
Organisms can significantly condense their genomes to fit within cells through various DNA-binding proteins (1, 2). Several traditional biochemical methods are available for studying DNA-protein interactions, including electrophoretic mobility shift assay (EMSA) (3), cross-linking (4), footprinting (5), chromatin immunoprecipitation (6), and systematic evolution of ligands by exponential enrichment (SELEX) (7). These methods typically involve bulk experiments with vast numbers of molecules in solution. These ensemble measurements provide averaged-out information, often masking essential details and the heterogenous nature of DNA-protein interactions (8, 9). In contrast, single-molecule (SM) techniques enable the investigation of individual molecules in real-time, unveiling even the finest details of complex dynamic processes on a molecular level (8, 9).
Single-molecule approaches have opened new avenues for studying DNA-protein interactions. They have also provided insight into chromosome organization in greater details. They mainly fall into force- (or torque-) and fluorescence-based categories. Force-based experiments involve stretching or twisting DNA. They allow researchers to measure any changes in various observables, such as DNA end-to-end length in the presence of DNA-binding proteins. Fluorescence-based experiments directly visualize proteins on DNA to obtain their localization and distribution information.
Examples of force-based single-molecule techniques include tethered particle motion, optical tweezers, and magnetic tweezers. Tethered particle motion (TPM) analysis involves tracking movement of a metallic or polystyrene bead attached to one end of DNA while the other end is anchored to a surface (10). This immobilization restricts the bead’s Brownian motion in solution, with the extent of restriction correlating to the length of DNA molecule. Thus, changes in the bead’s motion provide information about DNA shortening events. TPM can detect DNA length changes of at least 10-15 nm (equivalent to 60-70 base pairs) for DNA molecules ranging from 100 to 4,000 base pairs in size (11). This makes TPM particularly suitable for studying proteins that cause DNA shortening through bending, looping, or translocation along the DNA strand (12, 13). Optical tweezer (OT) is another force-based single-molecule tool that offers the ability to measure both force and motion in real-time. In OT, a strong light intensity gradient is formed when a laser beam is focused on a tiny spot by a high numerical aperture microscope objective (14, 15). A dielectric bead near this focus is trapped by balancing scattering and gradient forces. In single-molecule studies of DNA-protein interactions, OT experiments often involve tethering one end of a DNA molecule to a trapped bead. The other DNA end is anchored to a glass surface (single OT) or another bead via a micropipette or a second laser beam (double OT). By moving either the surface (single OT) or one of the trapped beads (double OT) away from the other bead, tension is applied to the nucleoprotein complex (16). Magnetic tweezer (MT) experiments involve anchoring one end of a DNA molecule to a surface while attaching the other end to a micron-sized magnetic bead. A magnetic field created by permanent magnets or electromagnets positioned above the surface attracts magnetic beads vertically, stretching the DNA. By adjusting the distance between the magnet and the surface, researchers can control the force exerted on the DNA, ranging from 0.001 to 100 pN (17, 18), thus altering DNA extension as desired. Additionally, magnets can be rotated on the x-y plane to induce a torque on the DNA molecule (19, 20).
Förster resonance energy transfer (FRET) is a prominent example of a fluorescence-based single-molecule technique. It is based on nonradiative energy transfer from a donor fluorophore to an acceptor fluorophore (either within the same molecule or on another interacting molecule) via a dipole-dipole interaction (11, 21). Since FRET efficiency depends on distance between the donor and acceptor, smFRET allows observing dynamic conformational changes in a 2-8 nm range within a molecule or two nearby molecules (11, 21). DNA sensors with a FRET pair are typically used to study interactions with proteins that alter the distance between them, such as by DNA bending (22, 23). Alternatively, attaching the FRET acceptor to DNA and the donor to a DNA-interacting protein enables measurement of DNA-protein interactions with dynamic temporal resolutions at the single-molecule level (24-26).
Force- and fluorescence-based single-molecule approaches to DNA-protein interaction are not mutually exclusive. Optical or magnetic tweezers can be readily implemented with fluorescence microscopy (16, 27-29). In addition, applying force to multiple molecules is feasible in advanced versions of force-based OT and MT setups (30, 31). However, those multiplexed single-molecule tools require non-trivial instrumentation and calibrations (32). This mini review will illustrate a single-molecule flow-stretching assay as a versatile force-fluorescence hybrid tool. We will describe how this technique can be utilized in multiplexed DNA-protein interaction studies. DNA curtain assay, single-molecule DNA motion capture techniques, and protein-induced fluorescence enhancement (PIFE) are introduced as examples of DNA flow-stretching assay variants.
Single-molecule DNA flow-stretching assays (Fig. 1A) combine DNA stretching by hydrodynamic force with fluorescence imaging of DNA or DNA-binding proteins. DNA molecules are anchored onto a microfluidic sample chamber surface through strong non-covalent interactions, often utilizing streptavidin (or neutravidin)-biotin interaction with a high affinity (KD ∼10-15 M) (33). Upon applying laminar buffer flow, single-tethered DNA molecules experience hydrodynamic force, stretching them along the direction of the buffer flow (Fig. 1B). End-to-end DNA lengths, flexibility, and elasticity of flow-stretched DNAs, especially for long DNAs such as bacteriophage lambda DNA (48.5 kbp), have been well described with a worm-like chain (WLC) model (34-36). Theoretical estimates and experimental measurements have shown that increasing shear flow rates can lead to more stretching of surface-immobilized DNA molecules (up to about 90% of their full contour length) by overcoming entropic barriers (35, 37). Flowing buffer can also reduce lateral movement of immobilized DNA molecules. If the second end of surface-immobilized DNA is functionalized, buffer flow can be used to tether both DNA ends to the surface of the microfluidic sample chamber (37-39). Since flow-stretched DNAs are close to the cover glass surface (∼0.2 μm for bacteriophage lambda DNA) (40), DNA flow-stretching assays are readily implemented under a total internal reflection fluorescence (TIRF) microscope (Fig. 1A). Here, the excitation light source (typically laser light) guided either through a microscope objective (objective-type TIRF microscopy) or via a prism positioned above the flow cell (prism-type TIRF microscopy) is incident at a critical angle or above. When light source is reflected at the interface between the cover glass (or the quartz slide) and buffer, rapidly-decaying evanescent field is created and excites only fluorescent dyes close to the cover glass (or the quartz slide) surface (Fig. 1A). Thus, TIRF microscopy allows selective visualization of DNA-protein interactions at the single-molecule level with minimal background noise, even when fluorescently-labeled proteins diffuse in a solution. An advantage of using an objective-type TIRF microscope is that a microscope flow cell can be easily mounted on a high (> 1.4) numerical aperture objective and immersion oil. However, autofluorescence from the immersion oil and stray excitation rays sets limitations on achieving high signal-to-background ratio (41, 42), although new generations of objectives minimize this limitation (43). In general, when prism-type TIRF configuration is implemented, better signal-to-background ratios and an enhanced quality of the evanescent field are achieved (41, 42). Alternatively, highly inclined and laminated optical sheet (HILO) microscopy (44) can be employed instead of TIRF microscopy (32). However, we should mention that the selection depends on the specific experimental goals. When high sensitivity for surface interaction is demanded, TIRF can be a better choice. In addition, TIRF exhibits higher signal-to-noise (SNR) ratio with less photobleaching, and requires relatively simpler setup. On the other hand, HILO can be preferred where extreme resolution in a localized region is required. Especially, when imaging depth is required, HILO offers higher resolution and better contrast (45). Individual DNA molecules are typically labeled with fluorescent intercalating dyes to observe their full contour lengths, or their (untethered) free ends are tagged with a fluorescent probe such as fluorescent quantum dots or organic fluorophores (Fig. 1C). Proteins of interest are labeled with bright organic fluorophores or quantum dots.
DNA flow-stretching assays require immobilizing and stretching DNA molecules of various lengths. Longer DNA substrates are particularly useful as they can be stretched along a surface, making real-time visualization and tracking of protein movement on DNA convenient at the single-molecule level. Among long DNAs, bacteriophage λ-DNA, a 48,502 base pair molecule, is one of the favorite choices in single-molecule flow-stretching experiments for studying DNA-protein interactions due to its length, commercial availability, and ease of tagging other small molecules through its 12-nucleotide single-stranded overhangs on each end. For instance, annealing single-stranded overhangs with biotin- and digoxigenin-coupled complementary oligos allows tethering λ-DNA to flow-cell surface through biotin-streptavidin (neutravidin) interactions and labeling the λ-DNA with an anti-digoxigenin antibody-conjugated quantum dot, respectively (46-50). Alternatively, this single-molecule technique involves attaching a 2.8-μm super-paramagnetic bead to one end of a surface-tethered DNA molecule (32, 51). This DNA-bead combination is then subjected to a magnetic force generated by a magnet positioned above the experimental setup. Levitating the magnetic bead helps minimize unwanted interactions between the bead and the surface. When a laminar flow is applied, DNA stretching occurs due to a drag force (acting on the bead) proportional to the flow speed and bead size. The motion of the bead is often observed using dark field microscopy to achieve high signal-to-noise ratios (32). Spherical symmetry of the bead allows for precise tracking of its position, typically with an accuracy of 1-10 nanometers, using sub-pixel fitting techniques (52). Investigations of DNA-binding proteins often necessitate DNA substrates with specific protein-binding sequences or other chemical modifications internally incorporated. Although modified DNA substrates can be prepared through multistep ligations after restriction enzyme cleavages (28), their low yields (despite time-consuming) motivate the development of more efficient methodologies. Previous efforts have resulted in new approaches for site-specific modifications of long DNA based on a Red-recombineering system (53) and a single plasmid DNA (54). Additionally, a chromatinized DNA template can be easily prepared by assembling nucleosomes on λ-DNA using a commercially available kit (47).
Another critical consideration for a single-molecule DNA flow-stretching assay is minimizing or eliminating non-specific adsorption of proteins, DNAs, and fluorescent probes onto microfluidic sample chamber surfaces. Various blocking proteins such as bovine serum albumin (BSA) and casein or surface chemistry can passivate the glass surface and minimize non-specific biomolecule adsorption (Fig. 1D) (55). Among the surface passivation methods, PEGylation is most commonly employed, where the cover glass surface is first amino-silanized and then treated with methoxy-PEG (polyethylene glycol) and a small percentage of biotin-PEG molecules (48, 55, 56). Biotin-PEGs are docking sites for streptavidin (or neutravidin) where biotinylated DNAs can bind. When experiments need to be performed under an environment similar to the one inside cells, constructing supported lipid bilayers on streptavidin-affixed flow cell surface is an attractive choice of surface passivation (Fig. 1D, 2A) (37). Although different surface passivation methods (57) are continuously developed, a technique that works for a protein of interest is not necessarily the best approach for other proteins. Thus, readers should empirically determine each experimental system’s most suitable surface passivation method. If non-specific surface-bound fluorescently-labeled proteins disrupt proper single-molecule observations of proteins on DNA, elevating the height of the DNA using DNA tightropes or skybridge is a way of getting around such difficulties. For DNA tightropes, also known as elevated DNA platforms, ends of individual DNA molecules are suspended on top of two micron-sized beads using a hydrodynamic flow (58). These beads are usually coated with poly-L-lysine, ensuring strong DNA binding. This setup eliminates the need for various immobilization methods based on surface chemistry and continuous buffer flow after tightropes formation (12). Importantly, DNA molecules elevated above the glass surface can reduce background noise and improve discrimination of quantum dot-tagged proteins on DNA (59). Elevated DNA height is also implemented with DNA skybridge modality through a 4 μm-high 3D thin quartz structure (Fig. 1D) (38). Notably, the Gaussian light sheet beam in the parallel direction of DNA produced by the 3D structure itself enables a high signal-to-noise imaging of even single fluorescently-labeled proteins on DNA independent of the surface condition (38).
DNA flow-stretching assays have been utilized to investigate a variety of DNA-binding proteins at the single-molecule level. These assays enable real-time observation and assessments of individual protein binding events, dwell times, and movement of proteins along elongated DNA molecules (49, 60, 61). In the case of a DNA mismatch initiation protein MutS study, single-molecule Förster resonance energy transfer (smFRET) can be combined with DNA flow-stretching to provide insights into its binding and diffusion dynamics (62).
Single-molecule DNA flow-stretching methods offer profound insight into various biological processes performed by DNA-binding proteins, especially when any DNA length changes are involved. Movements of magnetic beads or quantum dots labeled at the end of flow-stretched DNAs manifest DNA end-to-end length changes. For instance, structural maintenance of chromosomes (SMC), microchidia (MORC), and ParB proteins themselves are capable of compacting flow-stretched DNAs. Supplementing those proteins with ATP (for SMC and MORC) or CTP (for ParB) can lead to altered DNA compaction rates (47, 49, 50). Experiments with those proteins (both wild-type and mutant versions) provide invaluable information on underlying working mechanisms of these proteins. Intriguingly, single- and double-stranded DNAs exhibit different elastic properties. Under low force, the end-to-end distance for an ssDNA is shorter than that for a dsDNA (32, 63). Length changes during ssDNA-dsDNA conversions are particularly powerful in studying DNA replication dynamics (64-69).
How DNA-binding proteins efficiently locate specific DNA sequences for binding amidst numerous nonspecific sequences is one of the open questions in biology (70-72). Single-molecule studies offer a promising avenue to delve into this question, with DNA flow-stretch assays proving to be distinctly powerful for this purpose, as exemplified in a study of Blainey et al. (40).
Although single-molecule DNA flow-stretching assays are robust, they have limitations. For instance, while they are multiplexed single-molecule approaches, there are limits for the number of tethered DNAs in a field-of-view to prevent overlaps between DNAs. In the case of doubly-tethered DNA experiments, finding the tethered DNA orientation (5’ to 3’ versus 3’ to 5’) is elusive. Due to those limitations and the growing needs for specialized experimental requirements, various DNA flow-stretching assay variants such as DNA curtains, DNA motion capture assay, and protein-induced fluorescence enhancement (PIFE) assay have been developed.
DNA curtain technique involves aligning long DNA strands in rows along lipid diffusion barriers on a surface, enhancing throughput and improving the spatial resolution of fluorescent proteins along the DNA sequence. The spacing between diffusion barriers is optimized to maximize the number of curtain rows visible in one field-of-view. Techniques that construct lipid diffusion barriers have evolved. Manual etching with a diamond-tipped drill on a quartz (or fused silica) microscope slide is one of the earliest methods. It is simple but lacks control over barrier width, depth, and arrangement (37). Nanofabrication provides better solutions (73-75). Here, a slide is coated with a polymer film, patterned with an electron beam, and then covered with metal (chromium). This electron beam lithography-based approach enables precise surface patterning and creates durable barriers (76, 77). Similarly, nanoimprint lithography utilizes inductively coupled plasma etching to achieve physical barriers (78). However, both techniques have low throughput due to the need for raster scanning and expensive equipment. A recent UV lithography-based method offers a rapid, cost-effective alternative for high-throughput manufacturing of chromium-based barriers (75, 79). A flow cell is then set atop the quartz (or fused silica) slide with lipid bilayers formed on its interior surface. Subsequently, streptavidin (or neutravidin) molecules are introduced, binding tightly to biotin-phospholipids and serving as a platform for anchoring biotinylated DNAs. Upon buffer flow, lipid-tethered DNAs migrate until diffusion barriers stop them. Aligned DNAs then form “DNA curtains” (Fig. 2B). The DNA curtain methodology has been pivotal in DNA-protein interaction studies (74, 80-89) and is typically implemented on a prism-type TIRF microscope due to a higher signal-to-background (or signal-to-noise) ratio.
DNA curtains can be configured in various ways to fulfil diverse requirements of single-molecule experiments. The most common way is using single-tethered DNA curtains, where individual DNA molecules are anchored on a lipid bilayer at one end via biotin-streptavidin linkage (Fig. 2B) (74). TIRF imaging of these DNA molecules requires a continuous buffer flow that stretches them along the flow cell surface (Fig. 2B, C). This configuration can temporarily stop the fluid flow as a control to ensure that observed DNA and protein molecules are not non-specifically bound to the surface. Constant buffer flow maintains consistent tension on all anchored DNA molecules. Single-tethered DNA curtains can be assembled using linear or zigzag diffusion barriers. While linear barriers are simpler, they can cause an overlap between tethered DNA molecules. On the other hand, zigzag barriers ensure a spatial separation between each DNA molecule, avoiding overlaps (90).
Single-tethered DNA curtains have been extensively used to study real-time dynamics of various DNA-binding proteins. For instance, biological functions of Abo1 (a histone chaperone in the fission yeast) have been elucidated through DNA curtains of this configuration (91). Additionally, combining single-tethered DNA curtains with biochemical assays has provided deeper molecular insights into Abo1-mediated histone loading onto DNA and its interactions with DNA (92). Single-tethered DNA curtains could facilitate real-time fluorescence imaging of the self-assembly kinetics of artificial virus-like nucleocapsids and behavior of quantum-dot-labeled DNA translocating proteins (81, 93). These singly-tethered DNA curtains have been used to probe a crowding effect on protein translocation (80). Other prominent examples include single-molecule studies of proteins involved in DNA end resection, such as the human resectosome, DNA-dependent protein kinase, poly(ADP-ribose) polymerase-1, and exonuclease 1 (86, 94-96). Additionally, single-tethered DNA curtains have enabled efficient investigations of interactions between DNA and Rad51 recombinase, a crucial component of eukaryotic homologous recombination machinery (87, 97, 98).
Some single-molecule experiments are best conducted without continuous buffer flow due to the use of expensive reagents or the need to observe individual protein molecules on surface-immobilized DNA. The hydrodynamic force from buffer flow can interfere with protein behavior, disrupting interactions or introducing bias, particularly during target search or protein movement along DNA. However, double-tethered DNA curtains offer a solution. Here, DNA molecules are immobilized at both ends, allowing visualization of the entire DNA length without fluid flow (Fig. 2D) (82). This setup involves linear or zigzag barriers and downstream pedestals to anchor DNA. One end of the DNA attaches to the lipid bilayer via biotin-streptavidin interaction, aligning with barriers under buffer flow. The other end modified with digoxigenin binds to a pedestal coated with digoxigenin antibodies. This double configuration ensures uniform orientation of surface-fixed DNA molecules concerning their nucleotide sequence.
Single- and double-tethered configurations are the two main types of DNA curtains commonly used for detailed single-molecule studies of DNA-protein interactions. However, other configurations can also be designed for specialized experiments. For instance, DNA curtains can be assembled in a parallel array of double-tethered isolated (PARDI) molecule patterns. This array maintains a large spacing between adjacent DNA molecules to study individual DNA-protein interactions without the influence of a high local DNA concentration (99, 100). Conversely, crisscrossed DNA curtains create high local DNA concentrations by crossing double-tethered DNA molecules, allowing investigation of protein movement between closely spaced DNA substrates (99-101).
Conventional DNA curtains typically use double-stranded DNA (dsDNA) as a substrate. However, single-stranded DNA (ssDNA) also plays a crucial role in many DNA repair and replication processes. To delve deeper into these processes, ssDNA curtains have been developed, allowing fluorescence imaging of stretched ssDNA molecules at the single-molecule level (Fig. 2E) (79, 83, 88). During assembly, long ssDNA molecules are synthesized via rolling circle replication using ϕ29 DNA polymerase (DNAP), a circular ssDNA template (such as M13mp18), and a biotinylated primer (77). These ssDNA products are immobilized on a lipid bilayer through biotin-streptavidin interaction. They are aligned parallel to diffusion barriers under buffer flow. However, imaging ssDNA is challenging due to its compactness and inability to be fluorescently labeled with intercalating dyes. Replication protein A (RPA) fused with fluorescent proteins can be used to overcome these limitations by binding to ssDNA and straightening it. RPA-ssDNA complexes are then stretched under buffer flow for visualization using TIRF microscopy (Fig. 2E) (83, 88). Additionally, ssDNA curtains can be assembled in a double configuration, where RPA-ssDNA complexes are non-specifically adsorbed downstream of lipid diffusion barriers (83, 88). Another extension of ssDNA curtains is low-complexity ssDNA curtains, enabling real-time synthesis and visualization of individual ssDNA molecules. These curtains can minimize secondary structures in ssDNA, facilitating its stretching with minimal force. Low-complexity ssDNA curtains are beneficial for studying physical changes induced by interactions with other nucleic acids or ssDNA-binding proteins (102).
DNA curtain techniques have made substantial contributions in visualizing various events during DNA damage repair and helped elucidating mechanisms of associated proteins. For instance, doubly-tethered DNA curtain technique was employed in the study of quantum dot-labeled human XPC (XPC-RAD23B) protein, and revealed that it displays diffusive, immobile, and constrained motions on DNA (103). Second, it has been known that R-loop (three-stranded nucleic acid structures composed of a DNA-RNA hybrid and a single-stranded DNA) accumulation can lead to genomic instability. A recent study based on DNA curtain assay demonstrated that TonEBP (tonicity-responsive enhancer binding protein) searches and preferentially binds R-loops via 3D collision and 1D diffusion (104). Third, homologous recombination (HR) and non-homologous end joining (NHEJ) are two main DNA double-strand break repair pathways. The single-molecule technique showed that DNA-dependent protein kinase (DNA-PK) and Ku70/80 are released from broken DNA ends by MRN (MRE11, RAD50, and NBS1)-CtIP complex and helps our understanding on sequential transition from NHEJ to HR (95). Fourth, during HR, the broken DNA ends need to be processed to generate single-stranded DNA. Multiple proteins such as helicases/nuclease 2 (Dna2), BLM/Sgs1 (Bloom syndrome protein in budding yeast), Rad52, and phosphorylated RPA (pRPA) are involved in this end processing and resection steps. Extensive studies on mechanisms of these proteins were performed using DNA curtain-based approaches (105-108).
In summary, DNA curtains allow simultaneous real-time imaging of hundreds or even thousands of distinct fluorescently labeled DNA-protein complexes (73). Its high-throughput capability, especially when coupled with total internal reflection fluorescence (TIRF) microscopy, can facilitate the gathering of extensive statistical data in a single experiment. Like other DNA flow-stretching assays, DNA curtains offer comprehensive insights into single-molecule dynamics of DNA-protein interactions during replication (109-112), DNA damage repair, and other critical cellular processes.
In the single-molecule DNA flow-stretching assay, laminar flow through a microfluidic sample chamber applies a differential hydrodynamic force along flow-stretched DNAs (Fig. 3A) (36). Tether-point-proximal DNA regions experience the strongest force, while the free end of the DNA region experiences the lowest tension. This leads to a nonuniform DNA stretching along its length (Fig. 3A) (36, 113). Due to the lowest tension and the minimal degree of DNA stretching at its free end, DNA-bridging proteins initiate a DNA compaction preferentially from the free DNA end. The compaction continues sequentially until reaching the DNA tether point (Fig. 3B). On the contrary, DNA bending (or wrapping) proteins compact a flow-stretched DNA simultaneously along its DNA length as they are relatively insensitive to tension variations along the DNA. Only incomplete DNA compaction (not all the way to the tether point) is achieved by DNA benders due to a limited number of DNA bending sites and modest decreases of DNA length caused by each DNA bender (Fig. 3B) (46, 49). Single-molecule DNA motion capture techniques have emerged to address different DNA compaction patterns between DNA bridgers and benders (46). First, bacteriophage λ-DNAs contain five EcoRI-binding sites. Five anti-His antibody-conjugated quantum dots can specifically bind those five binding sites through catalytically inactive His6-EcoRI(E111Q) mutant proteins (114) that recognize and bind EcoRI-binding sites without cleaving them. Simultaneously tracking all labeled quantum dots on flow-stretched DNA in real-time in the presence of a protein of interest allows one to distinguish different DNA compaction mechanisms. Applications of DNA motion capture assay have indicated that Bacillus subtilis ParB (Spo0J) and Caenorhabditis elegans MORC-1 proteins can compact DNA by a DNA bridging (loop trapping) mechanism while HBsu protein bends or wraps DNAs (46, 47). Interestingly, DNA motion capture assay has implied that Bacillus subtilis SMC (structural maintenance of chromosomes) protein adopts both bending and bridging DNA compaction mechanisms (49).
Protein-induced fluorescence enhancement (PIFE) is a photophysical phenomenon involving enhanced fluorescence signal when a DNA-binding protein is in close proximity to a fluorescent probe (Fig. 3C) (115, 116). When a Cy3 dye is excited, it can form a nonradiative cis isomer derived by the rotation of its carbon-carbon double bonds and thus compete with fluorescence emission. However, the presence of a protein and its interaction with the Cy3 dye can reduce the cis-trans isomerization and increase the singlet-state lifetime. It also enhances fluorescence intensity (115, 116). In addition to Cy3, PIFE has been shown with Cy5, DY547, and Alexa dyes, where two rings interconnected by carbon-carbon double bonds can rotate with respect to each other by cis-trans isomerization (115, 117). On the contrary, no PIFE has been observed with Cy3B due to the lack of cis-trans isomerization (117).
The PIFE phenomenon can be combined with a single-molecule DNA flow-stretching assay where DNAs are sparsely labeled with fluorescent dyes (for example, one dye per kilobase on average). Once proteins arrive on flow-stretched DNAs, integrated fluorescent intensity along each DNA increases due to PIFE (Fig. 3D). Any associated DNA compaction events are directly visible due to the presence of fluorescent probes on DNA (Fig. 3D) (113, 118). Another critical benefit of employing PIFE-based single-molecule DNA flow-stretching is that, since proteins of interest lack any fluorescent probes or modifications, conducting experiments with high protein concentrations is readily achievable (113). In summary, simultaneously monitoring of protein binding and DNA end-to-end length changes provides invaluable information on protein-DNA interactions.
A single-molecule DNA flow-stretching assay is a potent hybrid technique that can be used to investigate how DNA-binding proteins function and interact with DNA substrates by combining DNA-stretching force with fluorescence imaging of DNA or proteins. Many single-molecule biophysical techniques previously developed can only manipulate one biomolecule at a time. However, DNA flow-stretching assays allow observation of multiple individual DNA molecules simultaneously under fluid flow using a fluorescence microscope. This increased throughput facilitates efficient studies of dynamic DNA-protein interactions at the single-molecule level. One of the most attractive aspects of DNA flow-stretching method is that the technique and experimental setup can be readily modified for specific experimental needs, leading to the developments of variants such as DNA curtain, DNA motion capture assay, and protein-induced fluorescence enhancement (PIFE)-based DNA flow-stretching assay.
Although the single-molecule DNA flow-stretching approach and its variants are mature DNA-protein study modalities, there is still room for future improvements. First, expanding the assay to multi-color imaging will significantly enhance the quality of information that we can acquire. For example, labeling the free end of DNAs and proteins with spectrally distinct quantum dots or fluorescent dyes will enable us to correlate protein distribution on DNA with its function (DNA compaction) (49). Labeling multiple proteins with different color probes can reveal how different proteins interact with each other on a flow-stretched DNA. Second, our recent single-molecule DNA flow-stretching work has shown that small amino acid tags introduced for labeling could quantitatively and qualitatively alter protein properties (50). Whenever modifications of proteins are needed, their properties should be carefully validated to avoid any misinterpretation. Third, developments of improved DNA substrate preparation methods will help us implement more physiologically relevant experiments. Single-molecule DNA flow-stretching assay will remain one of the preferred methodology choices in the future. It will help us answer many open questions in the field of chromosome dynamics.
We thank Aleksey Aleshintsev and Brandon Shields for their critical reading and comments. We apologize to colleagues whose work was not discussed here due to space limitations. This work was supported by National Institutes of Health (NIH/NIGMS) R35GM143093.
The authors have no conflicting interests.
![]() |
![]() |