The synapse is densely packed with proteins involved in various highly regulated processes. Synaptic protein copy numbers and their stoichiometric distribution have a drastic influence on neuronal integrity and function. Therefore, the molecular analysis of synapses is a key element to understand their architecture and function. The overall structure of the synapse has been revealed with an exquisite amount of details by electron microscopy. However, the molecular composition and the localization of proteins are more easily addressed with fluorescence imaging, especially with the improved resolution achieved by super-resolution microscopy techniques. Notably, the fast improvement of imaging instruments has not been reflected in the optimization of biological sample preparation. During recent years, large efforts have been made to generate affinity probes smaller than conventional antibodies adapted for fluorescent super-resolution imaging. In this review, we briefly discuss the current views on synaptic organization and necessary key technologies to progress in the understanding of synaptic physiology. We also highlight the challenges faced by current fluorescent super-resolution methods, and we describe the prerequisites for an ideal study of synaptic organization.

Synapses share a characteristic structure

When first mentioned in 1897 by Charles S. Sherrington, the synapse was seen as “(…) a surface which might restrain diffusion, bank up osmotic pressure, restrict the movement of ions, accumulate electric charges, support a double electric layer, alter in shape and surface-tension (…)” [1].

Although today's view of the synapse became considerably more complex, this early description already acknowledges the various features and functions currently known of a synapse. Microscopy techniques, originally based on histological staining protocols using metal and salt precipitates [2], have been fundamental to study neuronal structures such as synapses. However, it was only in the 1950s that electron microscopy (EM) allowed a detailed view of subcellular synaptic structures [3,4]. In subsequent decades, several technical EM improvements resulted in more confidence in describing synaptic anatomy and morphology. Characteristic structures such as the postsynaptic density and the synaptic vesicle pool were described and remain prevailing research topics. Owing to the numerous EM studies on synaptic morphology, the ultrastructure of the synapse is now revealed in exquisite detail, including positioning of organelles and cytoskeletal elements [5,6]. Despite the high spatial resolution obtained in EM images, it has been difficult to use this microscopy technique to study the precise molecular composition of cellular structures. One major obstacle for this is the difficulty to use antibodies as protein reporters in EM preparations.

Conventional EM experiments typically require strongly reactive chemicals during the staining protocols. Chemicals such as glutaraldehyde and osmium tetroxide (OsO4) are commonly used in EM to fix the biological specimen and to preserve the biological ultrastructure. However, the treatment with dialdehydes like glutaraldehyde has been reported to hamper antigen recognition by specific affinity probes such as antibodies, which typically results in suboptimal detection of the proteins of interest [7,8]. Glutaraldehyde, as a cross-linking agent, can modify epitopes directly and can additionally generate a dense proteinaceous network that limits the diffusion of large molecules, such as antibodies [9]. It is important to note that these undesired features of glutaraldehyde are generally relevant when the proteins of interests are revealed with affinity probes, regardless of the microscopy technique used.

Nevertheless, immunolabeling using antibodies coupled with gold particles has been used in EM studies for more than four decades [10]. However, it has been challenging to achieve the flexibility which can be obtained in samples imaged with fluorescence light microscopy (such as live staining, cell surface vs. total cell labeling or simultaneous detection of three different proteins). The problems associated with the use of glutaraldehyde in immunostainings have been addressed by the implementation of rapid freezing or cryofixation, which deep freezes the sample in milliseconds, preserving more details of the synaptic ultrastructure [11,12]. By using cryofixation, the application of chemical fixatives can be entirely avoided, especially if cryo-fixed samples are subjected to freeze-substitution techniques [13]. Still, these methods require sophisticated equipment and immunogold labeling is still typically performed after resin embedding in sections, which results in the detection of epitopes exposed to the surface of the sections [14]. In contrast, pre-embedding immunogold staining enhances the number of epitopes detected by antibodies, thus achieving a higher labeling density. However, this compromises the ultrastructure preservation because the sample needs to be chemically fixed and permeabilized with detergents to allow large antibody molecules to enter the cell [7].

Notably, several new EM techniques keep appearing to enhance the ability of EM to determine the localization of specific proteins at high resolutions. For instance, a recent approach uses an engineered peroxidase (APEX) that can be fused to a protein of interest. This enzymatic tag oxidizes a substrate (diaminobenzidine), thereby generating an electron-dense precipitate that allows the precise localization of the tagged protein at high resolution [15,16]. Another major effort has been correlated microscopy techniques, which generally consist of combining the excellent spatial resolution of EM with a second microscopy technique that provides more specific molecular information and overlaps the images obtained by the two methods [17,18].

Overall, many details of synaptic structure can be revealed by EM in outstanding detail. However, during the last two decades, the progress of light microscopy techniques with enhanced resolutions has proceeded at a very high speed. Fluorescence-based techniques with spatial resolutions that are not limited by the diffraction of light are collectively known as super-resolution microscopy. One of the most attractive features of such fluorescence super-resolution techniques is the simplicity and flexibility of sample preparation, allowing conventional immunofluorescence protocols. Moreover, they allow one to perform experiments in living cells and to use the widespread fluorescent proteins to genetically tag the targets under investigation. Neuroscientists were one of the first in adopting these newer fluorescence microscopy techniques. However, the continuous and fast technical progress in super-resolution methods requires adjustments and optimization of the sample preparation to exploit the full benefits that such instruments and techniques can provide. Therefore, this review focuses on the progress and challenges to study synapses using fluorescence super-resolution microscopy techniques and other related methods.

Molecular organization of the synapse is addressed by fluorescence microscopy

The function of a synapse is strongly linked to its molecular composition. Therefore, it is vital to introduce a detectable label to study the subcellular localization of synaptic proteins by fluorescence microscopy [19]. This can be achieved in several ways, but the two most commonly used methods are the recombinant expression of proteins of interest fused to a fluorescent reporter, such as the green fluorescent protein (GFP) [20], or the application of affinity probes that specifically recognize the protein of interest. Predominantly, immunoglobulins (IgGs; or antibodies) raised against a target protein are used as specific molecular probes. Several proteins can be labeled in parallel by using antibodies coupled with different fluorophores or by recombinant fusion to different GFP derivatives. In contrast with EM, fluorescence microscopy therefore constitutes a straightforward method for the parallel detection of multiple targets, both in fixed tissue and in vivo, albeit at the cost of resolution. A good example for the power of fluorescence microscopy are early colocalization studies that revealed the accumulation of several neuronal proteins at synapses [21,22], such as the so-called soluble N-ethylmaleimide-sensitive fusion protein-attachment protein receptor (SNARE) proteins, required for vesicle fusion to membranes. The integration of functional and anatomical information for individual protein species triggered the generation of models that described different synaptic regulations and organizations as shown in Figure 1 [23]. The synaptic vesicle cycle was found to be a sophisticated molecular mechanism composed of various tightly regulated steps, which have been described in much detail [24,25]. This requires a finely tuned molecular organization of the participating proteins to ensure their function at the appropriate subcellular location. Moreover, different pools of synaptic vesicles have been described, showing varying responses to electric stimulation [26,27]. Only a minor fraction of the overall vesicle pool was found to be readily releasable upon moderate stimulation [26]. Recycling vesicles can be tracked by application of labeled affinity probes binding the luminal domain of synaptic vesicle proteins [28]. Taken together, these and other findings based on fluorescence microscopy have helped to change the view of the synapse from a rigid diffusion barrier toward a well-regulated and dynamic platform. In this context, the regulation of complex molecular interactions is essential to maintain synaptic integrity. This includes synaptic vesicle recycling, which constitutes one fundamental element of synaptic function. Although there is consensus on the major aspects of the typical synaptic vesicle cycle, several features of the endocytic process and neurotransmitter release mechanisms are still highly debated [25]. The question is which approach might lead us to an accurate view of synaptic organization and, thus, synaptic function?

Illustration and electron micrograph of presynaptic structures.

Figure 1.
Illustration and electron micrograph of presynaptic structures.

(A) A simplified schematic model of the synaptic vesicle cycle. Synaptic vesicles are decorated with vesicular associated membrane protein 2 (VAMP2 in blue) required for SNARE complex formation with syntaxin-1a (green) and synaptosome associated protein 25 (SNAP25 in red). Vesicles fuse with the plasma membrane at the active zone and vesicle material is subsequently internalized via endocytosis. (B) A more detailed model of synaptic vesicle cycling based on super-resolution microscopy observation [43]. High copy numbers of SNARE proteins are present on both vesicle and plasma membrane. Syntaxin-1a and SNAP25 were found to localize and cluster along the entire plasma membrane [46]. Synaptic vesicle pools were classified depending on their varying response to stimulation [26,27]. (C) Electron micrograph showing the ultrastructure of a presynapse. Note the accumulation of mitochondria (white arrowheads) in the vicinity of the synaptic vesicles (black arrowheads). Fusion of synaptic vesicle takes place at the active zone (red arrowheads). (D) 3D reconstruction of the presynapse considering the copy numbers and localization of 60 synaptic proteins. Modified model generated from previous data [52].

Figure 1.
Illustration and electron micrograph of presynaptic structures.

(A) A simplified schematic model of the synaptic vesicle cycle. Synaptic vesicles are decorated with vesicular associated membrane protein 2 (VAMP2 in blue) required for SNARE complex formation with syntaxin-1a (green) and synaptosome associated protein 25 (SNAP25 in red). Vesicles fuse with the plasma membrane at the active zone and vesicle material is subsequently internalized via endocytosis. (B) A more detailed model of synaptic vesicle cycling based on super-resolution microscopy observation [43]. High copy numbers of SNARE proteins are present on both vesicle and plasma membrane. Syntaxin-1a and SNAP25 were found to localize and cluster along the entire plasma membrane [46]. Synaptic vesicle pools were classified depending on their varying response to stimulation [26,27]. (C) Electron micrograph showing the ultrastructure of a presynapse. Note the accumulation of mitochondria (white arrowheads) in the vicinity of the synaptic vesicles (black arrowheads). Fusion of synaptic vesicle takes place at the active zone (red arrowheads). (D) 3D reconstruction of the presynapse considering the copy numbers and localization of 60 synaptic proteins. Modified model generated from previous data [52].

Just a decade ago, detailed proteomic analyses revealed the complexity of the synaptic vesicle proteome, generating a comprehensive list containing its molecular components [29]. Using EM and mass spectrometry, Takamori et al. [29] presented a quantitative molecular model of an average synaptic vesicle, emphasizing the importance of studies on the molecular stoichiometry of cellular processes. The approach to determine the copy number of a synaptic protein to better understand its physiological role has caused a paradigm change on the current view of synaptic function [30]. It is important to bear in mind that those results would not have been possible using conventional fluorescence microscopy because the direct visualization of individual synaptic vesicles or protein complexes cannot be achieved due to limited resolution, imposed by the diffraction of light. It is thus not surprising that the latest technology impulse to the field of synaptic studies is promoted by diffraction-unlimited fluorescent microscopy or super-resolution microscopy.

The revolution in fluorescence microscopy

Several far-field microscopy techniques that have been introduced in recent years now allow the discrimination of biological objects that are located closer than the diffraction of light, dramatically improving the level of observable details. Up to this point, the diffraction of light described by Ernst Abbe in the 19th century limited the possible attainable resolution. New microscopy techniques overcome this limit in several ways. In stimulated emission depletion (STED) microscopy [31], the fluorescence signal is modulated directly by targeted partial depletion of the fluorescence signal. For this, a doughnut-shaped laser beam is overlaid with the excitation beam, which results in the emission of fluorescence only from the center of the doughnut-shaped beam. This finally results in modulation of the point-spread function and thus improves the resolution. Alternatively, super-resolution can be achieved by stochastic imaging of individual fluorophores [stochastic optical reconstruction microscopy (STORM)/photo-activated localization microscopy (PALM)] [32,33]. In these techniques, several hundreds of images are recorded, in which only a sparse subset of fluorophores are emitting photons at every imaged frame. Thus, STORM/PALM requires specific photoswitchable fluorophores, which can alternate between a dark and a fluorescent state. The final image is reconstructed by determining the fluorophore positions after hundreds of localizations. Yet an alternative method, termed structured illumination microscopy, uses illumination patterns to enhance resolution by superimposing information components from which the super-resolution image can be created by reverse Fourier transformation [34,35]. These techniques, collectively known as super-resolution microscopy, are currently becoming the new standard in fluorescence microscopy in a range of laboratories worldwide. Further technical details on super-resolution techniques are available in many reviews [3639] and thus are not discussed here. In biological applications, STED and STORM/PALM have been shown to achieve a lateral resolution down to 20 and 9 nm, respectively [33,4042], resolving structures with notably better precision than diffraction-limited microscopy (i.e. ∼200 nm maximal lateral resolution).

Interestingly, one of the first biological applications of super-resolution microscopy was the investigation of synaptic structures [43]. Individual synaptic vesicles have an approximate diameter of 40 nm, escaping their individual visualization by conventional scanning confocal microscopy. Consequently, there were also mostly speculations on the quantitative distribution of synaptic vesicle proteins and their relative positioning. In contrast, STED-based imaging not only distinguished individual synaptic vesicles, but also allowed the study of their molecular composition [43]. Fluorescence super-resolution microscopy for the first time revealed synaptic vesicle trafficking in real time at nanoscale resolution [44]. Additionally, the subcellular organization of various processes, including receptor assembly, vesicle motility and SNARE complex structure, was revealed in more detail using super-resolution technologies [4548]. Surprisingly, several protein species were found in unexpected locations. One prominent example is the distribution of the SNARE proteins synaptosome-associated protein of 25 kDa (SNAP25) and syntaxin-1a, which are required for vesicle fusion at the active zone of the synapse. Although one might expect them to predominantly localize at the active zone, these proteins were reported to form clusters all over the cellular membrane [46]. It has been argued that such assemblies might serve as a molecular platform for SNARE complex formation or that they might modulate membrane integrity [49]. However, the physiological relevance and molecular function of SNARE clustering is still under investigation.

Super-resolution microscopy also had a notable impact on studies of neuronal cytoskeleton organization. In axons, periodic patterns of actin and spectrin were described [50]. These assemblies are assumed to take a major part in subcellular organization and protein recruitment. Furthermore, lattice-like patches of filamentous actin were observed at presynaptic compartments in vivo, probably serving as a scaffold for synaptic integrity [51]. These examples highlight that descriptive investigations of molecular organization can provide essential information from which physiological conclusions can be derived. From such data, models on synaptic function as well as on docking, fusion and retrieval of synaptic vesicles have been proposed and constantly refined [25]. When taking a closer look at the molecular environment within the synapse, the large amount of synaptic proteins suggests a densely packed gel-like structure rather than a freely diffusing environment [52]. In 2014, a molecular model of an average synapse was published, following the line of quantitative proteome analysis, this time also combined with STED microscopy [52]. This new model integrates the relative distribution and copy number of individual proteins present in a synapse and highlights the impressive complexity of the synaptic proteome (Figure 1D). Subsequently, similar studies have also been carried out for other neuronal compartments, such as the axon initial segment, which provide another important step toward the comprehensive molecular view of the neuron [53].

We believe that such observations from descriptive microscopy approaches are required to further advance our understanding of synaptic function and organization. Improving microscopy techniques constitutes a key element to gain access to a new level of subcellular information. However, most of these new super-resolution techniques also impose challenging requirements on the preparation of biological samples. Therefore, to exploit the full potential of the current diffraction-unlimited microscopy, we must also optimize our conventional labeling methods.

Affinity probes: a key to reveal synaptic organization

As discussed above, proteins need to be visualized in sufficient detail to obtain useful results for understanding synaptic function. Nowadays, a plethora of detection strategies are available which can be used in fluorescence imaging [36]. Additionally, quantitative imaging by fluorescence super-resolution microscopy becomes a new benchmark to analyze the subcellular protein composition [38,54]. Sample preparation and detection strategies turned out to be increasingly important to obtain an accurate and reproducible result in super-resolution microscopy [36]. Recombinant expression of proteins fused to a fluorescent reporter is used routinely to follow protein localization [55]. Besides fluorescent proteins, a group of genetically encoded self-labeling tags, such as the SNAP, HALO or CLIP tags, can be used as fusion domains in recombinant overexpression [56]. Their main feature is that the desired fluorophores are applied externally after expression and specifically incorporated into the SNAP, HALO or CLIP tags. As a large variety of fluorescent molecules for incorporation are available, these tags provide a flexible platform for specific protein labeling in multiple spectrally separable imaging channels. All these approaches require the introduction of foreign genes, which in the case of primary neuronal cultures might require some optimization. Still, this method is widely used in microscopy and can be carried out with comparably little effort [57]. However, sensitive cell lines and primary neuronal cultures can display unnatural morphologies after overexpression, indicating an impaired physiology. Moreover, the overexpression levels are typically different among individual cells, resulting in significant variations of the fluorescence signal and experimental conclusions. As the protein copy number is substantially increased beyond endogenous levels, molecular distribution of protein complexes or cellular function might also be affected. Mislocalization or artificial cluster formation of proteins can be induced by the fluorescent domain [58,59]. Finally, endogenous proteins already present in the cell cannot be visualized by this method, but are still taking part in the physiological processes of the cell. The latter limitation can partially be overcome by editing the cellular genome using molecular toolkits such as the clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated 9 (Cas9) system to knock in or knock out specific genes in cell cultures [60]. However, this approach is not widely used yet and cannot be applied on human samples and biopsies. Generating suitable fluorescent fusion constructs requires relatively little effort and allows observation of proteins in vivo and in their direct biological context. Still, these systems mostly rely on overexpression and consequently entail several challenges, as described above.

An alternative to detect endogenous proteins by fluorescence microscopy is the use of affinity probes conjugated to fluorophores. The application of such probes typically does not affect the expression levels of their target protein. For several, but mainly cytoskeletal, proteins naturally occurring specific interactors are known, which can be labeled fluorescently and utilized to visualize their target. Such ligands or toxins have classically been used to track the localization of their target protein and investigate cytoskeletal elements [61,62]. Drugs, for example phalloidin and taxol, bind and stabilize the actin filaments and microtubules, respectively, preserving the organization of the cytoskeleton. Staining cells with fluorescently labeled phalloidin or taxol is compatible with super-resolution microscopy methods, and thus these probes have been valuable tools to investigate the synaptic cytoskeleton [50,63]. However, the number of natural ligands to different target proteins remains very limited. Therefore, the majority of synaptic proteins requires yet alternative labeling strategies to investigate synaptic protein organization.

To detect endogenous proteins, IgGs are the most commonly used affinity probes for both EM and super-resolution microscopy [35,64]. Antibodies are raised by animal immunization and the subsequent extraction from blood serum for polyclonal antibodies or by generation of hybridoma cell lines for monoclonal antibody production [65]. The large biological diversity of antigens that the antibodies can recognize is due to their highly variable loops responsible for epitope-binding, termed complementarity-determining regions (CDRs). The CDRs are located at the periphery of the variable heavy-chain and variable light-chain domains of the antibody (VH and VL, respectively). Antibodies are widely used tools to detect proteins in their cellular context in both fixed and living samples. The number of antibodies commercially available is large and steadily growing. Often several different antibodies that bind different epitopes of the same target molecule can be obtained.

Fluorescent stainings based on antibody detections constitute a standard procedure for imaging biological samples. Generally, the sample is incubated with a primary antibody that binds to the target molecule. Subsequently, a fluorescently or by other means labeled secondary antibody is applied to detect the primary antibody. This results in a flexible system with several possible combinations of fluorophores. Typically, secondary antibodies are polyclonal, which results in amplification of the fluorescence signal due to binding of multiple secondary antibodies per primary antibody (Figure 2). However, this signal amplification also impairs quantitative imaging, as the decoration with secondary antibodies is not homogeneous [37]. Moreover, the number of dye moieties coupled with a secondary antibody typically ranges from zero to four dye molecules, creating an unknown distribution of fluorophores, which results in additional signal bias and thus hampers true quantitative analyses [36]. Furthermore, the application of multiple antibodies for parallel detection of multiple targets occasionally causes cross-reactivity of non-specific secondary antibodies against the different primary antibody species [64].

Molecular models comparing conventional antibodies and small monovalent affinity probes in the recognition of transmembrane target proteins.

Figure 2.
Molecular models comparing conventional antibodies and small monovalent affinity probes in the recognition of transmembrane target proteins.

(AC) A piece of membrane with a fictive arrangement of integral membrane proteins was generated using structural information from the Protein Data Bank (PDB accession codes: 1HZH, 3OGO, 3KG2, 2BG9, 4PE5, 2VT4, 4PAS, 2RH1, 1EWK, 3LUT, 1EWK, 2E4U, 3KS9, 4OR2 and 5CM4). All elements are displayed in correct relative size scale. The binding interface is highlighted in magenta, and fluorescent molecules conjugated with the affinity probes are colored red. (A) The large size of an antibody (top panel, green molecules) might not allow binding to all epitopes available, which can result in incomplete detection or labeling. Small affinity probes (lower panel, light green molecules) can access buried epitopes due to their reduced size. (B) The application of conventional antibodies (top panel) can induce clustering of the target protein. The use of secondary antibodies can additionally cluster target-bound primary antibodies, which might result in artificial staining patterns. In contrast, monovalent probes (lower panel) should not induce such clustering artifacts. (C) The average size of 12–15 nm of a single antibody delocalizes the source of the fluorescent signal from the intended epitope (top panel). However, small probes directly conjugated to fluorescent molecules bring the fluorophores closer to the target protein, which increases the actual accuracy obtained using super-resolution techniques. For simplification, a potassium ion channel is shown as an exemplary target, but the principle also applies to globular proteins in crowded environments such as synapses. Scale bar represents 10 nm. (D) Molecular models of small affinity probes also described in Table 1. The variable portion of the probe (paratope) is colored red. Example sequences were obtained from the Protein Data Bank (affibody: 2KZI; affitin: 4CJ1; alphabody: 4OE8; DARPIN: 4YDW; monobody: 5DC9; nanobody: 3K1K; aptamer: 4R8I).

Figure 2.
Molecular models comparing conventional antibodies and small monovalent affinity probes in the recognition of transmembrane target proteins.

(AC) A piece of membrane with a fictive arrangement of integral membrane proteins was generated using structural information from the Protein Data Bank (PDB accession codes: 1HZH, 3OGO, 3KG2, 2BG9, 4PE5, 2VT4, 4PAS, 2RH1, 1EWK, 3LUT, 1EWK, 2E4U, 3KS9, 4OR2 and 5CM4). All elements are displayed in correct relative size scale. The binding interface is highlighted in magenta, and fluorescent molecules conjugated with the affinity probes are colored red. (A) The large size of an antibody (top panel, green molecules) might not allow binding to all epitopes available, which can result in incomplete detection or labeling. Small affinity probes (lower panel, light green molecules) can access buried epitopes due to their reduced size. (B) The application of conventional antibodies (top panel) can induce clustering of the target protein. The use of secondary antibodies can additionally cluster target-bound primary antibodies, which might result in artificial staining patterns. In contrast, monovalent probes (lower panel) should not induce such clustering artifacts. (C) The average size of 12–15 nm of a single antibody delocalizes the source of the fluorescent signal from the intended epitope (top panel). However, small probes directly conjugated to fluorescent molecules bring the fluorophores closer to the target protein, which increases the actual accuracy obtained using super-resolution techniques. For simplification, a potassium ion channel is shown as an exemplary target, but the principle also applies to globular proteins in crowded environments such as synapses. Scale bar represents 10 nm. (D) Molecular models of small affinity probes also described in Table 1. The variable portion of the probe (paratope) is colored red. Example sequences were obtained from the Protein Data Bank (affibody: 2KZI; affitin: 4CJ1; alphabody: 4OE8; DARPIN: 4YDW; monobody: 5DC9; nanobody: 3K1K; aptamer: 4R8I).

Table 1
Selection of affinity probes commonly used in molecular biology

All listed affinity probes can achieve affinities in the nanomolar range, comparable with the affinities obtained for conventional IgG antibodies (kD = 10−7–10−9 M). aa, amino acids; MW, molecular mass.

Name [reference] Composition Origin Sequence length MW (kDa) Size (nm) S–S bridges 
IgG antibodies [65Two heavy chains and two light chains Immunization/hybridomas ∼1450 aa ∼150 10 >4 
Fab fragments [85Truncated heavy chain + light chain IgG, cleaved by papain ∼450 aa ∼50 >2 
Single chain variable fragments (scFvs) [84VH + VL domain of IgG antibody Recombinant variable IgG domains ∼220 aa ∼25 >1 
VHH fragments/nanobodies [86Variable domain of heavy chain antibody Camelidae heavy chain IgG subtype 70–110 aa ∼15 <2 
Affibody [75α-Helical structure Staphylococcus aureus Protein A ∼58 aa ∼6 ∼2 
Affitin [87Sac7d (DNA-binding protein) Sulfolobus acidocaldarius ∼66 aa ∼7 ∼2 
Alphabody [88,89Three α-helices Artificial peptides 70–120 aa ∼10 
DARPIN [90Three to five ankyrin repeats Membrane adaptor protein ∼160 aa ∼15–18 
Monobody [91Seven β-sheets Human fibronectin ∼90 aa ∼10 
Aptamers [68Strands of DNA or RNA Synthetic sequence 15–60 bp 5–15 3–5 
Name [reference] Composition Origin Sequence length MW (kDa) Size (nm) S–S bridges 
IgG antibodies [65Two heavy chains and two light chains Immunization/hybridomas ∼1450 aa ∼150 10 >4 
Fab fragments [85Truncated heavy chain + light chain IgG, cleaved by papain ∼450 aa ∼50 >2 
Single chain variable fragments (scFvs) [84VH + VL domain of IgG antibody Recombinant variable IgG domains ∼220 aa ∼25 >1 
VHH fragments/nanobodies [86Variable domain of heavy chain antibody Camelidae heavy chain IgG subtype 70–110 aa ∼15 <2 
Affibody [75α-Helical structure Staphylococcus aureus Protein A ∼58 aa ∼6 ∼2 
Affitin [87Sac7d (DNA-binding protein) Sulfolobus acidocaldarius ∼66 aa ∼7 ∼2 
Alphabody [88,89Three α-helices Artificial peptides 70–120 aa ∼10 
DARPIN [90Three to five ankyrin repeats Membrane adaptor protein ∼160 aa ∼15–18 
Monobody [91Seven β-sheets Human fibronectin ∼90 aa ∼10 
Aptamers [68Strands of DNA or RNA Synthetic sequence 15–60 bp 5–15 3–5 

The complex composed of primary and secondary antibodies can cause further potential problems for immunohistochemistry assays. First, recent reports show that both antibody quality and reproducibility of assays are very inconsistent [66,67]. Variation may occur in both mono- and poly-clonal antibody preparations, giving researchers a hard time to identify reliable functional antibodies for biological studies [64]. Secondly, antibodies can induce aggregation and artificial clusters of their target molecules due to their bivalent nature. This might create artifacts in living or poorly fixed samples investigated by fluorescence imaging, which consequently can distort the interpretation or lead to false conclusions [67,68]. A possible way to partially compensate for the artificial antibody cluster formation is to directly couple the primary antibodies with fluorophores. This has indeed been shown to be beneficial, particularly in super-resolution microscopy [69]. However, this strategy is rather inefficient, as antibody production is expensive and different fluorophores would need to be coupled individually for every application. Furthermore, the chemical reaction used for coupling the fluorophore to the immunoglobulin, for instance N-hydroxysuccinimide ester coupling, additionally often results in nonfunctional antibodies. Finally, the primary–secondary antibody complex has been shown to limit the best possible resolution the super-resolution microscope can achieve. A single antibody has an approximate linear size of 12–15 nm [70], resulting in an overall distance of the fluorophore from the target of approximately 20–25 nm for a primary–secondary antibody complex. The fluorophores are commonly coupled with the constant domains of the heavy chain (Fc domain) of the secondary antibody. Consequently, the fluorescence signal is delocalized from the actual position of the target protein, as shown in Figure 2. This effect is barely noticeable if epifluorescence microscopy and scanning confocal microscopy are used. However, super-resolution microscopy provides nanoscale resolution at which the spatial dimension of the antibody–antibody complexes becomes a limiting factor for localization precision [69,71]. This not only holds true for fluorescence super-resolution microscopy, but also can be generalized for each kind of high-resolution microscopy in which antibodies are used to detect target molecules, including EM and cryo-EM. In addition to the delocalization of the fluorescence signal, the volume of an antibody complex allows only a limited subset of epitopes to be detected due to steric hindrance [72] (Figure 2). This probably results in a poor labeling density, hence reducing the sensitivity of the assay and the quantity of localized target molecules [35,73]. Antibodies therefore might readily create staining patterns, which do not match the real biological structures. Therefore, antibody stainings for super-resolution microscopy should be used with caution and several controls are needed to minimize possible artifacts and to obtain biologically relevant information.

Consequently, replacing conventional antibodies with smaller probes is required to exploit the full potential of super-resolution imaging and to gain more detailed and precise information on synaptic organization and other fields in biology. In recent years, various alternative affinity probes have been developed to detect endogenous proteins. The prominent characteristics of these new probes are their reduction in size and their monovalent binding. At the same time, small affinity probes maintain the capability to specifically recognize their target molecules.

Searching for specific and smaller monovalent probes with high affinity

Predominantly, small affinity probes consist of a constant stable scaffold and variable regions that determine the specificity for the target molecule (Table 1). Several of these scaffolds originate from prokaryotic organisms and hence can easily be produced in conventional recombinant expression systems. Unlike antibodies, most of these small probes do not require extensive post-translational modifications and are translated as a single peptide chain, which greatly facilitates expression and proper folding [74]. Compared with conventional antibodies, the compact structure of small probes generally provides higher stability against elevated temperatures and extreme pH conditions [75,76]. At the same time, their affinities to the target are comparable with the range of affinities found for classical antibodies (typical affinities range from 10−7 to 10−9 M). Another advantage of the small probes for biomedical research is their more reliable penetration into biological tissues, compared with more bulky molecules, such as antibodies (Figure 3) [74]. A large number of these affinity scaffolds are generally derived from molecular interactions found in nature. These scaffolds then have been further engineered to constitute a new class of synthetic or in vitro affinity probes. One example is the adaptor protein ankyrin, which was used as a backbone for designed ankyrin repeat proteins (DARPINs). Variable loops connecting three to five ankyrin repeats mediate the binding to the target molecule [77]. Another kind of small probe, termed affibody, is derived from the prokaryotic Protein A, which has a strong affinity for IgG molecules [75]. Affibodies consist of three helices of variable composition, which form a stable scaffold. Alternatively, nucleic acids may also serve as affinity probes. Aptamers are short stretches of single-stranded DNA or RNA that bind their target epitope via their three-dimensional structure with high specificity [78]. As a large variety of aptamers can be synthesized with comparably little effort, they became a true alternative to protein-based affinity probes. To develop novel probes directed against a target protein, comprehensive libraries are generated by randomizing the variable region of the probe or via targeted mutagenesis. Subsequently, molecular display assays such as phage or ribosome display are used to screen and select specific affinity probes [7982]. Typically, the selected probes need to be further validated for application in immunofluorescence to ensure high specificity and affinity of the probe. As molecular displays are commonly performed in vitro and on recombinant purified antigens, they might cross-react with other proteins present in the cell. Hence, careful characterization of the selected candidates is required, using biochemical arrays such as ELISA, immunoblots or pull-down experiments [82,83].

Experimental comparison of conventional antibodies and small probes in fluorescence microscopy.

Figure 3.
Experimental comparison of conventional antibodies and small probes in fluorescence microscopy.

(A) Confocal and STED images of primary rat hippocampal neurons stained for tubulin using a rabbit polyclonal antibody (Cat. No. 302 203, Synaptic Systems GmbH, Germany) detected with secondary antibodies coupled to atto647N or with a recombinant scFv directly coupled to atto647N [95]. Although samples were treated equally, the STED microscopy images reveal a qualitative difference in the labeling density of the microtubule filaments. The scFv shows a more detailed and continuous pattern. Scale bar represents 2 µm. (B) Fluorescent staining of α-synuclein in mouse brain tissue. Paraformaldehyde-fixed rat brain slices of approximately 30 µm thickness were incubated overnight using primary antibodies (Cat. No. 128 002, Synaptic Systems GmbH, Germany) and 6 h with fluorescently labeled secondary antibodies or using directly labeled nanobodies against α-synuclein described previously [105]. The sample was imaged in a confocal microscope. The optical sections of approximately 600 nm of thickness demonstrate the limited tissue penetration of the larger antibody complex. Scale bar represents 20 µm.

Figure 3.
Experimental comparison of conventional antibodies and small probes in fluorescence microscopy.

(A) Confocal and STED images of primary rat hippocampal neurons stained for tubulin using a rabbit polyclonal antibody (Cat. No. 302 203, Synaptic Systems GmbH, Germany) detected with secondary antibodies coupled to atto647N or with a recombinant scFv directly coupled to atto647N [95]. Although samples were treated equally, the STED microscopy images reveal a qualitative difference in the labeling density of the microtubule filaments. The scFv shows a more detailed and continuous pattern. Scale bar represents 2 µm. (B) Fluorescent staining of α-synuclein in mouse brain tissue. Paraformaldehyde-fixed rat brain slices of approximately 30 µm thickness were incubated overnight using primary antibodies (Cat. No. 128 002, Synaptic Systems GmbH, Germany) and 6 h with fluorescently labeled secondary antibodies or using directly labeled nanobodies against α-synuclein described previously [105]. The sample was imaged in a confocal microscope. The optical sections of approximately 600 nm of thickness demonstrate the limited tissue penetration of the larger antibody complex. Scale bar represents 20 µm.

Table 1 summarizes a selection of prominent affinity probes for molecular biology and light microscopy. As a comprehensive overview of all available classes of affinity probes is beyond the scope of this review, the reader may refer to detailed descriptions of alternative affinity probes [74,76,84].

Most novel probes are derived from synthetic libraries made by randomizing the variable binding domain of naturally existing molecules. Such libraries need to be carefully optimized to obtain high sequence diversity in their variable domains for the subsequent in vitro screening via molecular display assays [92]. Until now, the number of commercially available in vitro generated probes is still limited to a few target proteins. A potential reason is that libraries of in vitro generated probes might not have sufficient diversity to bind any possible epitope with the desired affinity. In contrast, antibodies are generated by an animal immune system, which has stringent quality control checkpoints and paratope maturation steps resulting in great versatility and specificity for target detection. Countless excellent antibodies with both high specificity and affinity for synaptic proteins are currently available and used on a regular basis. This evidences the power of the animal immune system, which is employed to generate antibodies for research. To combine the high specificity and affinity achieved by immune selection with the required features of a small probe, researches have engineered conventional antibodies to obtain small IgG-derived affinity probes. Those probes lack most of the conserved elements of a full IgG molecule, and therefore are significantly reduced in size, but maintain their specificity and affinity to their targets. Generally, the variable domains of antibodies are mediating the epitope-binding. The constant Fc domain of an antibody can thus be cleaved off using enzymes, such as the cysteine protease papain, to generate Fab fragments [93]. However, the generation of Fab fragments requires large amounts of purified antibodies as starting material, their detection by secondary antibodies is difficult and direct conjugation to dye moieties typically results in nonfunctional fragments. Therefore, Fab fragments are smaller (Table 1), but a costly alternative to antibodies.

A convenient way to produce antibody-based affinity probes is the recombinant expression of both the heavy-chain and light-chain variable regions, fused by a peptide linker in bacteria. Such single-chain variable fragments (scFvs) have been developed for more than two decades and only recently have been shown to have some advantages in imaging over classical antibodies [69,94,95]. However, the alignment of the light and heavy variable domain requires an optimized peptide linker and the protein yield of functional scFvs is usually poor, mainly because they also require the formation of disulfide bridges. Disulfide bond formation is not promoted in the reducing cytoplasm of bacteria, which imposes special challenges for recombinant proteins whose functionality depends on disulfide bridge formation [96]. This problem can be tackled by directing the expression of the recombinant probes to the more oxidizing periplasmic space, or by the use of bacterial strains engineered to facilitate disulfide bridge formation within the cytosol [96,97]. Still, scFvs have by default limited solubility and their two individual variable domains (VH and VL) need to be arranged properly to achieve efficient binding to the target. A new class of antibody-derived affinity probes may overcome these problems, VHH fragments.

VHH fragments: minimal affinity probes derived from the immune system of camelids

VHH fragments, also known as nanobodies®, represent the variable fragment of IgG subtypes that are exclusively found in Camelidae (camels, dromedaries, llamas or alpacas). These camelid IgG subtypes are devoid of light chains and thus were originally termed heavy-chain antibodies [86,98100]. Consequently, the variable fragment of the heavy chain (i.e. VHH fragment) solely contains three variable CDRs mediating the binding to the epitope. The reduced area of the nanobody–antigen interaction due to the lack of a light chain can partially be compensated for by an extended CDR3 loop [101,102]. In fact, extended nanobody CDR loops were reported to facilitate the binding of structural epitopes, which are not accessible for conventional antibodies [86,103]. Regardless of their reduced interaction surface, nanobodies can achieve affinities in the low nanomolar range, which is comparable with those of conventional IgG antibodies [104,105]. In contrast, nanobodies can be easily expressed as recombinant molecules, purified in large-scale and coupled with a large palette of molecules (such as toxins, fluorophores and isotopes) using standard techniques. This makes them invaluable as tools for various affinity assays [82,83,106]. Interestingly, nanobodies have recently been used in several super-resolution imaging approaches. They have been shown not only to bring the fluorophore closer to the target protein, but also to enhance tissue penetration and labeling density which is beneficial to study densely packed protein compartments, as illustrated in Figures 2 and 3 [69,71,107]. However, it should be noted that immunofluorescence protocols with these smaller probes still require the permeabilization of the biological specimen by detergents to allow their penetration into the cells. Detergent permeabilization extracts not only lipids but also not-fixed cytosolic or transmembrane proteins, and thus, this procedure can generate serious artifacts regardless of the affinity probe used [36].

A nanobody binding to GFP was described already 10 years ago [108], and since then, it has been refined and studied extensively [82,109]. However, the prerequisite to overexpress GFP-tagged fusion proteins hampers the analysis on endogenous protein levels, as discussed above. Considering the dense protein packaging in the synapse, the detection of endogenous proteins is far more favorable to accurately depict synaptic function. Consequently, alternative nanobodies that specifically bind to synaptic proteins would be of major help in studying the anatomy of the synapse with super-resolution techniques. Until now, the number of commercially available nanobodies is very limited compared with conventional antibodies, even though it is rising constantly [82,83,110]. Using several directly labeled nanobodies in parallel is an attractive way to multiplex fluorescent labeling, i.e. to detect several antigens simultaneously. For example, small probes could improve methods like exchange-point accumulation for imaging in nanoscale topography (PAINT), in which conventional antibodies were coupled with small oligonucleotides and detected by fluorescently labeled complementary oligonucleotides. This is an elegant strategy to detect several targets on the same sample [111]. Therefore, the utilization of small probes binding several synaptic targets is likely to provide a new level of information on molecular organization, as the nanobody against GFP has proven [71]. Additionally, sophisticated and potentially fast protocols to screen for novel nanobodies have recently been reported [82,83]. However, our personal experience suggests that the selection and functional validation of nanobodies for microscopy of all potential candidates is laborious and has to be tailored individually for each target protein.

Concluding, nanobodies and any other small probes are interesting alternatives to conventional IgG molecules due to their small size and biochemical properties. In super-resolution approaches, small probes significantly improve both the actual resolution and the labeling density, which provides better descriptions of subcellular structures, as highlighted by our self-generated examples depicted in Figure 3 [71]. Additionally, these novel classes of affinity molecules might even allow the study of orphan proteins of yet unknown function, as their scaffolds allow alternative molecular arrangements compared with conventional antibodies. Considering the complex and closely packed proteome of the synapse, small probes combined with the latest imaging technologies seem to be an excellent tool to investigate synaptic proteins in their physiological numbers and organization. We believe that small probes against synaptic proteins will help to provide a more comprehensive view on the molecular composition of the synapse and to understand its function in the near future.

Perspectives on imaging techniques and molecular tools

Our view on synaptic organization has been influenced by improvements in both microscopy techniques and molecular tools to detect individual proteins. Super-resolution microscopy revolutionized far-field imaging and is becoming the new standard in neurobiology. As we discussed above, in the field of molecular imaging, the requirement for improved and reliable affinity probes is constantly rising. Hence, small affinity probes are expected to reveal great potential in nanoscale and quantitative imaging [112]. Interestingly, new technologies are constantly being developed and improved to allow molecular detection of endogenous targets in direct physiological context.

In particular, mass spectrometry is a commonly used technique providing complex biochemical information. In recent years, the impact of secondary ion mass spectrometry (SIMS) for imaging biological samples has increased. Several ion species can be detected in parallel with high spatial resolution, allowing the reconstruction of nanostructures in atomic detail [113,114]. Recently, a correlative microscopy technique using STED microscopy and nanoSIMS has been introduced [115]. This allowed one to detect specific proteins with isotopic and fluorescent labeling [116]. This approach uses a label that can be detected in both SIMS and fluorescence super-resolution microscopy. NanoSIMS can provide a 50-fold increase in axial resolution compared with fluorescence imaging and also allows to follow protein turnover by metabolic labeling using heavy isotopes [116]. Similarly, direct imaging using matrix-assisted laser desorption/ionization has been introduced into biology [117]. Pure mass spectrometry imaging of the molecular composition of the sample would make the use of any affinity probe obsolete. However, this technique cannot yet compete with the spatial resolution achieved by fluorescence super-resolution microscopy [118]. Still, label-free mass spectrometry imaging might eventually complement fluorescent super-resolution techniques in future molecular imaging approaches.

Additionally, novel techniques to achieve subdiffraction imaging in the neurosciences are flourishing. For example, PAINT can be used to study small structures such as synaptic vesicles [119]. PAINT is a pointillistic approach to achieve super-resolution: it takes advantage of the diffusion of individual affinity molecules, in this case fluorescently labeled oligonucleotides. Recently, the introduction of quantitative PAINT (qPAINT) shows the ability of this technique to count molecules in a broad dynamic range [111]. The development of novel affinity probes might have direct implications on localization accuracy of biological components. Another super-resolution technique, called image reconstruction by integrating exchangeable single-molecule localization (IRIS) [120], uses affinity probes that are rapidly exchanged by diffusion and thus improve labeling density. These techniques may also be used to multiplex fluorescent labeling of multiple targets detected in individual channels, as mentioned above. Owing to their reduced size, the small probes used in IRIS show high diffusion rates and proved valuable for this technique. However, such imaging techniques also require extensive screening for suitable probes, as well as long acquisition times for individual images. Yet another recent imaging technique modulates the specimen itself rather than the detection probe. In this approach, called expansion microscopy, the sample is embedded into a polymer gel that gets cross-linked with the proteins of the sample and is subsequently swelled up in all dimensions, increasing its original volume up to 125-fold. The subsequent imaging with a conventional epifluorescence or confocal scanning microscope provides enhanced resolution, similar to the one obtained with super-resolution microscopy [121]. Nevertheless, target molecules need to be detected as in a conventional immunofluorescence protocol before swelling of the sample. Although expansion microscopy promises a novel super-resolution capability at relatively low cost, it still requires high penetration and high labeling density of the affinity probes used. Therefore, small affinity probes are still necessary for an optimal staining in this emerging technology.

Finally, molecular imaging is a crucial and indispensable method to reveal neuronal and synaptic function. Our latest view of the synapse and its molecular components has been inferred or confirmed from data acquired by fluorescence microscopy. Moreover, super-resolution nanoscopy notably extended the model of the synaptic vesicle cycle and synaptic organization in general. By combining the latest imaging techniques with small affinity probes, the actual empirical resolution obtained from biological samples can be substantially improved. Especially in densely packed compartments such as synapses, small probes show several advantages in molecular imaging compared with conventional antibodies. This includes better labeling density, enhanced tissue penetration, access to buried epitopes and accurate positioning of the fluorescent reporter. We therefore conclude that descriptive research using novel small and monovalent affinity probes is becoming essential to further complement our understanding of the synapse and to refine both structural and physiological synaptic models.

Abbreviations

CDRs, complementarity-determining regions; DARPINs, designed ankyrin repeat proteins; EM, electron microscopy; GFP, green fluorescent protein; IgGs, immunoglobulins; IRIS, image reconstruction by integrating exchangeable single-molecule localization; PAINT, point accumulation for imaging in nanoscale topography; PALM, photo activated localization microscopy; scFvs, single-chain variable fragments; SIMS, secondary ion mass spectrometry; SNAP25, synaptosome-associated protein of 25 kDa; SNARE, soluble N-ethylmaleimide-sensitive fusion protein-attachment protein receptor; STED, stimulated emission depletion; STORM, stochastic optical reconstruction microscopy.

Acknowledgments

We thank Burkhard Rammner for generating all molecular models present in this article. We also thank Sven Truckenbrodt, Sebastian Jähne and Sabryna Junker for helpful comments on the manuscript. We apologize to all those authors whose work could not be cited due to space limitations.

Competing Interests

The Authors declare that there are no competing interests associated with the manuscript.

References

References
1
Foster
,
M.
and
Sherrington
,
C.
(
1897
)
A Textbook of Physiology, Part Three: The Central Nervous System
,
MacMillan Co. Ltd
,
London
2
De Robertis
,
E.D.P.
(
2013
)
Histophysiology of Synapses and Neurosecretion: International Series of Monographs on Pure and Applied Biology: Modern Trends in Physiological Sciences
,
Elsevier
3
Palay
,
S.L.
and
Palade
,
G.E.
(
1955
)
The fine structure of neurons
.
J. Biophys. Biochem. Cytol.
1
,
69
88
doi:
4
Palade
,
G.E.
and
Palay
,
S.L.
(
1954
)
Electron microscopical observations of interneuronal and neuromuscular synapses
.
Anat. Rec.
118
,
335
336
5
Harris
,
K.M.
and
Weinberg
,
R.J.
(
2012
)
Ultrastructure of synapses in the mammalian brain
.
Cold Spring Harb. Perspect. Biol.
4
,
a005587
doi:
6
Völgyi
,
K.
,
Gulyássy
,
P.
,
Háden
,
K.
,
Kis
,
V.
,
Badics
,
K.
,
Kékesi
,
K.A.
et al. 
(
2015
)
Synaptic mitochondria: a brain mitochondria cluster with a specific proteome
.
J. Proteomics.
120
,
142
157
PMID:
[PubMed]
7
Zhong
,
L.
,
Brown
,
J.C.
,
Wells
,
C.
and
Gerges
,
N.Z.
(
2013
)
Post-embedding immunogold labeling of synaptic proteins in hippocampal slice cultures
.
J. Vis. Exp.
doi:
8
Ayache
,
J.
,
Beaunier
,
L.
,
Boumendil
,
J.
,
Ehret
,
G.
and
Laub
,
D.
(
2010
)
Sample Preparation Handbook for Transmission Electron Microscopy
,
Springer New York
,
New York, NY
9
Hopwood
,
D.
(
1972
)
Theoretical and practical aspects of glutaraldehyde fixation
.
Histochem. J.
4
,
267
303
doi:
10
Roth
,
J.
,
Bendayan
,
M.
and
Orci
,
L.
(
1978
)
Ultrastructural localization of intracellular antigens by the use of protein A-gold complex
.
J. Histochem. Cytochem.
26
,
1074
1081
doi:
11
Adrian
,
M.
,
Dubochet
,
J.
,
Lepault
,
J.
and
McDowall
,
A.W.
(
1984
)
Cryo-electron microscopy of viruses
.
Nature
308
,
32
36
doi:
12
Zhao
,
S.
,
Studer
,
D.
,
Graber
,
W.
,
Nestel
,
S.
and
Frotscher
,
M.
(
2012
)
Fine structure of hippocampal mossy fiber synapses following rapid high-pressure freezing
.
Epilepsia
53
,
4
8
doi:
13
Feder
,
N.
and
Sidman
,
R.L.
(
1958
)
Methods and principles of fixation by freeze-substitution
.
J. Biophys. Biochem. Cytol.
4
,
593
600
doi:
14
Mielanczyk
,
L.
,
Matysiak
,
N.
,
Michalski
,
M.
,
Buldak
,
R.
and
Wojnicz
,
R.
(
2014
)
Closer to the native state. Critical evaluation of cryo-techniques for transmission electron microscopy: preparation of biological samples
.
Folia Histochem. Cytobiol.
52
,
1
17
doi:
15
Lee
,
S.-Y.
,
Kang
,
M.-G.
,
Park
,
J.-S.
,
Lee
,
G.
,
Ting
,
A.Y.
and
Rhee
,
H.-W.
(
2016
)
APEX fingerprinting reveals the subcellular localization of proteins of interest
.
Cell Rep.
15
,
1837
1847
doi:
16
Lam
,
S.S.
,
Martell
,
J.D.
,
Kamer
,
K.J.
,
Deerinck
,
T.J.
,
Ellisman
,
M.H.
,
Mootha
,
V.K.
et al. 
(
2014
)
Directed evolution of APEX2 for electron microscopy and proximity labeling
.
Nat. Methods
12
,
51
54
doi:
17
de Boer
,
P.
,
Hoogenboom
,
J.P.
and
Giepmans
,
B.N.G.
(
2015
)
Correlated light and electron microscopy: ultrastructure lights up!
Nat. Methods
12
,
503
513
doi:
18
Watanabe
,
S.
,
Punge
,
A.
,
Hollopeter
,
G.
,
Willig
,
K.I.
,
Hobson
,
R.J.
,
Davis
,
M.W.
et al. 
(
2011
)
Protein localization in electron micrographs using fluorescence nanoscopy
.
Nat. Methods
8
,
80
84
doi:
19
Griffiths
,
G.
and
Lucocq
,
J.M.
(
2014
)
Antibodies for immunolabeling by light and electron microscopy: not for the faint hearted
.
Histochem. Cell Biol.
142
,
347
360
doi:
20
Miesenböck
,
G.
,
De Angelis
,
D.A.
and
Rothman
,
J.E.
(
1998
)
Visualizing secretion and synaptic transmission with pH-sensitive Green fluorescent proteins
.
Nature
394
,
192
195
doi:
21
Oyler
,
G.A.
,
Higgins
,
G.A.
,
Hart
,
R.A.
,
Battenberg
,
E.
,
Billingsley
,
M.
,
Bloom
,
F.E.
et al. 
(
1989
)
The identification of a novel synaptosomal-associated protein, SNAP-25, differentially expressed by neuronal subpopulations
.
J. Cell Biol.
109
,
3039
3052
doi:
22
Bennett
,
M.K.
,
Calakos
,
N.
and
Scheller
,
R.H.
(
1992
)
Syntaxin: a synaptic protein implicated in docking of synaptic vesicles at presynaptic active zones
.
Science
257
,
255
259
doi:
23
Pocklington
,
A.J.
,
Armstrong
,
J.D.
and
Grant
,
S.G.N.
(
2006
)
Organization of brain complexity—synapse proteome form and function
.
Brief. Funct. Genomic. Proteomic.
5
,
66
73
doi:
24
Südhof
,
T.C.
(
2004
)
The synaptic vesicle cycle
.
Annu. Rev. Neurosci
27
,
509
547
doi:
25
Rizzoli
,
S.O.
(
2014
)
Synaptic vesicle recycling: steps and principles
.
EMBO J.
33
,
788
822
doi:
26
Rizzoli
,
S.O.
and
Betz
,
W.J.
(
2005
)
Synaptic vesicle pools
.
Nat. Rev. Neurosci.
6
,
57
69
doi:
27
Denker
,
A.
and
Rizzoli
,
S.O.
(
2010
)
Synaptic vesicle pools: an update
.
Front. Synaptic Neurosci.
2
,
135
doi:
28
Denker
,
A.
,
Bethani
,
I.
,
Kröhnert
,
K.
,
Körber
,
C.
,
Horstmann
,
H.
,
Wilhelm
,
B.G.
et al. 
(
2011
)
A small pool of vesicles maintains synaptic activity in vivo
.
Proc. Natl Acad. Sci. USA
108
,
17177
17182
doi:
29
Takamori
,
S.
,
Holt
,
M.
,
Stenius
,
K.
,
Lemke
,
E.A.
,
Grønborg
,
M.
,
Riedel
,
D.
et al. 
(
2006
)
Molecular anatomy of a trafficking organelle
.
Cell
127
,
831
846
doi:
30
Südhof
,
T.C.
(
2006
)
Synaptic vesicles: an organelle comes of age
.
Cell
127
,
671
673
doi:
31
Hell
,
S.W.
and
Wichmann
,
J.
(
1994
)
Breaking the diffraction resolution limit by stimulated emission: stimulated-emission-depletion fluorescence microscopy
.
Opt. Lett.
19
,
780
782
doi:
32
Betzig
,
E.
,
Patterson
,
G.H.
,
Sougrat
,
R.
,
Lindwasser
,
O.W.
,
Olenych
,
S.
,
Bonifacino
,
J.S.
et al. 
(
2006
)
Imaging intracellular fluorescent proteins at nanometer resolution
.
Science
313
,
1642
1645
doi:
33
Rust
,
M.J.
,
Bates
,
M.
and
Zhuang
,
X.
(
2006
)
Sub-diffraction-limit imaging by stochastic optical reconstruction microscopy (STORM
).
Nat. Methods
3
,
793
795
doi:
34
Gustafsson
,
M.G.L.
(
2005
)
Nonlinear structured-illumination microscopy: wide-field fluorescence imaging with theoretically unlimited resolution
.
Proc. Natl Acad. Sci. USA
102
,
13081
13086
doi:
35
Huang
,
B.
,
Bates
,
M.
and
Zhuang
,
X.
(
2009
)
Super-resolution fluorescence microscopy
.
Annu. Rev. Biochem.
78
,
993
1016
doi:
36
Fornasiero
,
E.F.
and
Opazo
,
F.
(
2015
)
Super-resolution imaging for cell biologists: concepts, applications, current challenges and developments
.
Bioessays
37
,
436
451
doi:
37
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
doi:
38
Sigrist
,
S.J.
and
Sabatini
,
B.L.
(
2012
)
Optical super-resolution microscopy in neurobiology
.
Curr. Opin. Neurobiol.
22
,
86
93
doi:
39
Schermelleh
,
L.
,
Heintzmann
,
R.
and
Leonhardt
,
H.
(
2010
)
A guide to super-resolution fluorescence microscopy
.
J. Cell Biol.
190
,
165
175
doi:
40
Harke
,
B.
,
Keller
,
J.
,
Ullal
,
C.K.
,
Westphal
,
V.
,
Schönle
,
A.
and
Hell
,
S.W.
(
2008
)
Resolution scaling in STED microscopy
.
Opt. Express
16
,
4154
4162
doi:
41
Göttfert
,
F.
,
Wurm
,
C.A.
,
Mueller
,
V.
,
Berning
,
S.
,
Cordes
,
V.C.
,
Honigmann
,
A.
et al. 
(
2013
)
Coaligned dual-channel STED nanoscopy and molecular diffusion analysis at 20 nm resolution
.
Biophys. J.
105
,
L01
L03
doi:
42
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
doi:
43
Willig
,
K.I.
,
Rizzoli
,
S.O.
,
Westphal
,
V.
,
Jahn
,
R.
and
Hell
,
S.W.
(
2006
)
STED microscopy reveals that synaptotagmin remains clustered after synaptic vesicle exocytosis
.
Nature
440
,
935
939
doi:
44
Westphal
,
V.
,
Rizzoli
,
S.O.
,
Lauterbach
,
M.A.
,
Kamin
,
D.
,
Jahn
,
R.
and
Hell
,
S.W.
(
2008
)
Video-rate far-field optical nanoscopy dissects synaptic vesicle movement
.
Science
320
,
246
249
doi:
45
Kellner
,
R.R.
,
Baier
,
C.J.
,
Willig
,
K.I.
,
Hell
,
S.W.
and
Barrantes
,
F.J.
(
2007
)
Nanoscale organization of nicotinic acetylcholine receptors revealed by stimulated emission depletion microscopy
.
Neuroscience
144
,
135
143
doi:
46
Lang
,
T.
and
Rizzoli
,
S.O.
(
2010
)
Membrane protein clusters at nanoscale resolution: more than pretty pictures
.
Physiology
25
,
116
124
doi:
47
Kamin
,
D.
,
Lauterbach
,
M.A.
,
Westphal
,
V.
,
Keller
,
J.
,
Schönle
,
A.
,
Hell
,
S.W.
et al. 
(
2010
)
High- and low-mobility stages in the synaptic vesicle cycle
.
Biophys. J.
99
,
675
684
doi:
48
Nair
,
D.
,
Hosy
,
E.
,
Petersen
,
J.D.
,
Constals
,
A.
,
Giannone
,
G.
,
Choquet
,
D.
et al. 
(
2013
)
Super-resolution imaging reveals that AMPA receptors inside synapses are dynamically organized in nanodomains regulated by PSD95
.
J. Neurosci.
33
,
13204
13224
doi:
49
Milovanovic
,
D.
and
Jahn
,
R.
(
2015
)
Organization and dynamics of SNARE proteins in the presynaptic membrane
.
Front. Physiol.
6
,
89
doi:
50
Xu
,
K.
,
Zhong
,
G.
and
Zhuang
,
X.
(
2013
)
Actin, spectrin, and associated proteins form a periodic cytoskeletal structure in axons
.
Science
339
,
452
456
doi:
51
Willig
,
K.I.
,
Steffens
,
H.
,
Gregor
,
C.
,
Herholt
,
A.
,
Rossner
,
M.J.
and
Hell
,
S.W.
(
2014
)
Nanoscopy of filamentous actin in cortical dendrites of a living mouse
.
Biophys. J.
106
,
L01
L03
doi:
52
Wilhelm
,
B.G.
,
Mandad
,
S.
,
Truckenbrodt
,
S.
,
Kröhnert
,
K.
,
Schäfer
,
C.
,
Rammner
,
B.
et al. 
(
2014
)
Composition of isolated synaptic boutons reveals the amounts of vesicle trafficking proteins
.
Science
344
,
1023
1028
doi:
53
Leterrier
,
C.
,
Potier
,
J.
,
Caillol
,
G.
,
Debarnot
,
C.
,
Rueda Boroni
,
F.
and
Dargent
,
B.
(
2015
)
Nanoscale architecture of the axon initial segment reveals an organized and robust
scaffold
.
Cell Rep.
13
,
2781
2793
doi:
54
Dani
,
A.
,
Huang
,
B.
,
Bergan
,
J.
,
Dulac
,
C.
and
Zhuang
,
X.
(
2010
)
superresolution imaging of chemical synapses in the brain
.
Neuron
68
,
843
856
doi:
55
Chalfie
,
M.
,
Tu
,
Y.
,
Euskirchen
,
G.
,
Ward
,
W.
and
Prasher
,
D.
(
1994
)
Green fluorescent protein as a marker for gene expression
.
Science
263
,
802
805
doi:
56
Dean
,
K.M.
and
Palmer
,
A.E.
(
2014
)
Advances in fluorescence labeling strategies for dynamic cellular imaging
.
Nat. Chem. Biol.
10
,
512
523
doi:
57
Karra
,
D.
and
Dahm
,
R.
(
2010
)
Transfection techniques for neuronal cells
.
J. Neurosci.
30
,
6171
6177
doi:
58
Krasowska
,
J.
,
Olasek
,
M.
,
Bzowska
,
A.
,
Clark
,
P.L.
and
Wielgus-Kutrowska
,
B.
(
2010
)
The comparison of aggregation and folding of enhanced green fluorescent protein (EGFP) by spectroscopic studies
.
Spectroscopy
24
,
343
348
doi:
59
Opazo
,
F.
and
Rizzoli
,
S.O.
(
2010
)
The fate of synaptic vesicle components upon fusion
.
Commun. Integr. Biol.
3
,
427
429
doi:
60
Jinek
,
M.
,
Chylinski
,
K.
,
Fonfara
,
I.
,
Hauer
,
M.
,
Doudna
,
J.A.
and
Charpentier
,
E.
(
2012
)
A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity
.
Science
337
,
816
821
doi:
61
Kumari
,
S.
,
Borroni
,
V.
,
Chaudhry
,
A.
,
Chanda
,
B.
,
Massol
,
R.
,
Mayor
,
S.
et al. 
(
2008
)
Nicotinic acetylcholine receptor is internalized via a Rac-dependent, dynamin-independent endocytic pathway
.
J. Cell Biol.
181
,
1179
1193
doi:
62
Herschman
,
H.R.
,
Simpson
,
D.L.
and
Cawley
,
D.B.
(
1982
)
Toxic ligand conjugates as tools in the study of receptor-ligand interactions
.
J. Cell. Biochem.
20
,
163
176
doi:
63
D'Este
,
E.
,
Kamin
,
D.
,
Göttfert
,
F.
,
El-Hady
,
A.
and
Hell
,
S.
(
2015
)
STED nanoscopy reveals the ubiquity of subcortical cytoskeleton periodicity in living neurons
.
Cell Rep.
10
,
1246
1251
doi:
64
Marx
,
V.
(
2013
)
Finding the right antibody for the job
.
Nat. Methods
10
,
703
707
doi:
65
Köhler
,
G.
and
Milstein
,
C.
(
1975
)
Continuous cultures of fused cells secreting antibody of predefined specificity
.
Nature
256
,
495
497
doi:
66
Baker
,
M.
(
2015
)
Reproducibility crisis: Blame it on the antibodies
.
Nature
May 21
,
274
276
67
Bradbury
,
A.
and
Plückthun
,
A.
(
2015
)
Standardize antibodies used in research
.
Nature
518
,
27
29
doi:
68
Opazo
,
F.
,
Levy
,
M.
,
Byrom
,
M.
,
Schäfer
,
C.
,
Geisler
,
C.
,
Groemer
,
T.W.
et al. 
(
2012
)
Aptamers as potential tools for super-resolution microscopy
.
Nat. Methods
9
,
938
939
doi:
69
Mikhaylova
,
M.
,
Cloin
,
B.M.C.
,
Finan
,
K.
,
van den Berg
,
R.
,
Teeuw
,
J.
,
Kijanka
,
M.M.
et al. 
(
2015
)
Resolving bundled microtubules using anti-tubulin nanobodies
.
Nat. Commun.
6
,
7933
doi:
70
Harris
,
L.J.
,
Skaletsky
,
E.
and
McPherson
,
A.
(
1998
)
Crystallographic structure of an intact IgG1 monoclonal antibody
.
J. Mol. Biol.
275
,
861
872
doi:
71
Ries
,
J.
,
Kaplan
,
C.
,
Platonova
,
E.
,
Eghlidi
,
H.
and
Ewers
,
H.
(
2012
)
A simple, versatile method for GFP-based super-resolution microscopy via nanobodies
.
Nat. Methods
9
,
582
584
doi:
72
Kent
,
S.P.
,
Ryan
,
K.H.
and
Siegel
,
A.L.
(
1978
)
Steric hindrance as a factor in the reaction of labeled antibody with cell surface antigenic determinants
.
J. Histochem. Cytochem.
26
,
618
621
doi:
73
Sauer
,
M.
(
2013
)
Localization microscopy coming of age: from concepts to biological impact
.
J. Cell Sci.
126
,
3505
3513
doi:
74
Škrlec
,
K.
,
Štrukelj
,
B.
and
Berlec
,
A.
(
2015
)
Non-immunoglobulin scaffolds: a focus on their targets
.
Trends Biotechnol.
33
,
408
418
doi:
75
Löfblom
,
J.
,
Feldwisch
,
J.
,
Tolmachev
,
V.
,
Carlsson
,
J.
,
Ståhl
,
S.
and
Frejd
,
F.Y.
(
2010
)
Affibody molecules: engineered proteins for therapeutic, diagnostic and biotechnological applications
.
FEBS Lett.
584
,
2670
2680
doi:
76
Gebauer
,
M.
and
Skerra
,
A.
(
2009
)
Engineered protein scaffolds as next-generation antibody therapeutics
.
Curr. Opin. Chem. Biol.
13
,
245
255
doi:
77
Binz
,
H.K.
,
Stumpp
,
M.T.
,
Forrer
,
P.
,
Amstutz
,
P.
and
Plückthun
,
A.
(
2003
)
Designing repeat proteins: well-expressed, soluble and stable proteins from combinatorial libraries of consensus ankyrin repeat proteins
.
J. Mol. Biol.
332
,
489
503
doi:
78
Lakhin
,
V.
,
Tarantul
,
V.Z.
and
Gening
,
L.V.
(
2013
)
Aptamers: problems, solutions and prospects
.
Acta Nat.
5
,
34
43
PMID:
[PubMed]
79
Rothe
,
A.
,
Hosse
,
R.J.
and
Power
,
B.E.
(
2006
)
In vitro display technologies reveal novel biopharmaceutics
.
FASEB J.
20
,
1599
1610
doi:
80
Carmen
,
S.
and
Jermutus
,
L.
(
2002
)
Concepts in antibody phage display
.
Brief. Funct. Genomic. Proteomic.
1
,
189
203
doi:
81
Plückthun
,
A.
(
2012
)
Ribosome display: a perspective
.
Methods Mol. Biol.
805
,
3
28
doi:
82
Fridy
,
P.C.
,
Li
,
Y.
,
Keegan
,
S.
,
Thompson
,
M.K.
,
Nudelman
,
I.
,
Scheid
,
J.F.
et al. 
(
2014
)
A robust pipeline for rapid production of versatile nanobody repertoires
.
Nat. Methods
11
,
1253
1260
doi:
83
Pardon
,
E.
,
Laeremans
,
T.
,
Triest
,
S.
,
Rasmussen
,
S.G.F.
,
Wohlkönig
,
A.
,
Ruf
,
A.
et al. 
(
2014
)
A general protocol for the generation of nanobodies for structural biology
.
Nat. Protoc.
9
,
674
693
doi:
84
Holliger
,
P.
and
Hudson
,
P.J.
(
2005
)
Engineered antibody fragments and the rise of single domains
.
Nat. Biotechnol.
23
,
1126
1136
doi:
85
Hudson
,
P.J.
and
Souriau
,
C.
(
2003
)
Engineered antibodies
.
Nat. Med.
9
,
129
134
doi:
86
Muyldermans
,
S.
(
2013
)
Nanobodies: natural single-Domain antibodies
.
Annu. Rev. Biochem.
82
,
775
797
doi:
87
Krehenbrink
,
M.
,
Chami
,
M.
,
Guilvout
,
I.
,
Alzari
,
P.M.
,
Pécorari
,
F.
and
Pugsley
,
A.P.
(
2008
)
Artificial binding proteins (Affitins) as probes for conformational changes in secretin PulD
.
J. Mol. Biol.
383
,
1058
1068
doi:
88
Desmet
,
J.
,
Verstraete
,
K.
,
Bloch
,
Y.
,
Lorent
,
E.
,
Wen
,
Y.
,
Devreese
,
B.
et al. 
(
2014
)
Structural basis of IL-23 antagonism by an Alphabody protein scaffold
.
Nat. Commun.
5
,
5237
doi:
89
Desmet
,
J.
and
Lasters
,
I.
(
2015
)
Non-natural proteinaceous scaffold made of three non-covalently associated peptides. U.S. Pat. US9217011 B2, December 22
.
90
Bennett
,
V.
and
Stenbuck
,
P.
(
1979
)
Identification and partial purification of ankyrin, the high affinity membrane attachment site for human erythrocyte spectrin
.
J. Biol. Chem.
254
,
2533
2541
[PubMed]
91
Varadamsetty
,
G.
,
Tremmel
,
D.
,
Hansen
,
S.
,
Parmeggiani
,
F.
and
Plückthun
,
A.
(
2012
)
Designed armadillo repeat proteins: library generation, characterization and selection of peptide binders with high specificity
.
J. Mol. Biol.
424
,
68
87
doi:
92
Sidhu
,
S.S.
,
Li
,
B.
,
Chen
,
Y.
,
Fellouse
,
F.A
,
Eigenbrot
,
C.
and
Fuh
,
G.
(
2004
)
Phage-displayed antibody libraries of synthetic heavy chain complementarity determining regions
.
J. Mol. Biol.
338
,
299
310
doi:
93
Nelson
,
A.L.
(
2010
)
Antibody fragments: hope and hype
.
MAbs
2
,
77
83
PMID:
[PubMed]
94
Dimitrov
,
A.
,
Quesnoit
,
M.
,
Moutel
,
S.
,
Cantaloube
,
I.
,
Poüs
,
C.
and
Perez
,
F.
(
2008
)
Detection of GTP-tubulin conformation in vivo reveals a role for GTP remnants in microtubule rescues
.
Science
322
,
1353
1356
doi:
95
Nizak
,
C.
,
Martin-Lluesma
,
S.
,
Moutel
,
S.
,
Roux
,
A.
,
Kreis
,
T.E.
,
Goud
,
B.
et al. 
(
2003
)
Recombinant antibodies against subcellular fractions used to track endogenous Golgi protein dynamics in vivo
.
Traffic
4
,
739
753
doi:
96
de Marco
,
A.
(
2009
)
Strategies for successful recombinant expression of disulfide bond-dependent proteins in Escherichia coli
.
Microb. Cell Fact.
8
,
26
doi:
97
Olichon
,
A.
and
Surrey
,
T.
(
2007
)
Selection of genetically encoded fluorescent single domain antibodies engineered for efficient expression in Escherichia coli
.
J. Biol. Chem.
282
,
36314
36320
doi:
98
Hamers-Casterman
,
C.
,
Atarhouch
,
S.
,
Muyldermans
,
S.
,
Robinson
,
G.
,
Hammers
,
C.
,
Songa
,
E.B.
et al. 
(
1993
)
Naturally occurring antibodies devoid of light chains
.
Nature
363
,
446
448
doi:
99
Greenberg
,
A.S.
,
Avila
,
D.
,
Hughes
,
M.
,
McKinney
,
E.C.
and
Flajnik
,
M.F.
(
1995
)
A new antigen receptor gene family that undergoes rearrangement and extensive somatic diversification in sharks
.
Nature
374
,
168
173
doi:
100
Nguyen
,
V.
,
Su
,
C.
,
Muyldermans
,
S.
and
van der Loo
,
W.
(
2002
)
Heavy-chain antibodies in Camelidae; a case of evolutionary innovation
.
Immunogenetics
54
,
39
47
doi:
101
Bond
,
C.J.
,
Marsters
,
J.C.
and
Sidhu
,
S.S.
(
2003
)
Contributions of CDR3 to VHH domain stability and the design of monobody scaffolds for naive antibody libraries
.
J. Mol. Biol.
332
,
643
655
doi:
102
Vincke
,
C.
and
Muyldermans
,
S.
(
2012
)
Single Domain Antibodies.
Methods in Molecular Biology 911
(
Saerens
,
D.
and
Muyldermans
,
S.
, eds.), pp.
15
26
,
Humana Press
,
Totowa, NJ
103
Braun
,
M.B.
,
Traenkle
,
B.
,
Koch
,
P.A.
,
Emele
,
F.
,
Weiss
,
F.
,
Poetz
,
O.
et al. 
(
2016
)
Peptides in headlock—a novel high-affinity and versatile peptide-binding nanobody for proteomics and microscopy
.
Sci. Rep.
6
,
19211
doi:
104
Arbabi Ghahroudi
,
M.
,
Desmyter
,
A.
,
Wyns
,
L.
,
Hamers
,
R.
and
Muyldermans
,
S.
(
1997
)
Selection and identification of single domain antibody fragments from camel heavy-chain antibodies
.
FEBS Lett.
414
,
521
526
doi:
105
Guilliams
,
T.
,
El-Turk
,
F.
,
Buell
,
A.K.
,
O'Day
,
E.M.
,
Aprile
,
F.A
,
Esbjörner
,
E.K.
et al. 
(
2013
)
Nanobodies raised against monomeric α-synuclein distinguish between fibrils at different maturation stages
.
J. Mol. Biol.
425
,
2397
2411
doi:
106
Olichon
,
A.
,
Marco
,
A.
,
DeVincke
,
C.
and
Muyldermans
,
S.
(
2012
)
Single Domain Antibodies.
Methods in Molecular Biology 911
(
Saerens
,
D.
and
Muyldermans
,
S.
, eds.), pp.
65
78
,
Humana Press
,
Totowa, NJ
107
Kaplan
,
C.
and
Ewers
,
H.
(
2015
)
Optimized sample preparation for single-molecule localization-based superresolution microscopy in yeast
.
Nat. Protoc.
10
,
1007
1021
doi:
108
Rothbauer
,
U.
,
Zolghadr
,
K.
,
Tillib
,
S.
,
Nowak
,
D.
,
Schermelleh
,
L.
,
Gahl
,
A.
et al. 
(
2006
)
Targeting and tracing antigens in live cells with fluorescent nanobodies
.
Nat. Methods
3
,
887
889
PMID:
17060912
109
Kirchhofer
,
A.
,
Helma
,
J.
,
Schmidthals
,
K.
,
Frauer
,
C.
,
Cui
,
S.
,
Karcher
,
A.
et al. 
(
2010
)
Modulation of protein properties in living cells using nanobodies
.
Nat. Struct. Mol. Biol.
17
,
133
138
doi:
110
Hassanzadeh-Ghassabeh
,
G.
,
Devoogdt
,
N.
,
De Pauw
,
P.
,
Vincke
,
C.
and
Muyldermans
,
S.
(
2013
)
Nanobodies and their potential applications
.
Nanomedicine (Lond)
8
,
1013
1026
PMID:
[PubMed]
111
Jungmann
,
R.
,
Avendaño
,
M.S.
,
Dai
,
M.
,
Woehrstein
,
J.B.
,
Agasti
,
S.S.
,
Feiger
,
Z.
et al. 
(
2016
)
Quantitative super-resolution imaging with qPAINT
.
Nat. Methods
13
,
439
442
doi:
112
Ta
,
H.
,
Keller
,
J.
,
Haltmeier
,
M.
,
Saka
,
S.K.
,
Schmied
,
J.
,
Opazo
,
F.
et al. 
(
2015
)
Mapping molecules in scanning far-field fluorescence nanoscopy
.
Nat. Commun.
6
,
7977
doi:
113
Boxer
,
S.G.
,
Kraft
,
M.L.
and
Weber
,
P.K.
(
2009
)
Advances in imaging secondary ion mass spectrometry for biological samples
.
Annu. Rev. Biophys.
38
,
53
74
doi:
114
Klitzing
,
H.A.
,
Weber
,
P.K
. and
Kraft
,
M.L.
(
2013
)
Nanoimaging
950
,
483
501
PMID:
[PubMed]
115
Saka
,
S.K.
,
Vogts
,
A.
,
Kröhnert
,
K.
,
Hillion
,
F.
,
Rizzoli
,
S.O.
and
Wessels
,
J.T.
(
2014
)
Correlated optical and isotopic nanoscopy
.
Nat. Commun.
5
,
3664
PMID:
[PubMed]
116
Vreja
,
I.C.
,
Kabatas
,
S.
,
Saka
,
S.K.
,
Kröhnert
,
K.
,
Höschen
,
C.
,
Opazo
,
F.
et al. 
(
2015
)
Secondary-ion mass spectrometry of genetically encoded targets
.
Angew. Chem. Int. Ed. Engl.
54
,
5784
5788
PMID:
[PubMed]
117
McDonnell
,
L.
and
Heeren
,
R.M.
(
2007
)
Imaging mass spectrometry
.
Mass Spectrom. Rev.
606
643
doi:
118
Römpp
,
A.
and
Spengler
,
B.
(
2013
)
Mass spectrometry imaging with high resolution in mass and space
.
Histochem. Cell Biol.
139
,
759
783
doi:
119
Sharonov
,
A.
and
Hochstrasser
,
R.M.
(
2006
)
Wide-field subdiffraction imaging by accumulated binding of diffusing probes
.
Proc. Natl Acad. Sci. USA
103
,
18911
18916
doi:
120
Kiuchi
,
T.
,
Higuchi
,
M.
,
Takamura
,
A.
,
Maruoka
,
M.
and
Watanabe
,
N
. (
2015
)
Multitarget super-resolution microscopy with high-density labeling by exchangeable probes
.
Nat. Methods
12
,
743
746
PMID:
[PubMed]
121
Chen
,
F.
,
Tillberg
,
P.W.
and
Boyden
,
E.S.
(
2015
)
Optical imaging. Expansion microscopy
.
Science
347
(
6221
),
543
548
doi: