Development of cyanine probes with dinitrobenzene quencher for rapid fluorogenic protein labelling

Abstract

A multicolour protein labelling technique using a protein tag and fluorogenic probes is a powerful approach for spatio-temporal analyses of proteins in living cells. Since cyanine fluorophores have attractive properties for multicolour imaging of proteins, there is a huge demand to develop fluorogenic cyanine probes for specific protein labelling in living cells. Herein, we develop fluorogenic cyanine probes for labelling a protein tag by using a dinitrobenzene fluorescence quencher. The probes enhanced fluorescence intensity upon labelling reactions and emitted orange or far-red fluorescence. Intramolecular interactions between the cyanine fluorophores and the dinitrobenzene quencher led not only to fluorescence quenching of the probes in the free state but also to promotion of labelling reactions. Furthermore, the probes successfully imaged cell-surface proteins without a washing process. These findings offer valuable information on the design of fluorogenic cyanine probes and indicate that the probes are useful as novel live-cell imaging tools.

This article is part of the themed issue ‘Challenges for chemistry in molecular imaging’.

1. Introduction

Multicolour protein labelling using synthetic fluorophores offers precise spatio-temporal information on subcellular localization and translocation of proteins in living cells [1–3]. To date, various labelling methods based on specific combinations of a protein tag and fluorescent probes have been reported [4–13]. There have been recent advances in this field in the development of fluorogenic probes, which are non-fluorescent in the free state and enhance fluorescence intensity in protein labelling reactions [3,10–16]. These probes enable quick high-contrast imaging of proteins without a washing process to remove non-fluorescent free probes. In multicolour imaging analyses using fluorogenic probes, it is necessary to use two or more fluorophores emitting fluorescence at different wavelengths. Among reported fluorophores, cyanine dyes have unique structural and optical features. Their fluorescence wavelengths are easily modulated by extending a methine chain so that cyanine dyes cover a broad range of fluorescence wavelengths from green to near-infrared regions [17]. Particularly, Cy3, Cy5 and their derivatives have been widely used in various studies including single-molecule and super-resolution imaging because of their high photostability or brightness [18]. Until now, fluorogenic cyanine probes for labelling the SNAP-tag have been created using the principle of Förster resonance energy transfer (FRET) [19]. FRET-type probes have an advantage in the availability of quenchers for various fluorophores. However, FRET quenchers are relatively large and cause a reduction in labelling kinetics probably due to steric hindrance [19,20]. Thus, a novel strategy is required to design fluorogenic cyanine probes.

We previously developed a photoactive yellow protein tag (PYP-tag) labelling system using fluorogenic probes [3,13,16]. This tag is a small protein (14 kDa) that originated in Halorhodospira halophila [21,22] and covalently bound to a coumarin/cinnamic acid ligand through transthioesterification with Cys69 [23]. One of the probes previously developed by our group, FCANB, consists of a modular structure with a 4-hydroxycinnamic acid ligand, a fluorophore, a linker and a nitrobenzene (NB) quencher [13]. The attractive feature of the modular scaffold is that various fluorophores emitting fluorescence at different wavelengths can be introduced into the probe. More importantly, the NB quencher effectively reduces the fluorescence intensity of the fluorophore through an intramolecular interaction between the quencher and the fluorophore. The advantage of the NB-type probes over FRET-type probes is that NB is smaller than FRET-type quenchers; thus, steric hindrance could be suppressed in labelling reactions. In this study, to overcome the current limitation in probe design based on FRET, we took advantage of the NB-type quencher. We developed fluorogenic cyanine probes emitting orange and far-red fluorescence by using an NB derivative, 3,5-dinitrobenzene, as a quencher (figure 1). These probes enhanced fluorescence intensity in labelling reactions with PYP-tag and quickly imaged cell-surface proteins without a washing process. Furthermore, this study shows that intramolecular interactions between the cyanine fluorophores and the dinitrobenzene quencher are important not only for a fluorogenic response but also for labelling kinetics. A list of acronyms used in this paper may be found in the electronic supplementary material.

Figure 1.

(a) Principle of fluorogenic PYP-tag labelling system using Cy5DNB2. (b) Structures of long-wavelength probes with fluorescence OFF–ON switches.

2. Experimental section

(a). Materials and instruments

General chemicals for synthesis were of the best grade available and supplied by Wako Pure Chemical Industries, Tokyo Chemical Industries, Sigma-Aldrich Chemical Co. and Kishida Chemical Co., and used without further purification. Plasmids, pcDNA-PYP-EGFR and pcDNA-EGFR, were constructed as described previously [13]. The recombinant protein (PYP-tag) was expressed and purified as described previously [3]. Compounds 1, 8, 11 and 14 were synthesized as reported previously [13], to which the reader is referred for details.

Nuclear magnetic resonance (NMR) spectra were measured with a JEOL JNM-AL400 spectrometer for ¹H-NMR (400 MHz) and ¹³C-NMR (100 MHz) or a Bruker AVANCE III 500 HD spectrometer for ¹H-NMR (500 MHz) and ¹³C-NMR (126 MHz) using tetramethylsilane as an internal standard. Electrospray ionization time-of-flight mass spectrometry (ESI-TOF MS) was carried out on a Waters LCT-Premier XE mass spectrometer. High-resolution mass spectra (HRMS) were obtained on a JEOL JMS-700 MStation mass spectrometer. Analytical thin-layer chromatography was performed on 60F254 silica plates (Merck) and visualized under UV light. Silica-gel column chromatography was performed using BW-300 (Fuji Silysia Chemical Ltd). High-performance liquid chromatography (HPLC) purification was carried out with a WP300 C4 column (4.6 or 10.0 mm × 250 mm, GL-Science, Inc.) or an Inertsil ODS-3 column (4.6 or 10.0 mm × 250 mm, GL-Science, Inc.). Gel permeation chromatography (GPC) purification was performed with JAIGEL-1H (Japan Analytical Industry). Fluorescence spectra were measured using a Nanolog (Horiba Scientific) or an F-7000 spectrometer (Hitachi) with a photomultiplier voltage of 400 or 700 V. UV–Vis absorption spectra were obtained with a V-650 spectrophotometer (Jasco). Fluorescence lifetime was measured using a fluorescence lifetime system TemPro (Horiba Scientific). Fluorescence images of gel were obtained using a Typhoon FLA 9500 (GE Healthcare). Fluorescence microscopic analyses were carried out using a confocal laser-scanning microscope (Olympus, FLUOVIEW FV10i) equipped with a 60× lens.

(b). Chemical syntheses

(i). Synthesis of Cy3NB (2)

Compound 1 (10.2 mg, 0.01 mmol) in DCM (5 ml) was mixed with TFA (0.5 ml), and the reaction was conducted for 2.5 h at room temperature. After removing the solvent, the residue was dissolved in DMF (1 ml) together with Cy3 (9.6 mg, 0.01 mmol), PyBOP (15.8 mg, 0.03 mmol), HOBt (4.1 mg, 0.03 mmol) and TEA (5 ml, 0.04 mmol) and stirred for 3.5 h at room temperature under an Ar atmosphere. After removing the solvent, the residue was purified by reversed-phase HPLC and eluted with H₂O/acetonitrile containing 0.1% formic acid using a C4 column to yield Cy3NB (2) as a red powder (3.8 mg; yield, 26%). ¹H-NMR (400 MHz, CD₃OD): δ 1.27–1.44 (m, 9H), 1.74 (s, 12H), 2.16 (t, J = 7.6 Hz, 2H), 3.13 (m, 2H), 3.46 (t, J = 5.6 Hz, 2H), 3.57–3.68 (m, 10H), 4.08 (t, J = 8.0 Hz, 2H), 4.17 (q, J = 4.4 Hz, 2H), 4.47 (s, 2H), 4.61 (s, 2H), 6.45 (d, J = 13.6 Hz, 1H), 6.48 (d, J = 13.6 Hz, 1H), 6.70 (d, J = 15.6 Hz, 1H), 6.72 (d, J = 15.6 Hz, 1H), 6.84 (d, J = 2.4 Hz, 1H), 7.33–7.39 (m, 10H), 7.61 (d, J = 8.4 Hz, 1H), 7.88–7.94 (m, 4H), 7.97 (d, J = 15.6 Hz, 1H), 8.15 (d, J = 8.4 Hz, 2H), 8.35 (t, J = 13.6 Hz, 1H); ¹³C-NMR (100 MHz, DMSO): δ 12.7, 24.2, 26.3, 27.9, 34.7, 36.7, 37.8, 42.4, 43.9, 45.0, 46.1, 47.0, 56.6, 57.5, 62.6, 68.8, 69.8, 70.0, 70.7, 103.2, 103.4, 110.9, 111.2, 115.9, 117.0, 117.6, 120.3, 120.4, 122.1, 123.7, 126.0, 126.8, 127.9, 129.6, 130.4, 130.5, 134.9, 138.1, 139.1, 140.6, 140.7, 141.1, 142.0, 142.4, 143.9, 144.8, 146.1, 148.5, 150.5, 160.5, 170.9, 172.6, 174.3, 174.6, 187.5; HRMS (FAB⁺) m/z: calcd for [M + Na]⁺ 1244.3936, found 1244.3959.

(ii). Synthesis of Cy5NB (3)

Compound 1 (8.2 mg, 0.01 mmol) in DCM (10 ml) was mixed with TFA (0.5 ml), and the reaction was conducted for 30 min at room temperature. After removing the solvent, the residue was added to DMF (500 ml) together with Cy5-NHS (5.0 mg, 0.01 mmol) and TEA (5 ml) and the reaction was conducted for 3.5 h at room temperature under an Ar atmosphere. After removing the solvent, the residue was purified by reversed-phase HPLC and eluted with H₂O/acetonitrile containing 0.1% formic acid using a C4 column to yield Cy5NB (3) as a blue powder (3.8 mg; yield, 26%). ¹H-NMR (400 MHz, CD₃OD): δ 1.19–1.29 (m, 9H), 1.61 (s, 12H), 2.05 (t, J = 7.6 Hz, 2H), 3.12 (m, 2H), 3.37 (t, J = 5.6 Hz, 2H), 3.46–3.58 (m, 10H), 3.95 (t, J = 8.0 Hz, 2H), 4.01 (q, J = 4.4 Hz, 2H), 4.38 (s, 2H), 4.52 (s, 2H), 6.17 (m, 2H), 6.50 (m, 2H), 6.66 (d, J = 15.6 Hz, 1H), 6.70 (d, J = 15.6 Hz, 1H), 6.75 (d, J = 2.4 Hz, 1H), 7.17–7.32 (m, 10H), 7.54 (d, J = 8.4 Hz, 1H), 7.77–7.79 (m, 4H), 7.96 (d, J = 15.6 Hz, 1H), 8.06 (d, J = 8.4 Hz, 2H), 8.19 (t, J = 13.6 Hz, 1H); ¹³C-NMR (100 MHz, DMSO): δ 12.57, 24.3, 26.1, 27.4, 34.6, 35.6, 37.8, 42.9, 43.9, 45.0, 46.0, 47.0, 54.4, 57.5, 62.6, 69.6, 69.7, 70.0, 70.2, 103.6, 103.9, 110.4, 110.7, 115.9, 116.9, 120.4, 120.4, 122.0, 123.6, 123.9, 125.8, 126.3, 126.6, 128.6, 129.7, 130.4, 134.9, 138.2, 139.1, 140.9, 141.0, 141.1, 142.0, 142.5, 145.5, 145.6, 146.9, 147.1, 148.1, 154.7, 154.8, 160.5, 170.6, 172.5, 173.1, 173.4, 187.6; HRMS (MALDI) m/z: calcd for [M]⁺ 1246.4193, found 1246.4182.

(iii). Synthesis of N-(4-nitrobenzyl)acetamide (4)

Acetic anhydride (420 ml, 4.45 mmol), (4-nitrophenyl)methanamine (201 mg, 1.06 mmol) and caesium carbonate (1036 mg, 3.18 mmol) were dissolved in DMF (10 ml) and the reaction was conducted for 5 h at room temperature. After removing the solvent, the residue was mixed with ethylacetate, washed with aqueous NaHCO₃ and dried with Na₂SO₄. By evaporating the solvent, compound 4 was acquired as a solid with pale brown colour (201 mg; yield, 98%). ¹H-NMR (400 MHz, CD₃CN-d₆): δ 4.02 (m, 2H), 8.57 (d, J = 2.0 Hz, 2H), 8.76 (t, J = 2.0 Hz, 1H); ¹³C-NMR (100 MHz, DMSO): δ 23.0, 42.1, 124.0, 128.6, 146.9, 148.3, 170.0; HRMS (EI⁺) m/z: calcd for [M]⁺ 194.0691, found 194.0692.

(iv). Synthesis of 5-acetamido-2-nitrobenzoic acid (5)

Acetic anhydride (454 ml, 4.61 mmol), 5-amino-2-nitrobenzoic acid (202 mg, 1.09 mmol) and caesium carbonate (873 mg, 2.68 mmol) were dissolved in DMF (10 ml) and stirred for 9.5 h at room temperature. The solvent was removed and the residue was added to ethylacetate, washed with 10% citric acid and water, and treated with Na₂SO₄. After removal of solvent by evaporation, purification was conducted by silica-gel column chromatography using hexane/ethylacetate/acetic acid (27/70/3) to yield compound 5 as a light yellow solid (214 mg; yield, 88%). ¹H-NMR (400 MHz, DMSO-d₆): δ 2.12 (s, 3H), 7.83 (dd, J = 8.8 Hz, J = 2.4 Hz, 1H), 7.97 (d, J = 2.4 Hz, 1H), 8.03 (d, J = 8.8 Hz, 1H); ¹³C-NMR (100 MHz, DMSO): δ 21.5, 24.69, 118.5, 120.2, 126.1, 141.7, 144.3, 167.1, 169.9; HRMS (EI⁺) m/z: calcd for [M]⁺ 224.0433, found 224.0435.

(v). Synthesis of N-(2,4-dinitrophenyl)acetamide (6)

Acetic anhydride (435 ml, 4.48 mmol), 2,4-dinitroaniline (200 mg, 1.09 mmol) and caesium carbonate (1071 mg, 3.29 mmol) were dissolved in DMF (10 ml) and stirred for 5 h at room temperature. After removing the solvent, the residue was mixed with ethylacetate, washed with aqueous NaHCO₃ and dried with Na₂SO₄. Following evaporation of the solvent, compound 6 was acquired as a solid with brown colour (228 mg; yield, 93%). ¹H-NMR (400 MHz, DMSO-d₆): δ 2.15 (s, 3H), 8.57 (d, J = 8.8 Hz, 1H), 8.52 (dd, J = 8.8 Hz, J = 2.8 Hz, 1H), 8.86 (d, J_ab = 2.8 Hz, 1H); ¹³C-NMR (100 MHz, DMSO): δ 24.2, 41.7, 121.6, 125.3, 129.0, 137.3, 141.2, 142.9, 169.5; HRMS (EI⁺) m/z: calcd for [M]⁺ 225.0386, found 225.0386.

(vi). Synthesis of N-(3,5-dinitrobenzyl)acetamide (7)

Acetic anhydride (60.4 ml, 0.64 mmol), (3,5-dinitrophenyl)methanamine (30.0 mg, 0.20 mmol) and caesium carbonate (148.8 mg, 0.46 mmol) were reacted in DMF (5 ml) for 75 min at room temperature. After removing the solvent, the residue was mixed with ethylacetate, washed with aqueous NaHCO₃ and treated with Na₂SO₄. Following removal of the solvent, purification was conducted by silica-gel column chromatography using hexane/ethylacetate (20/80) to yield compound 7 as a solid with pale brown colour (23.3 mg; yield, 64%). ¹H-NMR (400 MHz, CD₃CN-d₆): δ 1.92 (s, 3H), 4.49 (d, J = 6.4 Hz, 2H), 8.65 (s, 1H), 8.52 (d, J = 2.0 Hz, 2H), 8.72 (t, J = 2.0 Hz, 1H); ¹³C-NMR (100 MHz, DMSO): δ 23.0, 117.5, 128.1, 145.0, 148.5, 170.3; HRMS (CI⁺) m/z: calcd for [M + H]⁺ 240.0615, found 240.0621.

(vii). Synthesis of N-(3,5-dinitrobenzyl)-2-(4-(tritylthio)phenyl)acetamide (9)

Compound 8 (300 mg, 0.73 mmol), (3,5-dinitrophenyl)methanamine (264 mg, 0.64 mmol), DMAP (173 mg, 1.42 mmol) and WSCD/HCl (148 mg, 0.77 mmol) were dissolved in DMF (20 ml) and the reaction was conducted for 8 h at room temperature. After removing the DMF, the residue was mixed with ethylacetate, washed with water and treated using Na₂SO₄. Following removal of the solvent, purification was conducted by silica-gel column chromatography using DCM/methanol (97/3) to yield compound 9 as a colourless solid (287 mg; yield, 76%). ¹H-NMR (400 MHz, acetone-d₆): δ 3.51 (s, 2H), 4.63 (d, J = 6.0 Hz, 2H), 6.90 (d, J = 8.0 Hz, 2H), 7.05 (d, J = 8.0 Hz, 2H, g), 7.19–7.40 (m, 15H), 7.94 (s, 1H), 8.52 (d, J = 2.0 Hz, 2H), 8.80 (t, J = 2.0 Hz, 1H); ¹³C-NMR (100 MHz, acetone-d₆): δ 42.7, 43.0, 71.4, 117.9, 127.6, 128.4, 128.5, 129.9, 130.7, 133.2, 135.5, 136.8, 145.4, 145.6, 149.2, 171.3; HRMS (FAB⁺) m/z: calcd for [M + H]⁺ 590.1774, found 590.1753.

(viii). Synthesis of N-(3,5-dinitrobenzyl)-2-(4-mercaptophenyl)acetamide (10)

Compound 9 (264 mg, 0.45 mmol), TIS (2.5 ml) and TFA (1.0 ml) were dissolved in DCM (10 ml) and the reaction was conducted for 2 h at room temperature. After removing the solvent, the residue was mixed with ethylacetate, and washed with 10% citric acid and water, and then the mixture was dried with Na₂SO₄. Following evaporation of the solvent, purification was conducted by reversed-phase HPLC using a C18 column and H₂O/acetonitrile containing 0.1% formic acid to yield compound 10 as a colourless solid (1041 mg; yield, 67%). ¹H-NMR (400 MHz, acetone-d₆): δ 3.58 (s, 2H), 4.70 (d, J = 6.0 Hz, 2H), 7.25 (m, 4H), 7.98 (s, 1H), 8.55 (d, J = 2.0 Hz, 2H), 8.80 (t, J = 2.0 Hz, 1H); ¹³C-NMR (100 MHz, DMSO-d₆): δ 41.2, 41.6, 117.1, 127.4, 128.5, 129.7, 130.1, 132.7, 144.4, 150.0, 170.8; HRMS (EI⁺) m/z: calcd for [M + H]⁺ 348.0649, found 348.0654.

(ix). Synthesis of tert-butyl (4-((2-bromo-5-(methoxymethoxy)benzyl)oxy)butyl)carbamate (12)

Sodium hydride (46.6 mg, 1.9 mmol) was added to DMF (30 ml) and stirred for 1 h at 0°C, and then tert-butyl (4-hydroxybutyl)carbamate (366 mg, 1.9 mmol) was added to the solution. Compound 11 (500 mg, 1.6 mmol) dissolved in DMF (30 ml) was dropped into the mixture and reacted for 7 h at 0°C. After removing the solvent, the mixture was mixed with water and ethylacetate. The extract from the organic layer was mixed with brine and treated with Na₂SO₄. After evaporation of the solvent, purification was conducted by silica-gel column chromatography using hexane/ethylacetate (80/20) to yield compound 12 as a colourless oil (270 mg; yield, 40%). ¹H-NMR (400 MHz, CDCl₃): δ 1.39 (s, 9H), 1.55–1.71 (m, 4H), 3.09–3.12 (m, 2H), 3.43 (s, 3H), 3.60 (t, J = 6.0 Hz, 2H), 4.50 (s, 2H), 5.20 (s, 2H), 5.95 (s, 2H), 6.92 (dd, J = 8.0 Hz, J = 3.2 Hz, 1H), 7.22 (d, J = 3.2 Hz, 1H), 7.47 (d, J = 8.0 Hz, 1H); ¹³C-NMR (100 MHz, CDCl₃): δ 27.6, 27.7, 40.8, 56.1, 71.2, 72.3, 78.2, 95.1, 114.0, 117.5, 117.6, 133.8, 140.2, 156.6, 157.3; HRMS (FAB⁺) m/z: calcd for [M + H]⁺ 418.1224, found 418.1225.

(x). Synthesis of (E)-methyl-3-(2-((4-((tert-butoxycarbonyl)amino)butoxy)methyl)-4-(methoxymethoxy) phenyl)acrylate (13)

Compound 12 (250 mg, 0.6 mmol), triphenylphosphine (31.5 mg, 0.1 mmol), methylacrylate (258 mg, 3.0 mmol), palladium(II) acetate (13.5 mg, 0.10 mmol), potassium fluoride (69.7 mg, 1.2 mmol) and TEA (400 ml) were dissolved in DMF (20 ml) and microwaved for 1 h (170°C, 150 W). After removing the solvent, the residue was mixed with water and ethylacetate. The extract from the organic layer was dried using Na₂SO₄ and the removal of the remaining solvent was conducted by evaporation. Purification was conducted by silica-gel column chromatography using hexane/ethylacetate (70/30) to yield compound 13 as a colourless oil (43.2 mg; yield, 17%). ¹H-NMR (400 MHz, ACETN-d₆): δ 1.39 (s, 9H), 1.55–1.65 (m, 4H), 3.06–3.11 (m, 2H), 3.43 (s, 3H), 3.56 (t, J = 5.6 Hz, 2H), 3.75 (s, 3H), 4.61 (s, 2H), 5.26 (s, 2H), 5.91 (s, 2H), 6.38 (d, J = 16 Hz, 1H), 7.03 (dd, J = 8.0, J = 3.2 Hz, 1H), 7.12 (d, J = 3.2 Hz, 1H), 7.74 (d, J = 8.0 Hz, 1H), 7.97 (d, J = 16 Hz, 1H); ¹³C-NMR (100 MHz, ACETN-d₆): δ 18.0, 27.7, 28.6, 37.9, 40.7, 56.4, 70.7, 71.4, 95.0, 116.4, 117.7, 117.8, 129.0, 138.8, 140.7, 142.3, 152.5, 159.6, 177.2, 193.1; HRMS (FAB⁺) m/z: calcd for [M + H]⁺ 424.2330, found 424.2338.

(xi). Synthesis of (E)-S-(4-(2-((3,5-dinitrobenzyl)amino)-2-oxoethyl)phenyl)-3-(2-(14,14-dimethyl-12-oxo-2,5,8,13-tetraoxa-11-azapentadecyl)-4-(methoxymethoxy)phenyl)prop-2-enethioate (15)

Compound 14 (210.3 mg, 0.44 mmol) was added to DMF (2.0 ml) and mixed together with 2 N aqueous NaOH (1.0 ml) for 4.5 h at room temperature. Following the reaction, the mixture was acidified with 2 N aqueous HCl. The aqueous layer was diluted with water and the reaction product was extracted with ethylacetate. The solvent was removed and the residue (65.8 mg, 0.14 mmol) was added to DCM (5 ml). HBTU (63.8 mg, 0.16 mmol) and TEA (50 µl) were mixed with the reaction mixture and the reaction was conducted for 1 h at room temperature under an Ar atmosphere. Then, the compound 10 was added to the reaction mixture and stirred for 4 h at room temperature. After removing the solvent, the residue was diluted with chloroform and was purified by GPC using JAIGEL to yield compound 15 as a colourless solid (91.7 mg; yield, 82%). ¹H-NMR (400 MHz, ACETN-d₆): δ 1.38 (s, 3H), 3.17 (s, J = 6 Hz, 2H), 3.44–3.71 (m, 15H), 4.69 (s, 2H), 4.71 (s, 2H), 5.28 (s, 2H), 5.80 (s, 1H), 6.82 (d, J = 16 Hz, 1H), 7.06 (dd, J = 1.6 Hz, J = 6.0 Hz, 1H), 7.16 (d, J = 1.6 Hz, 1H), 7.46 (m, 4H), 7.82 (d, J = 6.0 Hz, 1H), 8.00 (d, J = 16 Hz, 1H), 8.13 (s, 1H), 8.57 (d, J = 2.0 Hz, 2H), 8.81 (t, J = 2.0 Hz, 1H); ¹³C-NMR (100 MHz, ACETN-d₆): δ 28.6, 41.0, 42.8, 43.3, 56.2, 70.6, 70.7, 70.9, 71.2, 71.3, 71.5, 94.8, 116.5, 117.8, 117.9, 124.0, 127.1, 127.2, 128.4, 129.5, 130.1, 132.1, 135.4, 135.5, 138.3, 139.1, 141.4, 145.6, 149.4, 160.1, 171.3, 187.6; HRMS (FAB⁺) m/z: calcd for [M + Na]⁺ 821.2674, found 821.2685.

(xii). Synthesis of (E)-S-(4-(2-((3,5-dinitrobenzyl)amino)-2-oxoethyl)phenyl)-3-(2-((4-((tert- butoxycarbonyl)amino)butoxy)methyl)-4-(methoxymethoxy)phenyl)prop-2-enethioate (16)

Compound 13 (43.2 mg, 0.1 mmol) was dissolved in DMF (5.0 ml) and mixed together with 2 N aqueous NaOH (200 ml) for 3 h at room temperature. Following the reaction, the mixture was acidified with 2 N aqueous HCl. The aqueous layer was diluted with water and the reaction product was extracted with ethylacetate. The solvent was removed and the residue (65.8 mg, 0.14 mmol) was added to DCM (20 ml). HBTU (40.8 mg, 0.1 mmol) and TEA (50 µl) were mixed with the reaction mixture and the reaction was conducted for 1 h at room temperature. Then, the compound 10 was mixed with the reaction mixture and the reaction was conducted for 1.5 h at room temperature under an Ar atmosphere. After removing the solvent, the residue was mixed with chloroform, and was purified by GPC using JAIGEL to yield compound 16 as a colourless solid (21.2 mg; yield, 32%). ¹H-NMR (400 MHz, ACETN-d₆): δ 1.37 (s, 9H), 1.54–1.66 (m, 4H), 3.06–3.11 (m, 2H), 3.45 (s, 3H), 3.57 (t, J = 5.6 Hz, 2H), 3.72 (s, 2H), 4.56–4.70 (m, 4H), 5.26 (s, 2H), 5.91 (s, 1H), 6.80 (d, J = 16 Hz, 1H), 7.03 (dd, J = 8.0, J = 3.2 Hz, 1H), 7.12 (d, J = 3.2 Hz, 1H), 7.32–7.45 (m, 4H), 7.82 (d, J = 8.8 Hz, 2H), 7.98 (d, J = 16 Hz, 1H), 8.15 (s, 1H), 8.57 (m, 2H), 8.80 (t, J = 2.0 Hz, 1H); ¹³C-NMR (100 MHz, ACETN-d₆): δ 27.6, 27.7, 28.6, 40.7, 42.8, 43.2, 56.2, 70.9, 71.3, 79.3, 94.8, 116.5, 117.8, 117.9, 124.0, 127.2, 127.2, 128.4, 129.4, 130.8, 135.2, 135.5, 138.3, 139.1, 141.5, 145.5, 149.5, 160.0, 171.4, 187.6; HRMS (FAB⁺) m/z: calcd for [M + H]⁺ 739.2644, found 739.2650.

(xiii). Synthesis of Cy3DNB (17)

Compound 15 (91.7 mg, 0.12 mmol) in DCM (30 ml) was mixed with TFA (0.5 ml), and the reaction was conducted for 2 h at room temperature. After removing the solvent, the residue (17.7 mg, 0.03 mmol) was added to DMF (15 ml) together with Cy3 (21.0 mg, 0.03 mmol), PyBOP (28.1 mg, 0.05 mmol), HOBt (7.3 mg, 0.05 mmol) and TEA (11 µl, 0.08 mmol), and stirred for 1 h at room temperature under an Ar atmosphere. After removing the solvent, the residue was purified by reversed-phase HPLC and eluted with H₂O/acetonitrile containing 0.1% formic acid using a C4 column to yield Cy3DNB (17) as a red powder (15.8 mg; yield, 46%). ¹H-NMR (400 MHz, DMSO-d₆): δ 1.28–1.53 (m, 9H), 1.68 (s, 12H), 2.04 (t, J = 7.6 Hz, 2H), 3.13 (m, 2H), 3.33 (t, J = 5.6 Hz, 2H), 3.54–3.60 (m, 10H), 4.07 (q, J = 4.4 Hz, 2H), 4.14 (d, J = 8.0 Hz, 2H), 4.52 (s, 2H), 4.55 (s, 2H), 6.48–6.53 (m, 2H), 6.73–6.82 (m, 2H), 7.36–7.39 (m, 7H), 7.65–7.83 (m, 6H), 8.33 (t, J = 13.6 Hz, 1H), 8.49 (d, J = 2.0 Hz, 2H), 8.68 (s, 1H), 8.94 (t, J = 2.0 Hz, 1H); ¹³C-NMR (100 MHz, ACETN-d₆): δ 12.3, 24.9, 25.7, 26.7, 27.3, 27.4, 35.0, 41.8, 48.9, 49.0, 66.7, 69.1, 69.2, 69.5, 69.6, 69.7, 70.2, 102.8, 102.9, 110.5, 110.7, 115.3, 116.4, 117.1, 120.0, 121.5, 123.1, 125.5, 126.3, 127.5, 129.2, 129.9, 130.0, 134.5, 134.7, 137.6, 138.5, 140.1, 140.2, 140.5, 141.4, 141.9, 144.3, 145.6, 147.9, 149.9, 160.0, 170.4, 172.0, 173.8, 174.1, 186.9; HRMS (FAB⁻) m/z: calcd for [M − H]⁻ 1265.3960, found 1265.3905.

(xiv). Synthesis of Cy5DNB (18)

Compound 15 (91.7 mg, 0.12 mmol) in DCM (30 ml) was mixed with TFA (0.5 ml), and the reaction was conducted for 30 min at room temperature. After removing the solvent, the deprotected product (9.9 mg, 0.02 mmol), Cy5-NHS (10 mg, 0.01 mmol) and TEA (2 ml, 0.1 mmol) were dissolved in DMF and stirred for 3.5 h at room temperature under an Ar atmosphere. The solvent was removed and purification was conducted by reversed-phase HPLC using a C4 column and H₂O/acetonitrile containing 0.1% formic acid to yield Cy5DNB (18) as a blue powder (4.62 mg; yield, 28%). ¹H-NMR (400 MHz, DMSO-d₆): δ 1.18–1.44 (m, 9H), 1.61 (s, 12H), 1.90–2.11 (m, 2H), 3.07 (s, 2H), 3.48–3.60 (m, 12H), 4.07 (q, J = 4.4 Hz, 2H), 4.14 (d, J = 8.0 Hz, 2H), 4.52 (s, 2H), 4.55 (s, 2H), 6.20–6.23 (m, 2H), 6.41 (s, 1H), 6.70 (d, J = 15.6 Hz, 1H), 6.65–6.67 (m, 1H), 6.75 (d, J = 2.4 Hz, 1H) 7.36–7.39 (m, 6H), 7.65–7.85 (m, 7H), 8.33 (t, J = 13.6 Hz, 1H), 8.50 (d, J = 2.0 Hz, 2H), 8.70 (t, 1H), 8.87 (t, J = 2.0 Hz, 1H); ¹³C-NMR (100 MHz, DMSO-d₆): δ 12.1, 24.8, 25.6, 26.3, 26.9, 37.4, 38.4, 40.8, 41.3, 45.7, 48.8, 48.9, 65.5, 69.1, 69.2, 69.5, 69.6, 69.7, 70.2, 103.1, 103.4, 107.8, 109.9, 110.2, 113.0, 115.3, 117.1, 120.0, 120.7, 121.5, 123.0, 125.5, 125.7, 126.1, 127.5, 129.2, 129.9, 134.4, 137.6, 138.6, 140.4, 140.5, 140.6, 142.0, 144.3, 148.0, 149.4, 152.6, 154.3, 160.0, 170.4, 172.0, 172.6, 173.0, 186.9; HRMS (MALDI); calcd for [M]⁻ 1291.4043, found 1291.4072.

(xv). Synthesis of Cy3DNB2 (19)

Compound 16 (21.2 mg, 0.03 mmol) in DCM (5 ml) was mixed with TFA (0.5 ml), and the reaction was conducted for 10 min at room temperature. After removing the solvent, the deprotected product (8.55 mg, 0.01 mmol) was added to DMF (2 ml) together with Cy3 (10.9 mg, 0.02 mmol), PyBOP (15.0 mg, 0.03 mmol), HOBt (3.9 mg, 0.03 mmol) and TEA (43 ml, 0.04 mmol), and stirred for 4 h at room temperature under an Ar atmosphere. After removing the solvent, purification was conducted by reversed-phase HPLC using H₂O/acetonitrile containing 0.1% formic acid and a C4 column to yield Cy3DNB2 (19) as a red solid (7.38 mg; yield, 42%). ¹H-NMR (400 MHz, DMSO-d₆): δ 1.29–1.40 (m, 7H), 1.49–1.56 (m, 4H), 1.69 (s, 12H), 2.02 (t, J = 7.2 Hz, 2H), 3.00 (t, J = 5.6 Hz, 2H), 3.41–3.56 (m, 4H), 4.07 (t, J = 5.6 Hz, 2H), 4.14 (d, J = 5.6 Hz, 2H), 4.52 (s, 2H), 4.55 (s, 2H), 6.47 (d, J = 13.6 Hz, 1H), 6.50 (d, J = 13.6 Hz, 1H), 6.76–6.82 (m, 2H), 7.36–7.40 (m, 5H), 7.66–7.84 (m, 7H), 8.33 (t, J = 13.6 Hz, 1H), 8.51 (d, J = 2.0 Hz, 2H), 8.80 (s, 1H), 8.96 (t, J = 2.0 Hz, 1H); ¹³C-NMR (100 MHz, ACETN-d₆): δ 12.1, 24.9, 25.7, 25.8, 26.0, 26.7, 27.3, 27.4, 35.2, 38.2, 47.3, 48.9, 49.0, 55.6, 56.7, 69.5, 70.1, 102.9, 109.1, 110.8, 115.4, 116.3, 116.7, 117.0, 120.0, 123.2, 125.5, 126.3, 127.5, 129.0, 129.2, 130.0, 134.4, 134.5, 137.5, 138.7, 140.1, 140.7, 141.4, 141.8, 141.9, 144.4, 145.7, 145.8, 148.0, 154.9, 160.2, 164.8, 170.3, 171.9, 174.0, 174.1; HRMS (FAB⁺) m/z: calcd for [M + Na]⁺ 1229.3640, found 1229.3638.

(xvi). Synthesis of Cy5DNB2 (20)

Compound 16 (21.2 mg, 0.03 mmol) in DCM (5 ml) was mixed with TFA (0.5 ml), and the reaction was conducted for 10 min at room temperature. After removing the solvent, the deprotected product (8.55 mg, 0.01 mmol), Cy5-NHS (11.4 mg, 0.01 mmol) and TEA (43 ml, 0.04 mmol) were dissolved in DMF (2 ml) and the reaction was carried out for 3.5 h at room temperature under an Ar atmosphere. After removing the solvent, purification was conducted by reversed-phase HPLC using H₂O/acetonitrile containing 0.1% formic acid and a C4 column to yield Cy5DNB2 (20) as a blue powder (7.45 mg; yield, 41%). ¹H-NMR (400 MHz, DMSO-d₆): δ 1.21–1.51 (m, 11H), 1.69 (s, 12H), 1.98 (t, J = 7.2 Hz, 2H), 3.00 (t, J = 5.6 Hz, 2H), 3.42 (t, J = 5.6 Hz, 2H), 3.58 (m, 2H), 4.03 (t, J = 5.6 Hz, 2H), 4.15 (d, J = 5.6 Hz, 2H), 4.53 (s, 2H), 4.56 (s, 2H), 6.25 (d, J = 13.6 Hz, 1H), 6.28 (d, J = 13.6 Hz, 1H), 6.55 (t, J = 13.6 Hz, 1H), 6.70–6.80 (m, 2H), 7.27–7.37 (m, 6H), 7.61–7.83 (m, 7H), 8.33 (t, J = 13.6 Hz, 1H), 8.49 (d, J = 2.0 Hz, 2H), 8.68 (s, 1H), 8.96 (t, J = 2.0 Hz, 1H); ¹³C-NMR (100 MHz, ACETN-d₆): δ 12.0, 24.8, 25.7, 26.0, 26.5, 27.0, 27.1, 28.9, 35.2, 38.1, 45.4, 48.5, 48.9, 61.1, 62.7, 69.4, 70.0, 103.5, 109.4, 110.1, 114.0, 114.4, 115.2, 115.3, 116.4, 117.1, 119.7, 121.5, 123.2, 124.0, 125.5, 126.1, 127.5, 129.2, 130.0, 131.6, 132.2, 134.4, 136.7, 137.6, 138.6, 140.7, 141.6, 144.3, 145.2, 148.0, 154.2, 156.6, 160.0, 163.2, 170.5, 171.8, 172.7, 175.7, 183.1; HRMS (FAB⁺) m/z: calcd for [M]⁺ 1233.3978, found 1233.3927.

(c). Protein labelling reactions in vitro

PYP-tag (10 µM) was reacted with each probe (8 µM) in the assay buffer (20 mM HEPES buffer (pH 7.4) containing 150 mM NaCl) at 37°C for 30 min. The reaction mixtures were incubated at 95°C for 5 min, and subsequently analysed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). After a fluorescence image was obtained by Typhoon FLA 9500 with excitation at 532 nm (Cy3NB, Cy3DNB and Cy3DNB2) or 635 nm (Cy5NB, Cy5DNB and Cy5DNB2), the gel was stained with Coomassie Brilliant Blue.

(d). Fluorescence measurements

Fluorescence measurements were conducted after each probe (5 µM) was incubated with or without PYP-tag (6 µM) in the assay buffer at 37°C for 3 h. Fluorescence spectra were recorded at the excitation wavelength of 550 nm (Cy3NB, Cy3DNB and Cy3DNB2) or 649 nm (Cy5NB, Cy5DNB and Cy5DNB2). Relative fluorescence quantum yields of the probes in the assay buffer were determined using Cy3 or Cy5 as a reference [24]. The second-order rate constants, k₂, of the labelling reactions of PYP-tag (8, 16, 24, 32 and 40 µM) with each probe (500 nM) were determined as described previously [13].

(e). Live-cell imaging

HEK293T cells were transfected with pcDNA-EGFR or pcDNA-PYP-EGFR by using Lipofectamine 3000 (Invitrogen), according to the manufacturer's protocol. The transfected cells were incubated with each probe (1.0 µM) in DMEM at 37°C for 30 min. After the labelling reaction, the unreacted probe was removed by washing the cells with HBSS three times. Subsequently, imaging experiments were conducted using a confocal laser-scanning microscope. For no-wash imaging, the washing procedure was omitted. Microscopic images were obtained using a confocal laser-scanning microscope with excitation at 559 nm (Cy3NB, Cy3DNB and Cy3DNB2) or 635 nm (Cy5NB, Cy5DNB and Cy5DNB2) and a 570–670 nm or a 660–760 nm emission filter.

3. Results and discussion

(a). Probe design and synthesis

As long-wavelength fluorophores, we chose sulfonated cyanine dyes, Cy3 and Cy5, which emit fluorescence at the maximum wavelengths of 570 nm and 670 nm, respectively. We connected each of the fluorophores and an NB quencher with a PYP-tag ligand to create NB-type probes, Cy3NB and Cy5NB, respectively (figure 1; see electronic supplementary material, scheme S1). SDS-PAGE analyses showed that these probes covalently bind to PYP-tag (see electronic supplementary material, figure S1a,c). The labelling reactions led to increase in the fluorescence intensity of the probes (figure 2a,d). Both probes, however, exhibited low fluorescence OFF–ON ratios and considerable fluorescence in the absence of PYP-tag. These results indicate that the quenching ability of the NB quencher was not sufficient. Thus, we explored NB derivatives that efficiently quench Cy3 and Cy5. The intriguing feature of the NB derivatives is that they have a fluorescence quenching effect on various fluorophores [3,13,25–27]. The quenching experiments were conducted by recording the fluorescence spectra of each fluorophore in the presence or absence of excess NB derivatives. In these experiments, nitrobenzoic acid and two isomers of dinitrobenzene were analysed to verify their fluorescence quenching efficiency (see electronic supplementary material, scheme S2 and figure S2). As a result, all of the compounds quenched the fluorescence of both fluorophores when their concentrations were increased. Among them, 3,5-(dinitrobenzyl)acetamide (DNB) showed the most significant effect on fluorescence quenching. Judging from these results, DNB was employed as a quencher and introduced into the modular scaffold of the probe to develop DNB-type probes, Cy3DNB and Cy5DNB. We also envisioned that fluorescence quenching was promoted by configuring the quencher close to the fluorophore moiety. Based on the structure of PYP, Cy3DNB2 and Cy5DNB2 (DNB2-type probes) were designed by shortening the fluorophore–ligand linker, to the extent that the fluorophore moiety did not cause a steric clash with the PYP-tag. The probes were created by synthesizing a ligand moiety that was then coupled with N-(3,5-dinitrobenzyl)-2-(4-mercaptophenyl)acetamide and conjugated with Cy3 or Cy5 (figure 1b; see electronic supplementary material, schemes S3–S6).

Figure 2.

Fluorescence emission spectra of (a) Cy3NB, (b) Cy3DNB, (c) Cy3DNB2, (d) Cy5NB, (e) Cy5DNB and (f) Cy5DNB2 in the presence (red line) or absence (blue line) of PYP-tag. All the reactions were measured in 20 mM HEPES buffer (pH 7.4) containing 150 mM NaCl at 37°C. The concentrations of the probes and PYP-tag were 5 µM and 6 µM, respectively.

(b). Labelling reactions of PYP-tag with probes

SDS-PAGE analyses demonstrated that all the probes form covalent complexes with the PYP-tag (figure 3; see electronic supplementary material, figure S1). As shown in figure 2, all of the probes increased fluorescence intensity upon binding to the PYP-tag. Both DNB-type probes showed weaker fluorescence intensity in the free state than NB-type probes, as expected from the quenching experiments described above. The fluorescence OFF–ON ratios of Cy3DNB and Cy5DNB were improved 2.5- and 3.5-fold, respectively. Moreover, fluorescence enhancement of DNB2-type probes was more significant than NB-type or DNB-type probes. The fluorescence OFF–ON ratios of Cy3DNB2 and Cy5DNB2 were improved 2.8- and 5.8-fold, respectively. DNB lowered the molar extinction coefficients of the fluorophores of all the probes in the free state in comparison with the bound state (table 1; see electronic supplementary material, figure S3), suggesting that intramolecular interactions occurred between the fluorophores and the quenchers and caused a reduction in the fluorescence quantum yield and intensity. These results are consistent with previous reports showing that NB derivatives reduce the fluorescence intensity of fluorophores by static or dynamic quenching interactions [3,13,25–27]. Stern–Volmer plots showed upward curvature in the higher concentration of DNB, although the curvature of Cy3 was slight (see electronic supplementary material, figure S4). These results indicate that DNB has static and dynamic quenching effects on the fluorophores [28]. The results also suggest that the intramolecular interactions also occur in the ground and excited states of the free probes. The introduction of short linkers in DNB2-type probes decreased the molar extinction coefficients of the probes. These results suggest that the short linker promoted the fluorophore–quencher interactions that led to a reduction in the fluorescence from the unbound probes and the consequent improvement of the fluorescence OFF–ON ratios of DNB2-type probes.

Figure 3.

SDS-PAGE analyses of labelling reactions of PYP-tag with (a) Cy3DNB2 and (b) Cy5DNB2. The concentrations of probes and PYP-tag were 8 µM and 10 µM, respectively. CBB, Coomassie Brilliant Blue staining.

Table 1.

Photophysical properties of PYP-tag probes. All the measurements were made in triplicate.

probes	λabs (nm)	λem (nm)	ε (M−1 cm−1)	Φf	k2a (M−1 s−1)
FCANBb	501	521	44 400	0.04	125
Cy3NB	552	570	124 000	0.07	262
Cy3DNB	552	571	116 000	0.05	423
Cy3DNB2	553	572	112 000	0.05	489
Cy5NB	651	675	173 000	0.24	414
Cy5DNB	651	675	165 000	0.15	577
Cy5DNB2	651	675	160 000	0.12	715
Cy3DNB-PYP-tagc	551	570	143 000	0.09	—
Cy3DNB2-PYP-tagc	552	570	145 000	0.10	—
Cy5DNB-PYP-tagc	649	674	193 000	0.44	—
Cy5DNB2-PYP-tagc	649	674	199 000	0.46	—

^ak₂= second-order rate constant.

^bSpectroscopic data of FCANB cited from Hori et al. [13] are presented.

^cSpectroscopic measurements were conducted after PYP-tag is completely labelled with each probe.

(c). Kinetic analyses of probes

Next, the second-order rate constants, k₂, for the labelling reactions between each probe and the PYP-tag were determined (table 1; see electronic supplementary material, figure S5). All the probes labelled the PYP-tag more rapidly than FCANB. A possible reason for their superior kinetics is due to electrostatic factors of PYP-tag and the probes. This tag is known as an acidic protein and binds to an anionic probe slowly because of a repulsive electrostatic force [16]. While FCANB is dianionic, the cyanine probes carry only a net negative charge. Therefore, it is suggested that the less negative charge of the cyanine probes suppressed the repulsive interactions with PYP-tag. The k₂ values of DNB-type and DNB2-type probes were higher than those of NB-type probes. These results indicated that the introduction of the dinitrobenzene quencher resulted in the acceleration of labelling reactions. Moreover, comparisons between DNB-type and DNB2-type probes with the same fluorophores verified that probes with shorter linker bound to PYP-tag more quickly. These kinetic characteristics can be explained by considering the intramolecular interaction between a fluorophore and a quencher. We previously demonstrated that the fluorophore of FCANB interacts not only with a NB quencher, but also weakly with a cinnamic acid ligand [13]. The latter interaction generates steric hindrance between the probe and PYP-tag, and reduces the labelling kinetics. In this view, it is likely that dinitrobenzene interacts more favourably with cyanine dyes than NB does. In addition, the short linker in DNB2-type probes could keep the fluorophore close to the quencher, consequently promoting their intramolecular interactions and labelling reactions. Thus, it is suggested that the fluorophore–quencher interactions could prevent the steric hindrance around the ligand moiety by suppressing the interaction of the fluorophore with the ligand, resulting in the acceleration of labelling reactions. Interestingly, the fluorescence OFF–ON ratios of Cy3 or Cy5 probes have linear relationships with their corresponding k₂ values (see electronic supplementary material, figure S6). As discussed above, the fluorescence OFF–ON ratios are inferred to be dependent on intramolecular interactions between the fluorophore and the quencher. Therefore, it is suggested that the dinitrobenzene and the short linker in the probes strengthen the intramolecular interactions, leading to fast labelling kinetics as well as high fluorescence OFF–ON ratios.

(d). Multicolour imaging of cell-surface proteins

We conducted live-cell fluorescence imaging experiments using the cyanine probes. Human embryonic kidney (HEK293T) cells were transfected using a plasmid encoding PYP-tag-fused epidermal growth factor receptor (PYP-EGFR), which displays PYP-tag on the extracellular side of the plasma membrane. The cells were incubated with each probe for 30 min, washed to remove the excess probes and subsequently imaged using a confocal laser-scanning microscope. Fluorescence signals were obtained from the membrane of the PYP-EGFR-expressing cells, whereas no signal was detected in the cells expressing EGFR (figure 4; see electronic supplementary material, figure S7). The results demonstrated that all of the probes specifically labelled PYP-EGFR on the cell surface in living cells.

Figure 4.

Live-cell imaging of PYP-EGFR on cell surfaces with (a) 1.0 µM Cy3DNB2 and (b) 1.0 µM Cy5DNB2 in HEK293T cells. The imaging experiments were carried out with the excitation at (a) 559 nm or (b) 635 nm by using a (a) 570–670 nm or (b) 660–760 nm emission filter. Scale bar, 20 µm.

Finally, we carried out wash-free imaging of cell-surface PYP-EGFR in living cells using Cy3DNB2 and Cy5DNB2. In this experiment, transfected cells were directly imaged without washout procedures after the addition of each probe. Fluorescence labelling of PYP-EGFR with Cy3DNB2 was clearly detected, although slight fluorescence was also detected on the membrane of cells expressing EGFR (figure 5a). Cy5DNB2 also stained PYP-EGFR on the cell membrane, whereas no fluorescence was observed from culture media or any part of cells expressing EGFR (figure 5b). These results showed that cell-surface proteins were successfully labelled and imaged using a no-wash protocol.

Figure 5.

No-wash live-cell imaging of PYP-EGFR on cell surfaces with (a) 1.0 µM Cy3DNB2 and (b) 1.0 µM Cy5DNB2 in HEK293T cells. The imaging experiments were carried out with the excitation at (a) 559 nm or (b) 635 nm by using a (a) 570–670 nm or (b) 660–760 nm emission filter. Scale bar, 20 µm.

4. Conclusion

In conclusion, we developed fluorogenic probes that emitted long-wavelength fluorescence by exploring the combinations of cyanine dyes and NB derivatives. Fluorescence enhancement of the probes occurred in response to labelling reactions of PYP-tag. Kinetic properties of the probes were superior to those of the previously reported probe, FCANB. The dinitrobenzene quencher and the short linker in the cyanine probes promoted labelling reactions and enhanced fluorescence OFF–ON ratios. All the probes specifically labelled cell-surface proteins. In particular, Cy5DNB2 showed the best fluorogenic and kinetic properties among the probes and enabled no-wash imaging of cell-surface proteins with high contrast. Importantly, the long-wavelength fluorogenic probes were successfully designed to take advantage of the intramolecular interactions between the fluorophores and the dinitrobenzene quencher. This strategy offers a promising approach that allows versatile design of multicolour fluorogenic probes with non-FRET pairs of long-wavelength fluorophores and quenchers. These probes will be attractive tools for various biological applications, including studies of protein expression, translocation and degradation.

Data accessibility

All the data in the paper are available upon request.

Authors' contributions

S.H., Y.H. and K.K. designed the experiments and wrote the manuscript. S.H conducted experiments. Y.H. and K.K. devised the project.

Competing interests

We declare we have no competing interests.

Funding

This research was supported by the Strategic International Collaborative Research Program (SICORP), Japan Science and Technology Agency (JST), by MEXT of Japan (grants nos. 25220207, 16H00768, 16K13099, 16F16331 to K.K., 26282215, 16K13088, 16H01428 ‘Resonance Bio’ to Y.H. and 14J00755 to S.H.) and by the Program for Creating Future Wisdom, Osaka University, selected in 2014.