A synergistic strategy to develop photostable and bright dyes with long Stokes shift for nanoscopy

Super-resolution microscopy, particularly STED, demands fluorophores with high brightness, exceptional photostability, and long Stokes shifts to minimize re-excitation and improve signal-to-noise. Classic rhodamines are widely used but suffer from twisted intramolecular charge transfer (TICT) in the excited state, leading to nonradiative decay, photobleaching, and photobluing. Prior strategies either inhibit TICT (e.g., via ring-constrained or electron-withdrawing auxochromes) to boost brightness and photostability, or introduce vibronic structures to enlarge Stokes shift, but none address all three properties simultaneously. The authors hypothesize that combining vibronic structure with TICT inhibition in asymmetric rhodamines can synergistically and simultaneously enhance brightness, photostability, and Stokes shift, yielding superior probes for confocal and STED nanoscopy and enabling multicolor and live-cell applications.

• Rhodamines are popular for STED due to intrinsic brightness and photostability, but typical symmetric scaffolds yield small Stokes shifts (~20–30 nm), causing spectral overlap and self-quenching.
• TICT inhibition strategies: replace dialkylamino groups with rigid/less donating auxochromes (e.g., azetidine, aziridine, 7-azabicyclo[2.2.1]heptane) or use electron-withdrawing substituents (quaternary piperazine, sulfone-functionalized piperidine) to raise the TICT barrier and increase quantum yields and photostability.
• Vibronic-structure-based designs increase Stokes shifts markedly by promoting vibrational relaxation, but often compromise brightness in aqueous media.
• Large Stokes shift dyes for nanoscopy exist, yet achieving concurrent improvements in brightness, photostability, and Stokes shift has remained elusive prior to this work.

Design and synthesis:

• Designed asymmetric rhodamines (YL series) by introducing a quinoxaline motif bearing tunable electron-withdrawing substituents to create vibronic structure and lower electron density (to inhibit TICT). Dyes 1–7 synthesized via condensation of 2-(4-diethylamino-2-hydroxybenzoyl)benzoic acid with quinoxaline building blocks (S-1–S-6) prepared by nucleophilic substitution or amidation/reduction. Reaction conditions detailed (e.g., proline methyl ester mediated steps; acetonitrile or THF solvents; BF3·OEt2/NaBH4 reductions; methanesulfonic acid for final condensations).
• Varied substituents modulate electron-withdrawing strength; correlated Hammett constants with photophysical shifts.

Photophysical characterization:

• Measured absorption/emission spectra, quantum yields, extinction coefficients, fluorescence lifetimes, photostability under 530 nm irradiation, solvent dependence, and photobluing for 1–7 in aqueous buffers (PBS) and various solvents. Brightness defined as ε×Φ. Compared to benchmarks (RhB, JF549).
• Selected YL578 (dye 6; 2-(2,2,2-trifluoroethyl) octahydropyrrolo[1,2-a]pyrazine) as optimal compromise of brightness and Stokes shift.
• DFT calculations: assessed HOMO/LUMO distributions and potential energy surfaces to estimate TICT formation barriers; demonstrated asymmetric electron distribution and increased TICT barrier for YL578 relative to RhB.
• Additional analogs: YL-Az (azetidine replacing diethylamino) and symmetric bis-YL synthesized to probe roles of TICT inhibition and asymmetry on Stokes shift and brightness.

Probe development and cell imaging:

• Generated HaloTag ligands: YL578-Halo from carboxylated YL578 (compound 9); a fluorogenic variant 10-Halo via acyl 2,2,2-trifluoroethylamide formation (tuning spiro/zwitterion equilibrium; measured pK_cyl and D50). Assessed in vitro fluorogenicity upon HaloTag binding (absorbance and fluorescence turn-on).
• Prepared organelle probes: YL578-Mito (ester) and YL578-Lyso (amide) via one-step derivatizations; validated colocalization with commercial trackers.
• Live-cell confocal imaging in HeLa and U-2 OS: evaluated cell permeability, wash-free staining, nuclear labeling of H2B-Halo, photostability (continuous 560 nm irradiation), and background (Fnuc/Fcyt ratios).

STED microscopy benchmarking:

• Fixed and live-cell STED (775 nm depletion; also 595 nm for coumarin-derived probe) comparing YL578-Halo with photostable standards CPY-Halo, 580CP-Halo, and JF608-Halo. Quantified frame-to-bleach performance (frames retaining >50% initial intensity), FWHM resolution per frame, and maximal resolution under optimized settings. Demonstrated 3D STED (xzy scanning) of Tomm20-Halo.

Generality to other scaffolds and biosensing:

• Applied 2-(2,2,2-trifluoroethyl) octahydropyrrolo[1,2-a]pyrazine substitution to rhodol (11), pyronin (12), coumarin (13,14), and Boranil (15), measuring spectroscopic parameters (λab/λem, Φ, ε, brightness, Stokes shift) versus parental references (R-2–R-6).
• Developed a coumarin-derived Halo ligand (16-Halo) and demonstrated specific nuclear labeling and live-cell STED at 595 nm depletion.
• Created an ALP-responsive probe (11-ALP) by masking 11 with a phosphate group; evaluated in vitro turn-on, linear response (0–4 mU/mL), selectivity against ions and biomolecules, and cellular response in HeLa vs L02 and Na3VO4-treated controls.

Instrumentation and conditions:

• Spectroscopy on Shimadzu UV-1800 and Hitachi F-4600. Confocal on Nikon A1 plus. STED on Abberior Instruments systems with specified excitation and depletion lines. Detailed cell culture, transfection, labeling, fixation, and imaging parameters provided in Methods and Supplementary Information.

• Systematic EWG tuning in quinoxaline of asymmetric rhodamines (1–7) revealed a linear correlation of emission maxima with Hammett σ, and trade-offs among brightness, Stokes shift, and FWHM.
• YL578 (dye 6) performance in PBS: λab/λem = 578/634 nm, Stokes shift 56 nm; Φ = 0.74; ε = 89,700 M−1 cm−1; brightness (ε×Φ) ≈ 66,400, about 2× RhB’s brightness (RhB: Φ = 0.31; ε = 105,000; ε×Φ ≈ 32,500) and ~2× its Stokes shift (27 nm). Markedly higher photostability and reduced photobluing under 530 nm irradiation for 80 min than RhB and JF549.
• DFT indicates asymmetric HOMO distribution and increased TICT formation barrier for YL578, consistent with inhibited TICT and vibronic structure contributions.
• Cell imaging: YL578 stains cells rapidly with stronger fluorescence than RhB; negligible photobleaching under continuous 560 nm excitation versus RhB, comparable to JF549.
• HaloTag labeling: YL578-Halo yields bright nuclear labeling with lower cytoplasmic background vs RhB-Halo; Fnuc/Fcyt ≈ 18 without washing. Fluorogenic 10-Halo shows 23-fold absorbance and 490-fold fluorescence increases upon binding; affords very high contrast (Fnuc/Fcyt ≈ 106), albeit with some brightness reduction due to incomplete zwitterion recovery.
• Organelle probes YL578-Mito and YL578-Lyso provide fast, high-contrast, wash-free staining with good colocalization to commercial trackers.
• STED microscopy: Compared to 580CP-Halo, CPY-Halo, and JF608-Halo (providing only 2–3 frames >50% initial intensity; first-frame FWHM ≈ 116±6, 86±9, 83±10 nm), YL578-Halo delivers 9 frames >50% initial intensity with maintained resolution (FWHM 57±5 nm), and achieves 37±4 nm FWHM under optimized conditions. Enabled 3D STED of mitochondrial Tomm20-Halo with sequential xzy scans. Supported multicolor STED with a single 775 nm depletion beam.
• Generality: Substitution strategy improved multiple scaffolds. Rhodol (11): Φ 0.62, ε 51,000, brightness 31,620; λab/λem 548/612 nm, Stokes shift 64 nm (vs R-2: Φ 0.21, brightness 12,600, Stokes 29 nm). Pyronin (12): brightness 59,140; Stokes 46 nm. Coumarins and Boranil: 13 (Stokes 117 nm; brightness 17,820; 5.2× parent), 14 (Stokes 92 nm; brightness 17,550; 8.1× parent), 15 (Stokes 136 nm; brightness 10,000; 8.0× parent). 16-Halo exhibits large Stokes shift (110 nm) and specific nuclear labeling; supports live-cell STED with 595 nm depletion.
• Biosensing: 11-ALP shows strong turn-on upon ALP; linear response 0–4 mU/mL (R² = 0.989), high selectivity, and higher cellular signal in HeLa vs L02; Na3VO4 reduces signal consistent with ALP inhibition.

Combining vibronic structure with TICT inhibition in asymmetric rhodamines simultaneously enhances brightness, photostability, and Stokes shift, meeting stringent requirements of STED nanoscopy. Mechanistically, the quinoxaline-based 2-(2,2,2-trifluoroethyl) octahydropyrrolo[1,2-a]pyrazine motif lowers electron density to raise the TICT barrier while promoting vibrational relaxation that increases Stokes shift; DFT corroborates increased TICT barriers and asymmetric electronic distributions. This synergy translates into practical imaging gains: YL578-derived probes resist photobleaching under 775 nm STED, enabling substantially more frames and higher resolution, including 3D acquisitions typically limited by bleaching. The approach is general across diverse scaffolds, indicating that replacing dialkylamino groups with this motif is a broadly applicable route to high-performance, long–Stokes shift dyes. The ability to retain functional handles on the xanthene core facilitates creation of fluorogenic protein labels and enzyme-activated sensors, expanding utility in bioimaging and biosensing.

The study introduces a synergistic molecular design—combining vibronic structures with TICT inhibition via a quinoxaline-based 2-(2,2,2-trifluoroethyl) octahydropyrrolo[1,2-a]pyrazine motif—that converts classic fluorophores into bright, photostable, long–Stokes shift dyes. The lead dye YL578 doubles both brightness and Stokes shift relative to Rhodamine B and exhibits superior photostability in STED, enabling more frames, higher resolution, and 3D imaging. The strategy generalizes to multiple dye families (rhodol, pyronin, coumarin, Boranil), achieving up to 8-fold brightness increases and Stokes shifts up to 136 nm, and supports development of no-wash protein labels and selective enzymatic sensors. Future work could extend this substitution to additional fluorophore classes, optimize cellular uptake and targeting, tailor photophysics for specific STED wavelengths, and evaluate in vivo performance and long-term biocompatibility.

• While broadly effective, some design variants exhibit trade-offs: e.g., fluorogenic spirolactam probe 10 improves contrast and permeability but reduces brightness due to incomplete zwitterion recovery upon binding.
• Symmetric analog (bis-YL) increases brightness in organic solvent but shortens Stokes shift, underscoring dependence on asymmetry for large Stokes shift.
• Photophysical and imaging characterizations focus on selected cell lines and conditions; generalization to diverse biological environments and in vivo settings remains to be validated.
• Detailed long-term biocompatibility, metabolism, and potential transporter-mediated uptake/efflux effects were not exhaustively characterized.