A Bioorthogonally Applicable, Fluorogenic, Large Stokes-Shift Probe for Intracellular Super-Resolution Imaging of Proteins

Herein, we present the synthesis and application of a fluorogenic, large Stokes-shift (>100 nm), bioorthogonally conjugatable, membrane-permeable tetrazine probe, which can be excited at common laser line 488 nm and detected at around 600 nm. The applied design enabled improved fluorogenicity in the orange/red emission range, thus efficient suppression of background and autofluorescence upon imaging biological samples. Moreover, unlike our previous advanced probes, it does not require the presence of special target platforms or microenvironments to achieve similar fluorogenicity and can be generally applied, e.g., on translationally bioorthogonalized proteins. Live-cell labeling schemes revealed that the fluorogenic probe is suitable for specific labeling of intracellular proteins, site-specifically modified with a cyclooctynylated, non-canonical amino acid, even under no-wash conditions. Furthermore, the probe was found to be applicable in stimulated emission depletion (STED) super-resolution microscopy imaging using a 660 nm depletion laser. Probably the most salient feature of this new probe is that the large Stokes-shift allows dual-color labeling schemes of cellular structures using distinct excitation and the same detection wavelengths for the combined probes, which circumvents chromatic aberration related problems.


Introduction
Fluorescent methods are distinguished amongst the techniques dedicated to the sensitive and accurate detection of cellular and subcellular events in vivo. Due to recent hardware developments in super-resolution microscopy, it is rather the lack of suitable probes, which is considered as the major limitation of further improvements [1,2]. Therefore, there is a clear need for advanced probes suitable for selective targeting and in vivo super-resolution imaging of intracellular structures. Such membrane permeable, improved probes should address challenges such as selective conjugatability, autofluorescence, and background fluorescence [3]. Autofluorescence can be overcome by carefully selecting fluorescent frameworks that are either excitable towards the red range of the spectrum or possess large Stokes-shifts. Suppression of background fluorescence of non-specifically bound probes, on the other hand, is more challenging and often requires extensive washing cycles following staining. An alternative approach to minimize background fluorescence becomes possible with the use of fluorogenic probes that possess quenched fluorescence until a specific reaction with their target [4][5][6]. With respect to selective conjugatability, several strategies have evolved lately as a result of recent [4][5][6]. With respect to selective conjugatability, several strategies have evolved lately as a result of recent advances in gene technology and chemical biology [7]. Bioorthogonal ligations [8] in combination with genetic code expansion (GCE) [9] technology, is probably the most powerful method for the selective and site-specific installation of virtually non-perturbing, synthetic probes onto the protein of interest (POI). Such an approach is used routinely in the labeling schemes of various proteins, in cellulo or in vivo [6][7][8][10][11][12].
In terms of reaction kinetics, biocompatibility, synthetic accessibility, and compatibility with GCE, strain promoted azide-alkyne cycloaddition (SPAAC) of azides and strained alkynes [13] and the inverse electron demand Diels-Alder (IEDDA) reaction of tetrazines and strained alkenes/alkynes [14] are the most robust bioorthogonal reactions. Besides, as we and others have recently demonstrated, functional groups of the SPAAC and IEDDA reactions, i.e., the azide and the tetrazine moieties, are able to modulate (i.e., quench) fluorescence by various mechanisms (e.g., energy transfer [15][16][17], rotation-induced relaxation [18,19], or photoinduced electron transfer [20]). Upon transformation of the quenching bioorthogonal unit in a specific ligation reaction, the fluorescence reinstates (Scheme 1). This two-in-one combination of bioorthogonal handles was exploited in the design of coumarin [21], BODIPY [16], phenoxazine [17], rhodamine [22], and cyanine-based [23], bioorthogonally applicable fluorogenic probes. It was observed, however, that the fluorescence increase upon bioorthogonal ligation, i.e., fluorogenicity, dramatically decreases towards the biologically preferred red range of the spectrum (Figure 1). To date, the most remarkable example of such fluorogenic probes in terms of fluorescence increase, are ultra violet (UV)/blue excitable ultrafluorogenic HELIOS (hyperemissive ligation-initiated orthogonal sensing) probes reported by Weissleder et al. in 2014 [21]. Emission of these coumarin-based probes was practically fully quenched by the bioorthogonal handle tetrazine. Upon specific reaction with trans-cyclooctenes (TCOs), 3-4 orders of magnitude increase in fluorescence were observed, which allowed no-wash fluorescence imaging of epidermal growth factor receptors (EGFR) using TCO labeled anti-EGFR antibodies. To optimize probes for biological applications, it would be highly desirable to create fluorogenic dyes that are either excitable towards the red range or possess large Stokes-shift, while keeping the excellent fluorogenicity of UV/blue excitable systems. Our continuing efforts to address this problem resulted in the development of multiple fluorogenicity, as demonstrated on orange/far red excitable advanced probes [24,25]. Though these double fluorogenic probes solved the above-mentioned problems, their applicability is limited either to special sequences (i.e., double genetic incorporation of two non-canonical-cyclooctynylated-amino acids) [24] or microenvironments (i.e., the polar environment in the proximity of ligation) [25]. Thus, new approaches that could result in generally applicable, advanced probes are still needed. Since probes with large Stokes-shifts may also overcome the hampering effects of autofluorescence, we hypothesized that blue-green excitable, thus good fluorogenic probes having considerably large Stokes-shifts, may offer a more general solution to these problems.
Inspired by the outstanding fluorogenic performance of tetrazine-quenched coumarins, and the large Stokes-shifts of coumaryl-vinyl-pyridinium probes, we set forth a systematic study aimed at constructing small, π-extended fluorogenic probes on the grounds of HELIOS frames.

General
Unless otherwise indicated, all starting materials were obtained from commercial suppliers (Sigma-Aldrich (St. Louis, MO, USA), Merck (Darmstadt, Germany), Alfa Aesar (Haverhill, MA, USA), Reanal (Budapest, Hungary), Molar Chemicals (Halásztelek, Hungary), Fluorochem (Hadfield, Glossop, UK) and used without further purification. Analytical thin-layer chromatography (TLC) was performed on silica gel 60 F254 precoated aluminum TLC plates from Merck. Flash column chromatography was performed on Teledyne Isco CombiFlash ® Rf + (Teledyne Isco, Lincoln, NE, USA) automated flash chromatography apparatus by using silica gel (25-40 μm) from Zeochem (Rüti, Switzerland). Microwave experiments were carried out on an AntonPaar (Graz, Austria) Monowave 300 microwave reactor (maximum power 850 W). NMR spectra were recorded on Varian Inova 500 MHz NMR spectrometer (Palo Alto, CA, USA). Chemical shifts (δ) are given in parts per million (ppm) by using solvent signals as the reference. Coupling constants (J) are reported in Hertz (Hz). Splitting patterns are designated as s (singlet), bs (broad singlet), d (doublet), dd (doublet of a doublet), t (triplet), q (quartet), or m (multiplet). Analytical reverse-phase highperformance liquid chromatography with ultra-violet/visible and mass spectrometry detection (RP- To optimize probes for biological applications, it would be highly desirable to create fluorogenic dyes that are either excitable towards the red range or possess large Stokes-shift, while keeping the excellent fluorogenicity of UV/blue excitable systems. Our continuing efforts to address this problem resulted in the development of multiple fluorogenicity, as demonstrated on orange/far red excitable advanced probes [24,25]. Though these double fluorogenic probes solved the above-mentioned problems, their applicability is limited either to special sequences (i.e., double genetic incorporation of two non-canonical-cyclooctynylated-amino acids) [24] or microenvironments (i.e., the polar environment in the proximity of ligation) [25]. Thus, new approaches that could result in generally applicable, advanced probes are still needed. Since probes with large Stokes-shifts may also overcome the hampering effects of autofluorescence, we hypothesized that blue-green excitable, thus good fluorogenic probes having considerably large Stokes-shifts, may offer a more general solution to these problems.
Inspired by the outstanding fluorogenic performance of tetrazine-quenched coumarins, and the large Stokes-shifts of coumaryl-vinyl-pyridinium probes, we set forth a systematic study aimed at constructing small, π-extended fluorogenic probes on the grounds of HELIOS frames.

Spectroscopic Measurements
Stock solutions of dyes were prepared in 1 mM concentration in DMSO. Reactions with (1R,8S,9s)-Bicyclo[6.1.0]non-4-yn-9-ylmethanol (BCN) were carried out in these stock solutions by adding an excess of solid BCN to the solution. The reaction was monitored by LC-MS until the reaction went to completion. Subsequent fluorescence measurements were carried out by diluting stock solutions to 5 µM concentrations with PBS. Spectroscopic measurements were performed on a Jasco FP 8300 spectrofluorometer (Halifax, NS, Canada) and a JASCO v750 spectrophotometer (Halifax, NS, Canada) in PBS (pH = 7.4) at room temperature (r.t.). The excitation and emission slits were set to 2.5 nm, scanning speed was set to 500 nm/min. All measurements were carried out in quartz cuvettes with a path length of 1 cm. Clear PBS (phosphate buffered saline) solution with additional DMSO was used for blank measurements. Each dye was excited at the respective excitation maxima and fluorescence was detected at the corresponding emission maxima. Turn-on values were calculated by dividing the start and end fluorescence intensity values of time-course measurements at the emission maxima of the products. Quantum yields were determined using coumarin 153 in ethanol (Φ = 0.55) [28] and rhodamine B in basic ethanol (Φ = 0.65) [29] as standard.

Actin Labeling
Sequential actin labeling was carried out with Cy3 [23] modified Phalloidin in TBS (tris-buffered saline) buffer for 40 min in the dark. Plates were washed and imaged.

Confocal and STED Imaging and Analysis
Confocal and STED images were acquired on a Leica TCS SP8 STED 3X microscope using the 488 or 552 nm laser for excitation; 660 nm STED (1.5 W, continuous wave) laser for depletion. The images were taken using a Leica HC PL APO 100×/1.40 oil immersion objective along with a Leica HyD detector. Using the Huygens STED Deconvolution Wizard (Huygens Software, Hilversum, The Netherlands), only a moderate degree of deconvolution was applied to the recorded STED images to avoid deconvolution artifacts. The deconvolution was based on theoretical point spread function (PSF) values. Images were analyzed using Leica (Weitzlar, Germany) Application Suite X and ImageJ software (NIH). Gaussian non-linear fitting and full-width at half maximum values (FWHM) were obtained with Origin Pro 9 software (Northampton, MA, USA).

Results and Discussion
Considering the possible means for the extension of the π-system of coumarins with a vinylpyridynium motif, either the 3rd or the 4th position seems an obvious choice for synthetic reasons. Literature examples show, however, that 4-(vinylpyridyl)coumarins lose their fluorescence [31], leaving 3-(vinylpyridyl)coumarins as possible π-extended frames. Such 3-(vinylpyridyl)coumarin probes possess remarkable Stokes-shifts (often referred to as megaStokes dyes, [32]) and red-shifted (ca. 80-100 nm) excitation and emission maxima with respect to plain coumarin [32,33]. However, direct modification of the original HELIOS frame is rather limited and does not allow ready extension of the π-system, for the 3rd position is already taken by the quenching phenyl-methyl tetrazine moiety. Thus, we first needed to examine whether a similar quenching effect of phenyl tethered tetrazines is exerted at the 4th position. Moreover, we also intended to study the quenching effect of a vinylene tethered tetrazine. To this end, we first explored the possibility of installing the tetrazine unit onto the 4th position of the coumarin scaffold via a vinylene or phenylene linker. Recently, we demonstrated the suitability of Suzuki-type of cross-coupling reactions for the direct installation of phenylene linked tetrazines onto various fluorescent cores [17,23]. Furthermore, Heck cross-coupling of in situ generated vinyltetrazines, a method developed by the Devaraj group, is also a feasible means for the introduction of the tetrazine motif [27].
Biomolecules 2020, 10, 397 7 of 12 direct installation of phenylene linked tetrazines onto various fluorescent cores [17,23]. Furthermore, Heck cross-coupling of in situ generated vinyltetrazines, a method developed by the Devaraj group, is also a feasible means for the introduction of the tetrazine motif [27].
To access such 4-substituted coumarins, we synthesized 7-(diethylamino)-2-oxo-2H-chromen-4-yl trifluoromethanesulfonate (2) and treated it either with boronate 3 or mesylate 4 in the presence of a Pd catalyst, furnishing the respective 4-functionalized coumarins (5 and 6) in medium yields (Scheme 2). With these HELIOS-like probes in hand, we tested their fluorogenic performance by comparing the fluorescence spectra of the parent tetrazines and their reaction products with cyclooctyne, BCN ((1R,8S,9s)-Bicyclo[6.1.0]non-4-yn-9-ylmethanol) (Scheme 1, Table 1, and Figure S1). We found that both 5 and 6 were practically fully quenched. BCN conjugate of 5, on the other hand, showed intense fluorescence. At the same time, 6-BCN was found to be poorly fluorescent. For both probes, we observed practically fully quenched spectra. Therefore, the determination of fluorescence enhancement values, FE (FE = I/I0, where I is the fluorescence intensity at λem of the respective BCN conjugate, and I0 is the intensity of the unconjugated probe at the λem of the BCN conjugate) was meaningless. Thus, we instead called this fluorogenicity infinite.
Knowing that 4-tethered phenylene-tetrazines also exert full quenching of the coumarin fluorescence, which can be reinstated upon reaction with a strained alkyne, we aimed at extending the π-conjugation of the fully quenched 5 scaffold by installing a 1-methyl-4-vinylpyridinyl at the 3rd position, to access red-shifted HELIOS-like fluorogenic probe. We proposed a synthetic routine, which utilizes Heck-cross coupling for the incorporation of the vinyl-pyridyl motif. To this end, 5 was treated with N-bromosuccinimide (NBS) to access 3-bromo functionalized intermediate 7, which was subsequently reacted with 4-vinylpyridine in the presence of a Pd catalyst. The resulting πextended coumarin was further treated with MeI to obtain a red-shifted Helios-like probe 8 in medium yield (Scheme 3).
Following synthesis, we have explored the spectral characteristics of probe 8 and its in situ formed BCN conjugate. To our delight, the excitation maximum in PBS buffer was found to be at around 488 nm, perfectly matching with one of the most frequently used microscope laser-line, while emission maximum was centered at 600 nm, in accordance with the large Stokes shift of related compounds [32,33]. Gratifyingly, a comparison of the fluorescence intensities of 8 and 8-BCN revealed a ca. 30-fold increase in quantum yield at 600 nm (for spectra see Figure S1). We did not observe any considerable changes in emission intensities between pH 4-9 either for 8 or 8-BCN ( Figure S3). With these HELIOS-like probes in hand, we tested their fluorogenic performance by comparing the fluorescence spectra of the parent tetrazines and their reaction products with cyclooctyne, BCN ((1R,8S,9s)-Bicyclo[6.1.0]non-4-yn-9-ylmethanol) (Scheme 1, Table 1, and Figure S1). We found that both 5 and 6 were practically fully quenched. BCN conjugate of 5, on the other hand, showed intense fluorescence. At the same time, 6-BCN was found to be poorly fluorescent. For both probes, we observed practically fully quenched spectra. Therefore, the determination of fluorescence enhancement values, FE (FE = I/I 0 , where I is the fluorescence intensity at λ em of the respective BCN conjugate, and I 0 is the intensity of the unconjugated probe at the λ em of the BCN conjugate) was meaningless. Thus, we instead called this fluorogenicity infinite.
Knowing that 4-tethered phenylene-tetrazines also exert full quenching of the coumarin fluorescence, which can be reinstated upon reaction with a strained alkyne, we aimed at extending the π-conjugation of the fully quenched 5 scaffold by installing a 1-methyl-4-vinylpyridinyl at the 3rd position, to access red-shifted HELIOS-like fluorogenic probe. We proposed a synthetic routine, which utilizes Heck-cross coupling for the incorporation of the vinyl-pyridyl motif. To this end, 5 was treated with N-bromosuccinimide (NBS) to access 3-bromo functionalized intermediate 7, which was subsequently reacted with 4-vinylpyridine in the presence of a Pd catalyst. The resulting π-extended coumarin was further treated with MeI to obtain a red-shifted Helios-like probe 8 in medium yield (Scheme 3).
Following synthesis, we have explored the spectral characteristics of probe 8 and its in situ formed BCN conjugate. To our delight, the excitation maximum in PBS buffer was found to be at around 488 nm, perfectly matching with one of the most frequently used microscope laser-line, while emission maximum was centered at 600 nm, in accordance with the large Stokes shift of related compounds [32,33]. Gratifyingly, a comparison of the fluorescence intensities of 8 and 8-BCN revealed a ca. 30-fold increase in quantum yield at 600 nm (for spectra see Figure S1). We did not observe any considerable changes in emission intensities between pH 4-9 either for 8 or 8-BCN ( Figure S3).
Next, we aimed at demonstrating the applicability of our new, fluorogenic large Stokes-shift probe in protein labeling schemes. First, we applied the probe to our fixed-cell labeling model. COS-7 cells were fixed and treated with a cyclooctynylated actin affinity tag, BCN-phalloidin, then washed. Subsequently, the tagged actin filaments were stained with probe 8, washed, and subjected to confocal microscopy. This experiment was repeated without the washing step to explore the fluorogenic nature of the probe (Supplementary Material, Figure S4). There was no substantial difference between the two image sets in terms of background fluorescence indicating that the extent of fluorescence increase is enough to differentiate between specific and non-specific staining. It is worth mentioning that labeling was efficient even at 0.1 μM labeling concentration. Next, we aimed at testing the probe in a more challenging environment, i.e., in live-cell labeling schemes. First, we assessed the toxicity of probe 8 on COS-7 cells. The presence of the probe did not affect the viability of the cells within a reasonable concentration range (0-30 μM) used for labeling experiments, as indicated by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay ( Figure  S6). To test probe 8 on live cells, a tailored, genetically bioorthogonalized cytoskeletal protein, vimentin, modified with a non-canonical, bioorthogonalized lysine (Lys( ε N-BCNendo) was expressed in COS-7 cells, using the orthogonal pyrrolysine tRNA Pyl /PylRS pair from Methanosarcina mazei. The construct also carried a C-terminal mOrange fusion tag for reference (Vim N116TAG -mOrange).
Live COS-7 cells were then treated with probe 8, washed, fixed, and subjected to confocal microscopy. Images of the as-treated COS-7 cells showed good co-localization of the probe with the reference mOrange, suggesting specific labeling at the position of the Lys-BCN (Figure 2). This experiment also indicated that probe 8 is membrane permeable and can stand the conditions of live cell labeling. We were curious whether the fluorogenic nature of 8 allows no-wash labeling schemes in live-cell labeling and repeated the labeling process without the washing and fixation steps. To our delight, images of live cells showed specific, low background labeling of vimentin filaments (Figure 2).
The spectral properties of 8 suggest that the probe is suitable for super-resolution imaging using STED microscopy with a 660 nm depletion laser (Supplementary Material, Figure S2). Therefore, we have subjected the samples to super-resolution imaging using STED microscopy ( Figure 3) and concluded that probe 8 is indeed suitable for STED microscopy. Again, we did not experience considerable change in the resolution between washed and non-washed samples. It should also be Scheme 3. Synthesis of π-extended HELIOS-like probe 8. Table 1.
Next, we aimed at demonstrating the applicability of our new, fluorogenic large Stokes-shift probe in protein labeling schemes. First, we applied the probe to our fixed-cell labeling model. COS-7 cells were fixed and treated with a cyclooctynylated actin affinity tag, BCN-phalloidin, then washed. Subsequently, the tagged actin filaments were stained with probe 8, washed, and subjected to confocal microscopy. This experiment was repeated without the washing step to explore the fluorogenic nature of the probe (Supplementary Material, Figure S4). There was no substantial difference between the two image sets in terms of background fluorescence indicating that the extent of fluorescence increase is enough to differentiate between specific and non-specific staining. It is worth mentioning that labeling was efficient even at 0.1 µM labeling concentration. Next, we aimed at testing the probe in a more challenging environment, i.e., in live-cell labeling schemes. First, we assessed the toxicity of probe 8 on COS-7 cells. The presence of the probe did not affect the viability of the cells within a reasonable concentration range (0-30 µM) used for labeling experiments, as indicated by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay ( Figure S6). To test probe 8 on live cells, a tailored, genetically bioorthogonalized cytoskeletal protein, vimentin, modified with a non-canonical, bioorthogonalized lysine (Lys( ε N-BCN endo ) was expressed in COS-7 cells, using the orthogonal pyrrolysine tRNA Pyl /PylRS pair from Methanosarcina mazei. The construct also carried a C-terminal mOrange fusion tag for reference (Vim N116TAG -mOrange).
Live COS-7 cells were then treated with probe 8, washed, fixed, and subjected to confocal microscopy. Images of the as-treated COS-7 cells showed good co-localization of the probe with the reference mOrange, suggesting specific labeling at the position of the Lys-BCN (Figure 2). This experiment also indicated that probe 8 is membrane permeable and can stand the conditions of live cell labeling. We were curious whether the fluorogenic nature of 8 allows no-wash labeling schemes in live-cell labeling and repeated the labeling process without the washing and fixation steps. To our delight, images of live cells showed specific, low background labeling of vimentin filaments (Figure 2). The spectral properties of 8 suggest that the probe is suitable for super-resolution imaging using STED microscopy with a 660 nm depletion laser (Supplementary Material, Figure S2). Therefore, we have subjected the samples to super-resolution imaging using STED microscopy ( Figure 3) and concluded that probe 8 is indeed suitable for STED microscopy. Again, we did not experience considerable change in the resolution between washed and non-washed samples. It should also be noted that no loss of signal-to-noise ratio was observed up to four cycles of STED imaging with~64% STED intensity. Further cycles, however, resulted in gradually fading images due to bleaching of the probe ( Figure S5).
Biomolecules 2020, 10, 397 9 of 12 noted that no loss of signal-to-noise ratio was observed up to four cycles of STED imaging with ~64% STED intensity. Further cycles, however, resulted in gradually fading images due to bleaching of the probe ( Figure S5).   noted that no loss of signal-to-noise ratio was observed up to four cycles of STED imaging with ~64% STED intensity. Further cycles, however, resulted in gradually fading images due to bleaching of the probe ( Figure S5).   Besides reducing autofluorescence, probes having large Stokes-shifts are also quite useful in terms of addressing a common problem present in multicolor labeling schemes and imaging. Lenses tend to fail to focus equally on all colors to the same points due to dispersion. This phenomenon, chromatic aberration [34], thus often compromises super-resolved images where more colors are used. Dyes that are excitable at distinct wavelengths but having similar emission windows can overcome this problem. Probe 8 seems ideal from this respect, as, due to its large Stokes-shift, it can be paired with other probes, having smaller Stokes-shift and emitting at around 600 nm. To demonstrate this, we applied a dual-color labeling scheme involving mitochondrial protein TOMM20 and actin filaments. Fixed HeLa cells were treated with anti-TOMM20 IgG and a secondary antibody tagged with 8, while actin filaments were stained with Cy3 [23] stained affinity tag, phalloidin. The two probes, 8 and Cy3, were excited at 488 nm and 552 nm, respectively, while detection was carried out at the same emission window. Images were also subjected to STED microscopy, giving rise to dual-color STED imaging ( Figure 4).
Biomolecules 2020, 10, 397 10 of 12 Besides reducing autofluorescence, probes having large Stokes-shifts are also quite useful in terms of addressing a common problem present in multicolor labeling schemes and imaging. Lenses tend to fail to focus equally on all colors to the same points due to dispersion. This phenomenon, chromatic aberration [34], thus often compromises super-resolved images where more colors are used. Dyes that are excitable at distinct wavelengths but having similar emission windows can overcome this problem. Probe 8 seems ideal from this respect, as, due to its large Stokes-shift, it can be paired with other probes, having smaller Stokes-shift and emitting at around 600 nm. To demonstrate this, we applied a dual-color labeling scheme involving mitochondrial protein TOMM20 and actin filaments. Fixed HeLa cells were treated with anti-TOMM20 IgG and a secondary antibody tagged with 8, while actin filaments were stained with Cy3 [23] stained affinity tag, phalloidin. The two probes, 8 and Cy3, were excited at 488 nm and 552 nm, respectively, while detection was carried out at the same emission window. Images were also subjected to STED microscopy, giving rise to dual-color STED imaging (Figure 4). . Confocal and STED microscopy images of IgG-8 labeled mitochondrial protein TOMM20 (magenta) and Cy3-phalloidin labeled actin filaments (cyan). Images were excited at 488 nm (8) and 552 nm (Cy3), while detected with the same detection window (580-650 nm). For STED images, a CW 660 nm depletion laser was used. Full-width at half maximum (FWHM) values for confocal (black) and STED images (red) were 224 ± 46 and 155 ± 10 nm for TOMM20 (upper-right) and 483 ± 76 and 124 ± 8 nm for actin (lower-right), respectively. Scale bar: 10 μm.

Conclusions
Herein, we presented the synthesis and application of a fluorogenic, large Stokes-shift, bioorthogonally applicable, membrane-permeable tetrazine probe. We studied the possibility of extending the π-electron system of ultra-fluorogenic HELIOS probes and concluded that the . Confocal and STED microscopy images of IgG-8 labeled mitochondrial protein TOMM20 (magenta) and Cy3-phalloidin labeled actin filaments (cyan). Images were excited at 488 nm (8) and 552 nm (Cy3), while detected with the same detection window (580-650 nm). For STED images, a CW 660 nm depletion laser was used. Full-width at half maximum (FWHM) values for confocal (black) and STED images (red) were 224 ± 46 and 155 ± 10 nm for TOMM20 (upper-right) and 483 ± 76 and 124 ± 8 nm for actin (lower-right), respectively. Scale bar: 10 µm.

Conclusions
Herein, we presented the synthesis and application of a fluorogenic, large Stokes-shift, bioorthogonally applicable, membrane-permeable tetrazine probe. We studied the possibility of extending the π-electron system of ultra-fluorogenic HELIOS probes and concluded that the quencher and bioorthogonal motif tetrazine can exert similar quenching of emission when it is placed at the 4th position of the coumarin frame. This finding enabled us to extend the conjugation of the tetrazine-coumarin via the 3rd position, giving rise to probe 8, excitable at common laser line 488 nm with emission range at around 600 nm. Cell labeling studies revealed that the probe is membrane permeable and can stand the challenging environment of live cells. Furthermore, the ca. 30-fold fluorogenicity allowed reduced background and consequently, no-wash imaging schemes, which is evidently advantageous from a practical perspective, but also, it might enable studies of proteins with rapid turnover times, such as many membrane receptors, nuclear proteins, or MAP kinases. The large Stokes-shift can also be exploited in multi-color labeling schemes and can address the problematics of chromatic aberration in combination with small Stokes-shift dyes emitting around 600 nm.