate further and maintain lateral and Resolution and localization precisions vary proteins. Biophys. J. 98, 158-163. Hell, S. W. (2003). Toward fluorescence nanoscopy. Nat. axial resolution when confronted with the light- between 10 and 100 nm, depending on the Biotechnol. 21, 1347-1355. scattering and absorption constraints imposed super-resolution technology. There is always a Hell, S. W. and Wichmann, J. (1994). Breaking the by animal tissue. need for more probes – the current color pallet is diffraction resolution limit by stimulated emission: stimulated-emission-depletion fluorescence microscopy. limited, which may restrict the pace at which Opt. Lett. 19, 780-782. Conclusion super resolution is adopted in biology. As super Hess, S. T., Girirajan, T. P. and Mason, M. D. (2006). The diffraction limit imposed by conventional resolution moves into 3D, the hardware for Ultra-high resolution imaging by fluorescence photoactivation localization microscopy. Biophys. J. 91, light microscopy no longer limits what can be ensemble techniques can become more 4258-4272. Journal of Cell Science 124 (10) 1611

Hess, S. T., Gould, T. J., Gudheti, M. V., Maas, S. A., Schermelleh, L., Carlton, P. M., Haase, S., Shao, L., thick biological samples. Proc. Natl. Acad. Sci. USA 105, Mills, K. D. and Zimmerberg, J. (2007). Dynamic Winoto, L., Kner, P., Burke, B., Cardoso, M. C., Agard, 20221-20226. clustered distribution of hemagglutinin resolved at 40 nm in D. A., Gustafsson, M. G. et al. (2008). Subdiffraction Watanabe, S., Punge, A., Hollopeter, G., Willig, K. I., living cell membranes discriminates between raft theories. multicolor imaging of the nuclear periphery with 3D Hobson, R. J., Davis, M. W., Hell, S. W. and Jorgensen, Proc. Natl. Acad. Sci. USA 104, 17370-17375. structured illumination microscopy. Science 320, 1332- E. M. (2011). Protein localization in electron micrographs Huang, B., Wang, W., Bates, M. and Zhuang, X. (2008). 1336. using fluorescence nanoscopy. Nat. Methods 8, 80-84. Three-dimensional super-resolution imaging by stochastic Schmidt, R., Wurm, C. A., Punge, A., Egner, A., Jakobs, Westphal, V. and Hell, S. W. (2005). Nanoscale resolution optical reconstruction microscopy. Science 319, 810-813. S. and Hell, S. W. (2009). Mitochondrial cristae revealed in the focal plane of an optical microscope. Phys. Rev. Lett. Kner, P., Chhun, B. B., Griffis, E. R., Winoto, L. and with focused light. Nano Lett. 9, 2508-2510. 94, 143903. Gustafsson, M. G. (2009). Super-resolution video Shannon, C. (1949). Communication in the presence of Westphal, V., Rizzoli, S. O., Lauterbach, M. A., Kamin, microscopy of live cells by structured illumination. Nat. noise. Proc. IRE 37, 10-21. D., Jahn, R. and Hell, S. W. (2008). Video-rate far-field Methods 6, 339-342. Shroff, H., Galbraith, C. G., Galbraith, J. A., White, H., optical nanoscopy dissects synaptic vesicle movement. Lee, H. L., Lord, S. J., Iwanaga, S., Zhan, K., Xie, H., Gillette, J., Olenych, S., Davidson, M. W. and Betzig, E. Science 320, 246-249. Williams, J. C., Wang, H., Bowman, G. R., Goley, E. D., (2007). Dual-color superresolution imaging of genetically Wildanger, D., Rittweger, E., Kastrup, L. and Hell, S. W. Shapiro, L. et al. (2010). Superresolution imaging of expressed probes within individual adhesion complexes. (2008). STED microscopy with a supercontinuum laser targeted proteins in fixed and living cells using Proc. Natl. Acad. Sci. USA 104, 20308-20313. source. Opt. Express 16, 9614-9621. photoactivatable organic fluorophores. J. Am. Chem. Soc. Shroff, H., Galbraith, C. G., Galbraith, J. A. and Betzig, Yildiz, A., Forkey, J. N., McKinney, S. A., Ha, T., 132, 15099-15101. E. (2008). Live-cell photoactivated localization microscopy Goldman, Y. E. and Selvin, P. R. (2003). Myosin V walks Manley, S., Gillette, J. M., Patterson, G. H., Shroff, H., of nanoscale adhesion dynamics. Nat. Methods 5, 417-423. hand-over-hand: single fluorophore imaging with 1.5-nm Hess, H. F., Betzig, E. and Lippincott-Schwartz, J. Shtengel, G., Galbraith, J. A., Galbraith, C. G., localization. Science 300, 2061-2065. (2008). High-density mapping of single-molecule Lippincott-Schwartz, J., Gillette, J. M., Manley, S., York, A. G., Ghitani, A., Vaziri, A., Davidson, M. W. and trajectories with photoactivated localization microscopy. Sougrat, R., Waterman, C. M., Kanchanawong, P., Shroff, H. (2011). Confined activation and subdiffractive Nat. Methods 5, 155-157. Davidson, M. W. et al. (2009). Interferometric fluorescent localization enables whole-cell PALM with genetically Nagerl, U. V., Willig, K. I., Hein, B., Hell, S. W. and super-resolution microscopy resolves 3D cellular expressed probes. Nat. Methods 8, 327-333. Bonhoeffer, T. (2008). Live-cell imaging of dendritic spines ultrastructure. Proc. Natl. Acad. Sci. USA 106, 3125-3130. by STED microscopy. Proc. Natl. Acad. Sci. USA 105, Subach, F. V., Patterson, G. H., Renz, M., Lippincott- 18982-18987. Schwartz, J. and Verkhusha, V. V. (2010). Bright Punge, A., Rizzoli, S. O., Jahn, R., Wildanger, J. D., monomeric photoactivatable red fluorescent protein for Cell Science at a Glance on the Web Meyer, L., Schonle, A., Kastrup, L. and Hell, S. W. two-color super-resolution sptPALM of live cells. J Am. (2008). 3D reconstruction of high-resolution STED Chem. Soc. 132, 6481-6491. Electronic copies of the poster insert are microscope images. Microsc. Res. Tech. 71, 644-650. Thompson, R. E., Larson, D. R. and Webb, W. W. (2002). available in the online version of this article Rust, M. J., Bates, M. and Zhuang, X. (2006). Sub- Precise nanometer localization analysis for individual at jcs.biologists.org. The JPEG images can diffraction-limit imaging by stochastic optical fluorescent probes. Biophys. J. 82, 2775-2783. be downloaded for printing or used as reconstruction microscopy (STORM). Nat. Methods 3, 793- Vaziri, A., Tang, J., Shroff, H. and Shank, C. V. (2008). slides. 795. Multilayer three-dimensional super resolution imaging of Journal of Cell Science