Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2024 Jul 2;128(28):6751–6759. doi: 10.1021/acs.jpcb.4c02065

Self-Quenched Fluorophore-DNA Labels for Super-Resolution Fluorescence Microscopy

Laurell F Kessler †, Ashwin Balakrishnan †, Tanja Menche †, Dongni Wang †, Yunqing Li †, Maximilian Mantel †, Marius Glogger ‡, Marina S Dietz †, Mike Heilemann †,*
PMCID: PMC11264260  PMID: 38955346

Abstract

graphic file with name jp4c02065_0004.jpg

Protein labeling through transient and repetitive hybridization of short, fluorophore-labeled DNA oligonucleotides has become widely applied in various optical super-resolution microscopy methods. The main advantages are multitarget imaging and molecular quantification. A challenge is the high background signal originating from the presence of unbound fluorophore-DNA labels in solution. Here, we report the self-quenching of fluorophore dimers conjugated to DNA oligonucleotides as a general concept to reduce the fluorescence background. Upon hybridization, the fluorescence signals of both fluorophores are restored. We expand the toolbox of fluorophores suitable for self-quenching and report their spectra and hybridization equilibria. We apply self-quenched fluorophore-DNA labels to stimulated emission depletion microscopy and single-molecule localization microscopy and report improved imaging performances.

Introduction

Fluorophore-labeled, short DNA oligonucleotides are extensively used in various super-resolution microscopy applications, including the single-molecule localization microscopy (SMLM)1 method, DNA points accumulation in nanoscale topography (DNA-PAINT),2 super-resolution optical fluctuation imaging (SOFI),3 and stimulated emission depletion (STED) microscopy.4,5 An advantage that applies to all these applications is multitarget imaging, which is enabled by repetitive rounds of imaging and washing of sequence-orthogonal DNA strands. An additional benefit that is particular to single-molecule DNA-PAINT is the opportunity for molecular quantification by kinetic analysis.6 Furthermore, weak-affinity DNA labels were shown to optimize the performance of neural networks that predict super-resolved images from high-density single-molecule data.7

Despite all of these opportunities, the use of weak-affinity DNA labels demands keeping them in the imaging buffer during a microscopy experiment, which results in a higher fluorescence background and a limit in acquisition speed. To address these challenges, FRET-quenched fluorophores attached to extended DNA oligonucleotides were introduced.8,9 Recently, DNA oligonucleotides labeled with two identical oxazine fluorophores on either end were shown to exhibit self-quenching through short-distance molecular interactions.10,11 These probes led to a lower background signal in the unbound state and consequently to a higher signal-to-background (SBR) ratio. An almost 2-fold higher photon yield in the bound state improved the localization precision (DNA-PAINT) and spatial resolution (DNA-PAINT and STED-PAINT).10 In the present work, we went beyond that proof of concept and explored whether fluorophores prominently used in DNA-PAINT (e.g., the carbocyanine Cy3B) and STED (e.g., the silicon-rhodamine SiR) exhibit fluorogenicity through dimerization when conjugated to short DNA oligonucleotides. Fluorogenic dimers of SiR and Cy3B have not yet been reported. Furthermore, we included the rhodamine fluorophore TMR, which was previously used as a protein configuration sensor exploiting dimer-induced fluorescence quenching.12 We show that the design of fluorogenic dimers indeed works for these familiar DNA-PAINT and STED dyes, thereby enabling a general strategy to construct fluorophore-labeled DNA probes for low-background super-resolution microscopy. In addition, we demonstrate multicolor imaging using these probes.

Methods

DNA Oligonucleotides

The sequences of the imager and docking strands used in this study are listed in Table S1. The sequences for DNA origami structures were purchased from Eurofins Genomics (Tables S6–S8).

Absorption and Fluorescence Spectroscopy

For absorption measurements, imager strands were diluted in 1× phosphate-buffered saline (PBS, diluted from a 10× stock, no. 14200067, Gibco, Thermo Fisher Scientific, USA), 0.5 M NaCl, and 1 mM ethylenediaminetetraacetic acid (EDTA) to a final concentration of 10 μM and were analyzed in the absence and presence of docking strands (100 μM). For fluorescence spectroscopy, an imager strand concentration of 100 nM and a docking strand concentration of 100 μM were used.

For absorption spectroscopy, a Specord S600 absorption spectrometer (Analytik Jena, Jena, Germany) was used. The average was taken over 10 spectra, and an integration time of 227 ms was used. Fluorescence spectra were acquired with a Cary Eclipse fluorescence spectrophotometer (Agilent Technologies, USA) with a slit size of 5 nm, mode set to slow, a gain dependent on the sample (SiR: 720; TMR: 780; Cy3B: 600), and an excitation wavelength dependent on the sample (SiR: 647 nm; TMR and Cy3B: 560 nm).

Dissociation constants for the different imager strands were determined by performing a titration series with a fixed concentration of the imager strand (1 nM imager strands) and different concentrations of the complementary strand (0–80 μM for SiR; 0–40 μM for TMR; 0–20 μM for Cy3B) and reading out the fluorescence. A home-built confocal setup described elsewhere13 was used for measuring the fluorescence. Briefly, a 647 nm (for SiR)/532 nm (for TMR and Cy3B) continuous-wave (CW) solid-state laser (both from Coherent, USA) passed an acousto-optic tunable filter and was coupled into a single-mode optical fiber. The excitation light was expanded by a collimator and reflected by a dichroic mirror into a water-immersion objective (60×, 1.2 NA, UPlanSApo, Olympus, Japan). In the detection path, out-of-focus light was filtered out by a 100 μm pinhole, and the fluorescence was filtered using a bandpass filter (ET 700/75 for SiR (AHF Analysentechnik, Germany) and HC 590/20 for TMR and Cy3B (AHF Analysentechnik, Germany)), which was then split using a beam splitter onto two avalanche photodiodes (APDs). The samples were warmed to 21 °C by a custom-built water-cooling system, and the fluorescence intensity was measured at a 50 μm depth from the glass surface. A laser intensity of about 500 μW (measured at the back focal plane of the objective) was used for excitation. For binding curves, each sample was recorded 3× for 60 s. The binding curves were measured 3× in independent experiments.

Fluorescence intensities I of the imager strands dependent on the docking strand concentration [DS] were analyzed with OriginPro (version 2024, OriginLab Corp., USA) by plotting the average intensity against the docking strand concentration. Binding curves were fitted with the Hill function14 to extract the dissociation constant KD:

graphic file with name jp4c02065_m001.jpg 1

Imin and Imax are the minimum and maximum fluorescence intensities observed for the free imager strand and for the complex of the imager and docking strand, respectively. The Hill coefficient n was set to 1 as the imager and docking strand associate in a 1:1 stoichiometry.

Fluorescence Correlation Spectroscopy

For FCS measurements, the above-described home-built confocal setup was used.13 The fluorescence signal detected from two APDs was cross-correlated with a real-time correlator card (Flex03lq, correlator.com; Bridgewater, USA). A 1 nM (9 nt) fluorophore-labeled imager strand was measured in the absence and in the presence of a saturating concentration of the docking strand (10–20 μM) at 21 °C. At this docking strand concentration, roughly all imager strands were bound to their complementary strands. The laser focus was positioned at a depth of 50 μm in the sample from the glass surface. The excitation intensity was about 500 μW at the back focal plane of the objective. Three measurements per sample were performed, each for 60 s. The correlation curves were analyzed in OriginPro. The three measurements were averaged, and a three-dimensional diffusion model with a single diffusing component and either three or two or one subdiffusional kinetics was fitted to the data.

graphic file with name jp4c02065_m002.jpg 2

Here, N is the average number of emitting fluorophores in the observation volume, while τD is the average diffusion time through the observation volume, and Inline graphicis the shape factor. The second term describes the photophysical kinetics with n ranging from 1 to 3 depending on the correlation curves, K describing the equilibrium constant, and τK the relaxation time of the process. For the subdiffusional kinetics arising from 10–7 s, the amplitude K can also be described in terms of the fraction D of the nonfluorescing species:

graphic file with name jp4c02065_m004.jpg 3

An equilibrium of the unbound imager strand between a fluorescing state F and a nonfluorescing, self-quenched species D with the rates kdark and kfluorescence is assumed.

graphic file with name jp4c02065_m005.jpg 4

The rates of the switching between the fluorescing and dark state can be described in terms of the equilibrium constant Inline graphic and the relaxation time τK = (kdark + kfluorescence)−1.15,16

graphic file with name jp4c02065_m007.jpg 5

Labeling of Antibodies with DNA Oligonucleotides

Unlabeled secondary antibodies (AffiniPure goat antimouse IgG, no. 115-005-003; AffiniPure donkey antimouse IgG, no. 715-005-150; AffiniPure donkey antirabbit IgG, no. 711-005-152 (Jackson ImmunoResearch, USA)) were self-labeled with azide-modified docking strands (P1 in the case of goat antimouse and donkey antimouse antibodies and P5 in the case of the donkey antirabbit antibody) according to a previously published protocol.17 In brief, the antibody was concentrated with an Amicon spin filter (MWCO 100 kDa, Merck, Germany) when the antibody concentration was below 2 mg/mL. The DBCO-sulfo-NHS ester linker (Jena Bioscience, Germany) was dissolved in dimethylformamide (DMF) (Sigma-Aldrich, Germany) and then diluted in 1× phosphate-buffered saline (PBS). The linker and the antibody were mixed in a 10:1 molar ratio and incubated for 90 min at 4 °C gentle shaking. The unbound linker was removed using Zeba desalting columns (Thermo Fisher Scientific, Germany). The azide-modified P1 docking strand and the linker-antibody conjugate were incubated in a 10:1 molar ratio overnight at 4 °C while slightly shaking. The next day, unbound DNA was removed using an Amicon spin filter (MWCO = 100 kDa). The docking strand-labeled antibody was stored at 4 °C.

Cell Culture

U-2 OS cells (CLS Cell Lines Service GmbH, Germany) were utilized for α-tubulin measurements, and MODE-K18 cells were used for TOM20 and KDEL measurements. U-2 OS cells were cultured in Dulbecco’s modified Eagle medium (DMEM)-F12 supplemented with 10% (v/v) fetal bovine serum (FBS), 100 U/mL penicillin, and 100 μg/mL streptomycin and 1% (v/v) GlutaMAX (all reagents from Gibco, Thermo Fisher, Germany), and MODE-K cells were cultured in Dulbecco’s modified Eagle medium (DMEM) supplemented with 10% (v/v) fetal bovine serum (FBS), 100 U/mL penicillin, and 100 μg/mL streptomycin and 1% (v/v) GlutaMAX (all reagents from Gibco, Thermo Fisher, Germany). Both cell lines were incubated at 37 °C with 5% CO2 in an automatic CO2 incubator (model C150, Binder GmbH, Germany). All cells were passaged every 3–4 days or upon reaching 80% confluency.

Immunofluorescence of Microtubules

For staining microtubules, U-2 OS cells were seeded on 8-well chambered coverglass (Sarstedt, Germany) that were coated with RGD-modified poly-l-lysine-grafted polyethylene glycol (PLL-PEG-RGD, self-synthesized as described by Harwardt et al.19) at a density of 15,000 cells per well and were incubated overnight at 37 °C and 5% CO2. The cells were either treated with a microtubule stabilization buffer (MTSB) (80 mM PIPES pH 6.8, 1 mM MgCl2, 5 mM EGTA, and 0.5% (v/v) Triton X-100, Sigma-Aldrich, Germany) for 30 s followed by fixation with 0.5% glutaraldehyde (v/v) (Sigma-Aldrich, Germany) in MTSB for 10 min at room temperature or directly fixed using 4% (v/v) formaldehyde (Sigma-Aldrich, Germany) and 0.1% (v/v) glutaraldehyde in 1× PBS for 20 min at 37 °C. The sample was washed once with 1× PBS then treated with NaBH4 solution (Roth, Germany) to reduce autofluorescence from glutaraldehyde followed by washing thrice with 1× PBS. The sample was then blocked using an immunofluorescence staining (IF) buffer (3% (w/v) BSA and 0.1% (v/v) Triton X-100 in 1× PBS) for 10 min after which it was incubated with primary antibody solution (anti-α-tubulin, no. T5168, Sigma-Aldrich, Germany) in an IF buffer diluted 1:500 for 1.5 h at room temperature while shaking followed by washing thrice with 1× PBS. The sample was then treated with a self-labeled secondary antibody (P1-goat antimouse) solution diluted 1:100 in an IF buffer for 1.5 h at room temperature followed by washing thrice with 1× PBS. The sample was then postfixed with 4% (v/v) formaldehyde for 10 min at room temperature and washed thrice with 1× PBS.

Immunofluorescence of NPC

For staining of NPC, U2OS-Nup96-EGFP cells were seeded on 8-well chambered coverglass (Sarstedt, Germany) coated with RGD-modified poly-l-lysine-grafted polyethylene glycol (PLL-PEG-RGD, self-synthesized as described by Harwardt et al.19) at a density of 30,000 cells per well and were incubated 1 day at 37 °C and 5% CO2. The cells were fixed using 4% (v/v) formaldehyde (Sigma-Aldrich, Germany) in 1× PBS for 20 min at 37 °C. The fixed cells were blocked for 30 min with a commercially available buffer (antibody incubation buffer, Massive Photonics, Germany) and then incubated for 1 h with a P1 docking strand-modified nanobody (Massive-tag-Q anti-GFP, Massive Photonics, Germany) diluted 1:100 in an antibody incubation buffer, with gentle shaking at room temperature. Afterward, cells were washed 3× with 1× PBS and postfixed with 4% (v/v) formaldehyde in 1× PBS for 10 min at room temperature. After washing again 3× with 1× PBS, gold beads (gold nanoparticles, 100 nm diameter, product no. A11-100-NPC-DIH-1-25, NanoPartz, USA) as fiducial markers were diluted 1:5 in 1× PBS and incubated for 8 min.

Immunofluorescence of TOM20 and KDEL

For staining of TOM20 and KDEL, MODE-K cells were seeded on an 8-well chambered coverglass (Sarstedt, Germany) coated with fibronectin at a density of 10,000 cells per well and were incubated 1–2 days at 37 °C and 5% CO2. The cells were then fixed using 4% (v/v) formaldehyde (Sigma-Aldrich, Germany) in 1× PBS for 30 min at 37 °C. The washing and fixing buffers were prewarmed to 37 °C. The fixed cells were blocked and permeabilized with a blocking buffer (3% BSA + 0.1% saponin in 1× PBS) at RT with shaking for 1 h. Then, they were incubated with primary antibodies (mouse anti-KDEL (Thermo Fisher, no. MA5-27581): 1:200 dilution in a blocking buffer; rabbit anti-TOM20 (11802-1-1AP, Proteintech, Germany): 1:200 dilution in a blocking buffer) for 2 h at RT with shaking. Afterward, cells were washed 3× with 1× PBS and labeled with secondary antibodies (donkey antimouse P1: 1:100; donkey antirabbit P5: 1:100) for 2 h at RT with shaking followed by washing thrice with 1× PBS. They were then postfixed with 4% (v/v) formaldehyde in 1× PBS for 10 min at RT. The sample was stored in 1× PBS for further imaging.

Folding and Purification of DNA Origami

Annealing of a rectangular DNA origami was performed in a mixture of 40 μL of a 1× TE (Tris-EDTA) buffer with 12.5 mM MgCl2 containing a 10 nM scaffold strand (M13mp18; tilibit nanosystems, Germany), 100 nM core staple strands, 1 μM biotinylated staple strands, and 1 μM DNA-PAINT handles. The mixture was incubated for 5 min at 65 °C and subsequently cooled to 25 °C over the course of 4 h. The DNA origami was purified using 100 kDa MWCO ultracentrifugal filter units (Amicon Ultra, Germany). The purified origami structure was stored in an FoB5 buffer (5 mM Tris pH 8, EDTA pH 8, 5 mM NaCl, and 5 mM MgCl2) at −20 °C. For strand sequences, see Tables S5–S7. The sequences were adapted from the Picasso design tool.20

Immobilization of DNA Origami

The channel of a glass slide (μ-Slide VI 0.5 glass bottom, Ibidi, Germany) was flooded with 1× PBS. Subsequently, biotinylated BSA (1 mg/mL, dissolved in 1× PBS) (Sigma-Aldrich, Germany) was added, and the solution was incubated for 15 min. After 3× washing with 1× PBS, streptavidin (0.2 mg/mL, dissolved in 1× PBS) (Sigma-Aldrich, Germany) was added and incubated for 15 min. The channel was washed 3× with 1× PBS and filled with purified DNA origami. After incubation for 20 min, the channel was washed 3× with a DNA-PAINT buffer (1× PBS, 500 mM NaCl). Gold beads (2 pM; 100 nm diameters were chosen with regard to the wavelength of the excitation laser; λexc = 647 nm, Nanopartz, USA) in a DNA-PAINT buffer were added and incubated for 10 min. The channel was washed once more with a DNA-PAINT buffer, and the imaging buffer was filled into the channel.

STED Imaging

STED imaging was performed on an Abberior expert line microscope (Abberior Instruments, Germany) with an Olympus IX83 body (Olympus Deutschland GmbH, Germany) by using a UPLXAPO 60× NA 1.42 oil immersion objective (Olympus Deutschland GmbH, Germany). For all acquired images, the imager strands were diluted to a concentration of 300 nM in a DNA-PAINT buffer (1× PBS, 500 mM NaCl). For image acquisition of microtubule samples, they were excited with either a 561/640 nm pulsed excitation laser (Table S9) and depleted using a 775 nm pulsed laser (Table S9) having a 2D doughnut point spread function and with a delay of 750 ps to 8 ns. Fluorescence was collected in the spectral range of 571–630 nm in the case of 561 nm excitation and 650–760 nm in the case of 640 nm excitation using an APD. All images were acquired with a pinhole of 0.81 airy unit (AU), a line accumulation/integration of 30, a pixel dwell time of 5 μs, and a pixel size of 15 nm.

For two-color image acquisition of TOM20 and KDEL, the sample was excited with a 640 nm excitation laser (10 μW at the back focal plane) and depleted using a 775 nm pulsed laser (55 mW at the back focal plane) having a 3D top-hat point spread function and with a delay of 750 ps to 8 ns. Fluorescence was collected in the spectral range of 571–630 (561 nm excitation) and 650–760 nm (640 nm excitation) using two APDs. The images were acquired with a pinhole of 0.81 AU, a line accumulation/integration of 20, a pixel dwell time of 5 μs, and a pixel size of 70 nm.

STED Image Analysis

The fluorescence intensity from microtubules was calculated by filtering for regions with microtubules using a skeletonized binary mask. The total signal intensity of the microtubules was measured and presented as signal/μm2. For background calculation, regions outside a cell were selected, and the signal intensity was measured and presented as signal/μm2. The resolution of the STED images was calculated using image decorrelation analysis,21 which is available as a plugin in Fiji22 with a min radius of 0 and a max radius of 1 and an Nr of 50 and an Ng of 10. All analysis was performed in Fiji.

DNA-PAINT Imaging

DNA-PAINT imaging of the cell and origami samples was carried out on a home-built widefield setup based on a Nikon Eclipse Ti microscope described before.10 Briefly, the excitation light was generated by a DPSS laser at 640 nm (LPX-640L-500-CSB-PPA, Oxxius SA, France) and a DPSS laser at 561 nm (SAPPHIRE 561-300 CW CDRH, Coherent, United States) with the required excitation power controlled by an acousto-optic tunable filter (AOTFnC-400.650-TN, AA Opto Electronic, France). To clean the beam profile, the laser was coupled by a fiber collimator (60FC-4-M6.2-33, Schäfter & Kirchhoff GmbH, Germany) into a polarization-maintaining single-mode optical fiber (PMC-E-400RGB, Schäfter & Kirchhoff GmbH, Germany) and subsequently recollimated to an FWHM diameter of 6 mm (60FC-T-4-M50L-01, Schäfter & Kirchhoff GmbH, Germany). The collinear beam was then directed through two telescope lenses (AC255-030-A-ML and AC508-150-A-ML, Thorlabs GmbH, Germany), which focused the beam onto the back focal plane of the objective (CFI Apochromat TIRF 100XC Oil, Nikon, Japan). A mirror mounted on a motorized translation stage (MTS50-Z8, Thorlabs GmbH, Germany) was used to vary the illumination angle between widefield, HILO, or TIRF. The excitation light was coupled into the microscope by means of a dielectric beamsplitter (zt405/488/561/640rpc, AHF Analysentechnik, Germany), which also transmitted the emission light into the detection beam path. The axial focus was maintained using an autofocus system (Ti-PFS, Nikon), and the lateral position was adjusted using a motorized stage (Ti-S-ER, Nikon) combined with a piezo stage (Nano-Drive, MadCityLabs, USA). After spectral filtering with a red bandpass filter (700/75 ET, Chroma Technology Corp., USA) respectively with an orange bandpass filter (610/60 ET, Chroma Technology Corp., USA), the emission light was projected onto an Andor Ixon Ultra EMCCD camera (DU-897U-CS0, Andor, North Ireland). Origami and NPC samples were measured in TIRF illumination. Parameters used in DNA-PAINT imaging are compiled together as Table S10.

DNA-PAINT Data Analysis

DNA-PAINT data were analyzed with Picasso software (v0.6.0).23 Single-molecule localization was performed with Picasso Localize using the following parameters: baseline, 200.4 photons; sensitivity, 4.32; quantum efficiency, 0.95; a min net gradient of 200,000 for NPCs, 100,000 for SiR, and 250,000 for Cy3B measurements of origamis. Super-resolved images were reconstructed by using Picasso Render. The image stacks for microtubules were drift corrected using redundant cross correlation (RCC) on Picasso Render with 1000–4000 frames. Origami and NPC image stacks were drift corrected using gold beads as fiducial markers (gold nanoparticles, 100 nm diameter, product no. A11-100-NPC-DIH-1-25, NanoPartz, USA). Localizations from DNA origami and NPCs were filtered for the ellipticity (0–0.1) and the width of the point spread function (0.8–1.2).

The photon numbers of DNA origami were extracted from single binding sites (9 nt P1-SiR: n = 1050 binding sites; 9 nt SiR-P1-SiR: n = 471 binding sites; 9 nt P1-Cy3B: n = 3732 binding sites; 9 nt Cy3B-P1-Cy3B: n = 2757 binding sites). The resulting histograms were fitted with one or two Gaussian functions depending on the shape of the frequency distribution. The nearest neighbor-based analysis (NeNA) localization precision24 was extracted directly from Picasso Render. In order to extract the time ton of single binding events, the localizations of single docking strands were linked with a radius of 4× NeNA and 6 transient dark frames. A binding time was extracted from each binding site, and a relative frequency histogram was plotted and fitted with a single log-normal fit.6

graphic file with name jp4c02065_m008.jpg 6

Here, A is the area, ω is the log standard deviation, tfit is the center of the fit, and the mode of the fit gives the average binding time, ton, of the imager strand and is given by

graphic file with name jp4c02065_m009.jpg 7

Both Gaussian and log-normal fits were performed in OriginPro.

The spatial resolution of SMLM images was calculated using rolling Fourier ring correlation (rFRC)25 available as a plugin in Fiji. Briefly, a single-frame rFRC (1/7 hard threshold) was run with a block size of 64 pixels, a background intensity of 20, a skip of 2 pixels, and a pixel size of 2 nm.

Results

We designed fluorophore-labeled DNA oligonucleotides (“imager strands”) using a well-established sequence frequently used for DNA-PAINT (P1; see Table S1)2 and carrying either two fluorophores at the 5′- and 3′-end (dye-P1-dye) or one fluorophore at the 3′-end (P1-dye) (Figure 1A). As fluorophores, we selected one representative out of three classes of organic fluorophores that are used in advanced fluorescence microscopy methods: a silicon-rhodamine (SiR), tetramethylrhodamine (TMR), and the carbocyanine Cy3B (Figure S1A–C and Table S1). We characterized these fluorophore-DNA probes as single-stranded oligonucleotides as well as hybridized them to a sequence-complementary DNA oligonucleotide (“docking strand”), using absorption and fluorescence spectroscopy (Figure 1B). The absorption spectra of dual-labeled, single-stranded Cy3B-P1-Cy3B and TMR-P1-TMR showed a clear blueshifted band, which indicated the formation of a fluorophore dimer; this band disappeared when these probes were hybridized to the docking strand (Figure 1Bi). For SiR-P1-SiR, only a faint dimer band was detected (Figure 1Bi). The fluorescence emission spectra of all three dual-labeled, single-stranded probes showed an effective increase in fluorescence upon hybridization of 3.3 (SiR-P1-SiR), 5.2 (TMR-P1-TMR), and 1.7 (Cy3B-P1-Cy3B) (Figure 1Bi). This indicates quenching of dual-labeled imager strands when they are freely diffusing in solution. In order to assess the kinetics of fluorescence quenching, we recorded fluorescence correlation spectroscopy (FCS) data of single- and dual-labeled probes and observed multiple subdiffusional kinetics (Figure S2 and Table S2). One of the subdiffusional kinetics, in the range of 10–7 s, was only observed for dual-labeled imager strands and was absent for single-labeled counterparts (Figure S3). For the dual-labeled probes, this fast kinetic component appears only for freely diffusing imager strands and completely disappears when hybridized to docking strands, hinting that it arises from the self-quenching process (Figure S3, red). We further quantified the kfluorescence and kdark rates of the self-quenching process in the range of 105–106 s–1 (Figure S4 and Table S2).

Figure 1.

Figure 1

Open in a new tab

Spectroscopic characterization and hybridization equilibria for single- and dual-labeled DNA oligonucleotides. (A) Schematic of conventional DNA-PAINT (blue) and DNA-PAINT using self-quenching fluorophores (red). The inlay shows fluorophores that exhibit self-quenching. (B) (i) Absorption and (ii) emission spectra of single- (3′) and dual-labeled (3′ and 5′) P1 imager strand sequences labeled with SiR, TMR, and Cy3B diluted in a DNA-PAINT imaging buffer with and without sequence-complementary docking strands. Excitation lasers used for emission spectra measurement are shown with dashed lines in part B, red: 647 nm; green: 560 nm. (C) Fluorogenicity plotted against the KD of dual-labeled (red) and single-labeled (blue) P1 imager strands of different fluorophores used in this study and that of ATTO 655.

Next, we compared the hybridization properties of single- and dual-labeled DNA oligonucleotides by measuring the dissociation constant (KD). We found an increased binding affinity (decreased KD) in the case of Cy3B-P1-Cy3B relative to P1-Cy3B and a slightly decreased affinity in the case of both SiR-P1-SiR and TMR-P1-TMR relative to P1-SiR and P1-TMR, respectively (Figure S5). We contextualized KD with the measured fluorogenicity and included the previously published value for ATTO 655-P1-ATTO 65510 (Figure 1C). From this, ATTO 655 (Figure S1D) and Cy3B exhibit a decreased KD (increase in binding affinity) when dual-labeled to the P1 imager strand, whereas SiR and TMR show the contrary. We next evaluated a shorter docking strand length to observe how KD could be influenced and whether it is possible for a given fluorophore to choose a strand length and sequence for an optimum binding–unbinding rate. For 8 nt P1 docking strand length, we observe an increase in KD by a factor of 10 for ATTO 655-P1-ATTO 655 (Figure S6). Since KD could also be influenced by the strand sequence itself, we repeated these experiments with the P5 imager strand Table S1). P5 dual-labeled with ATTO 655 (ATTO 655-P5-ATTO 655) showed a lower KD (82 nM) than ATTO 655-P1-ATTO 655 (179 nM)10 while similar to single-labeled P5 (P5-ATTO 655; 78 nM) (Figure S7A). The fluorogenicity is 4-fold for the ATTO 655-P5-ATTO 655 as compared to the 5-fold of ATTO 655-P1-ATTO 655 (Figure S7B).

We applied these dual-labeled, self-quenched imager strands to STED microscopy. We rationalized that the higher background signal arising from high concentrations of freely diffusing imager strands (300 nM)4 would be mitigated with self-quenched imager strands. In previous work, we have shown that ATTO 655-P1-ATTO 655 reduces the fluorescence background of unbound probes and additionally increases the fluorescence yield of bound probes.10 Here, we generalized this concept by investigating other fluorophores (SiR, TMR, and Cy3B) on dual-labeled imager strands with STED microscopy, enabling two-color applications.

First, we recorded STED images of microtubules in U-2 OS cells (indirect immunolabeling for α-tubulin; see Methods) using either single- or dual-labeled P1 with SiR, TMR, and Cy3B (Figure 2A). All dual-labeled imager strands exhibited a decrease in the background fluorescence for unbound probes (Figure S8A). Interestingly, the fluorescence intensity in the STED images is different for the three fluorophores. In the case of TMR-P1-TMR and Cy3B-P1-Cy3B, the fluorescence intensity is increased, whereas SiR-P1-SiR exhibits a decrease in fluorescence intensity when bound to its complement (Figure S8B). Still, the background fluorescence decrease of SiR-P1-SiR relative to P1-SiR is 3-fold, and the overall signal-to-background ratio (SBR) of SiR-P1-SiR is 2.3-fold higher than P1-SiR (Figure 2B and Figure S7C). In conjunction, TMR-P1-TMR exhibits an SBR increase of 2.3-fold over P1-TMR, and Cy3B-P1-Cy3B exhibits an SBR increase of 3.2-fold over P1-Cy3B. Upon initial observation, the SBR increase does not directly reflect the fluorogenicity from emission spectra but fits well when taken together with changes in KD, which determines the exchange of the fluorophore-labeled imager strands at the target site. SiR-P1-SiR shows a fluorogenicity of 3.3-fold and an increase in KD of 1.3-fold, thereby resulting in an overall SBR of 2.3-fold. TMR-P1-TMR shows a similar fluorogenicity of 5.2 and an increase in KD of 1.4-fold and hence exhibits an SBR of 2.3-fold. Cy3B-P1-Cy3B on the other hand exhibits only 1.7-fold fluorogenicity and shows a decrease in KD of 0.3-fold, which is observed here as an SBR of 3.2-fold. The overall increase in SBR among dual-labeled imager strands also corresponds to an overall improvement in spatial resolution for all dual-labeled imager strands (Figure 2C) (Table S3). Exploiting this new toolbox of dual-labeled imager strands, we performed simultaneous two-color STED imaging of mitochondria (indirect immunolabeling for TOM20) and the endoplasmic reticulum (ER) (indirect immunolabeling for KDEL, see Methods) using ATTO 655-P5-ATTO 655 and TMR-P1-TMR, respectively (Figure 2D).

Figure 2.

Figure 2

Open in a new tab

STED microscopy with single- and dual-labeled DNA oligonucleotides. (A) STED images of U-2 OS cells immunostained for α-tubulin with different single- and dual-labeled P1 imager strands (300 nM). Scale bar = 2 μm. (B) Percentage increase in SBR for all dual-labeled P1 imager strands over their single-labeled counterparts. (C) Percentage increase in resolution for all dual-labeled P1 imager strands relative to single-labeled P1 imager strands. For both B and C, n = 7 cells for P1-SiR and 6 cells for SiR-P1-SiR. For all images using single- and dual-labeled TMR and Cy3B, n = 5 cells. (D) Dual-color STED image of a MODE-K cell immunostained for TOM20 and KDEL with ATTO 655-P5-ATTO 655 (green) and TMR-P1-TMR (magenta), respectively; the scale is 5 μm.

We then explored the use of self-quenching imager strands for DNA-PAINT (Figure 3). We chose Cy3B and SiR for their photostability and spectral orthogonality. For Cy3B, we used 8 nt docking strands to compensate for the low KD measured for 9 nt Cy3B-P1-Cy3B bound to its 9 nt complementary strand (Figure S5) that would be less optimal for DNA-PAINT microscopy. We used DNA origami nanostructures as a platform for comparing single- and dual-labeled P1 imager strands. The DNA origami was constructed with three binding sites (9 and 8 nt docking strands for SiR and Cy3B, respectively) spaced ∼50 nm from each other (Figure 3A,C). For both SiR and Cy3B, we determined the photon yield per target site and observed one population for single-labeled and two populations for dual-labeled imager strands, respectively (Figure 3B,D). For dual-labeled imager strands, the appearance of two populations can be ascribed to photobleaching of one of the fluorophores while bound to the target or other mechanisms that deactivate one of the fluorophores (Figure S9). We next determined the localization precision and found a decrease for SiR-P1-SiR (6.6 nm) compared to P1-SiR (12.3 nm), as well as for Cy3B-P1-Cy3B (5.5 nm) as compared to P1-Cy3B (8.0 nm) (Figure 3B,D, inlays). We attribute the larger change in localization precision for SiR to a higher background signal of P1-SiR as compared to that of SiR-P1-SiR (see Figure 1Bii). In addition, we analyzed the binding times (ton) of the imager strands through hybridization with the docking strands. We observed a decrease in ton for SiR-P1-SiR (239 ms) over P1-SiR (410 ms) and a small change in ton for Cy3B-P1-Cy3B (351 ms) over P1-Cy3B (344 ms) (Figure S10).

Figure 3.

Figure 3

Open in a new tab

Self-quenching fluorophores for DNA-PAINT microscopy. (A) DNA-PAINT overview image and zoomed-in image of DNA origami with three binding sites (9 nt docking strands) spaced ∼50 nm from each other acquired with 10 nM P1-SiR (top) and SiR-P1-SiR (bottom). (B) Frequency distributions of photons from (A) acquired with P1-SiR (top, blue) and SiR-P1-SiR (bottom, red). The inlay shows the localization precision calculated by using the nearest neighbor analysis (NeNA) of both overview images in (A). (C) DNA-PAINT overview image and zoomed-in image of DNA origami with three binding sites (8 nt docking strands) spaced ∼50 nm from each other acquired with 10 nM P1-Cy3B (top) and Cy3B-P1-Cy3B (bottom). (D) Frequency distributions of photons from (C) acquired with P1-Cy3B (top, blue) and Cy3B-P1-Cy3B (bottom, red). The inlay shows the localization precision calculated using the nearest neighbor analysis (NeNA) of both overview images in part (C). (E) DNA-PAINT image of the nuclear pore complex protein Nup96 (NPC) and a zoomed-in image acquired using 1.5 nM P1-Cy3B or Cy3B-P1-Cy3B (the full-size DNA-PAINT images are shown in Figure S11). (F) Frequency distribution of photons measured from manually selected NPCs from (E) acquired with P1-Cy3B (top, blue) and Cy3B-P1-Cy3B (bottom, red). The inlay shows the localization precision calculated using the nearest neighbor analysis (NeNA) of both overview images in (E).

Next, we applied single- and dual-labeled imager strands for DNA-PAINT imaging in U-2 OS and visualized the nuclear pore complex (NPC) protein Nup96 (Figure 3E,F and Figure S11). We quantified the photon output for both P1-Cy3B and Cy3B-P1-Cy3B by choosing NPCs located in the same area of the camera field of view, in order to ensure equal illumination conditions and exhibiting a circular structure with an 8-fold symmetry. We found a two-population photon distribution (Figure 3F) and better localization precision (8.3 nm for Cy3B-P1-Cy3B as compared to 11.5 nm for P1-Cy3B) (Figure 3F, inlay) and resolution (26 and 31 nm, respectively) (Figure S12 and Table S4) in images generated with dual-labeled imager strands. This demonstrates that the extended repertoire of self-quenchable fluorophores achieves a higher photon yield and improved localization precision in DNA-PAINT microscopy.

Discussion

We report the self-quenching of fluorophore dimers attached to short DNA oligonucleotides as a general concept for DNA-based microscopy. Expanding the first reports on oxazine dimers,10,11 we report a fluorogenicity of DNA imager strands dual-labeled with a silicon-rhodamine (SiR), a rhodamine (TMR), or a carbocyanine (Cy3B), in a range of 1.7 to 5.2. Of note, we also tested other fluorophores, which either did not exhibit fluorogenicity or exhibit unspecific binding, making them unfit to be used as self-quenching imager strands (data not included, Table S5). Fluorescence quenching occurs through the formation of a short-distance H-dimer,15 and the fluorescence signal is restored upon hybridization to a sequence-complementary strand. This distinguishes the present concept from other reports that used, e.g., Förster resonance energy transfer (FRET), which occurs at longer distances and demands for extension of DNA probes.8,9,26 A second unique property of our concept is that target binding, i.e., hybridization to a sequence-complementary DNA probe, comes with an almost 2-fold increase in fluorescence intensity that is contributed by the two fluorophores. Furthermore, we observed that the hybridization affinity of dual-labeled probes is different from those of single-labeled probes and depends on the fluorophore and the DNA sequence, which opens the possibility of further optimization.

We demonstrate the application of self-quenched DNA probes in two major super-resolution microscopy techniques, STED and DNA-PAINT microscopy. For STED microscopy, we found an increase in signal-to-background (SBR) for all dual-labeled probes compared to single-labeled probes, ranging between 2- and 3-fold. The main parameters that determined the SBR were the fluorogenicity of the probes and the dissociation constant (KD). The increase in the SBR was accompanied by an improved spatial resolution (Table S3). Further optimization is possible by adapting the hybridization kinetics to the “imaging kinetics” in STED microscopy. The availability of dual-labeled probes with orthogonal spectral properties enabled two-color STED imaging (Figure 2D). For DNA-PAINT microscopy, we observe an increase in photon yield, accompanied by a reduced localization precision and, thus, an improved spatial resolution.

Beyond the application shown in this work, we envision the use of dual-labeled DNA probes for other microscopy techniques, such as SOFI3 and MINFLUX.27 If necessary, the hybridization kinetics can be modified by the sequence length and base content. These probes complement the concept of fluorogenic dimers that was reported for, e.g., membrane-binding labels.28 An extension to more weak-affinity fluorophore labels29 or protein tags30 and combinations thereof5 can be envisioned.

In conclusion, we show that the concept of self-quenching dimers could be extended to other classes of fluorophores and that these probes provide a higher SBR and improved spatial resolution in cutting-edge microscopy applications.

Acknowledgments

We thank Petra Freund for assistance with cell culture. We thank Prof. Florian Greten, Georg-Speyer-Haus, Frankfurt, for kindly providing MODE-K cells. We thank Prof. Josef Wachtveitl for access to the absorption spectrometer and Florian Hurter for experimental support. We gratefully acknowledge funding by the Deutsche Forschungsgemeinschaft (grants CRC 1177, CRC 1507, and INST 161/1020-1 FUGG) and the SubCellular Architecture of LifE (SCALE) consortium funded by the Goethe University Frankfurt, Germany.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.jpcb.4c02065.

  • Tables summarizing spectroscopic properties of single- and dual-labeled DNA probes, DNA oligonucleotide sequences, FCS fit parameters, and details of settings used in microscopy experiments; graphics showing chemical structures, FCS plots and analyses, binding curves of single- and dual-labeled DNA probes, and additional imaging data and further image analysis (PDF)

Author Contributions

# L.F.K. and A.B. contributed equally.

The authors declare no competing financial interest.

Special Issue

Published as part of The Journal of Physical Chemistry Bvirtual special issue “Advances in Cellular Biophysics”.

Supplementary Material

jp4c02065_si_001.pdf (2.5MB, pdf)

References

  1. Sauer M.; Heilemann M. Single-Molecule Localization Microscopy in Eukaryotes. Chem. Rev. 2017, 117 (11), 7478–7509. 10.1021/acs.chemrev.6b00667. [DOI] [PubMed] [Google Scholar]
  2. Schnitzbauer J.; Strauss M. T.; Schlichthaerle T.; Schueder F.; Jungmann R. Super-Resolution Microscopy with DNA-PAINT. Nat. Protoc. 2017, 12 (6), 1198–1228. 10.1038/nprot.2017.024. [DOI] [PubMed] [Google Scholar]
  3. Glogger M.; Spahn C.; Enderlein J.; Heilemann M. Multi-Color, Bleaching-Resistant Super-Resolution Optical Fluctuation Imaging with Oligonucleotide-Based Exchangeable Fluorophores. Angew. Chem., Int. Ed. Engl. 2021, 60 (12), 6310–6313. 10.1002/anie.202013166. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Spahn C.; Hurter F.; Glaesmann M.; Karathanasis C.; Lampe M.; Heilemann M. Protein-Specific, Multicolor and 3D STED Imaging in Cells with DNA-Labeled Antibodies. Angew. Chem., Int. Ed. Engl. 2019, 58 (52), 18835–18838. 10.1002/anie.201910115. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Glogger M.; Wang D.; Kompa J.; Balakrishnan A.; Hiblot J.; Barth H.-D.; Johnsson K.; Heilemann M. Synergizing Exchangeable Fluorophore Labels for Multitarget STED Microscopy. ACS Nano 2022, 16 (11), 17991–17997. 10.1021/acsnano.2c07212. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Jungmann R.; Avendaño M. S.; Dai M.; Woehrstein J. B.; Agasti S. S.; Feiger Z.; Rodal A.; Yin P. Quantitative Super-Resolution Imaging with qPAINT. Nat. Methods 2016, 13 (5), 439–442. 10.1038/nmeth.3804. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Narayanasamy K. K.; Rahm J. V.; Tourani S.; Heilemann M. Fast DNA-PAINT Imaging Using a Deep Neural Network. Nat. Commun. 2022, 13 (1), 5047. 10.1038/s41467-022-32626-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Chung K. K. H.; Zhang Z.; Kidd P.; Zhang Y.; Williams N. D.; Rollins B.; Yang Y.; Lin C.; Baddeley D.; Bewersdorf J. Fluorogenic DNA-PAINT for Faster, Low-Background Super-Resolution Imaging. Nat. Methods 2022, 19 (5), 554–559. 10.1038/s41592-022-01464-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Auer A.; Strauss M. T.; Schlichthaerle T.; Jungmann R. Fast, Background-Free DNA-PAINT Imaging Using FRET-Based Probes. Nano Lett. 2017, 17 (10), 6428–6434. 10.1021/acs.nanolett.7b03425. [DOI] [PubMed] [Google Scholar]
  10. Kessler L. F.; Balakrishnan A.; Deußner-Helfmann N. S.; Li Y.; Mantel M.; Glogger M.; Barth H.; Dietz M. S.; Heilemann M. Self-Quenched Fluorophore Dimers for DNA-PAINT and STED Microscopy. Angew. Chem., Int. Ed. Engl. 2023, 62 (39), e202307538 10.1002/anie.202307538. [DOI] [PubMed] [Google Scholar]
  11. Ghosh A.; Meub M.; Helmerich D. A.; Doose S.; Sauer M. Fast, Whole-Cell DNA-PAINT Imaging with Dual-Labeled Self-Quenched Imager Probes. bioRxiv 2023, 10.1101/2023.10.09.561472. [DOI] [Google Scholar]
  12. Grama L.; Somogyi B.; Kellermayer M. S. Global Configuration of Single Titin Molecules Observed through Chain-Associated Rhodamine Dimers. Proc. Natl. Acad. Sci. U. S. A. 2001, 98 (25), 14362–14367. 10.1073/pnas.191494098. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Dietz M. S.; Wehrheim S. S.; Harwardt M.-L. I. E.; Niemann H. H.; Heilemann M. Competitive Binding Study Revealing the Influence of Fluorophore Labels on Biomolecular Interactions. Nano Lett. 2019, 19 (11), 8245–8249. 10.1021/acs.nanolett.9b03736. [DOI] [PubMed] [Google Scholar]
  14. Hill A. V. The Possible Effects of the Aggregation of the Molecules of Hemoglobin on Its Dissociation Curves. J. Physiol. 1910, 40 (suppl), i–vii. [Google Scholar]
  15. Bollmann S.; Löllmann M.; Sauer M.; Doose S. Dimer Formation of Organic Fluorophores Reports on Biomolecular Dynamics under Denaturing Conditions. Phys. Chem. Chem. Phys. 2011, 13 (28), 12874–12882. 10.1039/c1cp21111k. [DOI] [PubMed] [Google Scholar]
  16. Doose S.; Neuweiler H.; Sauer M. Fluorescence Quenching by Photoinduced Electron Transfer: A Reporter for Conformational Dynamics of Macromolecules. ChemPhysChem 2009, 10 (9–10), 1389–1398. 10.1002/cphc.200900238. [DOI] [PubMed] [Google Scholar]
  17. Schnitzbauer J.; Strauss M. T.; Schlichthaerle T.; Schueder F.; Jungmann R. Super-Resolution Microscopy with DNA-PAINT. Nat. Protoc. 2017, 12 (6), 1198–1228. 10.1038/nprot.2017.024. [DOI] [PubMed] [Google Scholar]
  18. Vidal K.; Grosjean I.; Revillard J. P.; Gespach C.; Kaiserlian D. Immortalization of Mouse Intestinal Epithelial Cells by the SV40-Large T Gene. Phenotypic and Immune Characterization of the MODE-K Cell Line. J. Immunol. Methods 1993, 166 (1), 63–73. 10.1016/0022-1759(93)90329-6. [DOI] [PubMed] [Google Scholar]
  19. Harwardt M.-L. I. E.; Schröder M. S.; Li Y.; Malkusch S.; Freund P.; Gupta S.; Janjic N.; Strauss S.; Jungmann R.; Dietz M. S.; et al. Single-Molecule Super-Resolution Microscopy Reveals Heteromeric Complexes of MET and EGFR upon Ligand Activation. Int. J. Mol. Sci. 2020, 21 (8), 2803. 10.3390/ijms21082803. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Schnitzbauer J.; Strauss M. T.; Schlichthaerle T.; Schueder F.; Jungmann R. Super-Resolution Microscopy with DNA-PAINT. Nat. Protoc. 2017, 12 (6), 1198–1228. 10.1038/nprot.2017.024. [DOI] [PubMed] [Google Scholar]
  21. Descloux A.; Grußmayer K. S.; Radenovic A. Parameter-Free Image Resolution Estimation Based on Decorrelation Analysis. Nat. Methods 2019, 16 (9), 918–924. 10.1038/s41592-019-0515-7. [DOI] [PubMed] [Google Scholar]
  22. Schindelin J.; Arganda-Carreras I.; Frise E.; Kaynig V.; Longair M.; Pietzsch T.; Preibisch S.; Rueden C.; Saalfeld S.; Schmid B.; et al. Fiji: An Open-Source Platform for Biological-Image Analysis. Nat. Methods 2012, 9 (7), 676–682. 10.1038/nmeth.2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Schnitzbauer J.; Strauss M. T.; Schlichthaerle T.; Schueder F.; Jungmann R. Super-Resolution Microscopy with DNA-PAINT. Nat. Protoc. 2017, 12 (6), 1198–1228. 10.1038/nprot.2017.024. [DOI] [PubMed] [Google Scholar]
  24. Endesfelder U.; Malkusch S.; Fricke F.; Heilemann M. A Simple Method to Estimate the Average Localization Precision of a Single-Molecule Localization Microscopy Experiment. Histochem. Cell Biol. 2014, 141 (6), 629–638. 10.1007/s00418-014-1192-3. [DOI] [PubMed] [Google Scholar]
  25. Zhao W.; Huang X.; Yang J.; Qu L.; Qiu G.; Zhao Y.; Wang X.; Su D.; Ding X.; Mao H.; Jiu Y.; Hu Y.; Tan J.; Zhao S.; Pan L.; Chen L.; Li H.; et al. Quantitatively Mapping Local Quality of Super-Resolution Microscopy by Rolling Fourier Ring Correlation. Light Sci. Appl. 2023, 12 (1), 298. 10.1038/s41377-023-01321-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Deußner-Helfmann N. S.; Auer A.; Strauss M. T.; Malkusch S.; Dietz M. S.; Barth H.-D.; Jungmann R.; Heilemann M. Correlative Single-Molecule FRET and DNA-PAINT Imaging. Nano Lett. 2018, 18 (7), 4626–4630. 10.1021/acs.nanolett.8b02185. [DOI] [PubMed] [Google Scholar]
  27. Ostersehlt L. M.; Jans D. C.; Wittek A.; Keller-Findeisen J.; Inamdar K.; Sahl S. J.; Hell S. W.; Jakobs S. DNA-PAINT MINFLUX Nanoscopy. Nat. Methods 2022, 19 (9), 1072–1075. 10.1038/s41592-022-01577-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Aparin I. O.; Yan R.; Pelletier R.; Choi A. A.; Danylchuk D. I.; Xu K.; Klymchenko A. S. Fluorogenic Dimers as Bright Switchable Probes for Enhanced Super-Resolution Imaging of Cell Membranes. J. Am. Chem. Soc. 2022, 144 (39), 18043–18053. 10.1021/jacs.2c07542. [DOI] [PubMed] [Google Scholar]
  29. Albertazzi L.; Heilemann M. When Weak Is Strong: A Plea for Low-Affinity Binders for Optical Microscopy. Angew. Chem., Int. Ed. Engl. 2023, 62 (35), e202303390 10.1002/anie.202303390. [DOI] [PubMed] [Google Scholar]
  30. Kompa J.; Bruins J.; Glogger M.; Wilhelm J.; Frei M. S.; Tarnawski M.; D’Este E.; Heilemann M.; Hiblot J.; Johnsson K. Exchangeable HaloTag Ligands for Super-Resolution Fluorescence Microscopy. J. Am. Chem. Soc. 2023, 145 (5), 3075–3083. 10.1021/jacs.2c11969. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

jp4c02065_si_001.pdf (2.5MB, pdf)

Articles from The Journal of Physical Chemistry. B are provided here courtesy of American Chemical Society

RESOURCES