Advanced microscopy methods for visualizing chromatin structure

(1)

Review

Advanced microscopy methods for visualizing chromatin structure

Melike Lakadamyali

a,⇑

, Maria Pia Cosma

b,c,d,⇑

a

ICFO-Institut de Ciències Fotòniques, Mediterranean Technology Park, 08860 Barcelona, Spain

b

Centre for Genomic Regulation (CRG), Dr. Aiguader 88, 08003 Barcelona, Spain

c

Universitat Pompeu Fabra (UPF), 08003 Barcelona, Spain

d_{Institució Catalana de Recerca i Estudis Avançats (ICREA), Pg. Lluis Companys 23, 08010 Barcelona, Spain}

a r t i c l e

i n f o

Article history: Received 23 March 2015 Revised 30 March 2015 Accepted 1 April 2015 Available online 17 April 2015 Edited by Wilhelm Just Keywords:

Chromatin

Super resolution microscopy Nucleosome clutches

a b s t r a c t

In the recent years it has become clear that our genome is not randomly organized and its architec-ture is tightly linked to its function. While genomic studies have given much insight into genome organization, they mostly rely on averaging over large populations of cells, are not compatible with living cells and have limited resolution. For studying genome organization in single living cells, microscopy is indispensable. In addition, the visualization of biological structures helps to under-stand their function. Up to now, fluorescence microscopy has allowed us to probe the larger scale organization of chromosome territories in the micron length scales, however, the smaller length scales remained invisible due to the diffraction limited spatial resolution of fluorescence micro-scopy. Thanks to the advent of super-resolution microscopy methods, we are finally starting to be able to probe the nanoscale organization of chromatin in vivo and these methods have the potential to greatly advance our knowledge about chromatin structure and function relationship.

1. Introduction

Light microscopy is an enabling technology for biology. It allows non-invasive visualization of cellular and sub-cellular structures in multiple colors, in 3D and in living cells. The use of fluorescent probes such as fluorescent proteins ensures high contrast and high molecular specificity. However, the spatial resolution of light microscopy is inherently limited to a scale comparable to the wavelength of light due to diffraction. Therefore, it is not possible to resolve structures smaller than 200–300 nm using visible light and far field optical microscopy.

Chromatin is made of DNA wrapped around the nucleosomes, which are formed by the core histone proteins H2A, H2B, H3 and H4. Histone variants can also substitute some of the core histones. Nucleosomes are compacted together in the chromatin fiber and the histone H1 contributes to this packaging (for details on chro-matin structure please see[1]). Considering the small size and high compaction of chromatin inside the nucleus, the 200–300 nm level of spatial resolution achievable with far field optical micro-scopy is not sufficient for studying chromatin structure.

Nevertheless, light microscopy has played an important role in enhancing our understanding of chromatin assembly and dynam-ics. For example, by targeting inert fluorescent tracers of different sizes (such as fluorescent protein oligomers or fluorescently labeled dextrans) into the nucleus, it has been possible to monitor their dynamics using methods like fluorescence recovery after pho-tobleaching (FRAP), fluorescence correlation spectroscopy (FCS), fluorescence anisotropy and fluorescence photoactivation[2–4].

In addition, single particle tracking has also been used to follow the movement of ﬂuorescent tracers such as quantum dots[2]. The movement of these tracers mostly occurs via diffusion and can be hindered by the presence of obstacles created by highly compact chromatin structures and nuclear crowding. This type of studies was used to propose that the chromatin inside the nucleus may fol-low a fractal organization[2,5]since chromatin hinders the diffu-sion of inert tracers in a size-independent manner. It has also been shown that histone acetylation increases chromatin accessi-bility to ﬂuorescent tracers[3].

In addition to using inert tracers, the dynamics of chromatin-as-sociated proteins such as core and linker histones has also been monitored using methods like FRAP, FCS and ﬂuorescence aniso-tropy[6–10]. For example, it has been shown that the linker his-tone H1 is dynamic and exchanges between bound and unbound states both in heterochromatin and euchromatin, with a residence time of several minutes[7,8]. Core histones, on the other hand, are likely more stable and can reside on the chromatin for longer time periods[4]. Monitoring ﬁber associated proteins as well as DNA

http://dx.doi.org/10.1016/j.febslet.2015.04.012

manuscript.

⇑ Corresponding authors. Address: Centre for Genomic Regulation (CRG), Dr. Aiguader 88, 08003 Barcelona, Spain (M.P. Cosma).

E-mail addresses:[email protected](M. Lakadamyali),pia.cosma@crg. eu(M.P. Cosma).

(2)

can also give information on the global organization of chromatin. For example, it has been shown that mouse embryonic stem cells (mESCs) have a relatively spatially more uniform chromatin com-paction compared to differentiated cells, which show more hetero-geneity [9,11]. Likewise, the reprogramming of the somatic genome into pluripotency is associated with a removal of epige-netic barriers and open chromatin[12].

In parallel to these dynamic studies, fluorescence in situ hybridization (FISH) has been used to label and visualize specific DNA sequences. Using multiple fluorophores of different colors to simultaneously detect and visualize multiple targets (Chromosome Painting) it was shown that chromosomes are not randomly organized but occupy discrete territories in the cell nucleus (for a review see [13]). These methods, while valuable, provide only an indirect measure of chromatin organization inside the nucleus and their spatial resolution is limited by diffraction. In order to directly visualize chromatin structure, techniques with higher spatial resolution are needed, such as electron microscopy, electron spectroscopic imaging and small angle X-ray diffraction

[14–17]. However, these methods are invasive, not applicable in living cells and often lack contrast and molecular speciﬁcity. 2. Super-resolution methods

With the advent of super-resolution microscopy, diffraction limit is no longer an impenetrable barrier to light microscopy. In the last decade, several techniques have been developed that over-come the diffraction limit, extending the spatial resolution of light microscopy to length scales as small as 10–20 nm, an order of mag-nitude improvement over conventional ﬂuorescence microscopy. Super-resolution microscopy methods can be broadly divided into two categories: those that are based on patterning the illumination light, such as Saturated Structured Illumination Microscopy – (S)SIM[18,19]and Stimulated Emission Depletion (STED)[20]or those that are based on single molecule detection and localization, such as Stochastic Optical Reconstruction Microscopy – STORM

[21] and (Fluorescence) Photoactivation Localization Microscopy – PALM and fPALM [22,23]. Most of these methods have been applied to image chromatin and nuclear organization at high reso-lution[24–39]. Below, we summarize the concept behind each one of these methods as well as their strengths and weaknesses and overview how each method has been used to visualize nuclear organization.

2.1. Structured Illumination Microscopy (SIM)

SIM takes advantage of patterned illumination light to improve the spatial resolution of light microscopy[18,19]. Typically a sinu-soidal pattern of bright and dark stripes is obtained by interfering two counter-propagating excitation beams and this patterned light is used for exciting the sample. The sample will have a distribution of ﬂuorophores corresponding to the sample features that are labeled. In the frequency space, the small sample features that are below the diffraction limit correspond to high frequencies, which cannot be imaged with conventional optics. The patterned light multiplies with the labeled ﬂuorophore distribution (i.e. sam-ple features) and this multiplication leads to the appearance of moiré fringes. These moiré fringes have lower frequency than the original sample features and therefore they can be imaged by conventional optics. Knowledge of the original illumination pattern can then be used to solve the unknown sample features from the moiré fringes (Fig. 1A)[18].

Conventional SIM leads to only a 2-fold improvement in spatial resolution, since the patterned light itself is diffraction limited

[18,40]. It is possible to improve the spatial resolution further with

the help of saturation of fluorescence intensity using high laser powers, however, to date, saturated SIM has not been demon-strated in biological samples. SIM can also be extended to the third dimension by using three interfering beams to generate the patterned illumination along the z-axis in addition to the lateral (x, y) axis[31]. Despite its modest improvement in spatial resolu-tion, SIM is excellent for live-cell imaging applications since it pro-vides high temporal resolution. The low illumination intensity used in SIM also minimizes phototoxicity and its compatibility with a wide range of fluorescent probes makes it easy to use genet-ically encoded fluorescent proteins for labeling intracellular structures. 2.2. Stimulated Emission Depletion Microscopy (STED) and Reversible Saturable Optical Fluorescence Transition (RESOLFT)

STED uses a focused laser beam to excite fluorophores within a diffraction-limited volume. Once excited, a subset of the fluo-rophores is switched off by forcing them to return back to the ground state through stimulated emission by using a depletion beam (STED beam). Importantly, the depletion beam is shaped to resemble a doughnut, which leads to an effective reduction in the excitation area to a sub-diffraction region in the center of the doughnut (Fig. 1B)[20]. To acquire a high-resolution image of the sample, the excitation and the depletion beam are then scanned point by point across the sample. The spatial resolution of the final image depends on many parameters, but most impor-tantly on the intensity of the STED beam[20], and spatial resolu-tion in the order of 30–40 nm is typical in biological samples. The compatibility of STED with a wide range of fluorescent probes leads to flexibility in labeling. However, in practice, relatively high laser powers are required for the depletion beam. Therefore, pho-tobleaching and phototoxicity can become a potential problem and limit the overall choice of fluorophores to those that are bright and photostable (e.g. organic fluorophores such as Atto dyes).

Stimulated emission is not the only saturable optical transition that can be used for super-resolution imaging with focused light. The STED concept has been extended to other optical transitions with the use of photoswitchable fluorescent proteins, for example, and this more general approach is referred to as reversible sat-urable optical fluorescence transition (RESOLFT) microscopy[41]. In RESOLFT, fluorescent probes are switched off from long-lived states compared to the short excited lifetime used in STED. Therefore, the intensity requirement for the depletion beam is much lower compared to STED[41]. Recently, fatigue resistant flu-orescent proteins have been developed that can switch between bright and dark states over thousands of cycles[42,43]. These flu-orescent proteins work particularly well for RESOLFT requiring very low light intensities, albeit providing lower spatiotemporal resolution compared to STED. STED and RESOLFT can both be extended to 3D imaging and the axial resolution can be improved by creating a doughnut illumination also along the z-axis [44]. While it is possible to achieve isotropic resolution in all three dimensions with this approach, 3D STED is technically more chal-lenging to implement than 2D STED.

2.3. Single Molecule Localization Microscopy (SMLM)

Methods such as STORM and PALM[21–23]take advantage of single molecule detection and localization to break the diffraction limit. We will refer to these methods as Single Molecule Localization Microscopy (SMLM) since they share the same basic concept. The position of a single ﬂuorescent probe can be localized with very high precision (nanometer) determined mainly by the number of photons collected from that probe[45,46]. However, this concept by itself is not sufﬁcient to improve image resolution,

(3)

since biological samples are densely labeled with a high number of fluorescent probes whose images overlap and cannot be distin-guished. Therefore, in a densely labeled biological sample, it is nec-essary to separate the emission of many overlapping fluorescent probes over time. This separation can be achieved with the help of photoswitchable fluorescent probes[47,48]. These are probes whose fluorescence can be switched between bright and dark states using light illumination. At any given time, most of the flu-oropores are in a long-lived dark state[49]. Only a small subset of these fluorophores is stochastically activated into the fluorescent state by excitation with a given wavelength of light (often UV illu-mination)[49]. As a result, the single molecule images of this small subset of molecules do not overlap and their positions can thus be determined precisely. By repeating the process of activation, imag-ing and de-activation for several cycles, a super-resolution image can be reconstructed from molecule positions (Fig. 1C)[49].

Localization based methods, therefore, require the use of fluo-rescent probes that can be photoswitched between bright and dark states. Today, there is a wide choice of these fluorescent probes, ranging from photoactivatable, photoconvertible or photoswitch-able fluorescent proteins to photoswitchphotoswitch-able organic fluorophores

[50,51]. Fluorescent proteins are typically dimmer than small organic ﬂuorophores and lead to lower spatial resolution since the localization precision depends on the number of photons col-lected. Depending on the ﬂuorescent probe used, it is possible to achieve a spatial resolution in the order of 20–40 nm and a tempo-ral resolution of 0.5–10 s (in live cell imaging)[52]. In addition, it is also possible to improve the axial resolution. A typical way to extend SMLM imaging to the third dimension is by placing cylin-drical lens in the imaging path[53]. Due to astigmatism, molecules that are above and below the focal plane will appear elliptical and

the ellipticity will be either in the horizontal or vertical directions. With proper calibration, the ellipticity of each molecule can be mapped to the molecule’s z-position. This method can achieve an axial resolution of 50 nm over a range of 800 nm (400 nm above and below the focal plane)[53]. It is also possible to combine astigmatism with a dual-objective geometry in order to capture more photons and improve the z-resolution to about 20 nm – however, at the expense of imaging depth [54]. Astigmatism belongs to a subset of methods referred to as point spread function (PSF) engineering. Instead of engineering the PSF to be elliptical, it is also possible to generate other PSF types, such as a double helix

[55]. In this case, the pitch of the double helix changes based on the molecule’s distance from the focal plane. Finally, an interferomet-ric method, in which photons from a single molecule collected through two opposing objectives are allowed to interfere (iPALM), has also been used for 3D super-resolution imaging, pro-viding an impressive axial resolution of 10 nm but at the expense of imaging depth and technical simplicity[56].

3. Fluorescent labeling for visualizing the chromatin structure at high resolution

Sample preparation and labeling are in general highly impor-tant in ﬂuorescence microscopy but super-resolution microscopy puts further demands on the type of labeling required. The spatial resolution is directly linked to the labeling density through the Nyquist criterion [57,58]. The labeling density must be high enough such that the average distance between individual labels is half the desired spatial resolution (Fig. 2). In addition, while common ﬂuorescent probes are readily useable for SIM, other super-resolution methods such as STED and SMLM have more

Unknown sample features Patterned illumination Moiré fringes

A

SIM

_B

STED

+ =

Excitation beam STED beam Effective excitation beam

C

SMLM Diffraction limited image Reconstructed high-resolution image switch on localize switch off

Fig. 1. Working principle of the different super-resolution methods. (A) Structured Illumination Microscopy (SIM). The fine spatial details in a sample appear as higher frequency information in the Fourier space. If two fine patterns, one belonging to the sample features and one belonging to the patterned illumination, are superimposed, the two patterns multiply creating what is known as moiré fringes. These moiré fringes are much coarser (lower frequency) than the two patterns and will be observable with conventional optics. Since the details of the illumination pattern are known, the details of the sample features can be back calculated from the resulting moiré fringes. (B) Stimulated Emission Depletion Microscopy (STED). A focused excitation beam is superimposed with a doughnut shaped depletion beam (STED beam). The STED beam switches fluorophores off, and the fluorescence signal only comes from a small confined region in the center of the doughnut, therefore reducing the effective illumination spot. (C) Single Molecule Localization Microscopy (SMLM). Fluorophores are iteratively switched on, their positions are identified by finding the center position of their images and finally they are switched off. Repeated on-off switching and localization allows reconstruction of the underlying structure at high resolution.

(4)

specific requirements on the fluorescence properties of the probes. In the case of STED, bright, photostable fluorescent probes that have emission spectra matching the depletion laser wavelength are needed. SMLM requires the use of bright photoswitchable flu-orescent probes that have low duty cycle (i.e. spend most of their time in the dark state) [59]. These demands restrict the overall choice of fluorescent probes to only a few options, making multi-color super-resolution imaging more challenging[60].

3.1. Labeling of chromatin associated proteins

To visualize chromatin, one can either label chromatin-associ-ated proteins such as the core histone proteins or the DNA itself. Chromatin associated proteins can be tagged with the help of genetically encoded fluorescent proteins and this type of labeling provides the easiest option for visualizing chromatin especially in living cells. However, transient overexpression of tagged proteins may lead to a number of artefacts such as altered function and mis-localization of the tagged proteins. Importantly, this strategy is not quantitative since it does not allow visualizing the endogenous population of proteins. With the advent of gene editing technolo-gies, such as zinc fingers[61], TALENs (transcription activator-like effector nuclease)[61]and CRISPR/Cas9 (clustered regularly inter-spaced short palindromic repeats coupled with the endonuclease Cas9)[62], it is now easier to genetically label proteins by modify-ing their endogenous gene locus. One difficulty is the fact that flu-orescent proteins are dim and less photostable compared to small organic fluorophores, making them less desirable in particular for use in super-resolution microscopy.

It is also possible to label chromatin-associated proteins using small organic fluorophores in conjunction with antibodies. While this strategy allows the visualization of endogenous proteins, it is not compatible with imaging in living cells. A promising approach is to combine genetically encoded tags such as SNAP-, CLIP- and HALO-tags with small organic fluorophore labeling [63–66]. These tags, which can be fused to the protein of interest, form a covalent link with a fluorophore-labeled small peptide. In particu-lar, the recent development of membrane permeable, bright and photoswitchable small organic fluorophores[67]should make this approach broadly applicable to image chromatin associated pro-teins in living or fixed cells with super-resolution.

3.2. Labeling of DNA

Imaging the DNA itself at high resolution is more challenging than imaging chromatin-associated proteins. There are a range of DNA speciﬁc labels such as intercalating dyes (SYBR Green, TOTO-1, YOYO-1, SYTO), minor or major groove binding dyes (PicoGreen), most of which have been shown to be photoswichable and compatible with localization based super-resolution methods

[25,68]. However, most of these dyes are dim, not very photostable, and their photoswitching properties (e.g. duty cycle) are not opti-mal, degrading the spatial resolution that can be achieved. Furthermore, they are often not penetrant into living cells. An alternative approach is the use of thymidine analog, 5-ethynyl-20_{-deoxyuridine (EdU) or deoxycytodine analog, 5-ethyl-2}0 -deoxy-cytidine (EdC). These artificial nucleotides integrate into the DNA sequence during replication and a fluorophore of choice can be incorporated using click chemistry[38]. This approach allows the use of best super-resolution compatible fluorophores to label DNA globally but is more difficult to use in living cells. A recent exciting development in labeling specific gene loci is the use of TALEs, (DNA binding proteins without endonuclease activity) or the CRISPR/dCas9 (deactivated Cas9 without the endonuclease activity) fused with fluorescent proteins. While the TALEs have been shown to be efficient in labeling repetitive sequences[69], CRISPR/dCas9 has also been used for labeling non-repetitive sequences[70]. Both of these labeling approaches are compatible with super-resolution microscopy (STED and SMLM) depending on the fluorescent protein used. In addition, the fluorescence signal can potentially be increased via the use of a recently developed SunTag[71]. The SunTag system consists of a polypeptide fused to a protein of interest, which is recognized by a single-chain vari-able fragment (scFv) antibody. The scFv antibody is fused to a flu-orescent protein, therefore recruiting multiple copies of the fluorescent protein to the target via the polypeptide chain. This system can potentially be used very efficiently to label DNA in liv-ing cells.

Overall, with the development of new ﬂuorophores and differ-ent approaches for targeting these ﬂuorophores to the chromatin, super-resolution imaging is already starting to give new insights into the chromatin structure in different cells.

4. Imaging nuclear organization at high resolution

The nucleosomes are the smallest repeating unit of the chro-matin ﬁbers and have dimension of 11 nm, therefore well below the resolution limit of conventional ﬂuorescent microscopy. For this reason, super resolution microscopy techniques are extremely useful to study chromatin structure. Here, we will summarize some relevant examples where different super resolution micro-scopy approaches have been used.

4.1. Imaging nuclear structures with SIM

SIM was one of the first super-resolution methods to be used to visualize nuclear structures. Using a custom built 3D SIM micro-scope, which allowed resolution up to 100 nm, chromatin struc-tures on standard cytological preparations were visualized, i.e. using DAPI to stain DNA and antibodies conjugated with conven-tional dyes to label proteins of interest[31]. By this approach, a remarkable visualization of the chromatin, of the nuclear pore complexes (NPC) and of the nuclear lamina could already be achieved at that early time. Fixed mouse C2C12 myoblast cells were imaged after being stained with DAPI. The first interesting observation was that the DAPI signal was not homogenous (as nor-mally appeared before in wide-field conventional imaging) but

Increasing Label Density

Fig. 2. The impact of label density on spatial resolution. When the label density is low, only a few fluorophores will be localized leading to a sparse reconstructed image that does not reflect the underlying structure. As the label density and hence the localization density increases, the reconstruction will more faithfully capture the underlying structure. According to the Nyquist criterion to reach a given spatial resolution, the distance between individual labels (or localizations, the crosses in the figure) should be at least half the desired spatial resolution.

(5)

instead presented many well-deﬁned holes. These holes were basi-cally the regions in which DNA was excluded from the NPC. By imaging the nuclear lamina it was clear that the peripheral hete-rochromatin was surrounded by the lamina, which in turn was covered by nuclear pore complexes. The width of 3D-SIM-recorded NPC was 120 ± 3 nm. Furthermore, with 3D-SIM invaginations of the lamina structures could also be visualized. This work was one of the ﬁrst studies reporting precision of measurements of chromatin components at high resolution[31].

3D-SIM was also used to further study nuclear compartmental-ization and nuclear function. The distribution of replicating DNA, nascent RNA, transcribing RNA Polymerase II, histone modiﬁca-tions associated with active chromatin (such as H3K4me3 and H4K8Ac) was investigated within the nuclear space of murine mammary tumor C127 interphase nuclei by 3D-SIM and to some extent by a version of SMLM (Spectral Precision Distance Microscopy, SPDM) using the stochastic blinking of conventional ﬂuorophores like GFP. 3D-SIM showed an interchromatin compart-ment made of a network of channels and wider lacunas, which was separated from the compact chromatin domains by the perichro-matin region constituted of decondensed chroperichro-matin. The inter-chromatin compartment was considered to serve as storage of proteins, splicing speckles and factors necessary for the function of the perichromatin region. RNA polymerase II was described to form clusters in this study with some indication about their dimension. However, quantitative analysis of these results might provide in the future more mechanistic information linking func-tion with nuclear compartmentalizafunc-tion[28].

To translate these studies into the imaging of a concrete para-digm of a non-transcribed chromatin region, the organization of the Barr body, i.e. the inactive X chromosome, was visualized by 3D-SIM in human and mouse somatic cells[33]. In this work, a novel arrangement of the chromatin was postulated. A new chro-matin compartment called active nuclear compartment formed by the interchromatin and by perichromatin was found to form the Barr body. This active nuclear compartment was collapsed in the Barr body and it excluded RNA polymerase II. Xist RNA foci were enriched at the boundaries of the collapsed active nuclear compartments. Time course imaging also revealed the formation of Barr body during differentiation of undifferentiated female Embryonic Stem Cells (ESCs). Xist RNA were initially found in close association with RNA Polymerase II in decondensed chromatin. Progressively until day 9 of differentiation, upon spreading of Xist, RNA Polymerase II was excluded from Xist territories when ﬁnally the Barr body emerged[33].

4.2. SMLM and the visualization of the chromatin ﬁbers

Besides 3D-SIM, SMLM based super resolution microscopy methods have also recently been used to visualize nuclear organi-zation. For example, stochastic light switching of Green Fluorescent Protein (GFP) (SPDM) was used to analyze distribution of H2B in HeLa and human fibroblasts[24]. This study has been one of the first attempts to image histones at high resolution within intact nuclei. In both cell lines histones showed a nuclear distribu-tion significantly different from random on a scale of <100 nm. However, remarkable different distribution patterns were noted in this study depending on whether H2B fused to GFP was overex-pressed in HeLa cells from a plasmid or it was stably integrated. The authors did not discuss the biophysical differences between the two different GFPs used, Emerald Green Fluorescent Protein (emGFP), overexpressed as H2B-emGFP fusion protein versus GFP, stably integrated as H2B-GFP. emGFP is a GFP variant with improved photostability and brightness, thus different photo-bleaching properties are likely associated to the two proteins. Density fluctuations of localized histone proteins were also

analyzed and found to be in the range of >1–2

l

m, which was related to differences in large-scale chromatin compaction[24].

The use of EGFP instead of phoswitchable proteins for super res-olution microscopy was also shown by Matsuda and co-workers

[72]. Here, the authors demonstrated that EGFP can be efficiently converted from green to red when exposed to blue light and in the presence of riboflavin. A specific medium, called RiMOS, con-taining riboflavin, methionine, and an oxygen scavenger was opti-mized to favor efficient photoconversion. Using this medium, denoising/deconvolution conditions and EGFP fused to H2AvD (an abundant H2A variant), super-resolution reconstruction of pro-metaphase/metaphase chromosomes in Drosophila embryos was achieved. Clearly resolved clusters of localizations called ‘‘fibrous structures’’ were evident and shown to have cross-section diame-ter of 70 nm.

Super-resolution imaging of histones and other nuclear struc-tures in living cells is certainly one of the most interesting direc-tions to take in the next coming years, although it is very challenging to achieve both high spatial and high temporal resolu-tion while minimizing phototoxicity of cells to the laser lights. Nonetheless, live cell imaging of histone proteins at nanoscale res-olution has been demonstrated at the proof of concept level[66]. For this, living HeLa cells overexpressing H2B fused with Escherichia coli dihydrofolate reductase (H2B-eDHFR) were incu-bated with a specific chemical tag, trimethoprim (TMP) that has high affinity for eDHFR. TMP was linked to the organic fluorophore ATTO655, which has properties compatible with SMLM. Analysis of HeLa nuclei showed that clusters of localizations were resolved and these clusters were separated by distances of 100 nm. Here the authors describe their results by indicating that they could dis-criminate adjacent ‘‘histone core proteins’’. However, to extrapo-late how many core nucleosomes were present for each cluster was not possible without a quantitative analysis. By reconstructing super-resolved images from subsets of raw data frames using a sliding window algorithm, videos of movements of nucleosome clusters were also recorded. The histone clusters moved at 3 nm s1speed in the interphase nuclei[66]. This work repre-sents a pioneer study for the imaging of histones in living cells. However, the significance and relevance of the extracted parame-ters for chromatin organization and dynamics are not discussed in detail, but rather the emphasis is placed in the use of small tai-lored organic fluorophores for their use in live cell super resolution imaging.

Although an impressive level of resolution in the visualization of chromatin structures was reached in the above studies, the visu-alization of nucleosomes was not quantitative. Very recently, endogenous H2B was imaged by SMLM in a large collection of fixed and living cells [30]. The chromatin fibers were visualized at nanoscale resolution in human fibroblasts (hFbs) before and after transcriptional activation, in human induced pluripotent stem cells (hiPSCs), in mouse embryonic stem cells (ESCs), in a number of dif-ferent ESC mutants and in mouse neural precursor cells (mNPCs). Importantly, the nucleosomes across the fiber were quantified after building a SMLM-based calibration curve, which allowed relating the single molecule localization numbers to the number of nucleosomes. Using this kind of quantitative analysis, a new model of chromatin assembly was identified. Nucleosomes arranged in heterogenous groups, which the authors named ‘‘nucleosome clutches’’ which were separated by nucleosome-depleted regions (Fig. 3). The median number of nucleosomes per clutch and their density was cell specific. Pluripotent cells, such as mESCs and hiPSCs and transcriptionally activated hFbs had less dense clutches containing a lower number of nucleosomes with respect to somatic cells, such as mNPCs and hFbs. Furthermore, distribution of RNA Polymerase II and of histone H1 was also studied. RNA Polymerase II associated with the smallest clutches,

(6)

which formed the euchromatin, while H1 was enriched in the lar-ger ones, which were recognized to be the closed heterochromatin. Finally, computer simulations using parameters extracted from experimental data allowed to estimate the nucleosome occupancy of the DNA fiber, which was 60% in hFbs and 45% in transcription-ally active Fbs[30]. In this work, the ability to visualize and count the nucleosome number across the chromatin fiber allowed the depiction of a novel model of chromatin assembly and the correla-tion between structure of the fiber and cellular phenotype.

In addition to histones, SMLM was also used to image DNA in living cells. U2OS cells were labeled with the cyanine-based pico-green dye and imaging was performed with an optimized live cell imaging medium containing ascorbic acid and an oxygen scaveng-ing system[39]. This medium promoted efficient reversible photo-switching of the fluorophores and enabled a resolution of 70 nm. Time-lapse STORM video performed by taking frames each 10 min to allow picogreen fluorophores to recover from dark states showed local rearrangement of the DNA fibers[39]. Although pico-green has several advantages over other DNA dyes, such as its min-imal perturbation of the DNA structure and the possibility to be used in living cells, it is however relatively dim leading to lower resolution. Therefore, it is not possible to visualize single DNA fibers, which is highly desirable.

Super resolution imaging of DNA has been also performed in HeLa cells grown in a medium containing 5-ethynyl-20 -deoxyuri-dine (EdU), which is incorporated in the DNA in the place of thymi-dine[38]. The cells were fixed and subjected to click chemistry reaction of the EdU with an azide-group labeled fluorophore Alexa Fluor 647. Super resolved maps of DNA fibers were evident at an unprecedented resolution. High structural heterogeneity within different areas of the nucleus was shown. To image single fibers of DNA, pulse labeling of the nascent DNA was carried out and filamentous structures of an average length of 0.6

l

m were reveled[38]. EdU-based labeling and click chemistry could reach a very high resolution of DNA imaging, much higher than the one obtained with DNA dyes. However, although theoretically possible, click chemistry has not been used for living cells, which constitutes a limitation for this method.

Finally DNA within chromatin spreads isolated from interphase HeLa cells was visualized with a SMLM based method (Binding Activated Localization Microscopy, BALM) that uses the binding/ unbinding of YOYO-1 dye instead of photoswitching to achieve

the sparse labeling and imaging of single ﬂuorophores over time

[32]. Chromatin ﬁber of width ranging from 80 to 150 ± 40 nm were visualized when the chromatin spreads were prepared from cells cultured without or with serum, respectively. This difference in width was believed to be due to the different transcriptional states, active versus quiescent, of the cells. RNA Polymerase II was also imaged and found to be enriched in gap structures across the DNA spreads, which were considered to be the site of active transcription[37].

Besides chromatin, SMLM has also been used to visualize other nuclear structures, such as nuclear pore complexes, NPCs. SMLM provides a spatial resolution of 10–30 nm, which is comparable to the size of a NPC and does not allow visualizing the position of individual subunits with respect to one another within the NPC. To overcome this problem, Szynborska et al. combined many images of individual NPC and using single particle averaging, reached an impressive resolution of 1 nm[35]. This type of particle averaging in combination with SMLM was also previously shown to improve the spatial resolution and allow visualization of the eightfold symmetry of the NPC[27]. Visualizing the orientation of different NPC subunits allowed Szynborska et al. to propose ahead-to-tail arrangement of the Nup107–160 complex along the circumference of the pore in the cytoplasmic and nucleoplasmic rings[35]. The level of resolution reached in this work was possible thanks to the regular and symmetric structure of the NPC. 4.3. The bacterial chromosome

In the recent years, super-resolution microscopy has also played an important role in the visualization of various proteins and sub-structures in bacteria, including the bacterial chromosome binding proteins (nucleoid-associated proteins, NAPs) and the bacterial DNA. Thanks to these super-resolution studies, our view of the ‘‘unstructured’’ bacterial DNA is being continuously revised and modiﬁed. In particular, two studies have pointed towards a high level of organization of the bacterial genome.

Using the SMLM approach, Wang et al. studied the distribution of various fluorescent protein (mEos2) tagged NAPs inside the E. coli nucleoids[36]. They found that, while some NAPs such as HU appeared homogenously scattered in the nucleoids, H-NS, which is a global transcription silencer, formed compact clusters. The genes regulated by H-NS were sequestered inside these Fig. 3. Comparison of conventional fluorescence image (upper) and super-resolution (STORM) image (lower) of core histone H2B showing the arrangement of nucleosomes in a fibroblast nucleus. A zoom of the region inside the yellow rectangle is shown to the right. A cluster identification algorithm is used to automatically group localizations into clusters (colored crosses) and segment nucleosome clutches in the super resolution image.

(7)

clusters. They concluded that this clustering might create substan-tial folding and condensation of the bacterial DNA and might shape the 3D architecture of the bacterial chromosome.

The EdU labeling and subsequent click chemistry approach, which was shown to be very efﬁcient for labeling the eukaryotic DNA[29,38], was also demonstrated to be useful for labeling the DNA of E. coli[34]. Once again, combining this labeling strategy with SMLM, it was shown that the bacterial DNA is organized into cell-cycle dependent hetero-structures, including helices, horse-shoe structures and square structures [34]. These structures seemed to be attached to the bacterial membrane. The existence of these ordered structures can be taken as evidence for substantial spatiotemporal organization of the bacterial DNA at various stages of the cell cycle.

Similar to the eukaryotic genome, super-resolution methods will likely play an important role in the near future for unravelling the mechanisms of bacterial chromosome organization.

5. Conclusions and perspectives

The visualization of chromatin components, such as different histone types, histone variants, DNA, RNA polymerase II, factors belonging to the transcription machinery and RNAs at nanoscale resolution in single living cells will clearly provide us with much more understanding about how genes are activated or repressed. We are still in very early days, however, the advent of the SMLM has already made the visualization of these components at unprecedented resolution possible. We need to develop more quantitative methods, which can allow us to build predictive mod-els about gene function. What we have learnt from the genomic era is already quite advanced; instead the discoveries generated from the use of super-resolution microscopy methods are just in their infancy. As these methods and their application to chromatin visu-alization evolve, we predict that the emerging information will complement and enhance the genomic data. There are challenges ahead, such as improving the temporal resolution without induc-ing phototoxicity to use these methods in livinduc-ing cells, correlatinduc-ing information at the dynamic and super-resolution level [73,74], improving throughput to acquire large sets of data and developing ways to extract quantitative information from the images [75]. Despite these challenges, our prediction is that, in the near future, by literally viewing how chromatin ﬁbers change and how genes are activated/repressed we can better understand a number of pro-cesses associated to cellular physiology and dysfunctions. Acknowledgments

We thank Maria Aurelia Ricci for help with the ﬁgures. We are grateful for support from an European Research Council Grant (242630-RERE) (to M.P.C.) and (337191-MOTORS) (to M.L.), the Ministerio de Ciencia e Inovacion (SAF2011-28580) (to M.P.C.), Fundacio’ La Marato’ de TV3 (to M.P.C.), the AXA Research Fund (to M.P.C.), System’s Microscopy Network of Excellence consortium (call identiﬁer number FP-7-HEALTH.2010.2.1.2.2) to M.L., Spanish Ministry of Economy and Competitiveness (MINECO) and the ‘‘Fondo Europeo de Desarrollo Regional’’ (FEDER) through grant FIS2012-37753 (to M.L.).

References

[1] Kornberg, R.D. and Lorch, Y. (1999) Twenty-ﬁve years of the nucleosome, fundamental particle of the eukaryote chromosome. Cell 98, 285–294. [2] Bancaud, A., Huet, S., Daigle, N., Mozziconacci, J., Beaudouin, J. and Ellenberg, J.

(2009) Molecular crowding affects diffusion and binding of nuclear proteins in heterochromatin and reveals the fractal organization of chromatin. EMBO J. 28, 3785–3798.

[3] Gorisch, S.M., Wachsmuth, M., Toth, K.F., Lichter, P. and Rippe, K. (2005) Histone acetylation increases chromatin accessibility. J. Cell Sci. 118, 5825– 5834.

[4] Misteli, T. (2001) Protein dynamics: implications for nuclear architecture and gene expression. Science 291, 843–847.

[5] Bancaud, A., Lavelle, C., Huet, S. and Ellenberg, J. (2012) A fractal model for nuclear organization: current evidence and biological implications. Nucleic Acids Res. 40, 8783–8792.

[6] Banerjee, B., Bhattacharya, D. and Shivashankar, G.V. (2006) Chromatin structure exhibits spatio-temporal heterogeneity within the cell nucleus. Biophys. J. 91, 2297–2303.

[7] Lever, M.A., Th’ng, J.P., Sun, X. and Hendzel, M.J. (2000) Rapid exchange of histone H1.1 on chromatin in living human cells. Nature 408, 873–876. [8] Misteli, T., Gunjan, A., Hock, R., Bustin, M. and Brown, D.T. (2000) Dynamic

binding of histone H1 to chromatin in living cells. Nature 408, 877–881. [9] Talwar, S., Kumar, A., Rao, M., Menon, G.I. and Shivashankar, G.V. (2013)

Correlated spatio-temporal ﬂuctuations in chromatin compaction states characterize stem cells. Biophys. J. 104, 553–564.

[10] Weidemann, T., Wachsmuth, M., Knoch, T.A., Muller, G., Waldeck, W. and Langowski, J. (2003) Counting nucleosomes in living cells with a combination of ﬂuorescence correlation spectroscopy and confocal imaging. J. Mol. Biol. 334, 229–240.

[11] Meshorer, E., Yellajoshula, D., George, E., Scambler, P.J., Brown, D.T. and Misteli, T. (2006) Hyperdynamic plasticity of chromatin proteins in pluripotent embryonic stem cells. Dev. Cell 10, 105–116.

[12] Lluis, F. and Cosma, M.P. (2013) Resetting epigenetic signatures to induce somatic cell reprogramming. Cell. Mol. Life Sci. 70, 1413–1424.

[13] Cremer, T. and Cremer, C. (2001) Chromosome territories, nuclear architecture and gene regulation in mammalian cells. Nat. Rev. Genet. 2, 292–301. [14] Efroni, S., Duttagupta, R., Cheng, J., Dehghani, H., Hoeppner, D.J., Dash, C.,

Bazett-Jones, D.P., Le Grice, S., McKay, R.D., Buetow, K.H., Gingeras, T.R., Misteli, T. and Meshorer, E. (2008) Global transcription in pluripotent embryonic stem cells. Cell Stem Cell 2, 437–447.

[15] Fussner, E., Ching, R.W. and Bazett-Jones, D.P. (2011) Living without 30 nm chromatin ﬁbers. Trends Biochem. Sci. 36, 1–6.

[16] Joti, Y., Hikima, T., Nishino, Y., Kamada, F., Hihara, S., Takata, H., Ishikawa, T. and Maeshima, K. (2012) Chromosomes without a 30-nm chromatin ﬁber. Nucleus 3, 404–410.

[17] Nishino, Y., Eltsov, M., Joti, Y., Ito, K., Takata, H., Takahashi, Y., Hihara, S., Frangakis, A.S., Imamoto, N., Ishikawa, T. and Maeshima, K. (2012) Human mitotic chromosomes consist predominantly of irregularly folded nucleosome ﬁbres without a 30-nm chromatin structure. EMBO J. 31, 1644–1653. [18] Gustafsson, M.G. (2000) Surpassing the lateral resolution limit by a factor of

two using structured illumination microscopy. J. Microsc. 198, 82–87. [19] Gustafsson, M.G. (2005) Nonlinear structured-illumination microscopy:

wide-ﬁeld ﬂuorescence imaging with theoretically unlimited resolution. Proc. Natl. Acad. Sci. USA 102, 13081–13086.

[20] Klar, T.A., Jakobs, S., Dyba, M., Egner, A. and Hell, S.W. (2000) Fluorescence microscopy with diffraction resolution barrier broken by stimulated emission. Proc. Natl. Acad. Sci. USA 97, 8206–8210.

[21] Rust, M.J., Bates, M. and Zhuang, X. (2006) Sub-diffraction-limit imaging by stochastic optical reconstruction microscopy (STORM). Nat. Methods 3, 793– 795.

[22] Betzig, E., Patterson, G.H., Sougrat, R., Lindwasser, O.W., Olenych, S., Bonifacino, J.S., Davidson, M.W., Lippincott-Schwartz, J. and Hess, H.F. (2006) Imaging intracellular ﬂuorescent proteins at nanometer resolution. Science 313, 1642–1645.

[23] Hess, S.T., Girirajan, T.P. and Mason, M.D. (2006) Ultra-high resolution imaging by ﬂuorescence photoactivation localization microscopy. Biophys. J. 91, 4258– 4272.

[24] Bohn, M., Diesinger, P., Kaufmann, R., Weiland, Y., Muller, P., Gunkel, M., von Ketteler, A., Lemmer, P., Hausmann, M., Heermann, D.W. and Cremer, C. (2010) Localization microscopy reveals expression-dependent parameters of chromatin nanostructure. Biophys. J. 99, 1358–1367.

[25] Flors, C. (2011) DNA and chromatin imaging with super-resolution ﬂuorescence microscopy based on single-molecule localization. Biopolymers 95, 290–297.

[26] Flors, C. and Earnshaw, W.C. (2011) Super-resolution ﬂuorescence microscopy as a tool to study the nanoscale organization of chromosomes. Curr. Opin. Chem. Biol. 15, 838–844.

[27] Loschberger, A., van de Linde, S., Dabauvalle, M.C., Rieger, B., Heilemann, M., Krohne, G. and Sauer, M. (2012) Super-resolution imaging visualizes the eightfold symmetry of gp210 proteins around the nuclear pore complex and resolves the central channel with nanometer resolution. J. Cell Sci. 125, 570– 575.

[28] Markaki, Y., Gunkel, M., Schermelleh, L., Beichmanis, S., Neumann, J., Heidemann, M., Leonhardt, H., Eick, D., Cremer, C. and Cremer, T. (2010) Functional nuclear organization of transcription and DNA replication: a topographical marriage between chromatin domains and the interchromatin compartment. Cold Spring Harb. Symp. Quant. Biol. 75, 475–492.

[29] Raulf, A., Spahn, C.K., Zessin, P.J., Finan, K., Bernhardt, S., Heckel, A. and Heilemann, M. (2014) Click chemistry facilitates direct labelling and super-resolution imaging of nucleic acids and proteins. RSC advances 4, 30462– 30466, http://dx.doi.org/10.1039/c4ra01027b (Electronic supplementary information (ESI) available).

(8)

[30] Ricci, M.A., Manzo, M., Garcia-Parajo, M.F., Lakadamyali, M. and Cosma, M.P. (2015) Chromatin ﬁbers are formed by heterogeneous groups of nucleosomes in vivo. Cell 160, 1145–1158.

[31] Schermelleh, L., Carlton, P.M., Haase, S., Shao, L., Winoto, L., Kner, P., Burke, B., Cardoso, M.C., Agard, D.A., Gustafsson, M.G., Leonhardt, H. and Sedat, J.W. (2008) Subdiffraction multicolor imaging of the nuclear periphery with 3D structured illumination microscopy. Science 320, 1332–1336.

[32] Schoen, I., Ries, J., Klotzsch, E., Ewers, H. and Vogel, V. (2011) Binding-activated localization microscopy of DNA structures. Nano Lett. 11, 4008–4011. [33] Smeets, D., Markaki, Y., Schmid, V.J., Kraus, F., Tattermusch, A., Cerase, A., Sterr,

M., Fiedler, S., Demmerle, J., Popken, J., Leonhardt, H., Brockdorff, N., Cremer, T., Schermelleh, L. and Cremer, M. (2014) Three-dimensional super-resolution microscopy of the inactive X chromosome territory reveals a collapse of its active nuclear compartment harboring distinct Xist RNA foci. Epigenet. Chromatin 7, 8.

[34] Spahn, C., Endesfelder, U. and Heilemann, M. (2014) Super-resolution imaging of Escherichia coli nucleoids reveals highly structured and asymmetric segregation during fast growth. J. Struct. Biol. 185, 243–249.

[35] Szymborska, A., de Marco, A., Daigle, N., Cordes, V.C., Briggs, J.A. and Ellenberg, J. (2013) Nuclear pore scaffold structure analyzed by super-resolution microscopy and particle averaging. Science 341, 655–658.

[36] Wang, W., Li, G.W., Chen, C., Xie, X.S. and Zhuang, X. (2011) Chromosome organization by a nucleoid-associated protein in live bacteria. Science 333, 1445–1449.

[37] Wang, Y., Maharana, S., Wang, M.D. and Shivashankar, G.V. (2014) Super-resolution microscopy reveals decondensed chromatin structure at transcription sites. Sci. Rep. 4, 4477.

[38] Zessin, P.J., Finan, K. and Heilemann, M. (2012) Super-resolution ﬂuorescence imaging of chromosomal DNA. J. Struct. Biol. 177, 344–348.

[39] Benke, A. and Manley, S. (2012) Live-cell dSTORM of cellular DNA based on direct DNA labeling. ChemBioChem 13, 298–301.

[40] Gustafsson, M.G., Shao, L., Carlton, P.M., Wang, C.J., Golubovskaya, I.N., Cande, W.Z., Agard, D.A. and Sedat, J.W. (2008) Three-dimensional resolution doubling in wide-ﬁeld ﬂuorescence microscopy by structured illumination. Biophys. J. 94, 4957–4970.

[41] Hofmann, M., Eggeling, C., Jakobs, S. and Hell, S.W. (2005) Breaking the diffraction barrier in ﬂuorescence microscopy at low light intensities by using reversibly photoswitchable proteins. Proc. Natl. Acad. Sci. USA 102, 17565– 17569.

[42] Brakemann, T., Stiel, A.C., Weber, G., Andresen, M., Testa, I., Grotjohann, T., Leutenegger, M., Plessmann, U., Urlaub, H., Eggeling, C., Wahl, M.C., Hell, S.W. and Jakobs, S. (2011) A reversibly photoswitchable GFP-like protein with ﬂuorescence excitation decoupled from switching. Nat. Biotechnol. 29, 942– 947.

[43] Grotjohann, T., Testa, I., Leutenegger, M., Bock, H., Urban, N.T., Lavoie-Cardinal, F., Willig, K.I., Eggeling, C., Jakobs, S. and Hell, S.W. (2011) Diffraction-unlimited all-optical imaging and writing with a photochromic GFP. Nature 478, 204–208.

[44] Schmidt, R., Wurm, C.A., Jakobs, S., Engelhardt, J., Egner, A. and Hell, S.W. (2008) Spherical nanosized focal spot unravels the interior of cells. Nat. Methods 5, 539–544.

[45] Thompson, R.E., Larson, D.R. and Webb, W.W. (2002) Precise nanometer localization analysis for individual ﬂuorescent probes. Biophys. J. 82, 2775– 2783.

[46] Yildiz, A., Forkey, J.N., McKinney, S.A., Ha, T., Goldman, Y.E. and Selvin, P.R. (2003) Myosin V walks hand-over-hand: single ﬂuorophore imaging with 1.5-nm localization. Science 300, 2061–2065.

[47] Bates, M., Blosser, T.R. and Zhuang, X. (2005) Short-range spectroscopic ruler based on a single-molecule optical switch. Phys. Rev. Lett. 94, 108101. [48] Patterson, G.H. and Lippincott-Schwartz, J. (2002) A photoactivatable GFP for

selective photolabeling of proteins and cells. Science 297, 1873–1877. [49] Huang, B., Babcock, H. and Zhuang, X. (2010) Breaking the diffraction barrier:

super-resolution imaging of cells. Cell 143, 1047–1058.

[50] Fernandez-Suarez, M. and Ting, A.Y. (2008) Fluorescent probes for super-resolution imaging in living cells. Nat. Rev. Mol. Cell Biol. 9, 929–943. [51] Lippincott-Schwartz, J. and Patterson, G.H. (2009) Photoactivatable

ﬂuorescent proteins for diffraction-limited and super-resolution imaging. Trends Cell Biol. 19, 555–565.

[52] Lakadamyali, M. (2014) Super-resolution microscopy: going live and going fast. ChemPhysChem 15, 630–636.

[53] Huang, B., Wang, W., Bates, M. and Zhuang, X. (2008) Three-dimensional super-resolution imaging by stochastic optical reconstruction microscopy. Science 319, 810–813.

[54]Xu, K., Babcock, H.P. and Zhuang, X. (2012) Dual-objective STORM reveals three-dimensional ﬁlament organization in the actin cytoskeleton. Nat. Methods 9, 185–188.

[55]Pavani, S.R., Thompson, M.A., Biteen, J.S., Lord, S.J., Liu, N., Twieg, R.J., Piestun, R. and Moerner, W.E. (2009) Three-dimensional, single-molecule ﬂuorescence imaging beyond the diffraction limit by using a double-helix point spread function. Proc. Natl. Acad. Sci. USA 106, 2995–2999.

[56]Shtengel, G., Galbraith, J.A., Galbraith, C.G., Lippincott-Schwartz, J., Gillette, J.M., Manley, S., Sougrat, R., Waterman, C.M., Kanchanawong, P., Davidson, M.W., Fetter, R.D. and Hess, H.F. (2009) Interferometric ﬂuorescent super-resolution microscopy resolves 3D cellular ultrastructure. Proc. Natl. Acad. Sci. USA 106, 3125–3130.

[57]Lakadamyali, M., Babcock, H., Bates, M., Zhuang, X. and Lichtman, J. (2012) 3D multicolor super-resolution imaging offers improved accuracy in neuron tracing. PLoS One 7, e30826.

[58]Oddone, A., Vilanova, I.V., Tam, J. and Lakadamyali, M. (2014) Super-resolution imaging with stochastic single-molecule localization: concepts, technical developments, and biological applications. Microsc. Res. Tech. 77, 502–509. [59]Dempsey, G.T., Vaughan, J.C., Chen, K.H., Bates, M. and Zhuang, X. (2011)

Evaluation of ﬂuorophores for optimal performance in localization-based super-resolution imaging. Nat. Methods 8, 1027–1036.

[60] Tam, J., Cordier, G.A., Borbely, J.S., Sandoval Alvarez, A. and Lakadamyali, M. (2014) Cross-talk-free multi-color STORM imaging using a single ﬂuorophore. PLoS One 9, e101772.

[61]Wood, A.J., Lo, T.W., Zeitler, B., Pickle, C.S., Ralston, E.J., Lee, A.H., Amora, R., Miller, J.C., Leung, E., Meng, X., Zhang, L., Rebar, E.J., Gregory, P.D., Urnov, F.D. and Meyer, B.J. (2011) Targeted genome editing across species using ZFNs and TALENs. Science 333, 307.

[62]Hsu, P.D., Lander, E.S. and Zhang, F. (2014) Development and applications of CRISPR-Cas9 for genome engineering. Cell 157, 1262–1278.

[63]Jones, S.A., Shim, S.H., He, J. and Zhuang, X. (2011) Fast, three-dimensional super-resolution imaging of live cells. Nat. Methods 8, 499–508.

[64]Klein, T., Loschberger, A., Proppert, S., Wolter, S., van de Linde, S. and Sauer, M. (2011) Live-cell dSTORM with SNAP-tag fusion proteins. Nat. Methods 8, 7–9. [65]van de Linde, S., Heilemann, M. and Sauer, M. (2012) Live-cell super-resolution imaging with synthetic ﬂuorophores. Annu. Rev. Phys. Chem. 63, 519–540. [66]Wombacher, R., Heidbreder, M., van de Linde, S., Sheetz, M.P., Heilemann, M.,

Cornish, V.W. and Sauer, M. (2010) Live-cell super-resolution imaging with trimethoprim conjugates. Nat. Methods 7, 717–719.

[67]Grimm, J.B., English, B.P., Chen, J., Slaughter, J.P., Zhang, Z., Revyakin, A., Patel, R., Macklin, J.J., Normanno, D., Singer, R.H., Lionnet, T. and Lavis, L.D. (2015) A general method to improve ﬂuorophores for live-cell and single-molecule microscopy. Nat. Methods 12, 244–250.

[68]Flors, C. (2010) Photoswitching of monomeric and dimeric DNA-intercalating cyanine dyes for super-resolution microscopy applications. Photochem. Photobiol. Sci. 9, 643–648.

[69]Miyanari, Y., Ziegler-Birling, C. and Torres-Padilla, M.E. (2013) Live visualization of chromatin dynamics with ﬂuorescent TALEs. Nat. Struct. Mol. Biol. 20, 1321–1324.

[70] Chen, B., Gilbert, L.A., Cimini, B.A., Schnitzbauer, J., Zhang, W., Li, G.W., Park, J., Blackburn, E.H., Weissman, J.S., Qi, L.S. and Huang, B. (2013) Dynamic imaging of genomic loci in living human cells by an optimized CRISPR/Cas system. Cell 155, 1479–1491.

[71]Tanenbaum, M.E., Gilbert, L.A., Qi, L.S., Weissman, J.S. and Vale, R.D. (2014) A protein-tagging system for signal ampliﬁcation in gene expression and ﬂuorescence imaging. Cell 159, 635–646.

[72]Matsuda, A., Shao, L., Boulanger, J., Kervrann, C., Carlton, P.M., Kner, P., Agard, D. and Sedat, J.W. (2010) Condensed mitotic chromosome structure at nanometer resolution using PALM and EGFP-histones. PLoS One 5, e12768. [73]Balint, S., Verdeny Vilanova, I., Sandoval Alvarez, A. and Lakadamyali, M.

(2013) Correlative live-cell and superresolution microscopy reveals cargo transport dynamics at microtubule intersections. Proc. Natl. Acad. Sci. USA 110, 3375–3380.

[74]Tam, J., Cordier, G.A., Balint, S., Sandoval Alvarez, A., Borbely, J.S. and Lakadamyali, M. (2014) A microﬂuidic platform for correlative live-cell and super-resolution microscopy. PLoS One 9, e115512.

[75]Durisic, N., Cuervo, L.L. and Lakadamyali, M. (2014) Quantitative super-resolution microscopy: pitfalls and strategies for image analysis. Curr. Opin. Chem. Biol. 20, 22–28.