The explosive growth of single-molecule techniques is transforming our understanding of biology, helping to develop new physics inspired by emergent biological processes, and leading to emerging areas of nanotechnology. Key biological and chemical processes can now be probed with new levels of detail, one molecule at a time, from the nanoscopic dynamics of nature's molecular machines to an ever-expanding range of exciting applications across multiple length and time scales. Their common feature is an ability to render the underlying distribution of molecular properties that ensemble averaging masks and to reveal new insights into complex systems containing spatial and temporal heterogeneity. Small fluorescent probes are among the most adaptable and versatile for single-molecule sensing applications because they provide high signal-to-noise ratios combined with excellent specificity of labeling when chemically attached to target biomolecules or embedded within a host material. In this review, we examine recent advances in probe designs, their utility, and applications and provide a practical guide to their use, focusing on the single-molecule detection of nucleic acids, proteins, carbohydrates, and membrane dynamics. We also present key challenges that must be overcome to perform successful single-molecule experiments, including probe conjugation strategies, identify tradeoffs and limitations for each probe design, showcase emerging applications, and discuss exciting future directions for the community.

Single-molecule sensing techniques are revolutionizing our understanding of biological systems by enabling the molecular building blocks of life to be studied with extraordinary levels of detail,1 new soft-matter physics relevant to complex processes to be explored, and new physics theories to be developed.2 The last several decades have witnessed an explosive growth in the use of small nanoscale fluorescent probes for investigating biomolecular structure and function under a huge range of experimental conditions. It is now abundantly clear that a capability to measure fluorescence intensity, absorption, quantum yield, spectrum, lifetime, correlation time, and anisotropy from single probes in situ, in vitro, and in vivo, with associated growth in new analytical tools,3 enables researchers to access population distributions that are otherwise hidden by the ensemble average. While x-ray crystallography and electron microscopy tools have, for example, provided important structural details of biomolecular systems, they lack the ability to follow the time-sequence of corresponding dynamics and often miss critical interactions or transient conformations because of the requirement for static, frozen, or powdered samples. Fully characterizing biological dynamics and accessing micro-environmental distributions thus requires an ability to follow individual interactions without averaging over all steps in the process, enabling the study of the physics of life in effect one molecule at a time.4 In this respect, the emergence of single-molecule fluorescence methods has led to transformative insights into cases where static and/or dynamic heterogeneity is present, such as in a biological machine whose properties continuously and dynamically alter over multiple time scales. For all of nature's biomolecules, a variety of chemical and physical events trigger time-dependent conformational switching and interactions, and accessing how they dynamically operate is extremely attractive for researchers across the physical-life sciences interface.

Although some biomolecules contain weakly fluorescing units, for example in proteins the aromatic amino acids (tryptophan, tyrosine and phenylalanine), advances in modern synthesis techniques have enabled highly emissive fluorescent probes to be chemically tagged to biomolecular structures with high specificity. Since the earliest detection of single pentacene molecules at low temperatures,5,6 techniques which enable fluorescence detection from small probes tagged to target biomolecules under physiological environments have developed rapidly and been applied extensively, as approximated by the abundance of research papers with “single-molecule fluorescence” in the article. Although this is clearly a simple analysis, the number of such papers per year, shown in Fig. 1(a), demonstrates that the field has more than doubled in size since a similar analysis was carried out over a decade ago.7 

FIG. 1.

Small molecule fluorescent probes enable quantification of biological interactions and dynamics. (a) The number of papers indexed in the PubMed database with single molecule fluorescence in the article illustrates the growth of the field. [(b)–(j)] Single-molecule fluorescence techniques provide quantitative information on a wide range of parameters. For example, (b) by following raw image acquisition from small molecule fluorescent probes, the intensity profile of each molecule may be modeled to reveal the probe's location with nanometer precision if the number of emitted photons per probe is typically >107. (c) The binding of a protein to a small molecule probe may in some cases lead to fluorescence enhancement. The protein-induced enhancement is a particularly powerful approach for probing protein dynamics upon interaction with nucleic acids. (d) By capturing images as a function of time, 2D correlation maps documenting all detected probe locations can be used for particle tracking. (e) Under continuous excitation, each fluorescently tagged sub-unit within a complex photobleaches causing a stepwise fluorescence decrease in approximately equal magnitude steps. The number of active fluorophores, and therefore number of photobleaching steps, yields the complex stoichiometry. (f) Anti-correlations in the fluorescence time traces of donor- (green) and acceptor- (red) labeled complexes enable nanoscale molecular dynamics to be accessed via FRET. (g) Fluorescence decay curves obtained via time-correlated single photon counting techniques reveal lifetime distributions and interconversion rates reflecting biomolecular conformations and environments. (h) Fluorescence emission spectra of various probes shift in response to changes in the local biomolecular environment, enabling accurate determination of, for example, local ion concentrations. (i) Polarization microscopy enables rotational and orientational behaviors to be extracted from single probes. Here, the orientation of a probe's dipole moment rotating about an axis can be determined; the dipole is only efficiently excited when it rotates through the axis of the excitation polarization. (j) FRET analysis captures the structure of single biomolecules, with distance errors reported to be as low as only a few percent.

FIG. 1.

Small molecule fluorescent probes enable quantification of biological interactions and dynamics. (a) The number of papers indexed in the PubMed database with single molecule fluorescence in the article illustrates the growth of the field. [(b)–(j)] Single-molecule fluorescence techniques provide quantitative information on a wide range of parameters. For example, (b) by following raw image acquisition from small molecule fluorescent probes, the intensity profile of each molecule may be modeled to reveal the probe's location with nanometer precision if the number of emitted photons per probe is typically >107. (c) The binding of a protein to a small molecule probe may in some cases lead to fluorescence enhancement. The protein-induced enhancement is a particularly powerful approach for probing protein dynamics upon interaction with nucleic acids. (d) By capturing images as a function of time, 2D correlation maps documenting all detected probe locations can be used for particle tracking. (e) Under continuous excitation, each fluorescently tagged sub-unit within a complex photobleaches causing a stepwise fluorescence decrease in approximately equal magnitude steps. The number of active fluorophores, and therefore number of photobleaching steps, yields the complex stoichiometry. (f) Anti-correlations in the fluorescence time traces of donor- (green) and acceptor- (red) labeled complexes enable nanoscale molecular dynamics to be accessed via FRET. (g) Fluorescence decay curves obtained via time-correlated single photon counting techniques reveal lifetime distributions and interconversion rates reflecting biomolecular conformations and environments. (h) Fluorescence emission spectra of various probes shift in response to changes in the local biomolecular environment, enabling accurate determination of, for example, local ion concentrations. (i) Polarization microscopy enables rotational and orientational behaviors to be extracted from single probes. Here, the orientation of a probe's dipole moment rotating about an axis can be determined; the dipole is only efficiently excited when it rotates through the axis of the excitation polarization. (j) FRET analysis captures the structure of single biomolecules, with distance errors reported to be as low as only a few percent.

Close modal

The visualization of these single-molecule fluorescent probes has primarily enabled the localization and diffusion of single biomolecules to be observed directly,8 but more complex interactions such as protein dynamics,9–11 folding kinetics,12–14 and stoichiometry and kinetics of functional enzymes and molecular machines inside living cells,15–18 can also now be followed through changes to their spectroscopic fingerprints. Their utility has also extended to environmental sensing, enabling the organization and architecture of lipid membranes19–21 and signaling complexes on the surfaces of living cells22,23 to be explored. In addition to aiding visualization, these probes have become indispensable to the modern researcher because they also provide dynamic information concerning the quantity of the localized biomolecule in diffraction-limited volumes.

Another powerful application rests in Förster resonance energy transfer (FRET) experiments,24 where donor and acceptor probes are tagged to key molecular components of a biological system, and changes in their fluorescence properties (intensity and lifetime) report quantitatively on their separation distance with a ∼1–10 nm sensitivity.25 Widely considered as a spectroscopic nano-ruler, FRET has become a popular workhorse technique for identifying and characterizing real-time conformational changes within single nucleic acids26–28 proteins29 and enzymes,30 and recent probe developments25 have seen it combined with integrative modeling, giving rise to a new field of quantitative structural biology.31–33 

Fluorescent probes have also found utility in experiments tailored toward sub-millisecond temporal resolution,34 and they are providing new opportunities for linking heterogeneous transfer dynamics with thermal fluctuations in biological structures.35 Another interesting application is the real-time measurement of orientations of single molecules using polarized emission. Here, the probes emit polarized light along the axis of their transition dipole moments and if, for example, the polarized fluorescence emission intensity is measured as a function of the excitation polarization, then this allows for quantification of the probe's local orientation and rotational characteristics.36 The methodology has already been applied to the understanding of intercalators,37 probing biomolecular conformation under tension,38,39 following rotational dynamics within membranes40 and has facilitated investigations into tilting during processive motility,41 with new methods capable of correlating fluorescence polarization with super-resolved localization precision.42 

As demonstrated in Fig. 1(b), the demand for, and applications of, small fluorescent probes is growing considerably. In nearly all applications, the combination of photon statistics and physical laws quantitatively describes the behavior of single tagged biomolecules at work, enabling researchers to pinpoint key pathways and mechanisms which underpin the physics of life.4 This ability undoubtedly places single-molecule fluorescence techniques at the forefront of the physical-life sciences interface.43 

This review highlights advances in the development of fluorescent probes for single-molecule imaging and spectroscopy applications, focusing on those that can resolve spatial and temporal dynamics of individual biomolecules at work. As the number of successful fluorescent reporters increases, several design trends and considerations are becoming apparent. We highlight the most robust and adaptable of these designs and showcase their utility in the context of single-molecule fluorescence experiments.

Single-molecule fluorescence studies generally make use of extrinsic probes that are either specifically linked to a target biomolecule, nonspecifically intercalated into its structure, or embedded within a host matrix. They span the visible spectrum, as well as extending into the ultraviolet and near-infrared, and can be coupled to almost any biomolecule of interest, from nucleic acids and proteins, to carbohydrates, and lipids. A huge variety of probes now exist, and each has its own unique photophysical properties. However, great care must be taken to choose an appropriate label for each application since it is vital to maximize the signal-to-noise ratio detected from each individual molecule. The selection of a suitable probe with robust enough photophysical properties is, thus, a critical first step in the experimental design process.

In general, the probes should satisfy four major conditions. First, they should display excellent photophysics within the biomolecule's local environment. They should be water soluble, have an ability to strongly absorb excitation photons (extinction coefficient, ε, >50 000 M−1 cm−1) and be sufficiently emissive (quantum yield, ϕ, > 0.1) for the desired spatiotemporal resolution.24 Additionally, the probes should not aggregate in solution and remain photoactive across the duration of the measurement time window (photobleaching quantum yield, ϕpb < 10−6–10−7).44,45 Second, they should contain a linker for high specificity or demonstrate ease-of-delivery toward the biomolecule of interest. Third, they must also minimize any putative impairment to biological function due to steric hindrance effects. This is often carefully checked by comparing the activity of the labeled species with its unlabeled counterpart. Finally, and in the context of ratiometric FRET-based measurements, the chosen fluorophores should have large spectral separation between their emissions, have similar quantum yields, and must not exhibit time-dependent spectral shifts or intensity fluctuations.24,46 In this section, we focus the discussion on the most popular small molecule fluorescent probe designs which satisfy these conditions, including recent advances, emerging trends, their use, and relative performance.

Organic dyes are among the smallest in length scale (<1 nm) and most adaptable of all small molecule probe designs. When properly positioned, they are generally considered among the least invasive and perturbing. Their structure facilitates electron delocalization through a conjugated π-electron system, enabling the molecule to act as an efficient electric dipole. Organic dyes used for single-molecule fluorescence studies typically absorb and emit across the visible region of the electromagnetic spectrum; though, due to issues with photostability with dyes that absorb below 450 nm, and limitations regarding detection sensitivity at infrared wavelengths, most single-molecule fluorescence experiments involve organic dyes that cover the 480–750 nm window.47,48

Organic dyes are broadly classified into six major families: the cyanines, oxazines, boron-dipyrromethenes (BODIPYs), perylenes, diketopyrrolopyrroles (DPPs), and xanthenes, with the latter consisting of particularly popular fluorescein and rhodamine derivatives, such as the ATTO dyes.

The cyanines have the general structure shown in Fig. 2(a) and have found most utility for single-molecule applications involving proteins. Their lipophilic properties meant that they were initially employed as membrane stains, but sulfonated indocarbocyanine derivatives including Cy3, Cy5, and Cy7 [Figs. 2(d)–2(f)] are now widely used for biomolecular labeling. The cyanines are named according to the number of carbon atoms between the indoline moieties with the longer polymethine chains corresponding to longer emission wavelengths. In this context, Cy3, Cy5, and Cy7 exhibit peak emission at 568, 652, and 755 nm, respectively. While their quantum yields are typically lower than many other organic dyes (ϕCy3 ∼ 0.31; ϕCy5 ∼ 0.27; ϕCy7 ∼ 0.2), their high extinction coefficients (εCy3 ∼ 150 000 M−1 cm−1; εCy5 ∼ 250 000 M−1 cm−1; εCy7 ∼ 199 000) M−1 cm−1 position them among the brightest.49 

FIG. 2.

Chemical structures of commonly used organic dyes for single-molecule sensing applications. The general chemical structures of (a) cyanines, (b) fluoresceins and rhodamines, and (c) BODIPYs. Also shown are the chemical structures of (d) Cy3, (e) Cy5, (f) Cy7, (g) BDP-FL, (h) BDP-TMR, (i) AF488, (j) JF-549, and (k) AF647. Dotted lines indicate the peak emission wavelength of each probe design.

FIG. 2.

Chemical structures of commonly used organic dyes for single-molecule sensing applications. The general chemical structures of (a) cyanines, (b) fluoresceins and rhodamines, and (c) BODIPYs. Also shown are the chemical structures of (d) Cy3, (e) Cy5, (f) Cy7, (g) BDP-FL, (h) BDP-TMR, (i) AF488, (j) JF-549, and (k) AF647. Dotted lines indicate the peak emission wavelength of each probe design.

Close modal

Longer-chain dialkylcarbocyanines, such as DiO, DiI, and DiD, are often employed as membrane stains for live-cell,50 fixed tissue51 and model-membrane52 imaging. Due to their excellent lipophilic properties, incubation of biological membranes with solutions containing the probes for only a few minutes is sufficient to achieve uniform labeling via lateral diffusion [Fig. 3(a)]. In this way, high probe density is often obtained, with minimal reported effects on cell viability or physiology.50,53 An important point is that they do not tend to transfer from labeled to unlabeled membranes, except in the case of targeted fusion.54 The spectral characteristics of the dialkylcarbocyanines are determined by the heteroatoms in the terminal ring systems and length of the connecting bridge, as opposed to the chain length. DiO, for instance, absorbs strongly at 484 nm with peak emission at 501 nm, while DiI and DiD absorb and emit at 549/644 and 565/665 nm, respectively [Fig. 3(b)]. Much like the short-chain derivatives, the longer chains also have high extinction coefficients (εDiO ∼ 140 000 M−1 cm−1; εDil ∼ 148 000 M−1 cm−1; εDiD ∼ 193 000) and comparable excited state lifetimes (∼1 ns), though their quantum yields tend to be somewhat lower (∼0.07).55,56 Such probes offer a chance to probe the local dynamics and curvature of lipid bilayers via measurement of their rotational time trajectories.40,57 Both Dil and DiD, for example, have two hydrocarbon tails that mimic phospholipid tails, and their transition dipole moments lie along the plane of the membrane; thus, they are excellent candidates for probing rotational motion.

FIG. 3.

Spectral characteristics of lipophilic membrane stains. (a) Schematic illustration of a lipid bilayer stained with the FRET pair DiI and DiD (left panel). The structures of Dil (∼2.1 nm long) and DiD (∼2.0 nm long) (right panel) contain aliphatic tails which partition into the lipid bilayer, leaving the fluorophore on the external leaflet. FRET can occur between lipophilic donors and acceptors within a lipid mixture if their spatial separation is typically <8 nm. Absorption and normalized fluorescence emission spectra of membrane-bound (b) DiO (blue), DiI (orange), DiD (red) and (c) DiA (blue) and 4-Di-10-ASP (cyan).

FIG. 3.

Spectral characteristics of lipophilic membrane stains. (a) Schematic illustration of a lipid bilayer stained with the FRET pair DiI and DiD (left panel). The structures of Dil (∼2.1 nm long) and DiD (∼2.0 nm long) (right panel) contain aliphatic tails which partition into the lipid bilayer, leaving the fluorophore on the external leaflet. FRET can occur between lipophilic donors and acceptors within a lipid mixture if their spatial separation is typically <8 nm. Absorption and normalized fluorescence emission spectra of membrane-bound (b) DiO (blue), DiI (orange), DiD (red) and (c) DiA (blue) and 4-Di-10-ASP (cyan).

Close modal

Unlike the dialkylcarbocyanines, lipophilic aminostyryl probes, such as DiA and 4-Di-10-ASP, undergo substantial spectral shifts when incorporated within a membrane environment, and as such have also been employed for membrane staining, despite comparatively broad absorption and emission spectra [Fig. 3(c)].58 Derivatives including FAST DiI, in which the saturated tails have been replaced with diunsaturated alkyl groups, offer accelerated membrane diffusion and staining, while sulfonated, thiol-reactive variants, including CM-Dil, are suitable for staining after permabilization.59 Through FRET-based experiments, in which energy is transferred from membrane-embedded donors to acceptors [Fig. 3(a)], the incorporation of such dyes into model vesicles at relatively low molar percentages (∼0.1%) has been used to monitor vesicle fusion through lipid mixing assays56,60 and to reveal solubilization mechanisms and kinetics.21,61,62

While the cyanines represent one of the most versatile and adaptable of all small molecule probe designs, they are subject to photo-induced oxidation even in the presence of low levels of oxygen. As such, without the presence of oxygen scavengers in the local fluorophore environment, the dyes are subject to fast photobleaching rates, which often limits their use for long-term imaging.63 

The xanthenes represent another widely used group of probes, with the most popular being fluorescein- and rhodamine-based [Fig. 2(b)]. Synthesis is typically achieved via a simple condensation reaction, though traditional approaches in this regard were compatible only with the simplest functional groups. As a consequence, Pd-catalyzed cross coupling strategies beginning from simple fluorescein emerged, giving rise to a series of organic dyes commonly referred to as the Janelia Fluor (JF) dyes.64 The JF probes contain four-membered azetidine rings and have larger quantum yields and enhanced photostability relative to those reported for classic rhodamines and cyanines.65 JF 549 (εJF 549 = 101 000 M−1 cm−1, ϕJF 549 = 0.88) [Fig. 2(j)] which emits at 571 nm, has, for example, found particular utility in live-cell applications.66,67 Replacing the xanthene oxygen in JF 549 with a quaternary carbon gives rise to JF 648 (εJF 648 = 152 000 M−1 cm−1, ϕJF 648 = 0.54) with emission centered on a wavelength of 631 nm.68 Longer emission wavelengths of 664 nm (for example JF 646; εJF 646 = 152 000 M−1 cm−1, ϕJF 646 = 0.54) are also available through Si-Rhodamine synthesis, while shifts to shorter wavelengths have been achieved by replacing the azetidine group with an oxygen atom.69,70 Much like the cyanines, the xanthenes also suffer from limited photostability, though factors such as solubility, membrane permeability, cell compartmentalization and aggregation are dependent on the specific chemical structure of the probe and must be accounted for on a case-by-case basis.49 Nevertheless, their adaptability and comparable brightness to the cyanines also mean that they are commonplace among single-molecule research labs.

The boron-dipyrromethenes (BODIPYs) represent a relatively new class of probe. Much like the cyanines and xanthenes, they too exhibit narrowband absorption and emission spectra, but in contrast they have significantly higher quantum yields, often approaching unity. For example, the quantum yields of BODIPY 581/591, BDP TR, BDP TMR, and BDP FL are 0.83, 0.90, 0.95, and 0.97, respectively. However, strong intramolecular hydrogen bonding between hydroxyl and formyl groups within the BDP structure introduce backbone rigidity, and this leads to relatively small Stokes shifts of only a few nanometers.71 BODIPY FL [Fig. 2(g)] is a common substitute for fluorescein but with peak absorption and emission at wavelengths of 503 and 509 nm, respectively, only a 6 nm window exists for resolving the excitation and emission. On the other hand, BODIPY TMR [Fig. 2(h)], a derivative synthesized to match tetramethylrhodamine fluorescence, has peak absorption and emission at wavelength of 545 and 570 nm, offering improved flexibility. Despite favorable photophysical properties associated with the free dye in solution, BDP derivatives have reportedly suffered from reduced brightness upon conjugation to proteins, which ultimately places a constraint on single-molecule sensitivity.72 The relatively high hydrophobicity associated with BODIPYs also means care must be taken to ensure minimal nonspecific adhesion.73 Despite these concerns, BODIPYs have found utility in single-molecule applications involving protein folding74,75 and recently, photoswitchable versions have found utility in efficient FRET-based measurements.76–78 

Most organic dyes are available with functional groups for bio-conjugation, and many have been modified to include side chains and/or double bonds at specific locations in order to reduce flexibility, minimize cis-trans isomerization, and enhance the quantum yield. Cy3, for example, is capable of undergoing isomerization around the polymethine group and this can lead to spectral shifting and photoblinking. By incorporating three six-membered rings into its backbone, the derivative (Cy3B) is conformationally locked and exhibits a fourfold relative enhancement in quantum yield.44 Similarly, organic dyes with additional sulfo-groups help improve solubility, while charged sulfonate groups help decrease dye aggregation.46 For these reasons, organic dyes are also interchangeable. Cy5, for instance, is spectrally similar to Alexa Fluor 647 [Fig. 2(k)] but displays poor relative photostability. The incorporation of sulfonic acid groups into the Alexa Fluor 647 structure provides higher levels of solubility by comparison.79 

While the spectral characteristics of many organic dyes are similar, they also have many other unique attributes, and their performance must be carefully scrutinized for each purpose. For example, ATTO647N represents one of the most emissive and photostable red-emitters and has been used to achieve high spatial accuracy in super-resolution and localization experiments.80–82 However, it is comparatively hydrophobic when compared with many other organic dyes and can, depending on the environment, exhibit substantial spectral shifts.24 

Most organic dyes have found utility in protein and nucleic acid labeling, however to date there have been only a handful of reports on the single-molecule detection of carbohydrates. While the imaging of single Alexa Fluor 488 [Fig. 2(i)] labeled heparin sulfate disaccharides encapsulated within lipid vesicles has been reported,83 as has the detection of Alexa Fluor 647 tagged monosaccharides84,85 and Cy7-labeled maltose,86 this is clearly an area which demands further development, not least because carbohydrates underpin a wide-class of vital cellular functions.

A wide range of organic dyes also exist for the non-covalent labeling of DNA molecules. These can be broadly classified into three major classes: groove binders, intercalators, and cationic electrostatic/allosteric binders that bind to the negatively charged phosphate backbone via attractive ionic interactions [Fig. 4(a)].87 In the latter case, positive charges on the probe arise from the existence of an exocyclic ammonium or endocyclic pyridinium moiety. The intercalators, which can be cationic or neutral, bind to the DNA by inserting aromatic groups between adjacent base pairs, with representative examples including ellipticine, proflavine, acridine orange, methylene blue, ethidium bromide, thiazole orange, the SYTOX derivatives, and YOYO-1. In the case of ellipticine, molecular dynamics simulations indicate rapid intercalation timescales [Fig. 4(a)], and in nearly all cases, variations in the absorption and emission properties upon binding have enabled a multitude of imaging-based experiments.

FIG. 4.

Commonly used intercalators for visualization of single nucleic acids. (a) Schematic illustration of different DNA binders. The DNA backbone (PDB ID: 1BNA) and base pairs are shown in blue. Inset: ellipticine (represented by a ball and stick model) intercalates into the DNA bases (represented as rods) over ∼0.8 ns as suggested via targeted molecular dynamics simulations. (b) Chemical structure of YOYO-1. (c) Absorption (solid lines) and fluorescence emission spectra (dashed lines) of YOYO-1 (green) and SYTOX orange (orange). (d) Representative TIRFM images (lower panels) of single λ-DNA molecules immobilized onto a glass coverslip coated in polyethylene glycol (PEG) via biotin-streptavidin interactions in the presence of 1 nM SYTOX Orange under 20 μL/min flow conditions (top panel).

FIG. 4.

Commonly used intercalators for visualization of single nucleic acids. (a) Schematic illustration of different DNA binders. The DNA backbone (PDB ID: 1BNA) and base pairs are shown in blue. Inset: ellipticine (represented by a ball and stick model) intercalates into the DNA bases (represented as rods) over ∼0.8 ns as suggested via targeted molecular dynamics simulations. (b) Chemical structure of YOYO-1. (c) Absorption (solid lines) and fluorescence emission spectra (dashed lines) of YOYO-1 (green) and SYTOX orange (orange). (d) Representative TIRFM images (lower panels) of single λ-DNA molecules immobilized onto a glass coverslip coated in polyethylene glycol (PEG) via biotin-streptavidin interactions in the presence of 1 nM SYTOX Orange under 20 μL/min flow conditions (top panel).

Close modal

Visualization of single double-stranded DNA molecules undergoing conformational changes was initially achieved via fluorescence from 4′,6-diamidino-2-phenylindole (DAPI) upon interaction with adenine–thymine-rich regions.88 In such examples, we note that single intercalators and not visualized per se, but rather the nucleic acids are observed via multiple fluorescing labels. However, relatively low binding affinities and poor photophysical properties motivated the development of a new class of dyes that undergo quantum yield enhancements upon DNA binding. The most widely used of these probes for single-molecule research are YOYO-1 and SYTOX orange (SO).

YOYO-1 [Fig. 4(b)] is a cyanine derivative with peak absorption and emission at 491 and 509 nm, respectively [Fig. 4(c)], and binds to double stranded DNA with high affinity (kD ∼ 5–50 nM).89 Fluorescence enhancements of ∼1000-fold occur upon binding, leading to the high signal-to-noise ratios necessary for nucleic acid detection. YOYO-1 has, for example, helped researchers to identify conformational changes in single DNA molecules during replication,90 and has facilitated quantification of base-pair orientations via polarization microscopy.39 

SO, also a cyanine derivative with peak absorption at 547 nm and emission at 570 nm [Fig. 4(c)], intercalates into double-stranded DNA as a monomer91 and undergoes substantially greater emission intensity enhancements upon binding (>1000-fold). While the chemical structure of SO is proprietary, the reported dissociation constant of ∼10 nM is of similar magnitude to that observed by YOYO-1. The kinetic binding rates are, however, an order of magnitude faster, enabling labeling to be achieved almost immediately.92 As shown in Fig. 4(d), the most versatile application involving intercalators has been the DNA curtains technique,93 where single surface-tethered DNA molecules are stretched under flow containing nM concentrations of intercalator, and visualized by techniques such as total internal reflection fluorescence microscopy (TIRFM). The method allows for the parallel imaging of hundreds of aligned molecules and presents a robust experimental platform from which to investigate a multitude of protein- and enzyme-DNA interactions with millisecond timescale resolution.

The time constant of DNA intercalation depends on the number of intercalating moieties and the overall timescale required to reach equilibrium. The process can involve the insertion of a single moiety per probe, as is the case for ellipticine [Fig. 4(a)], two moieties per probe, such as for YOYO-1 [Fig. 4(b)] or multi-intercalating sub-units. The timescale for reaching the final equilibrium state ranges over six orders of magnitude; though, we note this is likely contingent upon the DNA template used, its structural conformation, the free energy landscape, and the accessibility of intercalation sites.94 The association rates and mechanisms of intercalation vary from probe-to-probe but in general, the association rates of traditional mono- and bi-intercalators, are several orders of magnitude faster than the association rates of unconventional groove binders. The mono-intercalator ethidium bromide displays association kinetics of only a few milliseconds in the ensemble, and bis-intercalators, including YOYO-1, intercalate with a typical time constant of a few seconds. On the other hand, unconventional binders, including actinomycin D, generally display slow association kinetics on the order of several thousand seconds, though destabilization of double stranded DNA by force has been shown to exponentially facilitate the on rates.95 Importantly, fast association rates are characteristic for common mono- and bis-cyanine-based intercalators and therefore one can suppose that structurally similar derivatives will display similar traits. The fast association rates achievable using mono- and bis-intercalators relative to the nominal timescale of typical DNA curtains experiments (∼10–100 s), thus, provide the basis for examining single nucleic acids in vitro. YOYO-1 has also been employed for high precision microscopy of single DNA molecules by utilizing its stochastic, reversible photolinking to generate super-resolved localized data of labeled DNA.96 

Despite indications that YOYO-1 and SO may subtly alter the mechanical and structural properties of DNA upon binding,97,98 other work has established that the persistence length and rigidity remain unaffected.99 In addition, the binding affinities are governed by a strongly tension-dependent but tunable dissociation rate, and optimization of this parameter can reduce the effect of the intercalators on strand separation and enzymatic function.100 

To complement the use of organic dyes for direct and indirect biomolecular labeling, several probes have been developed to sense the biomolecular environment. These include sensors which quantify the presence of metal ions and report the solution pH. Currently, there is no single dye that permits measurement of all environmental parameters directly within a single sample, but combinations of fluorescent organic dyes hold promise in this regard. Generally, these sensors are designed to measure free hydrated ions, while not engaging in competitive exchange.101 The majority of the probes undergo quenching between the metal binding domain and the fluorophore via photoinduced electron transfer (PET) in the absence of the ions, and undergo a fluorescence enhancement during binding due to PET-disruption. Derivatives of fluorescein, including the Zinpyr (ZP) family of dyes, have been employed as sensors of Zn2+ in live cells. ZP1, which contains a di-2-picolylamine Zn2+ chelator and a dichlorofluorescein emitter,102 is an established sensor of metalloneurochemistry,103 but next-generation probes, such as ZP2, ZP3, and ZP4, have since offered a sixfold increase in dynamic range, lower pKa values, simpler synthesis procedures, and enhanced signal-to-noise ratios. ZP1 derivatives, including ZnAF-1F and ZNAF-2F, in which the fluorine at the orthoposition of the phenolic hydroxyl group has been substituted, now offer a 69- and 60-fold fluorescence enhancement, respectively, when fully bound to Zn2+. However, the quantum yields of both probes in the absence of Zn2+ are relatively low (∼0.006), rendering their single-molecule detection challenging.104 

A number of Zn2+ probes have also been designed based on the structures of existing Ca2+ sensors, and of these, FluoZin-3 is one of the most widely used.105 Here, an acetate group on the Ca2+ chelator has reduced affinity for Ca2+, while offering a 200-fold fluorescence enhancement in the presence of Zn2+ (kD ∼ 15 nM). A growing number of similar small molecule sensors have been developed to measure vesicular Zn2+ pools, including Zinquin,106 ZincBy-1,107 SpiroZin1,108 and SpiroZin2,109 though differences in emission stability and nonspecific localization vary from probe-to-probe and must be taken into consideration.110 

In a similar way, the development of Ca2+ indicators111 has led to important insights into signaling pathways.112 Ca2+ sensors typically undergo either a fluorescent enhancement or decrease to reflect changes in the local Ca2+ concentration, though it is worth noting that indicator concentration, cytosolic location and pH may also contribute. These indicators are generally divided into single-color or ratiometric probes based on their response to Ca2+.113,114 Single-color probes, such as Fluo-4, display >100-fold increase in fluorescence intensity at 506 nm, whereas ratiometric indicators exhibit shifts in excitation and/or emission wavelengths upon binding. Consequently, single-color probes are generally used for qualitative estimates of Ca2+ levels. Recently, the Cal-520 probe, which undergoes a twofold increase in brightness when fully bound to Ca2+, has been used as probe protein- and surfactant-induced permeabilization in lipid vesicles and offers a promising alternative.21,115,116 Conversely, when Ca2+ binds to ratiometric indicators, such as Indo-1 and Fura-2, emission enhancements at shorter wavelengths concurrent with emission reductions at longer wavelengths typically occur, facilitating quantitative estimates of Ca2+ molarity. In the case of Fura-2, the requirement to perform alternating excitation at 340/380 nm can hinder data acquisition, though it has a higher dynamic range when compared with Indo-1.

It comes as no surprise that the design, synthesis, and characterization of a wide variety of organic dyes using an assortment of fluorogenic units have preceded a range of single-molecule applications. Yet, while the rhodamine and fluorescein derivatives generally exhibit high quantum yields and good photostability, their biological applications have been limited because their absorption and emission range extend only up to 600 nm. For single-molecule detection in living cells, autofluorescence is substantially reduced at wavelengths >600 nm, and thus red- and near-infrared wavelengths offer an overall improvement to the signal-to-noise ratio. Similarly, although cyanine derivatives have been extensively used, they are only moderately photostable. In this regard, rylene dyes formed via the linkage of naphthalene units in peri-positions, are known to have high quantum yields, some even as high as 0.8–0.9, are generally much more photostable than commercially available cyanines (typically by a factor of ∼100), with uncharged species showing particular promise in this area, and are available with emission >600 nm.117 Historically, a major limitation of rylene dyes for biological applications was their relatively poor solubility, but recently, the introduction of ionic sulfonyl, pyridoxy, polyethylene glycol, and peri-guanidine side groups have helped alleviate this issue and have facilitated a wide-range of single-molecule and live-cell experiments.117–119 Most can also be modified with functional groups for biomolecular labeling. Of particular note, the development of perylene diimides has shown particular promise for imaging membranes, biosensing in vitro, detecting antibodies, monitoring cellular uptake, and detecting gene/drug delivery in living cells.120 While solubility has vastly improved, minimizing undesirable self-aggregation properties of rylene dyes is still an area of concern, though clearly exciting prospects lie ahead for this class of organic probe.

In addition to the rylene dyes, water-soluble dyes containing diketopyrrolopyrroles (DPPs) have also offered attractive properties to the single-molecule community because of their excellent photostability and high quantum yields (0.4–0.9). The general DPP structure is synthesized by the reaction of aromatic nitrile with dialkyl succinate to produce a planar structure with strong intramolecular hydrogen bonding and π–π stacking between adjacent molecules, both of which are key to its chemical stability. Of particular note is the presence of a bicyclic lactam chromophoric unit containing three different functional groups (–C = C– double bonds, carbonyl and amine (NH) groups) that may be used as building blocks for further synthetic modification and derivatization, and as a platform for a vast array of functionalization possibilities. The biological applications, especially in living cells, have however been limited because most DPP derivatives absorb in the range 435–510 nm and emit <600 nm. That said, with moderately high extinction coefficients (∼25 000 M−1 cm−1) of solution and membrane-bound forms, considerable effort has been dedicated to their single-molecule application. For example, recent work involving structurally rigid L-shaped isoindoledione, produced via DPP synthesis, has enabled solvent-sensitive emission up to ∼630 nm with large Stokes shifts to be achieved while minimizing autofluorescence.121 In the same work, an N-alkylated isoindoledione containing a benzofuryl substituent was found to stain cell membranes exclusively, though a substantial reduction in quantum yield by comparison was noted. The use of DPP-based probes has also emerged as promising with respect to molecular imaging, and several studies have explored the two-photon absorption properties of DPP-conjugated dyes, demonstrating their potential utility for deep imaging. In this regard we refer the reader to an extensive review in this area.122 The application of ratiometric DPP-containing probes has also demonstrated specificity toward esterase in cells, even in the presence of other analytes.123 

Over the years, organic dyes have found vast utility in the context of in vitro single-molecule detection, imaging, and quantification. However, for many applications, especially those involving live-cell imaging, they have taken somewhat of a backseat, owing in part to the development of genetically encoded fluorescent proteins.

The purification of the green fluorescent protein (GFP) from Aequorea victoria jellyfish124 revolutionized the single-molecule field and led to methods of expressing translationally fused fluorescent proteins as labels of proteins of interest via genetic engineering, bypassing the need for any form of chemical attachment [Figs. 5(a) and 5(b)].125–127 Unlike quantum dots (QDs) and organic dyes, which require appropriately designed conjugation schemes, fluorescent proteins offer a valuable alternative for in cellulo and in vivo imaging. A second appealing characteristic rests in their ability to confer high cellular and sub-cellular specificity using promoters, enabling these probes to report from specific, often otherwise inaccessible regions. Additionally, fluorescent proteins are easily inserted into live cells by transfection or virus infection and can be upheld for timescales far surpassing 24 h prior to excretion.128 Fluorescent proteins have, thus, been applied extensively—for example, in live cell FRET biosensing experiments,129 as both donors and acceptors, in fluorescence lifetime imaging (FLIM) for the detection of protein–protein interactions,130 interrogating the dynamic interplay between proteins and lipids131 and to count the number of subunits in functional molecular machines.96 

FIG. 5.

Single-molecule imaging of fluorescent proteins. (a) Saccharomyces cerevisiae under glucose depletion. Here, the Mig1 repressor is fluorescently labeled with GFP (cyan). Representative fluorescence image obtained via slimfield microscopy141 of a single Mig1-GFP molecule (top panel) enables its spatial position within the cell to be evaluated when overlaid and compared against the corresponding brightfield image. (b) Crystal structures of GFP (top panel, PDB ID: 1GFL) and mCherry (bottom panel, PDB ID: 2H5Q) showing locations of the alpha helices, beta strands and coiled-coiled regions. (c) Absorption (dashed) and fluorescence emission (solid) spectra of EGFP and mCherry demonstrating the spectral overlap (yellow shaded region) necessary for compatibility with single-molecule FRET imaging.

FIG. 5.

Single-molecule imaging of fluorescent proteins. (a) Saccharomyces cerevisiae under glucose depletion. Here, the Mig1 repressor is fluorescently labeled with GFP (cyan). Representative fluorescence image obtained via slimfield microscopy141 of a single Mig1-GFP molecule (top panel) enables its spatial position within the cell to be evaluated when overlaid and compared against the corresponding brightfield image. (b) Crystal structures of GFP (top panel, PDB ID: 1GFL) and mCherry (bottom panel, PDB ID: 2H5Q) showing locations of the alpha helices, beta strands and coiled-coiled regions. (c) Absorption (dashed) and fluorescence emission (solid) spectra of EGFP and mCherry demonstrating the spectral overlap (yellow shaded region) necessary for compatibility with single-molecule FRET imaging.

Close modal

Among the most widely used FRET sensors is the blue–yellow mTurquoise2 (donor) and sEYFP (acceptor) pair, which provides a Förster radius of 5.9 and ∼2–8 nm sensitivity.130 Derivatives of sEYFP also exist (including mVenus, mCitrine, and YPet), and each has been tailored to accommodate minor pH switching.132–134 A major limitation of long-term FRET imaging, however, is their relatively poor photostability. The emission signal often dissipates rapidly over time, thereby affecting the ratio of donor to acceptor emission intensities and necessitating corrections for photobleaching. It follows that fluorescent proteins for FRET-based applications should be chosen based on high brightness, long-term photostability and insensitivity to pH fluctuations. Unfortunately, engineering fluorescent proteins with all desired properties remains a major experimental challenge, though progress has been made with the development of mClover3 and mRuby3. In such examples, oxygen access to the chromophores is limited.133,135 Pairs such as mClover3-mRuby3 or similarly, mNeongreen-mRuby3, therefore, hold promise for live-cell FRET imaging in the future. On the other hand, EYFP and mCitrine are strongly pH-sensitive and have potential for detecting activities, such as protein function, metabolic reactions, and autophagy, where pH regulation is critical.136 

Green–red FRET pairs, such as the EGFP-mCherry [Figs. 5(b) and 5(c)] and GFP-mRuby2 combination, overcome some of the limitations of blue–yellow pairs. For instance, excitation in the green generally reduces autofluorescence, the proteins are less phototoxic, they exhibit greater spectral separation, and they have extended distance sensitivity.137–139 Furthermore, unlike other fluorescent proteins, mCherry emission is only rarely interrupted by photoblinking. FRET pairs with spectra in the far-red, such as the mPlum-IFP1.4 duo,140 have the additional advantages of further reducing autofluorescence and offer the potential for deep-tissue imaging, though further developments in this area are required to improve overall brightness.

Although blue, green, yellow, and red fluorescent proteins have been extensively used, their complicated photophysics, coupled with photostability issues, mean that their application in single-molecule experiments is still challenging. To help bridge this gap, photoactivatable and photoswitchable fluorescent proteins have been engineered to aid with diffusion studies and to understanding pathways. While the former are induced to switch from a low-emissive dark state to an emissive bright state, the latter are stimulated to emit at shifted wavelengths.142 Among these derivatives, the photoactivatable variant of GFP, avGFP, exhibits an excitation spectrum with two distinct peaks (396 and 476 nm) corresponding to protonated and deprotonated chromophores. Upon UV excitation, the ratio of these peaks changes in favor of the deprotonated form.143 The photoactivatable and switchable properties of such fluorescent proteins allow the labeled biomolecule to be tracked without the need for continuous visualization, which goes some way to overcoming the issue of low photostability.

It comes as no surprise that many challenges remain in this area, not least of which is the need to engineer fluorescent proteins with higher quantum yields. One potential strategy to achieve this is to develop a suite of fluorescent proteins with improved maturation and folding attributes. Furthermore, one should be aware that self-assembling fluorescent proteins, caused by hydrophobic mutations can interfere with FRET-based distance conversions, though modifications of peptide linkers between fluorescent proteins and the sensing region could be a potential strategy to overcome this.144 The relatively large size of fluorescent proteins means that they can also interfere with, for example, kinase motion, though some fluorescent proteins bypass this by reporting on nucleo-cytoplasmic shuttling readouts.145 

Quantum dots (QDs) are nanoparticles composed of periodic groups of III-V, II-VI, or IV-VI semiconductor materials, such as CdS, CdSe, CdTe, ZnS, ZnSe, and InP with tunable physical dimensions as well as optoelectronic properties, which are not available from isolated molecules or bulk solids. They exhibit discrete energy levels, and their bandgap can be precisely modulated by varying their size. Their high emission intensities, large Stokes shift, narrow emission and broad absorption spectra, large molar extinction coefficients, high quantum yields, strong resistance to photobleaching, and long fluorescence lifetimes146–149 have made them particularly attractive across the single-molecule community as in vitro and in vivo biosensors,150 and their production has also led to substantial contributions toward the development of super-resolution imaging and single-particle tracking151 techniques. Furthermore, their electronic features are enabling the development of QD-based electrochemical152 and electroluminescent biosensing either as a catalyst or light-emitter, and recent developments have even seen them used as single particle drug delivery vehicles.153 

QDs are typically prepared using organometallic chemistry methods154 to yield emission wavelengths spanning across the UV, visible, and infrared. For biological applications, it is critical to render the QD soluble through surface-passivation, with the ideal water-soluble ligand (i) enhancing QD stability, (ii) maintaining resistance of the QD to photobleaching and degradation, (iii) containing functional groups for bio-conjugation, and (iv) minimizing the particle size. While the physical and optical properties of QDs have been extensively studied for single-molecule applications, such as multiplexed imaging,155in vivo bio-detection,156 and FRET,157 CdTe and CdSe derivatives have attracted particular attention owing to their versatility.

CdTe is a II-VI semiconductor with a bandgap energy of ∼1.5 eV at 300 K,158 corresponding to infrared emission, but as its size is reduced to the order of several nanometers via the quantum confinement of charge carriers, the fluorescence emission wavelength peak shifts through the visible range (500–750 nm). As the density of states near the conduction and valence bands reduces below 12 nm, discrete excitonic states form. Consequently, the bandgap increases, resulting in a peak shift of the spectrum159 (Fig. 6). In a similar way, the bandgap of CdSe QDs increases from 1.9 to 2.8 eV as the size decreases from 7 to 2 nm, enabling tailored emission in the range 450–650 nm.159 We note that on comparison to core-type CdTe QDs, core shell CdSe/ZnS particles exhibit narrower emission features (Fig. 6), though the precise range over which emission occurs and the spectral properties ultimately depend on the materials used, surface coatings and particle size. Though CdTe and CdSe are among the most widely used of all QDs, the implications of the Cd/Te and Cd/Se molar ratios on optical properties such as emission intensity, quantum yield and lifetime is a relatively new area that demands further exploration.160,161 Nevertheless, their high quantum yields (20%–80%) place them among the brightest of all available probes.162 Their extinction coefficients associated with the first excitonic absorption peak, though strongly size-dependent, are relatively large (ε = 1–8 × 105 M−1 cm−1)163 and their two-photon absorption cross sections are orders of magnitude larger than those associated with organic dyes.159 We note that the extinction coefficients at the first exciton peak are much lower than those at shorter wavelengths, in contrast to organic dyes which have their largest extinction coefficient at the peak of their absorption spectrum. Coupled with their broad excitation spectra which increase toward the UV, relatively long lifetimes (>10 ns) and resistance to chemical degradation, CdTe and CdSe QDs are excellent candidates for tracking time-dependent dynamic processes,151 biomedical imaging including in vivo tumor detection,164,165 deep tissue imaging,166 environmental sensing,167,168 and antibody detection.169 

FIG. 6.

Fluorescence emission of single quantum dots. Top panel: Size-dependent emission spectra associated with CdTe and CdSe/ZnS quantum dots. Shown are representative emission spectra for core-type CdTe-530 (solid green) (inset: absorption spectra), −580 (solid orange) and −680 (solid red) and CdSe/ZnS-530 (dashed green), −580 (dashed orange) and −650 (dashed red). Lower panel: Representative single particle fluorescence trajectory (left panel) and corresponding intensity histogram (right panel) indicating photoblinking from highly emissive on states to non-emissive “off” states obtained from TIRFM imaging of a single CdTe 580 QD (inset, scale bar = 500 nm). The dashed line corresponds to a threshold intensity level of six standard deviations above background used for differentiating between on and off states.

FIG. 6.

Fluorescence emission of single quantum dots. Top panel: Size-dependent emission spectra associated with CdTe and CdSe/ZnS quantum dots. Shown are representative emission spectra for core-type CdTe-530 (solid green) (inset: absorption spectra), −580 (solid orange) and −680 (solid red) and CdSe/ZnS-530 (dashed green), −580 (dashed orange) and −650 (dashed red). Lower panel: Representative single particle fluorescence trajectory (left panel) and corresponding intensity histogram (right panel) indicating photoblinking from highly emissive on states to non-emissive “off” states obtained from TIRFM imaging of a single CdTe 580 QD (inset, scale bar = 500 nm). The dashed line corresponds to a threshold intensity level of six standard deviations above background used for differentiating between on and off states.

Close modal

Core-type QDs, such as CdTe and CdSe, do however suffer from lower quantum yields and photostability,170 though this can be improved by passivation of the surface with semiconductors, such as CdS or ZnS. Some common examples of these so-called core-shell QDs include CdS on CdSe and ZnS on CdSe, the latter containing a larger fraction of brighter particles relative to the core-type case as a result of increased single-particle quantum yields.171 The application of such QDs in single-molecule imaging has been mainly directed toward mammalian cells, though there is an increasing tendency to apply them for intracellular tracking, diagnostics, in vivo imaging and therapeutic delivery172 and, for electrochemiluminescene assays where femtomolar detection of single particles is now possible.173,174 For example, the single-particle tracking of QD-conjugated membrane receptors175 and proteins176,177 in living cells has enabled their diffusion characteristics in response to environmental stimuli to be accessed for long timescales (>20 min) and temporal resolutions (<1 ms178) surpassing those conventionally accessible using organic dyes. Further QD tracking applications have included their use as tumor-targeting drug delivery vehicles179 and as encapsulated cargo within synaptic vesicles.180 QDs have also found utility as effective single-particle FRET acceptors, though their long fluorescence lifetimes dictate the need for donors with comparably long lifetimes, such as lanthanide dyes, as opposed to organic dyes, for detectable FRET.181 It follows that QD-based single-particle detection offers multifunctional and attractive opportunities for probing and manipulating biological systems, both in vitro and in vivo, but great care must be taken during their synthesis and integration with biological molecules to avoid perturbing function.

To harness the attractive optical properties of QDs, it would be ideal to minimize their size for biomolecular-labeling. Being comparably large, with often insoluble properties and being incapable of precise valency controlled labeling, they rely heavily on being passivated with organic ligands. While the ligands should improve QD solubility, they must also provide the QD surface with a chemical platform from which to enable effective and efficient biomolecular conjugation.182–184 CdTe QDs, for instance, are typically capped with mercaptopropionic acid or mercaptosuccinic acid terminated with –COOH for these reasons.185 Recently, QDs wrapped in functionalized oligonucleotides have shown promise in the context of single particle tracking,186,187 ligands covalently coupled to polyethylene glycols have helped to minimize nonspecific binding,188,189 and water-dispersible QDs comprising hydrophobic QDs and zwitterionic moieties have realized liposomal-like structures that preserve optical and colloidal stability.190 Amphiphilic polymers have also been used to improve bioconjugation,191 and although challenges still exist, strategies for producing monovalent QDs have been reported.191,192 We direct the reader to comprehensive reviews in these areas.172,189,191,193 Despite such advances, the QD size, solubility and valency of labeling are still significant hurdles that must be overcome when QDs are employed as single-particle sensors.

An interesting property of quantum dots is their fluorescent intermittency, whereby the fluorescence intensity from a single QD fluctuates between highly emissive “on” states and non-emissive “dark” states (Fig. 6). While the root cause of photoblinking is still debated, evidence points toward a mechanism in which electron transfer to trap states in the QD or surrounding matrix leads to photoinduced charging.194,195 QD photoblinking provides a simple way of achieving super-resolution localization via conventional fluorescence microscopy196 and blinking rates may be modulated in the presence of ions, offering environmental sensitivity.167,197,198 Taken together, understanding, suppressing, and manipulating the blinking characteristics of QDs are important lines of single-particle research.

Importantly, not all QDs are identical, and they cannot be considered as a uniform group. QD toxicity, for example, is closely linked to the intrinsic properties of the quantum dot, including material, shell type, ligand, surface chemistry, and size.199 A number of assays have been employed over the years to evaluate the influence of QDs on cellular organelles, protein expression, and clearance mechanisms, and in some cases, QD modifications have been made to mitigate against the effects, but it is important to note that while some examples of QD have demonstrable influence on biological function, others, have minimal impact.199–202 While it is outside the scope of this review to provide an exhaustive list of QD flavors and their reported toxicities, care must be taken to minimize or reduce toxicity, either through careful choice of QD or via QD modification, and this is especially true in the context of live-cell applications.197 

Fluorescent nanodiamonds (FNDs) are now emerging as promising biomarkers for single-molecule applications. While FNDs can be easily detected by conventional fluorescence microscopy, unlike QDs and organic dyes, their existence inside the cell does not typically induce cell death.203 FNDs have been reported to be over an order of magnitude brighter and more photostable than conventional organic dyes,204 and many display spectral shifts in response to changes in magnetic fields, electric fields and temperature gradients, making them useful nanosensors for high-resolution imaging.205 The tunable emission properties arise from the doping of nanodiamonds with defects such as nitrogen, europium, and silicon vacancies [Fig. 7(a)]. These behave like isolated atoms or molecules in a host matrix, with emission stemming from these locations as opposed to the bulk material206–211 [Fig. 7(b)]. Fluorescence arising from nitrogen-vacancy doped FNDs is photostable, even after months of continuous excitation.212 While they are also known for their biological inertness, successful FND-labeling of proteins213,214 and DNA215 has been achieved since their surface can be terminated with oxygen or hydrogen.

FIG. 7.

Fluorescence emission of single nanodiamonds. (a) Widefield TIRF image of single nanodiamonds (scale bar = 1 μm) and inset, structure of the nitrogen-vacancy defect. (b) Normalized fluorescence emission spectra obtained from 100 nm-sized nitrogen-vacancy center fluorescent nanodiamonds at a concentration of 1 mg/mL in ultra-pure water (λex = 532 nm). The emission spectrum displays zero phonon lines at 638 (red arrow) and 575 nm (blue arrow) corresponding to the presence of negatively charged and neutral defects, respectively.

FIG. 7.

Fluorescence emission of single nanodiamonds. (a) Widefield TIRF image of single nanodiamonds (scale bar = 1 μm) and inset, structure of the nitrogen-vacancy defect. (b) Normalized fluorescence emission spectra obtained from 100 nm-sized nitrogen-vacancy center fluorescent nanodiamonds at a concentration of 1 mg/mL in ultra-pure water (λex = 532 nm). The emission spectrum displays zero phonon lines at 638 (red arrow) and 575 nm (blue arrow) corresponding to the presence of negatively charged and neutral defects, respectively.

Close modal

Of all FNDs, those which are nitrogen-vacancy doped have shown most promise for single-molecule applications, though it is worth noting that only a fraction of elements in the periodic table have been incorporated as defects. Not only are they now routinely used for long-term particle tracking and localization in live cells,216–218 owing to their excellent photostability, they have also been used to sense magnetic fields through spectral shifts in their fluorescence emission, revealing FND orientation in the process.218 Their application has also extended to FRET-based sensing, where nitrogen-vacancy doped FNDs have acted as donors for black hole quenching dyes such as DY781,219 or as GFP acceptors for observing rotational motion in the F0F1 ATP synthase.216 

For future single-molecule applications where small-sized FNDs are required, it may be possible to prepare them with specific numbers of defects per nanodiamond, while maintaining photostability, and natural extensions are far reaching.

Coupling the probe of interest can be achieved by direct and indirect labeling methods, but challenges of controlling specificity and defining stoichiometry must be overcome for successful conjugation. In this section, we discuss various labeling methods, highlighting the advantages and limitations of each approach, such that careful selection of the best method can be chosen.

Techniques for chemically attaching organic fluorophores to target biomolecules have been critical for single-molecule imaging applications. Of all the techniques available, the site-specific covalent labeling of proteins with organic dyes has enabled the development of several single-molecule assays.220 Direct chemical attachment of purified proteins involves targeting the amino acid cysteine and amine groups [Fig. 8(a)]. In cysteine, a free sulfhydryl group can be rapidly cross-linked to an organic dye chemically engineered to contain a thiol-reactive agent such as maleimide, offering a highly specific and rapid labeling reaction under moderate conditions.47 Surface-accessible cysteines are particularly appealing for labeling because they are found in relatively low abundance. If necessary, they can also be introduced into an amino acid sequence using site-directed mutagenesis,221 though we again emphasize that care must be taken not to perturb the overall function of the target biomolecule.222,223 It may also be possible to selectively label a cysteine by inducing conformational changes to improve site accessibility224 or by manipulating the reversible protection of cysteines using metal ions.225 

FIG. 8.

Comparison of commonly used protein labeling methods. Schematic illustrations of labeling reactions involving (a) maleimide functionalized probe and surface accessible cysteine, (b) succinimidyl-ester functionalized probe and amine group and (c) azide functionalized probe and alkyne group on a surface accessible unnatural amino acid. Inset: chemical structure of the UAA propargyl lysine. (d) Accessible volumes of Alexa Fluor 546 (blue) and Alexa Fluor 647 (orange) tagged to Cys97 and Cys473 on the Rep helicase are illustrated as semi-transparent surfaces in the open (left) and closed (right) conformations.249 

FIG. 8.

Comparison of commonly used protein labeling methods. Schematic illustrations of labeling reactions involving (a) maleimide functionalized probe and surface accessible cysteine, (b) succinimidyl-ester functionalized probe and amine group and (c) azide functionalized probe and alkyne group on a surface accessible unnatural amino acid. Inset: chemical structure of the UAA propargyl lysine. (d) Accessible volumes of Alexa Fluor 546 (blue) and Alexa Fluor 647 (orange) tagged to Cys97 and Cys473 on the Rep helicase are illustrated as semi-transparent surfaces in the open (left) and closed (right) conformations.249 

Close modal

Under all conditions, the target protein must be maintained in a reduced form {using for example, dithiothreitol (DTT) or tris[2-carboxyethyl]phosphine (TCEP)} prior to the labeling reaction in order to prevent the formation of disulfide bridges and cysteine inactivation. Immediately prior to labeling, the reducing agents should be removed to prevent reoxidation and the thiol groups competing with the target thiols on the target biomolecule. Furthermore, the efficient removal of unreacted molecules prior to single-molecule imaging is key to avoid the presence of free dye within the measurement.

In a similar way, amine-reactive conjugates, such as N-hydroxysuccinimide (NHS) ester or isothiocyanates can be used for the specific labeling of lysine or N-terminal amines [Fig. 8(b)].226,227 However, unlike cysteines, lysines are found in relative abundance and can therefore be problematic when the aim is to directly attach a single probe.

Site-specific conjugation via encoded unnatural amino acids (UAAs) and highly specific biorthogonal reactions also provides a useful way of directly conjugating probes to a protein structure, and in general, this strategy overcomes some of the problems associated with cysteine labeling [Fig. 8(c)].228 UAAs generally containing ketone, azide, alkyne, or tetrazine groups can be encoded into the protein structure via modification of the cDNA sequence in response to a unique amber stop codon. These groups can then be coupled to functionalized dyes via high-yield click chemistry procedures.229,230 In general, if the protein only contains1-2 regular amino acids, then these can be replaced with the UAA during protein expression.231–233 Nonsense codons which encode the UAA selenocysteine (SeC) into the protein structure are particularly attractive due to their ease-of-conjugation toward organic dyes containing maleimide or α-haloketones. By far, the most common UAAs incorporated into protein structure are designed to undergo alkyne-azide click chemistry and useful examples include the labeling of azide-containing UAAs incorporated into the protein structure with Alexa Fluor 488-Alkyne.234 Here the chemical reaction uses copper as a catalyst and results in a highly selective and strong covalent bond formed between azide and alkyne chemical groups to form stable 1,2,3-triazoles. The UAA p-acetylphenylalanine, for example, can be incorporated into a protein structure in response to the TAG stop codon, and this reacts well with organic dyes containing hydroxylamine groups, though the reaction must be carried out at low pH.235 Propargyl lysine is an alternative option used to couple azide-modified fluorophores to the structure via copper catalyzed alkyne-azide cycloaddition [Fig. 8(c)].236 Efficient incorporation of UAAs into protein structures has enabled a variety of applications, including but not limited to single-molecule FRET studies on the T4 lysozyme,237 intracellular DNA-PAINT,238 and the super-resolution imaging of outer-membrane proteins in E. coli.239 To extend their utility further, unnatural fluorescent amino acids, such as Lys(BODIPYFL),240 4-cyanotryptophan,241 and dansyl alanine,242 have emerged for single-molecule studies of ion-channels and protein folding, and this is an exciting area that warrants further investigation.

For the site specific labeling of DNA and RNA molecules, short nucleic acid oligo inserts can be used. In this case, a nucleic acid sequence can be cut at specific locations by restriction endonucleases to enable short sequences of nucleic acids complementary to a specific oligo sequence to be inserted at that location. Incubation with the oligo will then result in binding to the complementary sequence.243–245 This is particularly useful since oligos can be modified to include a variety of chemical groups, including biotin, azide, and alkynes, to enable conjugation. Recently, bright and photostable fluorescent RNAs have also facilitated cellular RNA tracking experiments within living systems.246 

In most cases, and irrespective of the length of the conjugation linker, the fluorophore can conformationally diffuse within an accessible volume around the attachment site [Fig. 8(d)].32 In the case of FRET-based measurements, this can lead to uncertainties in accurate distance determination and hinder experiments where short distance ranges between the attachment points are required due to dye–dye interactions.247 It is, therefore, of utmost importance that in such applications, the positional distribution of the dye is assessed via geometric accessible volume simulations,31,248 provided that the local structure of the biomolecule is known, in order to obtain accurate quantitative details.33 

Direct protein labeling is often limited by low yield, high levels of impurities or situations where the direct attachment of large fluorophores alters the activity of the biomolecule.250,251 The use of protein tags, such as the polyhistidine (His) tag [Figs. 9(a) and 9(b)], has thus emerged as powerful tools for overcoming such issues. Here, the low molecular weight tag is attached to recombinantly expressed proteins, enabling downstream labeling to anti-His functionalized probes to be attached with high specificity.252 For live cell approaches, however, additional methods of labeling are required. Using a strategy complementary to immunostaining or antibody-labeling, a widely adopted strategy to perform site-specific fluorescent labeling of a protein of interest is to express that it is fused with a monovalent tag using a single genetic construct which enables the downstream attachment of a functionalized fluorophore. Among these include SNAP- [Fig. 9(c)] and CLIP-tags, which are derivatives of the 20 kDa DNA repair protein O6-alkylguanine-DNA alkyltransferase.251 The tags are specifically designed to irreversibly attach to O6-benzylguanine functionalized dyes via a stable thioether bond using a reactive cysteine in the tag.253 Protein tags have, thus, found applicability in the detection and quantitation of labeled proteins via conventional biochemical methods, such as in-gel fluorescence scanning of SDS-PAGE gels. For single-molecule work, the technique also has particular relevance for the labeling of membrane-bound receptor proteins,254,255 as well as proteins in sub-cellular compartments.256,257 A number of live-cell based applications have also highlighted protein tags and applications range from super-resolution imaging,258 measuring protein activity,259,260 determining interactions via FRET261 and particle tracking.262 

FIG. 9.

Comparison of protein labeling methods involving protein tags. Schematic illustrations of (a) conventional His-tag antibody coupling, (b) Ni-NTA-linker conjugation, (c) SNAP-tag, and (d) HALO-tag labeling.

FIG. 9.

Comparison of protein labeling methods involving protein tags. Schematic illustrations of (a) conventional His-tag antibody coupling, (b) Ni-NTA-linker conjugation, (c) SNAP-tag, and (d) HALO-tag labeling.

Close modal

An alternative to the SNAP/CLIP approach is the use of the HaloTag [Fig. 9(d)], a 33 kDa derivative of a bacterial haloalkane dehalogenase enzyme,263 which forms an irreversible covalent bond between the fused protein and the HaloTag ligand upon binding.264 Here, a transient alkyl-enzyme intermediate is formed during the displacement of a terminal chloride with Asp106, and since His272 does not catalyze the hydrolysis, a stable covalent bond is formed.263 Much like the SNAP- and CLIP-tags, only a single genetic construct is required and most are fused directly to the C- or N-terminus, but in contrast, the HaloTag can be used under relatively acidic conditions, opening possibilities for its utility in harsh microenvironments.265 

Alternative specific labeling strategies involving tetracysteine tags can be used if the methods listed above are unavailable. This approach involves the binding of the membrane permeable fluorescein derivative FlAsH or resofurin derivative ReAsH, to a peptide sequence of the form C–C–X–X–C–C, where C represents Cysteine and X denotes any amino acid.266,267 The recognition sequence is typically inserted into solvent accessible, looped or disordered regions on the protein of interest. Both fluorophores are non-emissive in the unbound state and become emissive upon conjugation. This technique has been applied extensively in vitro and in cellulo to a range of biological questions including but not limited to single-molecule protein dynamics268 and protein aggregation.269 

Two major problems for single-molecule studies utilizing fluorescent probes are photobleaching and photoblinking. In recent years, the community has made major advances toward solving these problems and a host of novel anti-fading agents have been tested and developed to decrease the rates of bleaching and blinking. Here, we discuss the most popular choices of oxygen scavengers and triplet state quenchers with emphasis placed on their relative benefits.

In the case of photobleaching, the fluorescent probe enters an irreversible non-emissive dark state which limits the time window over which individual molecules can be studied. When photoblinking occurs, reversible transitions to long-lived and non-radiative dark-states complicates, for example, particle tracking and FRET measurements. While the mechanisms by which photobleaching and photoblinking are not entirely defined, and are fluorophore-dependent, it is recognized that molecular oxygen plays a substantial role in the formation of non-emissive states, either via direct interaction with the fluorophore or indirectly by producing free radicals in solution. Consequently, many single-molecule investigations utilizing organic dyes have incorporated the use of oxygen scavengers such as chromatin, cyclooctatetraene, 4-nitrobenzyl alcohol and l-ascorbic acid, either covalently attached to the probe of interest or in free solution.270,271 However, effectively minimizing molecular oxygen through physical means is challenging and so enzymatic oxygen scavengers are commonly employed instead. For example, the combination of D-glucose, glucose oxidase, and catalase can reduce the molecular oxygen concentration via the coupled reactions though we note use of this system can lead to subtle pH reductions.272 An alternative strategy making use of protocatechuate-3,4-dioxygenase, protocatechuic acid, pyranose oxidase, catalase, and glucose can further reduce molecular oxygen levels.273 Furthermore, the addition of the vitamin E analogue 6-hydroxy-2,5,7,8-tetramethylchromane-2-carboxylic acid (Trolox) minimizes photoblinking through a mechanism that involves suppressing the triplet state through electron transfer and recovery of the resulting radical ion by the complementary redox reactions.274 

Unfortunately, the protection from photoblinking and photobleaching provided by this cocktail is non-existent for fluorescent proteins. To mitigate against this, researchers have often minimized light exposure and/or the sampling frequency, and typically choose fluorescent proteins which are most photostable.275,276 Careful selection of cell imaging media with selected vitamins (riboflavin and pyridoxal) removed has been found to decrease the photobleaching of GFP relative to conventional buffer solutions while maintaining cell function,277 and though a number of commercial anti-fade media exist, each is likely to be fluorophore- and cell-dependent.

Fluorescence intermittency has also been universally observed from quantum dots and this of course be a major limitation for long term imaging or tracking experiments. In specific cases, surface passivation of core–shell CdSe/ZnS quantum dots with short chain length thiol moieties was found to suppress photoblinking.198 Similarly, indium tin oxide nanoparticles suppressed the blinking of CdSeTe/ZnS core–shell quantum dots,278 though we note that these results are not generalizable to all types of quantum dot. Further experiments demonstrated that surface-modification of core-type CdTe quantum dots with gadolinium ions promotes blinking, which could find useful utility in super-resolution applications.168 

Other strategies of enhancing photostability, which do not rely on a chemical cocktail, include the fluorination of dyes at the synthesis stage. Polyfluorinated cyanine derivatives upon comparison with non-fluorinated analogues for instance, were found to display at least a tenfold reduction in aggregation, >10% increase in quantum yield, and fourfold greater resistance to photobleaching due to reduced reactivity toward singlet oxygen. Clearly, all are extremely attractive for single-molecule applications, and are generally attributed to the fluorination of the benzothiazole heterocycles.279 More recently, organic dyes containing a perfluorophenyl group have also attracted attention because of their enhanced photostability and optical properties. In this regard, ring-perfluorinated trimethine cyanines have been shown to display almost a twofold higher fluorescence quantum yield and lifetime, and 2.5-fold lower nonradiative deactivation rate constant compared with the nonfluorinated version.280 Other strategies to improve photostability involve introducing electron-deficient substituents to reduce the overall reactivity of the probe toward singlet oxygen,63 conjugation of the probe to a triplet state quencher,281 non-covalent encapsulation of the probe into a charged copolymer282 or nanoparticle,283 or through nanohybrid formation.284 

In most applications, removal of photobleaching is desirable. However we note that in the case of single-molecule stepwise photobleaching analysis, where the stoichiometry of a complex containing labeled sub-units is estimated based on the number of photobleaching events,285 both the labeling and detection efficiencies are critical for accurate estimations.286 

Single-molecule fluorescence methods are increasingly used across the chemical, biological, and medical sciences, and a growing interest is developing in near-infrared fluorophores as imaging probes, in part because they enable deeper imaging through organic material compared with shorter wavelength probes but also because they generally facilitate higher signal-to-noise ratios due to lower autofluorescence at longer wavelengths. Considering this, the incorporation of oligoglycerol dendrons into such probes has enabled additional improvements in photostability in vitro and within living cells relative to conventional cyanines, suggesting this approach may well also be applicable for improving the photostability of a wide range of hydrophobic aromatic probes.287 

Parallel developments in probe designs, experimental techniques, and computational methods have given rise to a new series of multiplexed, correlative technologies capable of tackling previously intractable biological questions.288 Improvements in the sensitivity and speed of detectors, the efficiency and stability of excitation sources and microfluidics have all played a role in the development of tools and techniques capable of detecting and manipulating single fluorescently labeled biomolecules at work. While it is out with the scope of this review to list and describe them all, these transformative technologies, which include, but are not limited to TIRF microscopy,289 confocal-based multi-parameter fluorescence detection,290 a suite of super-resolution methods141,291,292 and fluorescence-integrated optical- and magnetic-based manipulation tools,293 are contributing enormously to answering long-standing biological questions.294 At the heart of this is the development of single-molecule probes, without which these fundamental biological processes and functions would remain, for the most part, invisible.

Most single-molecule measurements are performed in biologically relevant solutions, with the target molecule either tethered to a surface or allowed to freely diffuse. In the former case, the most common immobilization strategy involves the tethering of the biomolecule of interest to a surface via biotin-streptavidin interactions, as schematically shown in Fig. 4(d). Here, mixing of a passivation reagent with its biotinylated counterpart and addition of Avidin allows the immobilization of biotinylated molecules via a biotin-avidin linkage with high specificity and affinity (kd ∼ 10−15 M).48 Alternative approaches include the use of covalent-bond-based click-chemistry,295 encapsulation within lipid vesicles,296 nonspecific electrostatic interactions with an adsorbed passivation layer297 or immobilization within a gel;298 however, in all cases, care must be taken to ensure the immobilization scheme does not hinder biological function and the probe does not interact with the surface. This issue is in part bypassed by detecting freely diffusing biomolecules, though the measurement window is limited by the time spent transiting through the excitation volume. For a typical confocal-based measurement, it is not uncommon for biomolecules to diffuse away from the excitation source within several milliseconds. On the flip side, this also affords an opportunity to gain access to diffusion coefficients and hydrodynamic radii via fluorescence fluctuation correlation spectroscopy and use of the Stokes–Einstein equation.299–301 

In nearly all cases, and irrespective of the detection scheme used, maximizing the available photon budget and the signal-to-noise ratio is critical for single-molecule detection, especially when sub-millisecond temporal resolution is required. Clearly, this goes hand-in-hand with the need to ensure abundance of the molecule of interest, and a protocol for high labeling efficiency. Nevertheless, this concern ultimately drives the need for brighter and more photostable probes to prolong observation time windows and increase the number of collected photons. While several analytical algorithms have been tailored toward improving the signal-to-noise ratio post-acquisition,302,303 and probe-free techniques, such as digital holographic304 and interferometric scattering305 microscopy, are emerging as promising tools for exploring biomolecular function, the requirement to obtain fluorescent probes with quantum yields approaching unity, while remaining minimally perturbing, remains an active and critical area of research.

Combining research efforts in a range of these different approaches, such as the engineering of new fluorescent probes with enhanced photostability, minimizing background fluorescence, and improving the sensitivity of detectors, will continue to aid maximization of the available photon signal. However, single-molecule experiments are not without limitations. For instance, low signal-to-noise ratios are often observed when fast acquisition rates are used and there is an intrinsic requirement for using the probes under dilute conditions. While a number of analytical tools3,301,306 have recently been developed to combat the signal-to-noise problem, essentially enabling otherwise hidden information to be obtained from millisecond sampled images, the latter limits the detection scheme to less than a few nanomolar, even when localized excitation is used. Indeed, localization microscopy and particle tracking perform badly if fluorophores are close together in the sample and the limiting concentration depends on the spatial dimensionality of the tagged biomolecule, its mobility, and ability to cluster. While labeling a sub-population can get the concentration of fluorescent species below this limit, the method yields a significant proportion of unlabeled species in the sample, and many of these may potentially have different physical properties to the tagged species. An alternative strategy involves delimited photobleaching whereby a fraction of the fluorophore population is photobleached for a given period prior to localization microscopy. Unfortunately, both methods produce a substantial dark population of the biomolecule under investigation. In this regard, the development of photoactivatable probes has been used in many cases to help alleviate the issue and extend the concentration range for analysis. Super-resolution methods, including photoactivated localization microscopy (PALM) and single-particle tracking, make use of such probes to facilitate investigations into the organization and localization of diffraction-limited fluorescent species in living cells with a density of molecules high enough to provide structural context. When the two methods are combined, information is gained on the positions of single-molecules by activating, localizing and bleaching subsets of for instance, photoactivatable fluorescent proteins.307 In this regard, we steer the reader to comprehensive reviews45,308–311 that discuss such approaches, mechanisms of photoactivation, and advantages and limitations of photoactivatable probe designs.

Carefully applied, the combination of single-molecule fluorescent probes, labeling strategy, and immobilization and detection scheme provides a strong platform for understanding the complexities of life's biomolecules in far greater detail than ever before.

New and exciting technical developments over the last few decades, even those currently at proof-of-principle stage, offer the modern multidisciplinary researcher a wealth of opportunity for characterizing single biomolecules in action, one-at-a-time, bypassing limitations associated with conventional ensemble averaging. Spectroscopic methods that employ single-molecule fluorescent probes, from particle tracking to FRET, not only complement structural and functional characterization methods, but also offer the ability to observe time-dependent, and often, heterogeneous interactions.

The ideal experimental toolbox allowing efficient applications of single-molecule techniques in complex biomolecular systems comprises two major factors: (i) the appropriately selected fluorescent probe(s), which should have sufficient spectroscopic and molecular properties, and (ii) a corresponding, appropriately designed, conjugation scheme. Developments in probe design, coupled with new site-specific labeling procedures, now provide the single-molecule community with a facile yet robust platform for quantitative analyses across a huge range of biomolecules and experimental situations.

Undoubtedly, there are many hurdles yet to overcome, for example, the general issues of increasing probe brightness and photostability; the broader application of single-molecule fluorescent probes for environmental sensing; the requirement to develop minimally invasive labeling schemes; the need for specific targeting of biomolecules within a biological soup; and the need to detect sensitive and transient interactions.

The potential of single-molecule techniques cannot be underestimated, and with new and emerging fluorescent probe designs, the promise of this field is vast. The opportunities presented by combining fluorescence sensing, imaging, and spectroscopic tools with each other, and indeed with others such as atomic force and cryo-electron microscopy, for example, could lead to revolutionary biological insights by unmasking currently intractable information and interactions. In short, fluorescent probes with optimum properties combined with appropriate conjugation strategies are essential for high-quality and insightful single-molecule fluorescence measurements. Small fluorescent probes enable access to the nanoworld and they are helping to resolve our understanding of the biological jigsaw puzzle piece-by-piece. As probe designs evolve further and their applications broaden, fresh ideas and questions will emerge, stimulating the next generation of modern single-molecule biophysical experiments.

M.C.L. was supported by BBSRC (Grant Ref. Nos. BB/R001235/1 and BB/P000746/1) and EPSRC (Grant Ref. No. EP/T002166/1). S.D.Q. was supported by Alzheimer's Research UK (Grant Ref. No. ARUK-RF2019A-001). The authors would like to thank Dr. S. Shashkova (University of York, UK) for providing the images shown in Fig. 5(a), Dr. G. D. Watson and Dr. A. Noy (both University of York, UK) for performing the molecular dynamics simulation shown in Fig. 4(a), and Dr. B. Ambrose (University of Sheffield, UK) for performing the modeling shown in Fig. 8(d).

The authors have no conflicts to disclose.

Mark Leake: Conceptualization (equal); Writing – original draft (equal); Writing – review & editing (equal). Steven D. Quinn: Conceptualization (equal); Writing – original draft (equal); Writing – review & editing (equal).

The data that support the findings of this study are available from the corresponding author upon reasonable request.

1.
H.
Miller
,
Z.
Zhou
,
J.
Shepherd
,
A. J. M.
Wollman
, and
M. C.
Leake
, “
Single-molecule techniques in biophysics: A review of the progress in methods and applications
,”
Rep. Prog. Phys.
81
(
2
),
024601
(
2018
).
2.
O. L. J.
Harriman
and
M. C.
Leake
, “
Single molecule experimentation in biological physics: Exploring the living component of soft condensed matter one molecule at a time
,”
J. Phys.: Condens. Matter
23
(
50
),
503101
(
2011
).
3.
M. C.
Leake
, “
Analytical tools for single-molecule fluorescence imaging in cellulo
,”
Phys. Chem. Chem. Phys.
16
(
25
),
12635
12647
(
2014
).
4.
M. C.
Leake
, “
The physics of life: One molecule at a time
,”
Philos. Trans. R. Soc. B
368
(
1611
),
20120248
(
2013
).
5.
W. E.
Moerner
and
L.
Kador
, “
Optical-detection and spectroscopy of single molecules in a solid
,”
Phys. Rev. Lett.
62
(
21
),
2535
2538
(
1989
).
6.
M.
Orrit
and
J.
Bernard
, “
Single pentacene molecules detected by fluorescence excitation in a para-terphenyl crystal
,”
Phys. Rev. Lett.
65
(
21
),
2716
2719
(
1990
).
7.
W. E.
Moerner
, “
New directions in single-molecule imaging and analysis
,”
Proc. Natl. Acad. Sci. U.S.A.
104
(
31
),
12596
12602
(
2007
).
8.
R. D.
Vale
,
T.
Funatsu
,
D. W.
Pierce
,
L.
Romberg
,
Y.
Harada
, and
T.
Yanagida
, “
Direct observation of single kinesin molecules moving along microtubules
,”
Nature
380
(
6573
),
451
453
(
1996
).
9.
T. J.
Ha
,
A. Y.
Ting
,
J.
Liang
,
W. B.
Caldwell
,
A. A.
Deniz
,
D. S.
Chemla
 et al, “
Single-molecule fluorescence spectroscopy of enzyme conformational dynamics and cleavage mechanism
,”
Proc. Natl. Acad. Sci. U.S.A.
96
(
3
),
893
898
(
1999
).
10.
H. P.
Lu
,
L. Y.
Xun
, and
X. S.
Xie
, “
Single-molecule enzymatic dynamics
,”
Science
282
(
5395
),
1877
1882
(
1998
).
11.
H.
Noji
,
R.
Yasuda
,
M.
Yoshida
, and
K.
Kinosita
, “
Direct observation of the rotation of F-1-ATPase
,”
Nature
386
(
6622
),
299
302
(
1997
).
12.
M. A.
Bopp
,
A.
Sytnik
,
T. D.
Howard
,
R. J.
Cogdell
, and
R. M.
Hochstrasser
, “
The dynamics of structural deformations of immobilized single light-harvesting complexes
,”
Proc. Natl. Acad. Sci. U.S.A.
96
(
20
),
11271
11276
(
1999
).
13.
H.
Hwang
and
S.
Myong
, “
Protein induced fluorescence enhancement (PIFE) for probing protein-nucleic acid interactions
,”
Chem. Soc. Rev.
43
(
4
),
1221
1229
(
2014
).
14.
Z.
Xie
,
N.
Srividya
,
T. R.
Sosnick
,
T.
Pan
, and
N. F.
Scherer
, “
Single-molecule studies highlight conformational heterogeneity in the early folding steps of a large ribozyme
,”
Proc. Natl. Acad. Sci. U.S.A.
101
(
2
),
534
539
(
2004
).
15.
M. C.
Leake
,
J. H.
Chandler
,
G. H.
Wadhams
,
F.
Bai
,
R. M.
Berry
, and
J. P.
Armitage
, “
Stoichiometry and turnover in single, functioning membrane protein complexes
,”
Nature
443
(
7109
),
355
358
(
2006
).
16.
R.
Reyes-Lamothe
,
D. J.
Sherratt
, and
M. C.
Leake
, “
Stoichiometry and architecture of active DNA replication machinery in escherichia coli
,”
Science
328
(
5977
),
498
501
(
2010
).
17.
A. H.
Syeda
,
A. J. M.
Wollman
,
A. L.
Hargreaves
,
J. A. L.
Howard
,
J. G.
Bruning
,
P.
McGlynn
 et al, “
Single-molecule live cell imaging of Rep reveals the dynamic interplay between an accessory replicative helicase and the replisome
,”
Nucl. Acids Res.
47
(
12
),
6287
6298
(
2019
).
18.
M.
Stracy
,
A. J. M.
Wollman
,
E.
Kaja
,
J.
Gapinski
,
J. E.
Lee
,
V. A.
Leek
 et al, “
Single-molecule imaging of DNA gyrase activity in living Escherichia coli
,”
Nucl. Acids Res.
47
(
1
),
210
220
(
2019
).
19.
C. W.
Hollars
and
R. C.
Dunn
, “
Probing single molecule orientations in model lipid membranes with near-field scanning optical microscopy
,”
J. Chem. Phys.
112
(
18
),
7822
7830
(
2000
).
20.
C. E.
Talley
and
R. C.
Dunn
, “
Single molecules as probes of lipid membrane microenvironments
,”
J. Phys. Chem. B
103
(
46
),
10214
10220
(
1999
).
21.
L.
Dresser
,
S. P.
Graham
,
L. M.
Miller
,
C.
Schaefer
,
D.
Conteduca
,
S.
Johnson
 et al, “
Tween-20 induces the structural remodeling of single lipid vesicles
,”
J. Phys. Chem. Lett.
13
,
5341
5350
(
2022
).
22.
Y.
Sako
,
S.
Minoguchi
, and
T.
Yanagida
, “
Single-molecule imaging of EGFR signalling on the surface of living cells
,”
Nat. Cell Biol.
2
(
3
),
168
172
(
2000
).
23.
S. Z.
Wang
,
R.
Vafabakhsh
,
W. F.
Borschell
,
T.
Ha
, and
C. G.
Nichols
, “
Structural dynamics of potassium-channel gating revealed by single-molecule FRET
,”
Nat. Struct. Mol. Biol.
23
(
1
),
31
36
(
2016
).
24.
R.
Roy
,
S.
Hohng
, and
T.
Ha
, “
A practical guide to single-molecule FRET
,”
Nat. Methods
5
(
6
),
507
516
(
2008
).
25.
L.
Wu
,
C.
Huang
,
B. P.
Emery
,
A. C.
Sedgwick
,
S. D.
Bull
,
X. P.
He
 et al, “
Forster resonance energy transfer (FRET)-based small-molecule sensors and imaging agents
,”
Chem. Soc. Rev.
49
(
15
),
5110
5139
(
2020
).
26.
L. E.
Baltierra-Jasso
,
M. J.
Morten
,
L.
Laflor
,
S. D.
Quinn
, and
S. W.
Magennis
, “
Crowding-induced hybridization of single DNA hairpins
,”
J. Am. Chem. Soc.
137
(
51
),
16020
16023
(
2015
).
27.
B. P.
Paudel
and
D.
Rueda
, “
Molecular crowding accelerates ribozyme docking and catalysis
,”
J. Am. Chem. Soc.
136
(
48
),
16700
16703
(
2014
).
28.
L.
Scharfen
and
M.
Schlierf
, “
Real-time monitoring of protein-induced DNA conformational changes using single-molecule FRET
,”
Methods
169
,
11
20
(
2019
).
29.
B.
Schuler
, “
Single-molecule FRET of protein structure and dynamics—A primer
,”
J. Nanobiotechnol.
11
,
S2
(
2013
).
30.
M.
Diez
,
B.
Zimmermann
,
M.
Borsch
,
M.
Konig
,
E.
Schweinberger
,
S.
Steigmiller
 et al, “
Proton-powered subunit rotation in single membrane-bound F0F1-ATP synthase
,”
Nat. Struct. Mol. Biol.
11
(
2
),
135
141
(
2004
).
31.
S.
Kalinin
,
T.
Peulen
,
S.
Sindbert
,
P. J.
Rothwell
,
S.
Berger
,
T.
Restle
 et al, “
A toolkit and benchmark study for FRET-restrained high-precision structural modeling
,”
Nat. Methods
9
(
12
),
1218
1225
(
2012
).
32.
M.
Dimura
,
T. O.
Peulen
,
C. A.
Hanke
,
A.
Prakash
,
H.
Gohlke
, and
C. A. M.
Seidel
, “
Quantitative FRET studies and integrative modeling unravel the structure and dynamics of biomolecular systems
,”
Curr. Opin. Struct. Biol.
40
,
163
185
(
2016
).
33.
B.
Hellenkamp
,
S.
Schmid
,
O.
Doroshenko
,
O.
Opanasyuk
,
R.
Kuhnemuth
,
S. R.
Adariani
 et al, “
Precision and accuracy of single-molecule FRET measurements-a multi-laboratory benchmark study
,”
Nat. Methods
15
(
9
),
669
(
2018
).
34.
S. J.
Holden
,
S.
Uphoff
,
J.
Hohlbein
,
D.
Yadin
,
L.
Le Reste
,
O. J.
Britton
 et al, “
Defining the limits of single-molecule FRET resolution in TIRF microscopy
,”
Biophys. J.
99
(
9
),
3102
3111
(
2010
).
35.
T.
Kondo
,
W. J.
Chen
, and
G. S.
Schlau-Cohen
, “
Single-molecule fluorescence spectroscopy of photosynthetic systems
,”
Chem. Rev.
117
(
2
),
860
898
(
2017
).
36.
D. Y.
Shroder
,
L. G.
Lippert
, and
Y. E.
Goldman
, “
Single molecule optical measurements of orientation and rotations of biological macromolecules
,”
Methods Appl. Fluoresc.
4
(
4
),
042004
(
2016
).
37.
C. A.
Valades Cruz
,
H. A.
Shaban
,
A.
Kress
,
N.
Bertaux
,
S.
Monneret
,
M.
Mavrakis
 et al, “
Quantitative nanoscale imaging of orientational order in biological filaments by polarized superresolution microscopy
,”
Proc. Natl. Acad. Sci. U.S.A.
113
(
7
),
E820
E828
(
2016
).
38.
J.
van Mameren
,
K.
Vermeulen
,
G. J. L.
Wuite
, and
E. J. G.
Peterman
, “
A polarized view on DNA under tension
,”
J. Chem. Phys.
148
(
12
),
123306
(
2018
).
39.
A. S.
Backer
,
A. S.
Biebricher
,
G. A.
King
,
G. J. L.
Wuite
,
I.
Heller
, and
E. J. G.
Peterman
, “
Single-molecule polarization microscopy of DNA intercalators sheds light on the structure of S-DNA
,”
Sci. Adv.
5
(
3
),
eaav1083
(
2019
).
40.
B. C.
Stevens
and
T.
Ha
, “
Discrete and heterogeneous rotational dynamics of single membrane probe dyes in gel phase supported lipid bilayer
,”
J. Chem. Phys.
120
(
6
),
3030
3039
(
2004
).
41.
Y.
Sun
,
O.
Sato
,
F.
Ruhnow
,
M. E.
Arsenault
,
M.
Ikebe
, and
Y. E.
Goldman
, “
Single-molecule stepping and structural dynamics of myosin X
,”
Nat. Struct. Mol. Biol.
17
(
4
),
485
491
(
2010
).
42.
J. W.
Shepherd
,
A. L.
Payne-Dwyer
,
J. E.
Lee
,
A.
Syeda
, and
M. C.
Leake
, “
Combining single-molecule super-resolved localization microscopy with fluorescence polarization imaging to study cellular processes
,”
J. Phys.-Photonics
3
(
3
),
034010
(
2021
).
43.
B. W.
Hoogenboom
and
M. C.
Leake
, “
The case for biophysics super-groups in physics departments
,”
Phys. Biol.
15
(
6
),
060201
(
2018
).
44.
Q. S.
Zheng
,
M. F.
Juette
,
S.
Jockusch
,
M. R.
Wasserman
,
Z.
Zhou
,
R. B.
Altman
 et al, “
Ultra-stable organic fluorophores for single-molecule research
,”
Chem. Soc. Rev.
43
(
4
),
1044
1056
(
2014
).
45.
T.
Ha
and
P.
Tinnefeld
, “
Photophysics of fluorescent probes for single-molecule biophysics and super-resolution imaging
,”
Annu. Rev. Phys. Chem.
63
,
595
617
(
2012
).
46.
J.
Lee
,
S.
Lee
,
K.
Ragunathan
,
C.
Joo
,
T.
Ha
, and
S.
Hohng
, “
Single-molecule four-color FRET
,”
Angew. Chem. Int. Ed.
49
(
51
),
9922
9925
(
2010
).
47.
C. P.
Toseland
, “
Fluorescent labeling and modification of proteins
,”
J. Chem. Biol.
6
(
3
),
85
95
(
2013
).
48.
A.
Gust
,
A.
Zander
,
A.
Gietl
,
P.
Holzmeister
,
S.
Schulz
,
B.
Lalkens
 et al, “
A starting point for fluorescence-based single-molecule measurements in biomolecular research
,”
Molecules
19
(
10
),
15824
15865
(
2014
).
49.
Y. H.
Fu
and
N. S.
Finney
, “
Small-molecule fluorescent probes and their design
,”
RSC Adv.
8
(
51
),
29051
29061
(
2018
).
50.
S. H.
Shim
,
C.
Xia
,
G.
Zhong
,
H. P.
Babcock
,
J. C.
Vaughan
,
B.
Huang
 et al, “
Super-resolution fluorescence imaging of organelles in live cells with photoswitchable membrane probes
,”
Proc. Natl. Acad. Sci. U.S.A.
109
(
35
),
13978
13983
(
2012
).
51.
N. A.
Staffend
and
R. L.
Meisel
, “
DiOlistic labeling of neurons in tissue slices: A qualitative and quantitative analysis of methodological variations
,”
Front. Neuroanat.
5
,
14
(
2011
).
52.
T. Y.
Yoon
,
B.
Okumus
,
F.
Zhang
,
Y. K.
Shin
, and
T.
Ha
, “
Multiple intermediates in SNARE-induced membrane fusion
,”
Proc. Natl. Acad. Sci. U.S.A.
103
(
52
),
19731
19736
(
2006
).
53.
T. P.
Lehmann
,
W.
Juzwa
,
K.
Filipiak
,
P.
Sujka-Kordowska
,
M.
Zabel
,
J.
Glowacki
 et al, “
Quantification of the asymmetric migration of the lipophilic dyes, DiO and DiD, in homotypic co-cultures of chondrosarcoma SW-1353 cells
,”
Mol. Med. Rep.
14
(
5
),
4529
4536
(
2016
).
54.
M. H.
Hofmann
and
H.
Bleckmann
, “
Effect of temperature and calcium on transneuronal diffusion of DiI in fixed brain preparations
,”
J. Neurosci. Methods
88
(
1
),
27
31
(
1999
).
55.
B. S.
Packard
and
D. E.
Wolf
, “
Fluorescence lifetimes of carbocyanine lipid analogues in phospholipid bilayers
,”
Biochemistry
24
(
19
),
5176
5181
(
1985
).
56.
Y.
Hu
,
Z.
Tian
, and
J.
Diao
, “
Single-molecule fluorescence measurement of SNARE-mediated vesicle fusion
,”
Methods Mol. Biol.
1860
,
335
344
(
2019
).
57.
R.
Machado
,
J.
Bendesky
,
M.
Brown
,
K.
Spendier
, and
G. M.
Hagen
, “
Imaging membrane curvature inside a FcepsilonRI-centric synapse in RBL-2H3 cells using TIRF microscopy with polarized excitation
,”
J. Imaging
5
(
7
),
63
(
2019
).
58.
I.
De Proost
,
I.
Pintelon
,
W. J.
Wilkinson
,
S.
Goethals
,
I.
Brouns
,
L.
Van Nassauw
 et al, “
Purinergic signaling in the pulmonary neuroepithelial body microenvironment unraveled by live cell imaging
,”
FASEB J.
23
(
4
),
1153
1160
(
2009
).
59.
T.
Van Haeften
and
F. G.
Wouterlood
, “
Neuroanatomical tracing at high resolution
,”
J. Neurosci. Methods
103
(
1
),
107
116
(
2000
).
60.
J.
Diao
,
Z.
Su
,
Y.
Ishitsuka
,
B.
Lu
,
K. S.
Lee
,
Y.
Lai
 et al, “
A single-vesicle content mixing assay for SNARE-mediated membrane fusion
,”
Nat. Commun.
1
,
54
(
2010
).
61.
P. A.
Dalgarno
,
J.
Juan-Colas
,
G. J.
Hedley
,
L.
Pineiro
,
M.
Novo
,
C.
Perez-Gonzalez
 et al, “
Unveiling the multi-step solubilization mechanism of sub-micron size vesicles by detergents
,”
Sci. Rep.
9
(
1
),
12897
(
2019
).
62.
J.
Juan-Colas
,
L.
Dresser
,
K.
Morris
,
H.
Lagadou
,
R. H.
Ward
,
A.
Burns
 et al, “
The mechanism of vesicle solubilization by the detergent sodium dodecyl sulfate
,”
Langmuir
36
(
39
),
11499
11507
(
2020
).
63.
Q. S.
Zheng
and
L. D.
Lavis
, “
Development of photostable fluorophores for molecular imaging
,”
Curr. Opin. Chem. Biol.
39
,
32
38
(
2017
).
64.
J. B.
Grimm
and
L. D.
Lavis
, “
Synthesis of rhodamines from fluoresceins using Pd-catalyzed C-N cross-coupling
,”
Org. Lett.
13
(
24
),
6354
6357
(
2011
).
65.
J. B.
Grimm
,
B. P.
English
,
J. J.
Chen
,
J. P.
Slaughter
,
Z. J.
Zhang
,
A.
Revyakin
 et al, “
A general method to improve fluorophores for live-cell and single-molecule microscopy
,”
Nat. Methods
12
(
3
),
244
250
(
2015
).
66.
E. E.
Swinstead
,
T. B.
Miranda
,
V.
Paakinaho
,
S.
Baek
,
I.
Goldstein
,
M.
Hawkins
 et al, “
Steroid receptors reprogram FoxA1 occupancy through dynamic chromatin transitions
,”
Cell
165
(
3
),
593
605
(
2016
).
67.
A. W.
Bisson
,
Y. P.
Hsu
,
G. R.
Squyres
,
E.
Kuru
,
F. B.
Wu
,
C.
Jukes
 et al, “
Treadmilling by FtsZ filaments drives peptidoglycan synthesis and bacterial cell division
,”
Science
355
(
6326
),
739
743
(
2017
).
68.
J.
Arden-Jacob
,
J.
Frantzeskos
,
N. U.
Kemnitzer
,
A.
Zilles
, and
K. H.
Drexhage
, “
New fluorescent markers for the red region
,”
Spectrochim. Acta, Part A
57
(
11
),
2271
2283
(
2001
).
69.
Y.
Koide
,
Y.
Urano
,
K.
Hanaoka
,
T.
Terai
, and
T.
Nagano
, “
Evolution of group 14 rhodamines as platforms for near-infrared fluorescence probes utilizing photoinduced electron transfer
,”
ACS Chem. Biol.
6
(
6
),
600
608
(
2011
).
70.
G.
Lukinavicius
,
K.
Umezawa
,
N.
Olivier
,
A.
Honigmann
,
G. Y.
Yang
,
T.
Plass
 et al, “
A near-infrared fluorophore for live-cell super-resolution microscopy of cellular proteins
,”
Nat. Chem.
5
(
2
),
132
139
(
2013
).
71.
A.
Dvivedi
,
P.
Rajakannu
, and
M.
Ravikanth
, “
meso-Salicylaldehyde substituted BODIPY as a chemodosimetric sensor for cyanide anions
,”
Dalton Trans.
44
(
9
),
4054
4062
(
2015
).
72.
M.
Sadoine
,
M.
Cerminara
,
M.
Gerrits
,
J.
Fitter
, and
A.
Katranidis
, “
Cotranslational incorporation into proteins of a fluorophore suitable for smFRET studies
,”
ACS Synth. Biol.
7
(
2
),
405
411
(
2018
).
73.
L. C.
Zanetti-Domingues
,
C. J.
Tynan
,
D. J.
Rolfe
,
D. T.
Clarke
, and
M.
Martin-Fernandez
, “
Hydrophobic fluorescent probes introduce artifacts into single molecule tracking experiments due to non-specific binding
,”
PLoS One
8
(
9
),
e74200
(
2013
).
74.
C. A.
Woolhead
,
P. J.
McCormick
, and
A. E.
Johnson
, “
Nascent membrane and secretory proteins differ in FRET-detected folding far inside the ribosome and in their exposure to ribosomal proteins
,”
Cell
116
(
5
),
725
736
(
2004
).
75.
W.
Holtkamp
,
G.
Kokic
,
M.
Jager
,
J.
Mittelstaet
,
A. A.
Komar
, and
M. V.
Rodnina
, “
Cotranslational protein folding on the ribosome monitored in real time
,”
Science
350
(
6264
),
1104
1107
(
2015
).
76.
F.
Schweighofer
,
L.
Dworak
,
C. A.
Hammer
,
H.
Gustmann
,
M.
Zastrow
,
K.
Ruck-Braun
 et al, “
Highly efficient modulation of FRET in an orthogonally arranged BODIPY-DTE dyad
,”
Sci. Rep.
6
,
28638
(
2016
).
77.
J. Y.
Shao
,
H. Y.
Sun
,
H. M.
Guo
,
S. M.
Ji
,
J. Z.
Zhao
,
W. T.
Wu
 et al, “
A highly selective red-emitting FRET fluorescent molecular probe derived from BODIPY for the detection of cysteine and homocysteine: An experimental and theoretical study
,”
Chem. Sci.
3
(
4
),
1049
1061
(
2012
).
78.
T.
Jin
,
V.
Garcia-Lopez
,
F.
Chen
,
J.
Tour
, and
G. F.
Wang
, “
Imaging single molecular machines attached with two BODIPY dyes at the air-solid interface: High probability of single-step-like photobleaching and nonscaling intensity
,”
J. Phys. Chem. C
120
(
46
),
26522
26531
(
2016
).
79.
J. E.
Berlier
,
A.
Rothe
,
G.
Buller
,
J.
Bradford
,
D. R.
Gray
,
B. J.
Filanoski
 et al, “
Quantitative comparison of long-wavelength Alexa Fluor dyes to Cy dyes: Fluorescence of the dyes and their bioconjugates
,”
J. Histochem. Cytochem.
51
(
12
),
1699
1712
(
2003
).
80.
J.
Chao
,
S.
Ram
,
E. S.
Ward
, and
R. J.
Ober
, “
Ultrahigh accuracy imaging modality for super-localization microscopy
,”
Nat. Methods
10
(
4
),
335
(
2013
).
81.
R. J.
Marsh
,
K.
Pfisterer
,
P.
Bennett
,
L. M.
Hirvonen
,
M.
Gautel
,
G. E.
Jones
 et al, “
Artifact-free high-density localization microscopy analysis
,”
Nat. Methods
15
(
9
),
689
692
(
2018
).
82.
R.
Kasper
,
B.
Harke
,
C.
Forthmann
,
P.
Tinnefeld
,
S. W.
Hell
, and
M.
Sauer
, “
Single-molecule STED microscopy with photostable organic fluorophores
,”
Small
6
(
13
),
1379
1384
(
2010
).
83.
C. E.
Dalton
,
S. D.
Quinn
,
A.
Rafferty
,
M. J.
Morten
,
J. M.
Gardiner
, and
S. W.
Magennis
, “
Single-molecule fluorescence detection of a synthetic heparan sulfate disaccharide
,”
ChemPhysChem
17
(
21
),
3442
3446
(
2016
).
84.
S.
Letschert
,
A.
Gohler
,
C.
Franke
,
N.
Bertleff-Zieschang
,
E.
Memmel
,
S.
Doose
 et al, “
Super-resolution imaging of plasma membrane glycans
,”
Angew. Chem. Int. Ed.
53
(
41
),
10921
10924
(
2014
).
85.
H.
Jiang
,
B. P.
English
,
R. B.
Hazan
,
P.
Wu
, and
B.
Ovryn
, “
Tracking surface glycans on live cancer cells with single-molecule sensitivity
,”
Angew. Chem. Int. Ed.
54
(
6
),
1765
1769
(
2015
).
86.
E.
Kim
,
S.
Lee
,
A.
Jeon
,
J. M.
Choi
,
H. S.
Lee
,
S.
Hohng
 et al, “
A single-molecule dissection of ligand binding to a protein with intrinsic dynamics
,”
Nat. Chem. Biol.
9
(
5
),
313
318
(
2013
).
87.
G.
Kaur
,
J. S.
Lewis
, and
A. M.
van Oijen
, “
Shining a spotlight on DNA: Single-molecule methods to visualise DNA
,”
Molecules
24
(
3
),
491
(
2019
).
88.
K.
Morikawa
and
M.
Yanagida
, “
Visualization of individual DNA-molecules in solution by light-microscopy—Dapi staining method
,”
J. Biochem.
89
(
2
),
693
696
(
1981
).
89.
M.
Reuter
and
D. T. F.
Dryden
, “
The kinetics of YOYO-1 intercalation into single molecules of double-stranded DNA
,”
Biochem. Biophys. Res. Commun.
403
(
2
),
225
229
(
2010
).
90.
S.
Stanojcic
,
N.
Kuk
,
I.
Ullah
,
Y.
Sterkers
, and
C. J.
Merrick
, “
Single-molecule analysis reveals that DNA replication dynamics vary across the course of schizogony in the malaria parasite plasmodium falciparum
,”
Sci. Rep.
7
,
4003
(
2017
).
91.
X. M.
Yan
,
R. C.
Habbersett
,
J. M.
Cordek
,
J. P.
Nolan
,
T. M.
Yoshida
,
J. H.
Jett
 et al, “
Development of a mechanism-based, DNA staining protocol using SYTOX orange nucleic acid stain and DNA fragment sizing flow cytometry
,”
Anal. Biochem.
286
(
1
),
138
148
(
2000
).
92.
X. M.
Yan
,
R. C.
Habbersett
,
T. M.
Yoshida
,
J. P.
Nolan
,
J. H.
Jett
, and
B. L.
Marrone
, “
Probing the kinetics of SYTOX orange stain binding to double-stranded DNA with implications for DNA analysis
,”
Anal. Chem.
77
(
11
),
3554
3562
(
2005
).
93.
B. E.
Collins
,
L. F.
Ye
,
D.
Duzdevich
, and
E. C.
Greene
, “
DNA curtains: Novel tools for imaging protein-nucleic acid interactions at the single-molecule level
,”
Methods Cell Biol.
123
,
217
234
(
2014
).
94.
A. A.
Almaqwashi
,
T.
Paramanathan
,
I.
Rouzina
, and
M. C.
Williams
, “
Mechanisms of small molecule-DNA interactions probed by single-molecule force spectroscopy
,”
Nucl. Acids Res.
44
(
9
),
3971
3988
(
2016
).
95.
T.
Paramanathan
,
I.
Vladescu
,
M. J.
McCauley
,
I.
Rouzina
, and
M. C.
Williams
, “
Force spectroscopy reveals the DNA structural dynamics that govern the slow binding of actinomycin D
,”
Nucl. Acids Res.
40
(
11
),
4925
4932
(
2012
).
96.
H.
Miller
,
Z. K.
Zhou
,
A. J. M.
Wollman
, and
M. C.
Leake
, “
Superresolution imaging of single DNA molecules using stochastic photoblinking of minor groove and intercalating dyes
,”
Methods
88
,
81
88
(
2015
).
97.
A.
Sischka
,
K.
Toensing
,
R.
Eckel
,
S. D.
Wilking
,
N.
Sewald
,
R.
Ros
 et al, “
Molecular mechanisms and kinetics between DNA and DNA binding ligands
,”
Biophys. J.
88
(
1
),
404
411
(
2005
).
98.
C. U.
Murade
,
V.
Subramaniam
,
C.
Otto
, and
M. L.
Bennink
, “
Interaction of oxazole yellow dyes with DNA studied with hybrid optical tweezers and fluorescence microscopy
,”
Biophys. J.
97
(
3
),
835
843
(
2009
).
99.
K.
Gunther
,
M.
Mertig
, and
R.
Seidel
, “
Mechanical and structural properties of YOYO-1 complexed DNA
,”
Nucl. Acids Res
38
(
19
),
6526
6532
(
2010
).
100.
A. S.
Biebricher
,
I.
Heller
,
R. F. H.
Roijmans
,
T. P.
Hoekstra
,
E. J. G.
Peterman
, and
G. J. L.
Wuite
, “
The impact of DNA intercalators on DNA and DNA-processing enzymes elucidated through force-dependent binding kinetics
,”
Nat. Commun.
6
,
7304
(
2015
).
101.
K. P.
Carter
,
A. M.
Young
, and
A. E.
Palmer
, “
Fluorescent sensors for measuring metal ions in living systems
,”
Chem. Rev.
114
(
8
),
4564
4601
(
2014
).
102.
G. K.
Walkup
,
S. C.
Burdette
,
S. J.
Lippard
, and
R. Y.
Tsien
, “
A new cell-permeable fluorescent probe for Zn2
+,”
J. Am. Chem. Soc.
122
(
23
),
5644
5645
(
2000
).
103.
E. M.
Nolan
and
S. J.
Lippard
, “
Small-molecule fluorescent sensors for investigating zinc metalloneurochemistry
,”
Acc. Chem. Res.
42
(
1
),
193
203
(
2009
).
104.
T.
Hirano
,
K.
Kikuchi
,
Y.
Urano
, and
T.
Nagano
, “
Improvement and biological applications of fluorescent probes for zinc, ZnAFs
,”
J. Am. Chem. Soc.
124
(
23
),
6555
6562
(
2002
).
105.
K. R.
Gee
,
Z. L.
Zhou
,
D.
Ton-That
,
S. L.
Sensi
, and
J. H.
Weiss
, “
Measuring zinc in living cells.: A new generation of sensitive and selective fluorescent probes
,”
Cell Calcium
31
(
5
),
245
251
(
2002
).
106.
P. D.
Zalewski
,
I. J.
Forbes
, and
W. H.
Betts
, “
Correlation of apoptosis with change in intracellular labile Zn(Ii) using zinquin [(2-methyl-8-P-toluenesulphonamido-6-quinolyloxy)acetic acid], a new specific fluorescent-probe for Zn(Ii)
,”
Biochem. J.
296
,
403
408
(
1993
).
107.
E. L.
Que
,
R.
Bleher
,
F. E.
Duncan
,
B. Y.
Kong
,
S. C.
Gleber
,
S.
Vogt
 et al, “
Quantitative mapping of zinc fluxes in the mammalian egg reveals the origin of fertilization-induced zinc sparks
,”
Nat. Chem.
7
(
2
),
130
139
(
2015
).
108.
P.
Rivera-Fuentes
and
S. J.
Lippard
, “
SpiroZin1: A reversible and pH-insensitive, reaction-based, red-fluorescent probe for imaging biological mobile zinc
,”
ChemMedChem
9
(
6
),
1238
1243
(
2014
).
109.
P.
Rivera-Fuentes
,
A. T.
Wrobel
,
M. L.
Zastrow
,
M.
Khan
,
J.
Georgiou
,
T. T.
Luyben
 et al, “
A far-red emitting probe for unambiguous detection of mobile zinc in acidic vesicles and deep tissue
,”
Chem. Sci.
6
(
3
),
1944
1948
(
2015
).
110.
Y.
Han
,
J. M.
Goldberg
,
S. J.
Lippard
, and
A. E.
Palmer
, “
Superiority of SpiroZin2 versus FluoZin-3 for monitoring vesicular Zn2+ allows tracking of lysosomal Zn2+ pools
,”
Sci. Rep.
8
,
15034
(
2018
).
111.
R. Y.
Tsien
,
T.
Pozzan
, and
T. J.
Rink
, “
T-cell mitogens cause early changes in cytoplasmic free Ca(2+) and membrane-potential in lymphocytes
,”
Nature
295
(
5844
),
68
71
(
1982
).
112.
M. D.
Bootman
,
K.
Rietdorf
,
T.
Collins
,
S.
Walker
, and
M.
Sanderson
, “
Ca2+-sensitive fluorescent dyes and intracellular Ca2+ imaging
,”
Cold Spring Harbor Protoc.
2013
(
2
),
83
99
.
113.
G.
Grynkiewicz
,
M.
Poenie
, and
R. Y.
Tsien
, “
A new generation of Ca2+ indicators with greatly improved fluorescence properties
,”
J. Biol. Chem.
260
(
6
),
3440
3450
(
1985
).
114.
A.
Minta
,
J. P. Y.
Kao
, and
R. Y.
Tsien
, “
Fluorescent indicators for cytosolic calcium based on rhodamine and fluorescein chromophores
,”
J. Biol. Chem.
264
(
14
),
8171
8178
(
1989
).
115.
P.
Flagmeier
,
S.
De
,
D. C.
Wirthensohn
,
S. F.
Lee
,
C.
Vincke
,
S.
Muyldermans
 et al, “
Ultrasensitive measurement of Ca2+ influx into lipid vesicles induced by protein aggregates
,”
Eur. Biophys. J.
56
,
7750
7754
(
2017
).
116.
S.
De
,
D. C.
Wirthensohn
,
P.
Flagmeier
,
C.
Hughes
,
F. A.
Aprile
, and
F. S.
Ruggeri
 et al, “
Different soluble aggregates of A beta 42 can give rise to cellular toxicity through different mechanisms
,”
Nat. Commun.
10
,
1541
(
2019
).
117.
M.
Davies
,
C.
Jung
,
P.
Wallis
,
T.
Schnitzler
,
C.
Li
,
K.
Mullen
 et al, “
Photophysics of new photostable rylene derivatives: Applications in single-molecule studies and membrane labelling
,”
ChemPhysChem
12
(
8
),
1588
1595
(
2011
).
118.
M. H.
Chien
,
M.
Brameshuber
,
B. K.
Rossboth
,
G. J.
Schutz
, and
S.
Schmid
, “
Single-molecule optical absorption imaging by nanomechanical photothermal sensing
,”
Proc. Natl. Acad. Sci. U.S.A.
115
(
44
),
11150
11155
(
2018
).
119.
R.
Korlacki
,
M.
Steiner
,
H.
Qian
,
A.
Hartschuh
, and
A. J.
Meixner
, “
Optical Fourier transform spectroscopy of single-walled carbon nanotubes and single molecules
,”
ChemPhysChem
8
(
7
),
1049
1055
(
2007
).
120.
M.
Sun
,
K.
Mullen
, and
M.
Yin
, “
Water-soluble perylenediimides: Design concepts and biological applications
,”
Chem. Soc. Rev.
45
(
6
),
1513
1528
(
2016
).
121.
O.
Vakuliuk
,
Y. W.
Jun
,
K.
Vygranenko
,
G.
Clermont
,
Y. J.
Reo
,
M.
Blanchard-Desce
 et al, “
Modified isoindolediones as bright fluorescent probes for cell and tissue imaging
,”
Chemistry
25
(
58
),
13354
13362
(
2019
).
122.
M.
Kaur
and
D. H.
Choi
, “
Diketopyrrolopyrrole: Brilliant red pigment dye-based fluorescent probes and their applications
,”
Chem. Soc. Rev.
44
(
1
),
58
77
(
2015
).
123.
J.
Wang
,
W.
Xu
,
Z.
Yang
,
Y.
Yan
,
X.
Xie
,
N.
Qu
 et al, “
New diketopyrrolopyrrole-based ratiometric fluorescent probe for intracellular esterase detection and discrimination of live and dead cells in different fluorescence channels
,”
ACS Appl. Mater. Interfaces
10
(
37
),
31088
31095
(
2018
).
124.
O.
Shimomura
,
F. H.
Johnson
, and
Y.
Saiga
, “
Extraction, purification and properties of aequorin, a bioluminescent protein from the luminous hydromedusan Aequorea
,”
J. Cell. Comp. Physiol.
59
,
223
239
(
1962
).
125.
T.
Lenn
and
M. C.
Leake
, “
Experimental approaches for addressing fundamental biological questions in living, functioning cells with single molecule precision
,”
Open Biol.
2
,
120090
(
2012
).
126.
S. W.
Chiu
and
M. C.
Leake
, “
Functioning nanomachines seen in real-time in living bacteria using single-molecule and super-resolution fluorescence imaging
,”
Int. J. Mol. Sci.
12
(
4
),
2518
2542
(
2011
).
127.
M. C.
Leake
, “
Shining the spotlight on functional molecular complexes: The new science of single-molecule cell biology
,”
Commun. Integr. Biol.
3
(
5
),
415
418
(
2010
).
128.
B. T.
Bajar
,
E. S.
Wang
,
S.
Zhang
,
M. Z.
Lin
, and
J.
Chu
, “
A guide to fluorescent protein FRET pairs
,”
Sensors
16
(
9
),
1488
(
2016
).
129.
R. R.
Duncan
, “
Fluorescence lifetime imaging microscopy (FLIM) to quantify protein-protein interactions inside cells
,”
Biochem. Soc. Trans.
34
(
Pt 5
),
679
682
(
2006
).
130.
S. V.
Koushik
,
H.
Chen
,
C.
Thaler
,
H. L.
Puhl
 III
, and
S. S.
Vogel
, “
Cerulean, venus, and venusY67C FRET reference standards
,”
Biophys. J.
91
(
12
),
L99
L101
(
2006
).
131.
A.
Nenninger
,
G.
Mastroianni
,
A.
Robson
,
T.
Lenn
,
Q.
Xue
,
M. C.
Leake
 et al, “
Independent mobility of proteins and lipids in the plasma membrane of Escherichia coli
,”
Mol. Microbiol.
92
(
5
),
1142
1153
(
2014
).
132.
O.
Griesbeck
,
G. S.
Baird
,
R. E.
Campbell
,
D. A.
Zacharias
, and
R. Y.
Tsien
, “
Reducing the environmental sensitivity of yellow fluorescent protein. Mechanism and applications
,”
J. Biol. Chem.
276
(
31
),
29188
29194
(
2001
).
133.
N. C.
Shaner
,
P. A.
Steinbach
, and
R. Y.
Tsien
, “
A guide to choosing fluorescent proteins
,”
Nat. Methods
2
(
12
),
905
909
(
2005
).
134.
B. T.
Bajar
,
E. S.
Wang
,
A. J.
Lam
,
B. B.
Kim
,
C. L.
Jacobs
,
E. S.
Howe
 et al, “
Improving brightness and photostability of green and red fluorescent proteins for live cell imaging and FRET reporting
,”
Sci. Rep.
6
,
20889
(
2016
).
135.
A.
Esposito
,
M.
Gralle
,
M. A.
Dani
,
D.
Lange
, and
F. S.
Wouters
, “
pHlameleons: A family of FRET-based protein sensors for quantitative pH imaging
,”
Biochemistry
47
(
49
),
13115
13126
(
2008
).
136.
H.
Tsutsui
,
S.
Karasawa
,
Y.
Okamura
, and
A.
Miyawaki
, “
Improving membrane voltage measurements using FRET with new fluorescent proteins
,”
Nat. Methods
5
(
8
),
683
685
(
2008
).
137.
B.
George Abraham
,
K. S.
Sarkisyan
,
A. S.
Mishin
,
V.
Santala
,
N. V.
Tkachenko
, and
M.
Karp
, “
Fluorescent protein based FRET pairs with improved dynamic range for fluorescence lifetime measurements
,”
PLoS One
10
(
8
),
e0134436
(
2015
).
138.
K. S.
Sarkisyan
,
A. S.
Goryashchenko
,
P. V.
Lidsky
,
D. A.
Gorbachev
,
N. G.
Bozhanova
,
A. Y.
Gorokhovatsky
 et al, “
Green fluorescent protein with anionic tryptophan-based chromophore and long fluorescence lifetime
,”
Biophys. J.
109
(
2
),
380
389
(
2015
).
139.
L.
Wang
,
W. C.
Jackson
,
P. A.
Steinbach
, and
R. Y.
Tsien
, “
Evolution of new nonantibody proteins via iterative somatic hypermutation
,”
Proc. Natl. Acad. Sci. U.S.A.
101
(
48
),
16745
16749
(
2004
).
140.
K. A.
Lukyanov
,
D. M.
Chudakov
,
S.
Lukyanov
, and
V. V.
Verkhusha
, “
Innovation: Photoactivatable fluorescent proteins
,”
Nat. Rev. Mol. Cell Biol.
6
(
11
),
885
891
(
2005
).
141.
M.
Plank
,
G. H.
Wadhams
, and
M. C.
Leake
, “
Millisecond timescale slimfield imaging and automated quantification of single fluorescent protein molecules for use in probing complex biological processes
,”
Integr. Biol.
1
(
10
),
602
612
(
2009
).
142.
A. F.
Bell
,
D.
Stoner-Ma
,
R. M.
Wachter
, and
P. J.
Tonge
, “
Light-driven decarboxylation of wild-type green fluorescent protein
,”
J. Am. Chem. Soc.
125
(
23
),
6919
6926
(
2003
).
143.
D. M.
Chudakov
,
V. V.
Verkhusha
,
D. B.
Staroverov
,
E. A.
Souslova
,
S.
Lukyanov
, and
K. A.
Lukyanov
, “
Photoswitchable cyan fluorescent protein for protein tracking
,”
Nat. Biotechnol.
22
(
11
),
1435
1439
(
2004
).
144.
S. L.
Spencer
,
S. D.
Cappell
,
F. C.
Tsai
,
K. W.
Overton
,
C. L.
Wang
, and
T.
Meyer
, “
The proliferation-quiescence decision is controlled by a bifurcation in CDK2 activity at mitotic exit
,”
Cell
155
(
2
),
369
383
(
2013
).
145.
S.
Regot
,
J. J.
Hughey
,
B. T.
Bajar
,
S.
Carrasco
, and
M. W.
Covert
, “
High-sensitivity measurements of multiple kinase activities in live single cells
,”
Cell
157
(
7
),
1724
1734
(
2014
).
146.
U.
Resch-Genger
,
M.
Grabolle
,
S.
Cavaliere-Jaricot
,
R.
Nitschke
, and
T.
Nann
, “
Quantum dots versus organic dyes as fluorescent labels
,”
Nat. Methods
5
(
9
),
763
775
(
2008
).
147.
O.
Chen
,
J.
Zhao
,
V. P.
Chauhan
,
J.
Cui
,
C.
Wong
,
D. K.
Harris
 et al, “
Compact high-quality CdSe-CdS core-shell nanocrystals with narrow emission linewidths and suppressed blinking
,”
Nat. Mater.
12
(
5
),
445
451
(
2013
).
148.
B.
Dubertret
,
P.
Skourides
,
D. J.
Norris
,
V.
Noireaux
,
A. H.
Brivanlou
, and
A.
Libchaber
, “
In vivo imaging of quantum dots encapsulated in phospholipid micelles
,”
Science
298
(
5599
),
1759
1762
(
2002
).
149.
X.
Michalet
,
F. F.
Pinaud
,
L. A.
Bentolila
,
J. M.
Tsay
,
S.
Doose
,
J. J.
Li
 et al, “
Quantum dots for live cells, in vivo imaging, and diagnostics
,”
Science
307
(
5709
),
538
544
(
2005
).
150.
K. D.
Wegner
and
N.
Hildebrandt
, “
Quantum dots: Bright and versatile in vitro and in vivo fluorescence imaging biosensors
,”
Chem. Soc. Rev.
44
(
14
),
4792
4834
(
2015
).