Quantitative single-molecule FLIM and PIE-FRET imaging of biomolecular systems

Background

The structural dynamics of proteins and nucleic acids are critical for their function in many biological processes but investigating these dynamics is often challenging with traditional techniques. Time-correlated single photon counting (TCSPC) coupled with confocal microscopy is a versatile biophysical tool that enables real-time monitoring of biomolecular dynamics in a variety of systems, across many timescales. Quantitative single-molecule time-resolved fluorescence methods are uniquely positioned to investigate transient interactions and structural changes, yet application in complex biological systems remains limited by technical and analytical challenges. Combining fluorescence lifetime imaging microscopy (FLIM) with pulsed interleaved excitation Förster resonance energy transfer (PIE-FRET) offers a robust approach to overcome these barriers, enabling accurate distance measurements and dynamic studies across diverse sample types.

Methods

We describe practical workflows for implementing FLIM/PIE-FRET for quantitative measurements of nanoscale distances and dynamic processes in various biomolecular systems on a commercial microscope. Benchmark DNA constructs, RNA/DNA hybrids, liposome-encapsulated enzymes, and live Saccharomyces cerevisiae strains were prepared and imaged. Correction factors for FRET efficiency recovery were determined from diffusion-based experiments, and results were validated by direct comparison of intensity- and lifetime-based analyses.

Results

FRET efficiencies from both intensity- and lifetime-based analyses were consistent across systems. DNA standards reproduced expected values, RNA/DNA hybrids reported on substrate dynamics, liposome encapsulation enabled single-enzyme conformational probing, and live-cell imaging revealed transient protein–protein interactions during ribosome biogenesis.

Discussion

This work establishes guidelines for implementing FLIM/PIE-FRET as an accessible method to interrogate nanoscale distances, conformational dynamics, and protein–protein interactions both in vitro and in live cells. The strategies outlined here facilitate broader adoption of quantitative single-molecule time-resolved fluorescence in structural and cell biology.

Supplementary Information

The online version contains supplementary material available at 10.1186/s44330-025-00048-1.

Keywords: Single-molecule, FLIM, PIE-FRET, RNA/DNA hybrids, Protein dynamics, Ribosome biogenesis, Liposome encapsulation

Time-resolved fluorescence microscopy and spectroscopy

Time-resolved fluorescence measurements were collected using a custom MicroTime 200 (PicoQuant - Berlin, Germany) with SymPhoTime64 software for data acquisition and analysis (Fig. 1A). The modular time-resolved confocal microscope system is built around an inverted Olympus IX-83 microscope body with a side port for optics and a piezoelectric z stage. It is equipped with three picosecond pulsed diode lasers (485, 531, and 636 nm) and a laser driver module (SEPIA II) capable of operating in pulsed (up to 80 MHz) and continuous wave (cw) modes. The lasers are cleaned up with narrowband filters, fiber-coupled into the main optical unit (MOU), directed to a main quad-band dichroic mirror, and focused onto the sample with a 60 ×ばつ 1.2 numerical aperture (NA) water immersion lens (Olympus UPlanSApo, Superachromat). A fast galvo beam scanning module (FLIMbee) with a 0.5 μs dwell time (lower limit) is used for laser beam scanning. Fluorescence from the sample is collected with the same objective and spatially filtered through an exchangeable circular confocal pinhole (100 μm used for these experiments) and directed to one or two single photon avalanche diodes (SPADs – Excelitas). For FRET experiments with 531 and 636 nm laser excitation, the fluorescence is spectrally filtered using a dichroic mirror (635 long pass) placed between the two detectors and bandpass filters in front of each detector to separate donor (582/64, ‘Channel 2’) and acceptor (690/70, ‘Channel 1’) emission for most experiments unless otherwise noted. For TCSPC, a multichannel event timer with 10 ps resolution and 650 ps deadtime (MultiHarp 150) in time-tagged time-resolved (TTTR) measurement mode applies time tags to individual photons detected at each SPAD, both relative to the laser pulse rate (nanotime) and relative to the start of the experiment (macrotime). Generating real-time histograms of the different time tags enables simultaneous fluorescence intensity, lifetime, and photon correlation data collection. Detected photons are also marked with a location in the beam scan of the sample to enable spatial mapping for FLIM. All fluorescence experiments were performed at room temperature.

Operating the lasers in PIE mode (Fig. 1B) enables selective excitation and detection of separate donor, acceptor, and FRET signals on the nanosecond (ns) time scale [34]. The laser excitation is synchronized in the SymPhoTime software such that the 531 and 636 nm lasers pulse in an alternating manner, successively exciting acceptor and donor molecules directly. Detected single photons are associated with the exciting laser pulse and the arrival detector to separate donor, acceptor, and FRET signals.

Benchmark DNA samples

DNA oligomers with sequences from a benchmark single-molecule FRET study [39] were purchased from Integrated DNA Technologies (IDT) with an internal amino modifier C6 deoxythymidine (dT) at the sites for fluorescent labeling (Supplementary Table S1 for sequences and Fig. 1C for schematic). For labeling, we incubated each donor (D) or acceptor (A) oligomer with a 20x molar excess of the appropriate Atto-NHS ester dye [Sigma-Aldrich 92835 (Atto 550 NHS ester) and 18373 (Atto 647 N NHS ester)] in a fresh sodium bicarbonate buffer in 18.2 MΩ double-deionized water (ddiH₂O), adjusted to pH 8.5 with 6 M HCl and reacted overnight at 4 °C. The oligomers were purified with a ZYMO DNA oligo purification kit (D4060), and the UV/Vis absorbance spectrum was measured to determine labeling efficiency (Supplementary Table S2). The labeled D and A oligomers were then annealed in equimolar amounts in 1x phosphate-buffered saline (PBS) by heating to 92 °C for 4 min in a Bio-Rad T100 thermal cycler and cooling to room temperature at − 1 °C per minute.

RNA/DNA hybrid substrate

All oligos were purchased from IDT. We designed an RNA/DNA hybrid substrate (Mid FRET U, MFU) labeled with a FRET pair (Atto550-Atto647N) to report cleavage and dynamics of the RNA substrate [46]. Once the oligos are annealed, both donor and acceptor are located on the substrate. The RNA/DNA hybrid strand has 15 RNA bases, including a single uridine (U) in the middle and 45 DNA bases [5’- (ATTO647N) rArGrA rArArA rArGrA rArAr(U or A) rArArG AGA ATA TCG GCA CGC TCG TGA GGT ATT TCA CAC CTT AAG CCA GCC − 3’, Fig. 3A]. The complementary DNA strand has 45 bases [5’- (Biosg) GGC TGG CTT AAG GTG TGA AAT ACC TCA CGA GCG TGC CGA TA/dT-Atto550/TCT − 3’]. The RNA/DNA hybrid oligo was purchased with the Atto647N modification. To produce the Atto550-labeled DNA complement, we ordered the oligo with an internal amino modifier C6 deoxythymidine (dT) at the site for fluorescent labeling and reacted it with Atto 647 N NHS ester (Sigma-Aldrich 18373) as described above for the benchmark DNA. The donor-labeled (Atto550) DNA strand was annealed to the acceptor-labeled (Atto647N) hybrid RNA/DNA strand by heating equimolar amounts of both strands in annealing buffer (10 mM Tris pH 7.5, 50 mM NaCl, 1 mM EDTA) to 95 °C for 5 min in a Bio-Rad T100 thermal cycler, followed by cooling to 25 °C over 70 min at −1 °C per minute, and then stored at 4 °C until further use. To prepare samples for attachment onto quartz slides, homebuilt flow cells were constructed and functionalized as described below. Immediately before use, each channel was flushed with 30 mM HEPES, 100 mM NaCl, and 0.1 mg/mL of streptavidin (in 30mM HEPES and 100 mM NaCl) was introduced to slide channels and incubated for 15 min. The channels were flushed again with 30 mM HEPES, 100 mM NaCl and 15 pM of MFU sample was injected into the channels and incubated for 15 min.

An image buffer was prepared to reduce blinking and photobleaching of the dyes. Image Buffer A (IA) was prepared by dissolving 200 mg of glucose in 10 mL of 30 mM HEPES,100 mM NaCl, 5mM DTT and adding 2 mL of cyclooctotetraene (COT). The solution was then filtered through a 0.22 Inline graphicm filter. Image buffer B (IB) was prepared by adding 14 mL of 95,000 U/mL catalase (CAT), 2 mL of 60 mg/mL glucose oxidase (GOX), and 2 mL of 14.3 M 2-Mercaptoethanol (βME) to 86 mL of the HEPES/NaCl/DTT buffer. Image buffer C (IC) was prepared by adding 2 mL of IB and 2 mL 14.3 M βME to 200 mL of IA. IC was flushed through the slide channels before viewing immobilized samples.

Protein expression and purification

Thermus aquaticus (Taq) MutL proteins (GenBank: AAB40601.2), with a 6x histidine tag and a single cysteine variation near the absolute N-terminal between the his-tag and the gene for fluorescent labeling, were overexpressed in E.coli BL21(DE3) cells with ampicillin selection and 1 mM IPTG induction at OD 0.7, followed by expression overnight at 25 °C. Cell pellets were stored at – 80 °C until ready for purification. Cell pellets were disrupted by tip ultrasonication, followed by cell lysate purification with cobalt-charged (TALON) affinity chromatography. Purified Taq MutL proteins were labeled with cysteine-maleimide attachment chemistry. Approximately one nanomole of purified protein (250 μL) was pipetted into a sterile 1.5 mL tube and 1 μL of 0.05 mM TCEP was added to reduce disulfide bonds for about 15 min. For double labeling, a mixture of two maleimide dye aliquots (Alexa Fluor 555 and Alexa Fluor 647, 20 nanomoles each) was dissolved in 1 μL of DMSO. The protein suspension was added to the dissolved dye, and the mixture reacted for 30 min at room temperature, followed by 2 h at 4 °C. Free dye was removed by purification with a P6 (Bio-Rad) gravity column. Three native cysteines were not accessible for labeling, therefore, a single cysteine site per monomer was labeled. SDS-PAGE gels confirmed the presence of MutL (MW_monomer ~ 60 kDa with the histidine tag, Supplementary Information, Section V).

Protein and fluorescent dye encapsulation

MutL proteins or fluorescent dyes (Alexa Fluor 647) were encapsulated in small/large unilamellar lipid vesicles (or liposomes) [47–49]. A solution of 1% biotinyl cap-phosphoethanolamine (Avanti Polar Lipids, 870277P) in Egg PC (Avanti Polar Lipids, 840051 C) was prepared in chloroform. The lipids were dried in a glass culture tube with an argon gas stream to evaporate the chloroform and then placed in a vacuum chamber until completely dry for several hours to overnight. Dried lipids were rehydrated by gentle vortex in a buffer containing 20 mM Tris HCl, 5 mM MgCl₂, 100 mM sodium acetate, 2% glucose, and 0.02% cyclooctatetraene for imaging (500 μL or 1 mL to a final lipid concentration of 1 mg/mL). The addition of glucose is for a glucose oxidase/catalase oxygen scavenging system, which was not used in these experiments. Protein or fluorescent dye was added to the lipid suspension to a final concentration of 5 nM. The lipid suspension was passed 21 times through an Avanti Polar Lipids mini extruder assembled with 200 nm diameter pore Whatman Nucleopore Track-Etch Membranes. A gravity Sepharose CL-4B (Sigma-Aldrich, 61970-08−9) column was used to separate free dye from the liposomes. The resulting liposomes were expected to be spherical with a diameter close to the membrane pore size [50] with encapsulated protein or dye (Fig. 3B).

Preparation of functionalized flow cells for single molecule surface attachment

The benchmark oligomers were purchased with a 5’ biotin modification on the acceptor strand (Supplementary Table S1) for attachment to homebuilt flow cells for imaging and spectroscopy. Quartz slides were predrilled prior to functionalization using 0.75 mm diamond-tipped drill bits (Triple Ripple, Arrowhead Lapidary) and a micro drill press. Slides and coverslips were cleaned by sonication in acetone, ethanol, and aqueous 1 M potassium hydroxide for 15 min each. Cleaned slides were stored in ddiH₂O. To prepare the mixture of methoxy and biotinylated PEG-silane (mPEG and bPEG, respectively) for slide and coverslip functionalization, ~ 20 mg of mPEG (Laysan Bio MPEG-SIL-2000, mean M_w of 2 kDa) was dissolved in 80 μL of ddiH₂O and ~ 2 mg of bPEG (Laysan Bio Biotin-PEG-SIL-3400, mean M_w of 3.4 kDa) was dissolved in 10 μL of ddiH₂O. After mixing 1 μL of the bPEG solution with the mPEG solution, 40 μL of bPEG/mPEG mixture was sandwiched between a dried slide and coverslip. The reaction proceeded overnight in a dark, humid environment at room temperature. The slides and coverslips were separated, rinsed with ddiH₂O, and dried with pressurized air. Once dried, another 40 μL of mPEG solution, prepared as above, was sandwiched between the slide and coverslip and allowed to react in a dark, humid environment at RT for ~ 30 min. The slides and coverslips were then rinsed and dried once more before assembling them into homebuilt flow cells for single-molecule measurements.

The flow cells were assembled by placing thin strips of double-sided tape between pairs of drilled holes on the slides to form individual channels of ~ 20 μL volume. The coverslip was then placed on the tape so that the mPEG/bPEG functionalized sides of the coverslip and slide faced one another. The cover slip was carefully pressed down to ensure the tape adhered fully to both sides and prevent leakage between neighboring channels. The edges of the chambers were sealed by applying epoxy and allowed to set for 1 h to overnight. Individual channels were washed three times with 100 μL of PBS. Next, 30 μL of 0.1 mg/mL streptavidin was injected into the chamber and allowed to bind to the biotinylated surface for 20 min. The chamber was washed again with PBS. Labeled, annealed biotinylated oligomers were added to the chamber at 10–15 pM and allowed to bind for 20 min. The chamber was rinsed three times to remove any unbound sample with 100 μL of PBS with added 2% glucose and 0.02% cyclooctatetraene for imaging. The addition of glucose is for a glucose oxidase/catalase oxygen scavenging system as described above.

Biosynthesis of DNA transformant

The sequence information for the template plasmid and primers used can be found in the Supplementary Information, Section VI. For seamless N-terminus tagging, we acquired yeast strains as a kind gift from Dr. Maya Schuldiner’s lab as a template for gene transfer into our background strain BY4741. Q5^® Hot Start High-Fidelity DNA Polymerase kit was used for all PCR reactions, and NEB’s Tm Calculator tool (https://tmcalculator.neb.com/#!/main) was used to calculate annealing temperatures from sequence inputs. For transformant generation, reactions were cycled 35x with a 4-minute extension time at 72 °C and purified using a QIAquick PCR purification kit. For gene transfer, parent strains were boiled in 20 mM NaOH at 95 °C, and 2 μL of boiled culture was used for PCR reactions.

Generation of S. cerevisiae strains

Homologous recombination was used to genetically engineer Saccharomyces cerevisiae strain BY4741 to endogenously tag the C-terminus of Nop7 and Rix7 with GFP, the C-terminus of Nsa1 with mCherry, and the N-terminus of Rea1 with mCherry. Briefly, the parent strain was grown at 30 °C at 220 rpm in 5 mL yeast extract peptone dextrose (YPD) media to log phase (OD 0.4–0.6), washed 1x with 1 mL ultrapure H₂O (UpH₂O), and the pellet resuspended in 100 μL of 0.1 M LiOAc in 10 mM Tris with 1 mM EDTA (Li-TE) at pH 7.5. Next, 15 μL of 10 mg/mL heat-denatured salmon sperm DNA and 10 μL of PCR generated transformant DNA (totaling 500–2500 ng) was added to the culture and incubated at 30 °C at 220 rpm for 30 min. Following that step, 1 mL of 40% PEG (MW 3500) in Li-TE buffer was added to cultures and incubated at 30 °C at 220 rpm for 30 min. Cultures were heat-shocked in a water bath at 42 °C for 1 h then outgrown overnight in 5 mL YPD at 30 °C at 220 rpm. Next day, the cultures were plated on selection plates and grown 3 days before individual colonies were selected and PCR verified. Nop7 and Rix7 strains were selected with Kanamycin (G418), Nsa1 was selected with hygromycin (Hyg), and Rea1 was selected with Nourseothricin (Nat1). For insertion validations, single yeast colonies were selected and boiled in 20 mM NaOH for 5 min at 95 °C, and 2 μL of boiled culture was cycled 25x with validation primers. Product size was checked on a 1% agarose gel.

Live S. cerevisiae cell imaging

S. cerevisiae BY4741 parent strains were genetically engineered to express fluorescent tags GFP and mCherry fused to proteins of interest as described above. For imaging, one colony was selected from the desired plate and grown in 5 mL YPD liquid medium, pH 5.8, at 30 °C and 220 rpm overnight (12–18 h). 100 μL was transferred into 5 mL of fresh media and incubated at 30 °C and 220 rpm for about 3–4 h. Cells were harvested in the log growth phase (OD 0.4–0.6) for optimal native protein expression by centrifugation (5 min at 4000 rpm and RT) of 1 mL of the liquid culture. The supernatant was discarded, and then the cells were washed three times with 1 mL of phosphate buffered saline (PBS) buffer (pH 7.4). The cells were then resuspended in 1 mL PBS.

20 μL of the resuspended cells were pipetted onto a coverslip and covered with a 1 cm x 1 cm agarose pad (1% (w/v) mixture of 1X PBS and agarose). 485 nm and 531 nm pulsed diode lasers were used for GFP and mCherry excitation, respectively. The power for each laser was set to 0.3–0.6 kW/cm² at a 20 MHz repetition rate for live cell imaging. The GFP and mCherry signals were split between ‘Channel 2’ and ‘Channel 1’ detectors with a 532 nm longpass dichroic filter. GFP emission was collected in ‘Channel 2’ with a 511/20 bandpass filter and mCherry emission was collected in ‘Channel 1’ with a 615/20 bandpass filter. FLIM images were captured using a 3–5 μs pixel dwell time and 100–150 frames were analyzed for each image obtained. The pixel size in each image was 100 nm.

Single-molecule FLIM and PIE-FRET imaging of benchmark DNA

The fast laser scanning confocal configuration of our microscope enables rapid imaging of single molecules attached to a surface, producing fluorescence lifetime imaging microscopy (FLIM) images that not only report the location of molecules, but also the fluorescence lifetime. Since FRET can be detected by calculating changes in the donor lifetime or fluorescence intensity, FLIM imaging can report on the energy transfer efficiency between two fluorescent reporter molecules that are attached to biomolecules of interest. Frequency and time-domain FLIM-FRET approaches have been utilized in live cell imaging for a variety of biological systems [51–57]. Acquiring sufficient photon counts to assign a fluorescence lifetime may limit its application to monitoring dynamic events. Here, we outline the combined use of FLIM and PIE-FRET signals (FLIM/PIE-FRET) that provide complementary information which can preserve spatial and dynamic information in vitro and in live cells.

To explore using FLIM/PIE-FRET in vitro, we attached annealed benchmark DNA samples to a biotinylated coverslip (Materials and Methods) and collected FLIM/PIE-FRET images, followed by single-molecule fluorescence intensity traces. After an initial FLIM acquisition, individual molecules in the images may be selected to probe single molecules or complexes over long periods of time and at high temporal resolution (until photobleaching occurs). The benchmark oligomers were purchased with a 5’ biotin modification (Supplementary Table S1) for surface attachment to homebuilt flow cells. We assembled the flow cells as described in the Materials and Methods section, and the annealed biotinylated DNA duplexes were attached to the microscope slide and imaged with FLIM in PIE mode. We obtain Inline graphic, Inline graphic (FRET), and Inline graphic images with a single acquisition. Figure 2A shows the single molecule images for annealed lo, mid, and hi FRET samples. The color scale of the pixels in a FLIM/PIE-FRET image indicates the fluorescence lifetime, which changes for the donor when FRET occurs. We selected individual DNA duplexes indicated by a bright spot in the image, and the fluorescence intensity time trace was collected at a single location. During the point acquisition, lifetime data is preserved as each detected photon is tagged with a macrotime and nanotime (Materials and Methods).

Each bright spot in Fig. 2A is a single DNA duplex (a few brighter spots may be 2–3 molecules). Due to the energy transfer that occurs in the FRET interaction between donor and acceptor dyes in an annealed substrate, we expect the donor intensity and fluorescence lifetime to be impacted by the distance from the acceptor dye. The brightness of each spot indicates the molecule fluorescence intensity, and the color of the pixels in each image indicate the fluorescence lifetime at that location, thus we can directly infer energy transfer from analyzing the set of FLIM/PIE-FRET images.

From left (lo FRET) to right (hi FRET) in Fig. 2A, the donor signal (Inline graphic) decreases in intensity (represented by photon counts on a pixel-by-pixel basis). The gray scale bar indicates pixel brightness, with darker areas being less intense and brighter more intense. The fluorescence lifetime of the molecules is represented by the Fast Lifetime in nanoseconds (see color bar), which is the average nanotime of the photons on a pixel-by-pixel basis. The Fast Lifetime represented the excited lifetime for a single exponential decay. From left (lo FRET) to right (hi FRET), the fluorescence lifetime of the donor molecules (Inline graphic) also decreases (green color to blue color). As the donor and acceptor dye become closer (left to right), we also expect the acceptor signal due to FRET (Inline graphic) to increase as is observed in the middle set of images. Direct excitation of the acceptor dye in PIE mode (Inline graphic) varies slightly across the three samples (lower images).

After collecting a FLIM/PIE-FRET image, we clicked on individual bright spots and collected single-molecule fluorescence intensity traces. A representative donor and acceptor intensity trace for each benchmark DNA sample is shown in Fig. 2B. As mentioned previously, FRET efficiency can be calculated with intensity or fluorescence lifetime-based measurements. Accurate recovery of the FRET efficiency from experimental donor and acceptor intensity data relies on the determination of three correction factors: Inline graphic, Inline graphic, and Inline graphic [33]. The correction factor Inline graphic accounts for spectral cross-talk from donor fluorescence leakage into the acceptor channel while the correction factor Inline graphic accounts for the direct excitation of the acceptor with the donor laser. The Inline graphic correction factor accounts for relative differences in the quantum yields and detection efficiencies of the donor and acceptor dyes [33, 39, 58]. While we used excited state decays to illustrate PIE (Fig. 1B), we determined the Inline graphic and Inline graphic correction factors with fluorescence lifetime and intensity data on diffusing molecules (Supplementary Information, Section IV). A previous smFRET benchmark study [39] outlined a robust correction procedure for all three factors and focused on intensity-based measurements where FRET efficiency was calculated from donor and acceptor photon counts after donor excitation. The FRET efficiency (Fig. 2B, black scatter) below each trace was calculated with Eq. 1 [39] using the Inline graphic and Inline graphic correction factors in Supplementary Table S5 determined from diffusing measurements (Supplementary Information, Section IV):

To calculate the gamma (Inline graphic) factor for immobilized molecules, the acceptor must photobleach before the donor. In each example trace, the FRET signal (green) vanishes after the acceptor (magenta) photobleaches in one step, and the donor (blue) signal recovers. We calculated the gamma correction factor, Inline graphic, for surface attached molecules [58–60] with Eq. 2:

, where Inline graphic is the average intensity of the acceptor molecule prior to the photobleaching step, Inline graphicis the average acceptor intensity after the bleach, and Inline graphic, Inline graphic are the same values for the donor molecule. The Inline graphic factors for the lo, mid, and hi FRET traces (Fig. 2B) were: 0.74, 0.59, and 0.64, respectively. The slight variation in the Inline graphic factor reflects the heterogeneity in single particles and their local environment. For comparison, the average gamma factor determined from the diffusing PIE-FRET measurements of each sample was 0.85 (Supplementary Information, Section IV, Table S6). The calculated average FRET efficiencies from single-molecule fluorescence intensity traces using Eq. 1 for the lo, mid, and hi FRET traces were: 0.11, 0.54, and 0.68, respectively. There is strong agreement between values obtained from the diffusing PIE-FRET experiments conducted on the same samples [(lo FRET: 0.14; mid FRET: 0.56; and hi FRET: 0.79), Supplementary Figure S3, and Table S6]. We also used the fluorescence lifetime data to calculate FRET for the same single molecules.

The photophysical process of relaxation from an excited state is influenced by radiative (photon-producing) and non-radiative processes. The observed average fluorescence lifetime (⟨τ⟩), usually in nanoseconds, depends on radiative Inline graphicand nonradiative Inline graphicrelaxation rates as Inline graphic [61]. The excited state decay process is also highly sensitive to the local molecular environment due primarily to influences on the available non-radiative relaxation pathways, and thus the non-radiative decay rate, Inline graphic. In principle, lifetime-based FRET can be calculated without any correction factors using Eq. 3 [14]:

, where Inline graphic is the fluorescence lifetime of the donor in the presence of the acceptor, andInline graphic is the fluorescence lifetime of the donor in the absence of the acceptor. The fluorescence lifetimes and diffusion coefficients for each benchmark DNA sample were also obtained from diffusing measurements, and the data are summarized in Supplementary Figure S1 and Table S4. For calculation of diffusion coefficients, we carefully calibrated the confocal volume of our three focused laser beams using two different methods (Supplementary Information, Section III). Using the single-molecule fluorescence intensity traces collected on individual surface-attached DNA duplexes offers a more direct method to calculate lifetime-based FRET efficiency compared to diffusing measurements.

To calculate the FRET efficiency for the individual benchmark DNA duplexes depicted in Fig. 2, we tail-fit the decays obtained from binning the nanotimes of photons for each signal (Inline graphic and Inline graphic), to a single exponential function (Supplementary Figure S5). For tail-fitting, we fit only the part of the decay just beyond the decay/instrument response function (IRF) peak, where the signal is mostly due to the fluorophore excited state decay. Tail fitting gives an appropriate lifetime value for single exponential decay behavior. The excited state decay of the donor signal (Inline graphic) was generated and tail-fit before and after the bleach step to obtain Inline graphicand Inline graphic, respectively (Supplementary Figure S5). The lifetime-based FRET efficiency calculated with Eq. 3 for the single DNA duplex traces in Fig. 2B are – lo FRET: 0.15, mid FRET: 0.52, and hi FRET: 0.66, consistent with the intensity-based FRET efficiency calculation on the same traces, and the values calculated for the diffusing measurements (Supplementary Table S6). For comparison, the FRET values for the Atto550/Atto647N labeled DNA samples reported in the previous benchmark study were- lo FRET: 0.15 ± 0.02; mid FRET: 0.56 ± 0.03; and hi FRET: 0.76 ± 0.015 [39]. The values obtained for the three traces in Fig. 2B are compared to the diffusing measurements in Table 1. Interestingly, a trend is observed in the Inline graphic fluorescence lifetime. This signal represents the fluorescence from the acceptor molecule due to energy transfer. With the donor and acceptor farthest apart in the lo sample, the lifetime of the acceptor excited state created from FRET is 5.2 ns, and it decreases to 4.65 and 3.92 for the mid and hi samples, respectively. The same trend in fluorescence lifetimes is not observed for the identical acceptor molecule under direct excitation (Inline graphic), suggesting that the quantum yield of the acceptor is influenced by the distance from the donor molecule in FRET [62]. When biomolecules are attached to a surface, a potential concern is introducing artifacts due to surface interactions. We have shown that with careful calibration and determination of correction factors, consistent lifetime and intensity-based FRET efficiency calculations can be obtained from surface-attached molecules.

FLIM/PIE-FRET of RNA/DNA hybrid molecules and MutL enzymes

Single-molecule imaging methods rely on surface attachment to study the molecules of interest. To investigate the conformation of individual RNA/DNA hybrid molecules, we prepared biotin-streptavidin functionalized homebuilt flow cells and flowed in the annealed biotinylated substrate diluted to 15 pM in 30 mM HEPES, 100 mM NaCl, plus a glucose oxidase/catalase oxygen scavenging system containing cyclooctatetraene to reduce photobleaching and blinking (Materials and Methods). RNA/DNA substrates were designed to report cleavage and dynamics of the RNA by SARS-CoV-2 endoribonuclease nonstructural protein 15 (Nsp15) [46], with the 45-base DNA duplex acting only as a surface anchor. Nsp15 is the 15th cleavage product of the SARS-CoV-2-encoded polyprotein pp1ab that is translated by host ribosomes and cleaved by viral proteases [63, 64]. Nsp15 structure [65] and function is highly conserved across coronaviruses. Nsp15 with its preference for cleaving 3’ of uridine (U) likely limits the length and abundance viral dsRNA intermediates to suppress the host immune response and contribute to pathogenesis [66]. Such activity makes Nsp15 an intriguing therapeutic target distinct from current treatment methods that target surface structural proteins to disable virus entry into the host. From high-resolution single particle cryo-EM reconstructions [65], we know the structure of Nsp15 before and after cleavage but the intermediate steps in the cleavage process have not been observed. Single-molecule measurements can provide key insights into how to inhibit the activity of Nsp15.

Figure 3A, left, shows the schematic of an RNA/DNA hybrid construct, Mid FRET U (MFU), one of several substrates we designed to explore cleavage by Nsp15 and dynamics of the RNA substrate [46]. We imaged a small region of the surface-attached sample and used a fast pattern matching algorithm [67] in SymPhoTime to separate the signals by excitation laser time gate and detection channel to create a single RGB image (Fig. 3A). Some colocalization of donor and acceptor signals is visible, however, the donor signal is weak due to FRET. In addition, due to incomplete annealing, some donor molecules may lack an acceptor, but we can distinguish different species with FLIM/PIE-FRET. For the annealed RNA/DNA hybrid substrates, there is a 15-base separation between donor and acceptor. The Förster distance for the Atto550-Atto647N FRET pair was estimated as 64.13 Å from FPbase (FPbase.org), therefore the expected separation between the donor and acceptor after annealing is approximately 60 Å, assuming a 0.34 nm separation per base. This donor-acceptor separation distance corresponds to a minimum FRET efficiency of approximately 0.6 for the MFU substrate. We note that while no Nsp15 was added in the surface-attached experiment in Fig. 3A, we have confirmed cleavage of a similar High FRET U (HFU) substrate in diffusing experiments, while the corresponding adenine-containing High FRET substrate (HFA) was un-cleaved [46]. The MFU substrate was designed specifically to explore substrate dynamics.

To observe individual substrates over the course of seconds to minutes, we clicked on a bright spot in the image and collected photons from only that location while separating the signal (Inline graphic, Inline graphic (FRET), and Inline graphic). The photon counts in each detection channel were collected into 100 ms bins to produce fluorescence intensity traces for the donor (blue), acceptor (magenta), and FRET (green) signals from an individual labeled RNA/DNA hybrid substrate (Fig. 3A, right). The FRET signal indicates relatively static FRET on the millisecond timescale, and the calculated FRET efficiency below in black, Inline graphic, is consistent with the expected value from the theoretical calculation described above. The calculated FRET was corrected for Inline graphic Inline graphic. We estimated the correction factors as described in the previous section and obtained: Inline graphic and Inline graphic (diffusing measurements) and Inline graphic for this individual trace. By comparison, the lifetime-based calculation yielded Inline graphic using the method described above for benchmark DNA, which is in excellent agreement with the intensity-based calculation. The ability to observe faster than millisecond dynamics is limited for surface-attached molecules due to necessary photon macrotime binning to achieve appropriate signal-to-noise. Indeed, when we measured the same substrates in a diffusing molecule experiment, we observed microsecond timescale dynamics of the RNA substrate using PIE-FRET and FCS in tandem [46]. We conclude that these dynamics are important for conferring the cleavage specificity of Nsp15 for uridine.

For detailed investigations of protein or nucleic acid conformational changes, it may be desired to allow the molecule to diffuse freely near the surface, in which case liposome encapsulation [47–49] can be employed. We expressed, purified, and fluorescently labeled Thermus aquaticus MutL (Materials and Methods) and encapsulated individual proteins in liposomes (Fig. 3B). MutL is an essential ATPase enzyme in the mismatch repair pathway that helps correct errors in newly synthesized DNA [68]. MutL is a dimer with two globular domains connected by flexible linker arms, with the N-terminal domains containing ATPase and DNA binding activity, and C-terminal domains containing dimerization and endonuclease functions. The partially intrinsically disordered linker arms allow MutL to adopt several distinct conformations [69] that are driven by a cycle of ATP binding and hydrolysis that is poorly understood. In addition, the partially intrinsically disordered nature of MutL preclude crystallization, making it challenging to explore with traditional methods, thus smFRET is an indispensable tool [23].

To study the dynamics of Taq MutL conformational changes, we encapsulated the enzymes labeled with AF555 and AF647 on the N-terminus between the Taq MutL gene and the His-tag and used PIE mode to image the surface-bound molecules in liposomes. Due to our stochastic labeling strategy of mixing both dyes together prior to the reaction, we obtain donor-only, acceptor-only, and double labeled subpopulations of Taq MutL, but they can be differentiated in the FLIM/PIE-FRET images and subsequent fluorescence intensity traces. The liposomes contain 1% biotinylated lipids that bind to a biotin-streptavidin functionalized flow cell coverslip (Materials and Methods) to hold the enzymes near the surface while allowing them to freely diffuse inside of the vesicle. The inner diameter of the extruded liposomes are ~ 190 nm since the lipid bilayer is about 5 nm wide. By comparison, the physical size of a Taq MutL monomer is estimated as 5–7 nm, while the dimer is 10–14 nm based on available structural models, although an experimentally obtained crystal structure does not exist for Taq MutL. Based on that estimate, we do not expect constrained motion of Taq MutL inside the liposomes. In initial control experiments, we encapsulated free Alexa Fluor 647 (AF647) dye molecules and AF647 single-labeled Taq MutL molecules (Supplementary Information, Section V), imaged, and probed individual molecules. For the encapsulated AF647 and AF647-MutL, we found that the average fluorescence lifetimes were 1.15 ± 0.08 ns (N = 127 molecules), and 1.49 ± 0.23 ns (N = 23 molecules), respectively. The lifetimes are consistent with the published fluorescence lifetime for AF647 (1.0 ns, Thermo Fisher), suggesting that the local environment of the fluorophore is not influenced significantly by the liposomes, however the presence of specific amino acid side chains in close proximity to the dye may lead to the observed increase in the AF647 fluorescence lifetime in the AF647-MutL conjugate. We obtained FLIM/PIE-FRET images simultaneously, and we used a fast pattern matching algorithm [67] in SymPhoTime to separate the signals by excitation laser time gate and detection channel to create a single RGB image (Fig. 3C). The Inline graphic and Inline graphic images are shown in Fig. 3D. We note that the dyes may be less photostable inside of the liposomes since the oxygen scavenging system used in the RNA/DNA hybrid experiment (Fig. 3A) is not accessible to the interior of the liposome after encapsulation, although we include cyclooctatetraene in the buffer during encapsulation to prevent triplet blinking (Materials and Methods).

The conformations of individual proteins were observed by first collecting the FLIM/PIE-FRET image and then selecting a bright spot to collect photons from a single encapsulated enzyme. The photon macrotimes were separated into 50 ms bins to produce a fluorescence intensity trace, and the intensity-based FRET was calculated as shown for 3 different enzymes in Fig. 3E. We used the methods outlined in the previous section to estimate the leakage, direct excitation, and gamma correction factors as: Inline graphic, Inline graphic (diffusing measurements) and Inline graphic as indicated for each trace. Using those values, we calculated the average FRET for each trace as: Inline graphic (for trace 1, FRET values above 0.95 were omitted as they may result from artifacts); Inline graphic; and Inline graphic. It is difficult to calculate accurate donor lifetimes with a high FRET signal (thus low donor signal) or a short duration of the FRET state; therefore, the corresponding lifetime-based calculations could not be performed. The high average FRET value for trace 1 is consistent with a condensed conformation of the MutL enzyme observed previously [69]. From those AFM experiments, MutL was observed to adopt four distinct conformations (Supplementary Figure S6D) in the absence of nucleotides, and it is suspected that MutL can spontaneously transition between the conformations, thus the distribution of FRET values observed in 3E may be due to spontaneous transitions to other conformations. Interestingly, trace 3 shows evidence of a transition between 2 or more FRET states. In diffusion-based PIE-FRET measurements detecting thousands of MutL enzymes in the absence of nucleotides, we observed a broad FRET distribution and evidence of conformational dynamics (Supplementary Figure S6E, F). The average FRET for each single MutL enzyme trace in 3E falls within the broad distribution obtained from diffusing measurements (Supplementary Figure S6E, F). Longer observation of individual enzymes in the presence of ATP and its analogs may reveal stable transitions between different conformational states of Taq MutL that are likely driven by a cycle of ATP hydrolysis. To further validate the FRET obtained from encapsulated proteins, we propose conducting studies of previously benchmarked proteins MalE and U2AF2 [41].

FLIM/PIE-FRET of live S. cerevisiae cells

Observing transient protein-protein interactions in their native cellular environment is important for mechanistic studies in cell biology. A variety of genetically encoded and endogenous fluorescent labeling strategies have been developed to track proteins or other subcellular structures in bacterial, yeast and mammalian cells [70, 71]. FLIM and FRET imaging are powerful tools for detecting protein-protein interactions in live cells [54–56, 72–76]. We imaged S. cerevisiae cells with fluorescent tags incorporated onto endogenous genes of interest. For this analysis, we selected two AAA(ATPases associated with various cellular activities)-ATPases essential for ribosome biogenesis, Rea1 and Rix7, and their respective interacting assembly factors Nop7 and Nsa1. Accurate and efficient production of the ribosomal subunits is critical for life and dysregulation of ribosome biogenesis is linked to many human diseases [77–80]. Rea1 is a large AAA-ATPase that plays a critical role in remodeling maturing pre-60 S particles by driving the release of assembly factors [81–83]. Nop7 is a pre-60 S specific factor that associates with pre-60 S particles during several of the same stages as Rea1 [83–85]. Rix7 catalyzes the release of the nucleolar protein Nsa1 [86]. In Rix7 mutants in live cells, Nsa1 cannot dissociate from pre-60 S particles, producing aberrant ribosomes, indicating that Rix7 is required for the release of Nsa1 and progression of 60 S synthesis [87]. We simultaneously acquired FLIM/PIE-FRET images of live S. cerevisiae cells expressing either Nop7-GFP/mCherry-Rea1 or Rix7-GFP/Nsa1-mCherry (Fig. 4). Imaging cells in PIE mode enables quasi-simultaneous excitation and detection of donor and acceptor molecules (compared to the timescale of molecular diffusion), thus protein-protein interactions can be monitored by performing a pixel-by-pixel FRET efficiency calculation using Eq. 1.

We determined the Inline graphic and Inline graphic correction factors [72] by analyzing images of many individual cells expressing only Nop7-GFP, Rix7-GFP, mCherry-Rea1, or Nsa1-mCherry (Supplementary Information, Section VII). Figure 4A shows the FLIM image of a cluster of double knock-in S. cerevisiae cells (Nop7-GFP/mCherry-Rea1), indicating the fluorescence lifetime at each pixel. Figure 4B is the PIE-FRET image. The green signal indicates the location of Nop7 proteins, while the red signal indicates the location of Rea1 enzymes. The bright Rea1 signal appears to mostly localize to the nucleus. The yellow color indicates pixelwise colocalization of the enzymes. Figure 4C is the pixel-wise FRET efficiency calculation, where the coloration of the pixels indicates the FRET efficiency. Figure 4D shows a cluster of S. cerevisiae Rix7-GFP/Nsa1-mCherry cells, with the corresponding PIE-FRET image (Fig. 4E), and spatial FRET efficiency calculation (Fig. 4F). The pixel coloration indicates low FRET (blue) localized in the nucleus, and a significant population of higher FRET in the nuclear periphery. Spatial FRET efficiency data can ultimately be converted to a physical distance with proper calibration using control samples, which will enable structural biology studies in live cells.

We acknowledge that some calculated FRET values in our complex system of interacting protein partners in a crowded nuclear environment approach a limiting value (low FRET), which is challenging to interpret. In our intermolecular protein-protein FRET experiments, the signals from donor-only, acceptor-only, and donor-acceptor species may confound the FRET efficiency calculation. To distinguish genuine protein-protein interactions in a cellular environment from false signals due to background fluorescence or high protein expression levels, fluorescent protein (FP) FRET standards [88, 89] can be used to benchmark the distance and concentration sensitivity in live cell experiments [51, 90–95]. In those experiments, cells are transfected with plasmids to express fusion proteins of FPs connected by amino acid linkers of varying lengths, generating intramolecular FRET standard constructs with known donor: acceptor stoichiometry and a range of FRET values. Determination of Inline graphic and Inline graphic correction factors with FP FRET standards with known stoichiometry in live cell imaging experiments enables calibration of FRET efficiency, E, and stoichiometry, S (Supplementary Information, Section IV), independently [96]. For our system, ideal FRET standard constructs will express GFP and mCherry [93] connected by amino acid linkers of varying lengths. Generating additional double knock-in S. cerevisiae strains with tagged non-interacting proteins as controls will also establish further confidence in our calculated FRET efficiency. Creating a three-dimensional representation of the fluorescence intensity data and calculating the pixel-based photon stoichiometry can establish a confidence index for spatial intensity-based FRET efficiency calculations [96]. Additional analyses of the lifetime data such as phasor analysis [27, 73, 97, 98] and global FLIM-FRET analysis [99] can also be performed to identify fluorophore species by their fluorescence lifetime. FLIM/PIE-FRET coupled with fast scanning enables high spatiotemporal resolution which can be interpreted to reveal dynamic structural information about biomolecular complexes in living cells.

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

• 1.Möckl L, Moerner WE. Super-resolution microscopy with single molecules in biology and Beyond–Essentials, current Trends, and future challenges. J Am Chem Soc. 2020;142(42):17828–44. 10.1021/jacs.0c08178. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 2.Berezin MY, Achilefu S. Fluorescence lifetime measurements and biological imaging. Chem Rev. 2010;110(5):2641–84. 10.1021/cr900343z. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 3.Becker W. Fluorescence lifetime imaging – techniques and applications. J Microsc. 2012;247(2):119–36. 10.1111/j.1365-2818.2012.03618.x. [DOI] [PubMed] [Google Scholar]
• 4.Becker W. Advanced Time-Correlated Single Photon Counting Applications. Springer; 2015. 10.1007/978-3-319-14929-5.
• 5.Kapusta P, Wahl M, Erdmann R. Advanced Photon Counting. Springer; 2015. 10.1007/978-3-319-15636-1.
• 6.Deniz AA, Dahan M, Grunwell JR, Ha TJ, Faulhaber AE, Chemla DS, et al. Single-pair fluorescence resonance energy transfer on freely diffusing molecules: observation of Forster distance dependence and subpopulations. Proc Natl Acad Sci USA. 1999;96(7):3670–5. 10.1073/pnas.96.7.3670. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 7.Ha T, Enderle T, Ogletree DF, Chemla DS, Selvin PR, Weiss S. Probing the interaction between two single molecules: fluorescence resonance energy transfer between a single donor and a single acceptor. Proc Natl Acad Sci USA. 1996;93(13):6264–8. 10.1073/pnas.93.13.6264. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 8.Roy R, Hohng S, Ha T. A practical guide to single-molecule FRET. Nat Methods. 2008;5(6):507–16. 10.1038/NMETH.1208. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 9.Eggeling C, Fries JR, Brand L, Gunther R, Seidel CAM. Monitoring conformational dynamics of a single molecule by selective fluorescence spectroscopy. Proc Natl Acad Sci USA. 1998;95(4):1556–61. 10.1073/pnas.95.4.1556. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 10.Schuler B, Lipman EA, Eaton WA. Probing the free-energy surface for protein folding with single-molecule fluorescence spectroscopy. Nature. 2002;419(6908):743–7. 10.1038/nature01060. [DOI] [PubMed] [Google Scholar]
• 11.Yang H, Luo G, Karnchanaphanurach P, Louie T-M, Rech I, Cova S, et al. Protein conformational dynamics probed by Single-Molecule electron transfer. Science. 2003;302(5643):262–6. [DOI] [PubMed] [Google Scholar]
• 12.Wozniak AK, Schröder GF, Grubmüller H, Seidel CAM, Oesterhelt F. Single-molecule FRET measures bends and Kinks in DNA. Proc Natl Acad Sci U S A. 2008;105:18337. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 13.Sisamakis E, Valeri A, Kalinin S, Rothwell PJ, Seidel CAM. Accurate single-molecule FRET studies using multiparameter fluorescence detection. In: Walter NG, editor. Methods in Enzymology, Vol 475: Single Molecule Tools, Pt B: Super-Resolution, Particle Tracking, Multiparameter, and Force Based Methods. San Diego: Elsevier Academic Press Inc; 2010. pp. 455–514. [DOI] [PubMed]
• 14.Lerner E, Barth A, Hendrix J, Ambrose B, Birkedal V, Blanchard SC, et al. FRET-based dynamic structural biology: Challenges, perspectives and an appeal for open-science practices. eLife. 2021;10:e60416. 10.7554/eLife.60416. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 15.Gurunathan K, Levitus M. FRET fluctuation spectroscopy of diffusing biopolymers: contributions of conformational dynamics and translational diffusion. J Phys Chem B. 2010;114:980. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 16.Sakon JJ, Weninger KR. Detecting the conformation of individual proteins in live cells. Nat Methods. 2010;7(3):203–U56. 10.1038/nmeth.1421. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 17.DeRocco VC, Anderson T, Piehler J, Erie DA, Weninger K. Four-color single-molecule fluorescence with noncovalent dye labeling to monitor dynamic multimolecular complexes. Biotechniques. 2010;49:807. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 18.Qiu R, DeRocco VC, Harris C, Sharma A, Hingorani MM, Erie DA, et al. Large conformational changes in MutS during DNA scanning, mismatch recognition and repair signalling. EMBO J. 2012;31(11):2528–40. 10.1038/emboj.2012.95. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 19.DeRocco V, Sass LE, Qiu R, Weninger KR, Erie DA. Dynamics of MutS-mismatched DNA complexes are predictive of their repair phenotypes. Biochemistry. 2014;53(12):2043–52. 10.1021/bi401429b. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 20.Qiu R, Sakato M, Sacho EJ, Wilkins H, Zhang X, Modrich P, et al. MutL traps MutS at a DNA mismatch. Proc Natl Acad Sci. 2015;112(35):10914–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 21.Gauer JW, LeBlanc S, Hao P, Qiu R, Case BC, Sakato M, et al. Single-Molecule FRET to measure conformational dynamics of DNA mismatch repair proteins. Methods Enzymol. 2016;581:285–315. 10.1016/bs.mie.2016年08月01日2. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 22.LeBlanc SJ, Gauer JW, Hao P, Case BC, Hingorani MM, Weninger KR, et al. Coordinated protein and DNA conformational changes govern mismatch repair initiation by MutS. Nucleic Acids Res. 2018;46(20):10782–95. 10.1093/nar/gky865. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 23.LeBlanc SJ, Kulkarni P, Weninger KR, Single Molecule FRET. A powerful tool to study intrinsically disordered proteins. Biomolecules. 2018;8(4):15. 10.3390/biom8040140. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 24.Nettels D, Hoffmann A, Schuler B. Unfolded protein and peptide dynamics investigated with Single-Molecule FRET and correlation spectroscopy from picoseconds to seconds. J Phys Chem B. 2008;112(19):6137–46. 10.1021/jp076971j. [DOI] [PubMed] [Google Scholar]
• 25.Sahoo H, Schwille P. FRET and FCS-Friends or foes? ChemPhysChem. 2011;12(3):532–41. 10.1002/cphc.201000776. [DOI] [PubMed] [Google Scholar]
• 26.Sun Y, Day RN, Periasamy A. Investigating protein-protein interactions in living cells using fluorescence lifetime imaging microscopy. Nat Protoc. 2011;6(9):1324–40. 10.1038/nprot.2011.364. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 27.Levitt JA, Poland SP, Krstajic N, Pfisterer K, Erdogan A, Barber PR, et al. Quantitative real-time imaging of intracellular FRET biosensor dynamics using rapid multi-beam confocal FLIM. Sci Rep. 2020;10(1):5146. 10.1038/s41598-020-61478-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 28.Vicidomini G, Ta H, Honigmann A, Mueller V, Clausen MP, Waithe D, et al. STED-FLCS: an advanced tool to reveal Spatiotemporal heterogeneity of molecular membrane dynamics. Nano Lett. 2015;15(9):5912–8. 10.1021/acs.nanolett.5b02001. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 29.Ghosh A, Karedla N, Thiele JC, Gregor I, Enderlein J. Fluorescence lifetime correlation spectroscopy: basics and applications. Methods. 2018;140. 10.1016/j.ymeth.201802009. 141:32 – 9; doi. [DOI] [PubMed]
• 30.Thiele JC, Helmerich DA, Oleksiievets N, Tsukanov R, Butkevich E, Sauer M, et al. Confocal Fluorescence-Lifetime Single-Molecule localization microscopy. ACS Nano. 2020;14(10):14190–200. 10.1021/acsnano.0c07322. [DOI] [PubMed] [Google Scholar]
• 31.Szalai AM, Siarry B, Lukin J, Giusti S, Unsain N, Cáceres A, et al. Super-resolution imaging of energy transfer by Intensity-Based STED-FRET. Nano Lett. 2021;21(5):2296–303. 10.1021/acs.nanolett.1c00158. [DOI] [PubMed] [Google Scholar]
• 32.Kapanidis AN, Lee NK, Laurence TA, Doose S, Margeat E, Weiss S. Fluorescence-aided molecule sorting: analysis of structure and interactions by alternating-laser excitation of single molecules. Proc Natl Acad Sci USA. 2004;101(24):8936–41. 10.1073/pnas.0401690101. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 33.Lee NK, Kapanidis AN, Wang Y, Michalet X, Mukhopadhyay J, Ebright RH, et al. Accurate FRET measurements within single diffusing biomolecules using alternating-laser excitation. Biophys J. 2005;88(4):2939–53. 10.1529/biophysj.104.054114. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 34.Muller BK, Zaychikov E, Brauchle C, Lamb DC. Pulsed interleaved excitation. Biophys J. 2005;89(5):3508–22. 10.1529/biophysj.105.064766. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 35.Lamb DC, Muller BK, Bruachle C. Applications of pulsed interleaved excitation. Biophys J. 2005;88(1):A373–A. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 36.Kudryavtsev V, Sikor M, Kalinin S, Mokranjac D, Seidel CAM, Lamb DC. Combining MFD and PIE for accurate Single-Pair Forster resonance energy transfer measurements. ChemPhysChem. 2012;13(4):1060–78. 10.1002/cphc.201100822. [DOI] [PubMed] [Google Scholar]
• 37.Hendrix J, Lamb DC. Implementation and application of pulsed interleaved excitation for Dual-Color FCS and RICS. In: Engelborghs Y, Visser A, editors. Fluorescence spectroscopy and microscopy: methods and protocols. Totowa: Humana Press Inc; 2014. pp. 653–82. [DOI] [PubMed] [Google Scholar]
• 38.Laurence TA, Kong X, Jäger M, Weiss S. Probing structural heterogeneities and fluctuations of nucleic acids and denatured proteins. Proceedings of the National Academy of Sciences. 2005;102(48):17348-53; 10.1073/pnas.0508584102 [DOI] [PMC free article] [PubMed]
• 39.Hellenkamp B, Schmid S, Doroshenko O, Opanasyuk O, Kuhnemuth R, Adariani SR, et al. Precision and accuracy of single-molecule FRET measurements-a multi-laboratory benchmark study. Nat Methods. 2018;15(9):669–. 10.1038/s41592-018-0085-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 40.Gotz M, Barth A, Bohr SSR, Borner R, Chen JX, Cordes T, et al. A blind benchmark of analysis tools to infer kinetic rate constants from single-molecule FRET trajectories. Nat Commun. 2022;13(1):12. 10.1038/s41467-022-33023-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 41.Agam G, Gebhardt C, Popara M, Mschtel R, Folz J, Ambrose B, et al. Reliability and accuracy of single-molecule FRET studies for characterization of structural dynamics and distances in proteins. Nat Methods. 2023. 10.1038/s41592-023-01807-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 42.Choi UB, Strop P, Vrljic M, Chu S, Brunger AT, Weninger KR. Single-molecule FRET-derived model of the synaptotagmin 1-SNARE fusion complex. Nat Struct Mol Biol. 2010;17(3):318–24. 10.1038/nsmb.1763. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 43.Kalinin S, Peulen T, Sindbert S, Rothwell PJ, Berger S, Restle T, et al. A toolkit and benchmark study for FRET-restrained high-precision structural modeling. Nat Methods. 2012;9(12):1218–U129. 10.1038/NMETH.2222. [DOI] [PubMed] [Google Scholar]
• 44.Dimura M, Peulen T-O, Sanabria H, Rodnin D, Hemmen K, Hanke CA, et al. Automated and optimally FRET-assisted structural modeling. Nat Commun. 2020;11(1):5394. 10.1038/s41467-020-19023-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 45.Peter MF, Gebhardt C, Machtel R, Munoz GGM, Glaenzer J, Narducci A, et al. Cross-validation of distance measurements in proteins by PELDOR/DEER and single-molecule FRET. Nat Commun. 2022;13(1):19. 10.1038/s41467-022-31945-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 46. Gordon K, Wright Z, Frazier M, Clark BS, Wilson I, Stephens CR, et al. A combined PIE-FRET and FCS assay to monitor RNA dynamics and cleavage by SARS-CoV-2 Nsp15. under consideration. 2025.
• 47.Boukobza E, Sonnenfeld A, Haran G. Immobilization in Surface-Tethered lipid vesicles as a new tool for single biomolecule spectroscopy. J Phys Chem B. 2001;105(48):12165–70. 10.1021/jp012016x. [Google Scholar]
• 48.Cisse I, Okumus B, Joo C, Ha T. Fueling protein–DNA interactions inside porous nanocontainers. Proc Natl Acad Sci. 2007;104(31):12646–50. 10.1073/pnas.0610673104. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 49.Okumus B, Arslan S, Fengler SM, Myong S, Ha T. Single molecule nanocontainers made porous using a bacterial toxin. J Am Chem Soc. 2009;131(41):14844–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 50.Olson F, Hunt CA, Szoka FC, Vail WJ, Papahadjopoulos D. Preparation of liposomes of defined size distribution by extrusion through polycarbonate membranes. Biochimica et biophysica acta (BBA). - Biomembr. 1979;557(1):9–23. 10.1016/0005-2736(79)90085-3. [DOI] [PubMed] [Google Scholar]
• 51. Sun Y, Hays NM, Periasamy A, Davidson MW, Day RN. Monitoring Protein Interactions in Living Cells with Fluorescence Lifetime Imaging Microscopy. Methods in Enzymology. 2012;504:371–91. 10.1016/B978-0-12-391857-4.00019-7. [DOI] [PMC free article] [PubMed]
• 52.Long Y, Stahl Y, Weidtkamp-Peters S, Postma M, Zhou W, Goedhart J, et al. In vivo FRET–FLIM reveals cell-type-specific protein interactions in Arabidopsis roots. Nature. 2017;548(7665):97–102. 10.1038/nature23317. [DOI] [PubMed] [Google Scholar]
• 53.Wang S, Shen B, Ren S, Zhao Y, Zhang S, Qu J, et al. Implementation and application of FRET–FLIM technology. J Innovative Opt Health Sci. 2019;12(05):1930010. 10.1142/S1793545819300106. [Google Scholar]
• 54.Kaufmann T, Herbert S, Hackl B, Besold JM, Schramek C, Gotzmann J, et al. Direct measurement of protein–protein interactions by FLIM-FRET at UV laser-induced DNA damage sites in living cells. Nucleic Acids Res. 2020;48(21):e122–e. 10.1093/nar/gkaa859. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 55.Spatola Rossi T, Pain C, Botchway SW, Kriechbaumer V. FRET-FLIM to determine protein interactions and membrane topology of enzyme complexes. Curr Protocols. 2022;2(10):e598. 10.1002/cpz1.598. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 56.Mandrou E, Thomason PA, Paschke PI, Paul NR, Tweedy L, Insall RH. A reliable system for quantitative G-Protein activation imaging in cancer cells. Cells. 2024. 10.3390/cells13131114. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 57.Bartels N, van der Voort NTM, Opanasyuk O, Felekyan S, Greife A, Shang X et al. Advanced multiparametric image spectroscopy and super-resolution microscopy reveal a minimal model of CD95 signal initiation. Sci Adv 10(35):eadn3238; 10.1126/sciadv.adn3238 [DOI] [PMC free article] [PubMed]
• 58.McCann JJ, Choi UB, Zheng L, Weninger K, Bowen ME. Optimizing methods to recover absolute FRET efficiency from immobilized single molecules. Biophys J. 2010;99(3):961–70. 10.1016/j.bpj.2010年04月06日3. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 59.Hildebrandt LL, Preus S, Birkedal V. Quantitative single molecule FRET efficiencies using TIRF microscopy. Faraday Discuss. 2015;184:131–42. 10.1039/c5fd00100e. [DOI] [PubMed] [Google Scholar]
• 60.Ha T, Ting AY, Liang J, Caldwell WB, Deniz AA, Chemla DS. Single-molecule fluorescence spectroscopy of enzyme conformational dynamics and cleavage mechanism. Proc Natl Acad Sci U S A. 1999;96:893. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 61.Lakowicz JR. Principles of fluorescence spectroscopy. Third ed. New York: Springer; 2006. [Google Scholar]
• 62.Kalinin S, Sisamakis E, Magennis SW, Felekyan S, Seidel CAM. On the origin of broadening of single-molecule FRET efficiency distributions beyond shot noise limits. J Phys Chem B. 2010;114:6197. [DOI] [PubMed] [Google Scholar]
• 63.Baker XDSC. An old protein with a new story: coronavirus endoribonuclease is important for evading host antiviral defenses. Virology. 2018. 10.1016/j.virol.2017年12月02日4. 517:157 – 63; doi. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 64.Hartenian E, Nandakumar D, Lari A, Ly M, Tucker JM, Glaunsinger BA. The molecular virology of coronaviruses. J Biol Chem. 2020;295(37):12910–34. 10.1074/jbc.REV120.013930. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 65.Frazier Meredith N, Dillard Lucas B, Krahn Juno M, Perera L, Williams Jason G, Wilson Isha M, et al. Characterization of SARS2 Nsp15 nuclease activity reveals it’s mad about U. Nucleic Acids Res. 2021;49(17):10136–49. 10.1093/nar/gkab719. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 66.Hackbart M, Deng X, Baker SC. Coronavirus endoribonuclease targets viral polyuridine sequences to evade activating host sensors. Proceedings of the National Academy of Sciences. 2020;117(14):8094 – 103; doi: 10.1073/pnas.1921485117 [DOI] [PMC free article] [PubMed]
• 67.Niehörster T, Löschberger A, Gregor I, Krämer B, Rahn H-J, Patting M, et al. Multi-target spectrally resolved fluorescence lifetime imaging microscopy. Nat Methods. 2016;13(3):257–62. 10.1038/nmeth.3740. [DOI] [PubMed] [Google Scholar]
• 68.Kunkel TA, Erie DA. DNA mismatch repair. Annu Rev Biochem. 2005;74:681–710. 10.1146/annurev.biochem.74.082803.133243. [DOI] [PubMed] [Google Scholar]
• 69.Sacho EJ, Kadyrov FA, Modrich P, Kunkel TA, Erie DA. Direct visualization of asymmetric adenine-nucleotide-induced conformational changes in MutL alpha. Mol Cell. 2008;29(1):112–21. 10.1016/j.molcel.2007年10月03日0. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 70.Tsien RY. The green fluorescent protein. Annu Rev Biochem. 1998;67. 10.1146/annurev.biochem.67.1.509. 67, 1998):509 – 44. [DOI] [PubMed]
• 71.Dean KM, Palmer AE. Advances in fluorescence labeling strategies for dynamic cellular imaging. Nat Chem Biol. 2014;10(7):512–23. 10.1038/nchembio.1556. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 72.Gordon GW, Berry G, Liang XH, Levine B, Herman B. Quantitative fluorescence resonance energy transfer measurements using fluorescence microscopy. Biophys J. 1998;74(5):2702–13. 10.1016/S0006-3495(98)77976-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 73.Day RN. Measuring protein interactions using Förster resonance energy transfer and fluorescence lifetime imaging microscopy. Methods. 2014;66(2):200–7. 10.1016/j.ymeth.2013年06月01日7. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 74.Long Y, Stahl Y, Weidtkamp-Peters S, Smet W, Du Y, Gadella TWJ, et al. Optimizing FRET-FLIM labeling conditions to detect nuclear protein interactions at native expression levels in living Arabidopsis roots. Front Plant Sci. 2018;9. 10.3389/fpls.2018.00639. [DOI] [PMC free article] [PubMed]
• 75.Parkkola H, Siddiqui FA, Oetken-Lindholm C, Abankwa D. FLIM-FRET analysis of Ras nanoclustering and Membrane-Anchorage. In: Rubio I, Prior I, editors. Ras activity and signaling: methods and protocols. New York, NY: Springer US; 2021. pp. 233–50. [DOI] [PubMed] [Google Scholar]
• 76.Petutschnig EK, Pierdzig L, Mittendorf J, Niebisch JM, Lipka V. A novel fluorescent protein pair facilitates FLIM-FRET analysis of plant immune receptor interaction under native conditions. J Exp Bot. 2024;75(3):746–59. 10.1093/jxb/erad418. [DOI] [PubMed] [Google Scholar]
• 77.Woolford JL Jr., Baserga SJ. Ribosome biogenesis in the yeast Saccharomyces cerevisiae. Genetics. 2013;195(3):643–81. 10.1534/genetics.113.153197. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 78.Klinge S, Woolford JL. Ribosome assembly coming into focus. Nat Rev Mol Cell Biol. 2019;20(2):116–31. 10.1038/s41580-018-0078-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 79.Farley-Barnes KI, Ogawa LM, Baserga SJ, Ribosomopathies. Old Concepts, new controversies. Trends Genet. 2019;35(10):754–67. 10.1016/j.tig.201907004. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 80.Vanden Broeck A, Klinge S. Eukaryotic ribosome assembly. Annu Rev Biochem. 2024;93:189–210. 10.1146/annurev-biochem-030222-113611. 93, 2024. [DOI] [PubMed] [Google Scholar]
• 81.Barrio-Garcia C, Thoms M, Flemming D, Kater L, Berninghausen O, Baßler J, et al. Architecture of the Rix1–Rea1 checkpoint machinery during pre-60S-ribosome remodeling. Nat Struct Mol Biol. 2016;23(1):37–44. 10.1038/nsmb.3132. [DOI] [PubMed] [Google Scholar]
• 82.Prattes M, Lo Y-H, Bergler H, Stanley RE. Shaping the nascent ribosome: AAA-ATPases in eukaryotic ribosome biogenesis. Biomolecules. 2019. 10.3390/biom9110715. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 83.Kater L, Mitterer V, Thoms M, Cheng J, Berninghausen O, Beckmann R, et al. Construction of the central protuberance and L1 stalk during 60S subunit biogenesis. Mol Cell. 2020;79(4):615–e285. 10.1016/j.molcel.2020年06月03日2. [DOI] [PubMed] [Google Scholar]
• 84.Thoms M, Ahmed YL, Maddi K, Hurt E, Sinning I. Concerted removal of the Erb1–Ytm1 complex in ribosome biogenesis relies on an elaborate interface. Nucleic Acids Res. 2016;44(2):926–39. 10.1093/nar/gkv1365. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 85.Mitterer V, Thoms M, Buschauer R, Berninghausen O, Hurt E, Beckmann R. Concurrent remodelling of nucleolar 60S subunit precursors by the Rea1 ATPase and Spb4 RNA helicase. eLife. 2023;12:e84877. 10.7554/eLife.84877. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 86.Lo Y-H, Romes EM, Pillon MC, Sobhany M, Stanley RE. Structural analysis reveals features of ribosome assembly factor Nsa1/WDR74 important for localization and interaction with Rix7/NVL2. Structure. 2017;25(5):762–e724. 10.1016/j.str.201703008. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 87.Gadal O, Strauss D, Braspenning J, Hoepfner D, Petfalski E, Philippsen P, et al. A nuclear AAA-type ATPase (Rix7p) is required for biogenesis and nuclear export of 60S ribosomal subunits. EMBO J. 2001;20(14):3695–704. 10.1093/emboj/20.14.3695. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 88.Koushik SV, Chen H, Thaler C, Puhl HL, Vogel SS. Cerulean, Venus, and VenusY67C FRET reference standards. Biophys J. 2006;91(12):L99–101. 10.1529/biophysj.106.096206. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 89.Evers TH, van Dongen EMWM, Faesen AC, Meijer EW, Merkx M. Quantitative Understanding of the energy transfer between fluorescent proteins connected via flexible peptide linkers. Biochemistry. 2006;45(44):13183–92. 10.1021/bi061288t. [DOI] [PubMed] [Google Scholar]
• 90.Hoppe A, Christensen K, Swanson JA. Fluorescence resonance energy Transfer-Based stoichiometry in living cells. Biophys J. 2002;83(6):3652–64. 10.1016/S0006-3495(02)75365-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 91.Zal T, Gascoigne NRJ, Photobleaching-Corrected FRET. Efficiency imaging of live cells. Biophys J. 2004;86(6):3923–39. 10.1529/biophysj.103.022087. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 92.Chen H, Puhl HL 3rd, Koushik SV, Vogel SS, Ikeda SR. Measurement of FRET efficiency and ratio of donor to acceptor concentration in living cells. Biophys J. 2006;91(5):L39–41. 10.1529/biophysj.106.088773. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 93.Albertazzi L, Arosio D, Marchetti L, Ricci F, Beltram F. Quantitative FRET Analysis With the E0GFP-mCherry Fluorescent Protein Pair. Photochemistry and Photobiology. 2009;85(1):287–97. 10.1111/j.1751-1097.2008.00435.x. [DOI] [PubMed]
• 94.Hochreiter B, Kunze M, Moser B, Schmid JA. Advanced FRET normalization allows quantitative analysis of protein interactions including stoichiometries and relative affinities in living cells. Sci Rep. 2019;9(1):8233. 10.1038/s41598-019-44650-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 95.Basak S, Saikia N, Dougherty L, Guo Z, Wu F, Mindlin F, et al. Probing interdomain linkers and protein supertertiary structure in vitro and in live cells with fluorescent protein resonance energy transfer. J Mol Biol. 2021;433(5):166793. 10.1016/j.jmb.2020.166793. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 96.Coullomb A, Bidan CM, Qian C, Wehnekamp F, Oddou C, Albigès-Rizo C, et al. QuanTI-FRET: a framework for quantitative FRET measurements in living cells. Sci Rep. 2020;10(1):6504. 10.1038/s41598-020-62924-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 97.Ma N, de Mochel NR, Pham PD, Yoo TY, Cho KWY, Digman MA. Label-free assessment of pre-implantation embryo quality by the fluorescence lifetime imaging microscopy (FLIM)-phasor approach. Sci Rep. 2019;9. 10.1038/s41598-019-48107-2. [DOI] [PMC free article] [PubMed]
• 98.Malacrida L, Ranjit S, Jameson DM, Gratton E. The phasor plot: A universal circle to advance fluorescence lifetime analysis and interpretation. Annual Rev Biophys. 2021;50. 10.1146/annurev-biophys-062920-063631. 50, 2021):575 – 93; doi. [DOI] [PubMed]
• 99.Daniela S, Frank D, Angelika CR, Bjorn von E, Cornelia S, Christine AFA. Spectrally resolved fluorescence lifetime imaging microscopy: Förster resonant energy transfer global analysis with a one- and two-exponential donor model. J Biomed Opt. 2011;16(2):026002. 10.1117/1.3533318. [DOI] [PubMed] [Google Scholar]

Supplementary Materials

Data Availability Statement

The datasets generated and/or analyzed during the current study are available in the Dryad data repository, https://doi.org/10.5061/dryad.zkh1893pv. No permissions are required to access the data.