A dual functional probe: sensitive fluorescence response to H₂S and colorimetric detection for SO₃²⁻

Abstract

Hydrogen sulfide (H₂S) and sulfite (SO₃²⁻) are two important sulfur-containing species that play different and important roles in industrial and biological processes. Accordingly, the development of efficient methods for simple, rapid, sensitive and selective monitoring of H₂S and SO₃²⁻ is of the utmost importance in both environmental and biological sciences. In this study, we developed a new dual functional probe NIR-DNP for discriminative detection of H₂S and SO₃²⁻. This probe can sense H₂S and SO₃²⁻ via two different approaches, a significant near-infrared fluorescence enhancement and color change from purple to cyan induced by H₂S as well as a visible color change from purple to colorless caused by SO₃²⁻. The detection limits of the probe NIR-DNP for H₂S and SO₃²⁻ in aqueous solutions were 36.53 nM and 33.33 nM, respectively. Competitive experiments demonstrated that the probe NIR-DNP had a high fluorescence selectivity for H₂S and excellent colorimetric selectivity for SO₃²⁻ over other analytes. The sensing mechanism of the probe toward H₂S and SO₃²⁻ was based on the H₂S-induced thiolysis of dinitrophenyl ether and SO₃²⁻-induced nucleophilic addition, respectively. Further investigation showed that the probe NIR-DNP could be used to develop an easy-to-prepare and easy-to-detect paper-based test strip for cheap and effective detection of SO₃²⁻. Also, the probe NIR-DNP has the potential to image exogenous and endogenous H₂S in living cells.

---

Introduction

Hydrogen sulfide (H₂S) and sulfite (SO₃²⁻) are two important members of sulfur-containing species that play vital roles in industrial, environmental and biological processes.^1–3 H₂S is the third endogenous gaseous signaling molecule following nitric oxide and carbon monoxide.⁴ Physiological levels of endogenous H₂S are known to be involved in a variety of physiological processes such as anti-inflammation,⁵ neuromodulation⁶ and apoptosis.⁷ Abnormal levels of H₂S are closely linked with various diseases such as liver cirrhosis⁸ and Alzheimer's.⁹ Sulfite (SO₃²⁻), because of its antioxidant and preservative properties, is widely used to preserve foods, beverages and pharmaceutical products from oxidation and microbial reactions.¹⁰ However, long-term and frequent exposure to high doses of sulfite can cause adverse reactions and acute symptoms including dermatitis, urticaria, flushing, hypotension, abdominal pain and diarrhea.¹¹ The development of efficient methods for simple, rapid, sensitive and selective monitoring of H₂S and SO₃²⁻ is very important in both environmental and biological sciences.

Currently, several techniques such as electrochemical assay,¹² gas chromatography¹³ and sulfide precipitation¹⁴ have been developed for the detection of H₂S and SO₃²⁻. However, these methods often require tedious sample and reagent preparation or complicated instruments and are not suitable for routine laboratory and on-site analyses. Fluorescent and colorimetric analyses provide good alternatives because of their low cost, easy-operation, simple instrumental implementation, fast response and excellent selectivity. Much effort has been devoted to the development of fluorescent and colorimetric probes for H₂S and SO₃²⁻. To date, several probes have been designed for H₂S detection by taking the advantages of some specific chemical reactions, including the reduction of azides/nitro groups,^15,16 demetallation of macrocyclic Cu(II) complexes,¹⁷ nucleophilic reaction¹⁸ and thiolysis of dinitrophenyl ether.¹⁹ Similarly, many probes for sulfite have also been designed on the basis of selective deprotection of levulinate groups,²⁰ reactions with aldehyde groups,²¹ Michael-type additions,²² and complexation with amines²³ in the presence of SO₃²⁻. However, most of these probes could only sense SO₃²⁻ or H₂S. There are very few probes that can simultaneously detect SO₃²⁻ and H₂S because of some of their common features, such as their nucleophilic and reducing properties.²⁴ In addition, most of the existing probes for H₂S have emissions in the ultraviolet or visible region, and the fluorescence imaging is easily interfered by cell auto-fluorescence.²⁵ Near-infrared (NIR, 650–900 nm) fluorescent probes have unique advantages for tracing molecules in vitro and in vivo, such as minimum photo-damage to biological samples, deep tissue penetration, and minimum interference from background auto-fluorescence by bio-molecules in living systems.^25–27 Therefore, it is necessary to develop new types of probes that can be used for the simultaneous detection of H₂S and SO₃²⁻, preferably with emissions located in the NIR region.

In this work, we designed and synthesized a new NIR fluorescence probe NIR-DNP (Scheme 1) that can distinguish the presence of H₂S and SO₃²⁻ by the naked eye. The probe was constructed by connecting 2,4-dinitrophenyl to a hemicyanine skeleton in which a propane sulfonate residue was introduced to the indolium group to improve its water solubility.²⁸ The probe displayed selective recognition of H₂S via a visible color change from purple to cyan and a significant fluorescence enhancement in the NIR region over other relevant species, and also exhibited a sensitive response to SO₃²⁻ through a visible color change from purple to colorless. In addition, further applications of NIR-DNP in a paper-based test strip for SO₃²⁻ and imaging exogenous and endogenous H₂S in living cells were studied.

Scheme 1 Synthesis of probe NIR-DNP. Reagents and conditions: (a) 1,3-propane sultone, toluene, reflux, 24 h, 89%; (b) DMF, CH₂Cl₂, POCl₃, reflux, 3 h, 73.9%; (c) sodium acetate, acetic anhydride, 70 °C, 1 h, 80.8%; (d) resorcinol, triethylamine, DMF, N₂ atmosphere, 50 °C, 4 h, 46.6%; (e) 1-fluoro-2,4-dinitrobenzene, K₂CO₃, DMF, N₂ atmosphere, 50 °C, 5 h, 67%.

Experimental section

Materials

2,3,3-Trimethylindolenine was purchased from J&K Chemical Technology (Shanghai, China). Cyclohexanone, resorcinol and 1,3-propanesulfonate were obtained from Aladdin (Shanghai, China). 1-Fluoro-2,4-dinitrobenzene was purchased from Xiya Reagent (Shandong, China). Whatman No. 1 filter paper was purchased from Fisher Scientific (Pittsburgh, PA). Dry solvents used in the synthesis were purified using standard procedures. All other chemicals were obtained from qualified reagent suppliers with analytical reagent grade.

Instruments

¹H-nuclear magnetic resonance (NMR) and ¹³C-NMR spectra of the products were recorded on a Bruker 500 MHz/125 MHz NMR spectrometer using tetramethylsilane (TMS) as the internal standard (chemical shifts in ppm). Electrospray ionization mass spectroscopy (ESI-MS) data were obtained using a Thermo Scientific LCQ FLEET mass spectrometer. High-resolution mass spectra (HRMS) data were recorded on a Waters Xevo® G2-XS QTOF mass spectroscopy (Waters, Manchester, UK) equipped with a Zpray™ ESI source operating in the negative ion mode. Fourier-transform infrared spectroscopy (FT-IR) was recorded on a Nicolet 5700 spectrometer at wavelengths of 400 cm⁻¹ to 4000 cm⁻¹ and a resolution of 3 cm⁻¹ over 32 scans. The pH values were measured using a digital pH meter (PHS-3C, Lei-ci, Shanghai, China). The UV-Vis absorption spectra were measured using Shimadzu UV1780 spectrometer (Shimadzu, Japan). The fluorescence spectra measurements were performed using Shimadzu RF-5301 fluorescence spectrometer (Shimadzu, Japan). The fluorescence images of the cells were taken using an inverted microscope (Olympus CKX41, Japan).

Synthesis of the probe NIR-DNP

The probe NIR-DNP was prepared as depicted in Scheme 1. For detailed information on the synthesis of the intermediates 1–4 see the ESI.† The probe NIR-DNP was synthesized by the nucleophilic substitution reaction between compound 4 and 1-fluoro-2,4-dinitrobenzene. To a solution of 4 (24.58 mg, 0.05 mmol) in N,N-dimethylformamide (DMF, 2 mL), potassium carbonate (13.8 mg, 0.1 mmol) was added and the mixture was stirred for 30 min under N₂ atmosphere. Then, 1-fluoro-2,4-dinitrobenzene (18.6 mg, 0.1 mmol) dissolved in DMF (2 mL) was added to the solution. The resulting mixture was stirred for 5 h at 50 °C. The solvent was evaporated and the crude product was purified by column chromatography on silica gel (CH₂Cl₂ : CH₃OH = 20 : 1, v/v) to afford the probe NIR-DNP as a blue-purple solid (22 mg, 67% yield). ¹H NMR (500 MHz, CDCl₃/CD₃OD): δ 8.93 (d, J = 2.7 Hz, 1H), 8.74 (d, J = 15.1 Hz, 1H), 8.49 (dd, J = 9.2, 2.8 Hz, 1H), 7.65 (d, J = 5.8 Hz, 1H), 7.59–7.53 (m, 4H), 7.48 (t, J = 7.4 Hz, 1H), 7.33 (d, J = 9.2 Hz, 1H), 7.26–7.24 (m, 2H), 7.06 (dd, J = 8.5, 2.2 Hz, 1H), 6.87 (d, J = 15.1 Hz, 1H), 4.65 (t, 2H), 3.02–2.99 (m, 2H), 2.82–2.79 (m, 4H), 2.37–2.31 (m, 2H), 2.00–1.93 (m, 2H), 1.81 (s, 6H) ppm; ¹³C NMR (125 MHz, CDCl₃/CD₃OD): δ 179.00, 159.80, 156.18, 154.38, 153.91, 147.83, 146.89, 142.67, 142.28, 141.02, 140.24, 130.73, 130.55, 129.33, 129.12, 128.01, 122.46, 121.96, 120.16, 120.07, 116.59, 115.92, 113.21, 107.65, 106.17, 51.06, 47.27, 44.16, 29.26, 27.46, 23.91, 23.57, 20.12 ppm; IR (KBr) ν 3040.96, 2923.43, 2854.68, 1603.70, 1578.15, 1534.07, 1499.74, 1455.64, 1346.55, 1319.87, 1260.96, 1207.56, 1152.62, 1115.21, 1061.72, 1034.94, 966.35, 922.66, 833.94, 761.06 cm⁻¹; HRMS (ESI, m/z): calcd for [C₃₄H₃₁N₃O₉S + H]⁺, 658.1815; found, 658.1951.

General procedure for spectra measurement

The stock solution of the probe NIR-DNP was prepared at 1.0 mM in DMSO. Stock solutions of various testing species in water or phosphate buffer saline (PBS) were prepared from Na₂S, Na₂SO₃, CaCl₂, KCl, NaCl, NaF, NaBr, NaNO₃, NaNO₂, NaN₃, Na₂SO₄, Na₂S₂O₃·5H₂O, CH₃COONa, Na₂CO₃, H₂O₂, NaHCO₃, NaH₂PO₃, cysteine and glutathione. A Na₂S (50 mM) aqueous solution was used as a H₂S source in all experiments, and a Na₂SO₃ (50 mM) aqueous solution was used as a SO₃²⁻ source.^17,29 A typical test solution was prepared by placing 20.0 μL of the NIR-DNP stock solution (1.0 mM), 180.0 μL of DMSO and an appropriate aliquot of each analyte stock solution into a 5.0 mL centrifugal tube and diluting the solution to 2.0 mL with PBS (20 mM, pH 7.4). For the detection of H₂S, the resulting solution was incubated with the appropriate testing species at ambient temperature for 45 min before recording the UV-Vis absorption and fluorescence spectra. For the detection of SO₃²⁻, the resulting solution was incubated with the appropriate testing species for 3 min before recording the UV-Vis absorption spectra. The 650 nm excitation wavelength was used for all the measurements.²⁶ The excitation and emission slit widths were all 5 nm.

Cell culture and fluorescence imaging

Human glioblastoma (U-251) cells were obtained from the Chinese Academy of Sciences (Shanghai, China). U-251 cells were routinely cultured with Dulbecco's modified Eagle's medium (DMEM, pH 7.4) and supplemented with 10% fetal bovine serum (FBS, Gibco Invitrogen, CA, USA), penicillin (100 U mL⁻¹), and streptomycin (100 U mL⁻¹) at 37 °C in a 5% CO₂ atmosphere. To maintain the cells in the exponential growth phase, they were normally passaged at a ratio of 1 : 3 every 72 h. Before using, the U-251 cells in the exponential growth phase were collected and re-seeded on 24-well plates at a density of 5 × 10⁴ cells per mL and cultured for 24 h. For exogenous H₂S imaging, the culture medium (the DMEM supplemented with 10% FBS) was removed, and the cells were washed twice with PBS (0.01 mol L⁻¹, pH 7.4). Then, the cells were treated with NIR-DNP (50 μM) at 37 °C for 60 min. After washing with PBS, the cells were incubated for another 60 min in the presence of Na₂S (500 μM). The controls were simultaneously performed by treating the cells only with the probe NIR-DNP for 60 min at 37 °C. After washing twice with PBS, the cells were incubated with a 4% (m/v) paraformaldehyde aqueous solution for 20 min at room temperature. Then, Hoechst 33258 (0.5 μg mL⁻¹) in PBS was added to stain the nuclei for 10 min. Fluorescence imaging was taken after washing the cells with PBS twice. For endogenous H₂S imaging, cells were first incubated with thiols (cysteine, 500 μM) for 60 min at 37 °C. After washing, the cells were then incubated with NIR-DNP (50 μM) for another 60 min. The medium was replaced by PBS, and cells were immediately imaged.

Results and discussion

Synthesis of probe NIR-DNP

The synthesis of the probe NIR-DNP is outlined in Scheme 1. The intermediates 1 and 2 were first prepared from 2,3,3-trimethylindolenin and cyclohexanone according to the literature procedures.^30,31 The intermediate 3 was obtained via a condensation reaction of intermediates 1 and 2 in acetic anhydride. The intermediate 4 was subsequently synthesized by treating 3 with resorcinol in the presence of triethylamine at 50 °C for 4 h. Finally, under basic conditions, the intermediate 4 was treated with 1-fluoro-2,4-dinitrobenzene to give the target probe NIR-DNP through a nucleophilic substitution reaction. The chemical structures of the compounds were well characterized by ¹H NMR, ¹³C NMR, HRMS and FT-IR. The detailed synthetic procedures and relevant spectral data are given in the Experimental section and ESI.†

Sensing property of probe NIR-DNP for H₂S

The sensing ability of probe NIR-DNP for H₂S was investigated in PBS (20 mM, pH 7.4) containing 10% DMSO (v/v). NIR-DNP (10 μM) showed a maximum absorption at 595 nm (ε = 2.96 × 10⁴ M⁻¹ cm⁻¹). Upon the addition of Na₂S (50.0 equiv.) to the NIR-DNP solution, the absorbance at 595 nm decreased, and a new absorption peaked at 685 nm appeared and gradually increased over time along with a distinct color change from purple to cyan (Fig. 1A and inset). The UV spectrum was close to the characteristic peak of compound 4 (ε = 2.95 × 10⁴ M⁻¹ cm⁻¹) (Fig. S1A, ESI†). The results showed that the probe allowed colorimetric detection of H₂S by the ‘naked eye’. Fluorescence analysis (Fig. 1B) exhibited that the NIR-DNP solution was almost non-fluorescent (Φ = 0.007, Table S1, ESI†). However, upon the addition of Na₂S, a large fluorescence enhancement at 707 nm was observed, and the resulting fluorescence spectrum was consistent with compound 4 (Φ = 0.031) (Fig. S1B and Table S1, ESI†). Kinetic studies of the probe NIR-DNP (10 μM) with the addition of Na₂S (50.0 equiv.) showed that the emission intensity at 707 nm gradually increased against time until it reached a plateau at about 45 min (t_1/2 ≈ 7 min) under the test conditions (Fig. 1B, inset). The observed first order rate constant k_obs was determined to be about 0.0718 min⁻¹ (Fig. S2, ESI†), and the emission intensity was found to increase about 53-fold at 707 nm. The relatively fast and distinct fluorescence signal change in the NIR region indicated that the probe NIR-DNP can be used as a sensitive NIR fluorescent sensor for ‘turn-on’ detection of H₂S in aqueous solutions.

Fig. 1 (A) Time-dependent absorbance spectra of the probe NIR-DNP (10 μM) upon the addition of Na₂S (500 μM) in PBS (20 mM, pH 7.4) containing 10% DMSO (v/v). Inset: plot of the absorbance at 595 nm and 685 nm as a function of time; the color change of NIR-DNP (10 μM) before (i) and after (ii) the addition of Na₂S (500 μM) under visible light. (B) Time-dependent fluorescent spectra of NIR-DNP (10 μM) upon addition of Na₂S (500 μM) in PBS (20 mM, pH 7.4) with 10% DMSO (v/v). Inset: plot of the emission intensity at 707 nm as a function of time. (C) Fluorescence spectra of NIR-DNP (10 μM) upon addition of Na₂S (0–7.5 μM). Inset: plot of the emission intensity at 707 nm as a function of the concentrations of Na₂S. (D) Fluorescent intensity of NIR-DNP (10 μM) at 707 nm in the presence of various species (1 mM Cys, 5 mM GSH and 500 μM others).

To further evaluate the capability of NIR-DNP for highly sensitive detection of H₂S, the changes in the fluorescence spectra of NIR-DNP in the presence of different concentrations of Na₂S were also investigated (Fig. 1C and S3, ESI†). Upon treatment with increasing concentrations of Na₂S, a characteristic fluorescence emission at 707 nm was observed and gradually enhanced. The fluorescent intensity stabilized when the amount of Na₂S was more than 200 μM (Fig. S3, ESI†). Further analysis showed that the fluorescent intensity at 707 nm increased linearly with the concentration of Na₂S ranging from 0 to 7.5 μM (R = 0.9993) (Fig. 1C, inset). The detection limit of NIR-DNP for H₂S was experimentally determined to be 36.53 nM based on the signal-to-noise ratio (S/N) = 3 under the test conditions, which was much lower than those obtained by most of the existing small-molecule fluorescent probes for H₂S (Table S2, ESI†). This result further confirmed that NIR-DNP is highly sensitive to H₂S. Also, the results indicated that the probe NIR-DNP was favorable for imaging intracellular H₂S whose physiological relevant concentrations are estimated to be in the range from nanomolar to millimolar levels.³²

To investigate the specificity of the probe for H₂S, the effects of various analytes including typical anions, cations and other biologically relevant species on the fluorescent intensity of NIR-DNP were tested (Fig. 1D and S4, ESI†). The fluorescent intensity of the probe NIR-DNP did not show an observable change upon the addition of various analytes including anions (F⁻, Cl⁻, Br⁻, N₃⁻, CO₃²⁻, HCO₃⁻, CH₃COO⁻, SO₄²⁻, H₂PO₄⁻), cations (Na⁺, K⁺, Ca²⁺), and reactive oxygen species (H₂O₂), nitrogen species (NO₃⁻, NO₂⁻) and sulfur species (SO₃²⁻, S₂O₃²⁻). However, after reacting with Na₂S, the fluorescent intensity of the probe NIR-DNP enhanced remarkably (∼47-fold). It is worth mentioning that although the addition of millimolar concentrations of the biothiols (i.e., 1 mM for Cys or 5 mM for GSH) to the probe gave a limited increase in the fluorescence intensity, it was far weaker than that caused by Na₂S. The high selectivity of the probe NIR-DNP to H₂S could be attributed to the obvious distinction in terms of size and pK_a values between H₂S and biothiols.^33,34 To check the practical ability of NIR-DNP as a H₂S selective fluorescent probe, competition experiments were also carried out by adding Na₂S to the NIR-DNP solution in the presence of the above analytes. As shown in Fig. S5A (ESI†), most of the analytes did not produce an obvious change in the fluorescence intensity except for SO₃²⁻. When competing with SO₃²⁻, the fluorescence intensity was about 60% lower than that in its absence. Moreover, biothiol (Cys or GSH) did not show a significant influence on the detection of H₂S, even when its concentration was in the millimolar range (Fig. S5B, ESI†). All of these results indicated that the probe NIR-DNP had a high selectivity for H₂S.

Sensing property of probe NIR-DNP for SO₃²⁻

In the selectivity experiments of probe NIR-DNP for H₂S, we found that the color of the probe solution in the presence of SO₃²⁻ changed from purple to colorless. This phenomenon indicated that the probe NIR-DNP may have the ability to sense H₂S and SO₃²⁻ by different approaches. As we know, there are relatively few single probes that can simultaneously detect SO₃²⁻ and H₂S (Table S2, ESI†). Therefore, the sensing ability of the probe NIR-DNP for SO₃²⁻ was also investigated in PBS (20 mM, pH 7.4) containing 10% DMSO (v/v). Firstly, the time-dependent absorption spectra of NIR-DNP (10 μM) in the presence of SO₃²⁻ (50.0 equiv.) were studied (Fig. 2A). The results showed that upon the addition of SO₃²⁻ (50.0 equiv.), the absorption band at 595 nm decreased rapidly. Meanwhile, a new absorption band centered at 375 nm appeared, accompanied with the solution color changing from purple to colorless (Fig. 2A, inset). The isosbestic point at 419 nm indicated the formation of a new compound. Further studies indicated that the absorption peak at 595 nm decreased rapidly to the baseline over 0 to 180 seconds, and a new peak at 375 nm increased to the top of the wavelength range (Fig. 2A, inset). The A₅₉₅/A₃₇₅ ratio varied from 7.37 to 0.078 after complete conversion, which was about a 95-fold decrease (Fig. S6, ESI†). Such a rapid and significant change in the signal ratios at the two wavelengths indicated that the probe NIR-DNP could be used as a colorimetric sensor for highly sensitive detection of SO₃²⁻.

Fig. 2 (A) Time-dependent absorption spectra of NIR-DNP (10 μM) upon addition of SO₃²⁻ (500 μM) in PBS (20 mM, pH 7.4) with 10% DMSO (v/v). Inset: plot of the absorbance at 595 nm and 375 nm as a function of time; the color change of NIR-DNP (10 μM) before (i) and after (ii) the addition of Na₂SO₃ (500 μM) under visible light. (B) Absorption spectra of NIR-DNP (10 μM) upon addition of Na₂SO₃ (0–125 μM). Inset: plot of the A₃₇₅/A₅₉₅ ratio as a function of the concentrations of SO₃²⁻. (C) Absorption spectra of NIR-DNP (10 μM) upon addition of various species (1 mM Cys, 5 mM GSH and 500 μM others). (D) Absorption ratio A₃₇₅/A₅₉₅ of NIR-DNP (10 μM) in the presence of various species.

The changes in the absorption spectra of NIR-DNP with an increasing SO₃²⁻ concentration were further studied (Fig. 2B and S7, ESI†). Upon the addition of increasing concentrations of SO₃²⁻, the absorption intensity at 595 nm gradually decreased, while the absorption intensity at 375 nm gradually increased. The A₃₇₅/A₅₉₅ ratio was linearly proportional to the SO₃²⁻ concentration in the range of 0–125 μM (R = 0.9971) (Fig. 2B, inset). The detection limit was determined to be about 33.33 nM based on S/N = 3, suggesting that the probe was potentially useful for quantitative determination of SO₃²⁻ in a large dynamic range and with a high sensitivity.

Considering the coexistence of other competing species in real samples, the selectivity of NIR-DNP toward SO₃²⁻ was also studied in the presence of other analytes including F⁻, Cl⁻, Br⁻, N₃⁻, CO₃²⁻, HCO₃⁻, NO₃⁻, NO₂⁻, S₂O₃²⁻, CH₃COO⁻, SO₄²⁻, H₂PO₄⁻, Na⁺, K⁺, Ca²⁺, Na₂S, Cys, GSH and H₂O₂. The results (Fig. 2C) showed that no obvious responses were observed in the UV-vis spectra of NIR-DNP upon the addition of the above species. However, the addition of SO₃²⁻ induced a new absorption band centered at 375 nm, accompanied by decrease of the absorption band at 595 nm. Observation of the color changes of the probe NIR-DNP before and after the addition of sulfite and other species under visible light showed that only SO₃²⁻ caused a marked change from purple to colorless (Fig. S8A, ESI†). Quantitative analysis (Fig. 2D) exhibited that the A₃₇₅/A₅₉₅ ratio of the probe was remarkably enhanced when NIR-DNP was treated with SO₃²⁻. It is worth noting that during the period (3 min) of SO₃²⁻ detection, Na₂S could only induce a small decrease in the absorbance at 595 nm (Fig. 1A and 2C) and a neglectable change in the A₃₇₅/A₅₉₅ ratio (Fig. 2D). However, when the probe was treated with Na₂S for 45 min, the color of the probe solutions changed from purple to cyan (Fig. S8B, ESI†). This result indicated that NIR-DNP could be used for distinguishing SO₃²⁻ and H₂S by the naked eyes. Moreover, the addition of biothiols (1 mM for Cys and 5 mM for GSH) to the probe could not cause a change in the A₃₇₅/A₅₉₅ ratio, which might be ascribed to their steric effects and high pK_a values.³⁴ This result implies that the probe was chemoselective for SO₃²⁻ over these biothiols at physiological pH. The detection of SO₃²⁻ using NIR-DNP in the presence of other analytes remain effective (Fig. S9, ESI†). All of these investigations clearly indicated that the probe NIR-DNP possessed excellent selectivity for SO₃²⁻.

Sensing mechanism of NIR-DNP toward H₂S

To confirm that the fluorescence response of NIR-DNP was triggered by H₂S, the thiolysis product of NIR-DNP was characterized by using NMR and EI-MS analyses and theoretical calculations. ¹H NMR analysis showed that the product of NIR-DNP after treatment with Na₂S was identical to the authentic compound 4 (Fig. S10, ESI†), which was further confirmed by EI-MS analysis. EI-MS showed a peak at m/z = 490.32, which is the same as the theoretical molecular mass of compound 4 (calcd for [M − 1]⁻ = 490.18) (Fig. S11, ESI†). To better understand the photophysical properties of the probe NIR-DNP and the thiolysis product 4, theoretical calculations were also performed using the density functional theory (DFT) and time-dependent DFT (TDDFT) by using Gaussian 09 program (Fig. 3B and C and Table S3, ESI†). For compound 4, the π electrons on both the HOMO and LUMO were essentially distributed over the entire backbone of compound 4. By contrast, for NIR-DNP, the π electrons on the HOMO and LUMO+2 primarily resided on the electron-donating moiety of compound 4, whereas, those on the LUMO and LUMO+1 were mainly located on the electron-withdrawing group of 2,4-dinitrobenzene. This result indicated that the NIR-DNP could bear efficient photo-induced electron transfer (PET) from compound 4 to the 2,4-dinitrobenzene group, rendering the relatively weak fluorescence of the probe. By contrast, the nearly complete overlap of electrons on the transition orbitals may induce the strong fluorescence emission for compound 4. All of these investigations demonstrated that the mechanism of NIR-DNP for H₂S sensing was a thiolysis process of dinitrophenyl ether with the release of compound 4 (Fig. 3A).

Fig. 3 (A) The mechanism of probe NIR-DNP for discriminative sensing of sulfite and sulfide. (B) Frontier molecular orbital plots of compound 4 in water. (C) Frontier molecular orbital plots of probe NIR-DNP in water. The results showed that the fluorescence emission of compound 4 is quenched by d-PET process.

Sensing mechanism of NIR-DNP toward SO₃²⁻

The obvious blue-shifts in the absorbance spectra of NIR-DNP toward SO₃²⁻ are consistent with the addition of SO₃²⁻ to the electrophilic C=C bond in the NIR-DNP, which led to a short conjugation structure of the reaction product. To confirm the formation of the addition product NIR-DNP–SO₃Na, an ¹H NMR titration experiment was carried out by addition of Na₂SO₃ into NIR-DNP in CD₃OD. The result (Fig. S12, ESI†) showed that the resonance signal corresponding to the alkene proton H_a at 7.54 ppm disappeared after it was treated with Na₂SO₃, and a new peak at 3.82 ppm assigned to the proton H_a′ appeared. The addition of Na₂SO₃ to C=C resulted in the formation of a chiral center of C_a′ with the nonequivalent protons of the methylene group at C_b. Therefore, the signal for H_b at 8.72 ppm shifted to a high field and appeared as two peaks at 2.48 ppm and 3.02 ppm, respectively, which was similar to those previously reported.^22,35 EI-MS analysis further confirmed this result. A peak appeared at m/z = 760.24 in the mass spectrum, which is nearly identical to the theoretical molecular mass of NIR-DNP–SO₃Na (calcd for [NIR-DNP–SO₃Na − 1]⁻, 760.13) (Fig. S13, ESI†). These results suggested that the mechanism of NIR-DNP for SO₃²⁻ sensing was a SO₃²⁻-induced nucleophilic addition, which generates a short conjugation structure of the addition product NIR-DNP–SO₃Na (Fig. 3A).

Effect of pH

The pH adaptability is very important for the practical applications of a newly designed probe. In the current study, the pH adaptability of the probe NIR-DNP was also investigated (Fig. S14 and S15, ESI†). The results showed that the fluorescence intensity and the A₃₇₅/A₅₉₅ ratio of NIR-DNP had no change in the pH range from 6.0 to 9.4, suggesting that the probe was stable and insensitive to pH changes in this range. However, upon treatment with Na₂S (50 equiv.), a strong fluorescence emission was observed in the pH range from 6.5 to 9.4 (Fig. S14, ESI†), which indicated that the probe NIR-DNP could be used to recognize H₂S in the physiological pH region. Additionally, upon the addition of SO₃²⁻ (50 equiv.), a significant enhancement in the A₃₇₅/A₅₉₅ ratio was observed in the pH range of 6.75 to 8.5 (Fig. S15, ESI†), revealing that the probe NIR-DNP could also be used to detect SO₃²⁻ under physiological conditions. The pH cross-talk of the probe in the “on” state (compound 4) was also studied. The results suggested compound 4 was a double NIR-emissive ratiometric fluorescent probe, which exhibited pH-dependent optical responses (Fig. S16 and S17, ESI†) similar to the studies reported previously.^36,37 The pK_a value was calculated to be 6.7 and was derived from the titration curve of the emission ratio (I₇₀₆/I₆₆₉) (Fig. S18, ESI†).

Paper-based test strips

To demonstrate the practical applications of the probe NIR-DNP for the detection of SO₃²⁻, a paper-based test strip was developed by wetting a neutral filter paper with a NIR-DNP methanol solution (100 μM). After drying in the air, a blue-purple test paper was ready for use. For the detection of SO₃²⁻, a SO₃² – contained solution was dropped on the test paper using a glass capillary. As shown in Fig. 4A, a rapid color change occurred, even in the presence of other analytes. However, the mixture of other analytes without SO₃²⁻ could not cause any detectable changes. In addition, the test papers showed different color changes to different concentrations of SO₃²⁻ ranging from 10 μM to 50 mM. These results indicated that the probe NIR-DNP could be used to develop an easy-to-prepare and easy-to-detect paper-based test strip for a cheap and effective way to detect SO₃²⁻.

Fig. 4 (A) Detection of SO₃²⁻ in the absence and presence of other analytes by using paper-based test strips treated by the probe NIR-DNP (100 μM): 1, 0 μM SO₃²⁻; 2, other analyte mixture; 3, 10 μM SO₃²⁻; 4, 50 μM SO₃²⁻; 5, 100 μM SO₃²⁻; 6, 200 μM SO₃²⁻; 7, 500 μM SO₃²⁻; 8, 500 μM SO₃²⁻ and other analytes; 9, 1 mM SO₃²⁻; 10, 5 mM SO₃²⁻; 11, 10 mM SO₃²⁻; 12, 50 mM SO₃²⁻. Other analytes were a mixture of K⁺, Na⁺, F⁻, Cl⁻, Br⁻, NO₃⁻, NO₂⁻, N₃⁻, SO₄²⁻, S₂O₃²⁻, CO₃²⁻, HCO₃⁻, H₂PO₄⁻, AcO⁻ (500 μM of each analyte in PBS, pH 7.4). (B) Fluorescence images of the probe NIR-DNP in U-251 cells. The cell nuclei were stained with Hoechst 33258. From the left to right, the cells were incubated with the probe NIR-DNP (50 μM) for 60 min; cells were first incubated with the probe NIR-DNP (50 μM) for 60 min, and then incubated with Na₂S (500 μM) for another 60 min; cells were first incubated with Cys (500 μM) for 60 min, and then incubated with the probe NIR-DNP (50 μM) for another 60 min. From the top to bottom, bright-field images; fluorescence images; merged image of fluorescence and Hoechst 33258-stained images.

Cell imaging

To demonstrate the potential applications of the probe NIR-DNP in the fluorescence imaging of cells, live cell imaging experiments were also performed using U-251 cells. Firstly, the cytotoxicity of the probe was evaluated by MTT assay.³⁸ The result (Fig. S19, ESI†) showed that more than 80% of the cells were viable when up to 80 μM NIR-DNP was used, which indicated that the probe did not show significant cytotoxicity in the 0–80 μM range. As displayed in Fig. 4B, U-251 cells incubated with NIR-DNP (50 μM) for 60 min at 37 °C showed no intracellular fluorescence. Owing to the low level of H₂S in cells, an exogenous Na₂S was added to increase the intracellular H₂S level.^17,29 The cells were treated with NIR-DNP (50 μM) for 60 min, followed by Na₂S (500 μM) for another 60 min. As a result, red fluorescence was clearly observed in the cytoplasm, suggesting that NIR-DNP could be used for exogenous H₂S imaging in U-251 cells. To further test whether NIR-DNP could detect endogenous production of H₂S, cells were first treated with Cys (500 μM) for 60 min. The results of the MTT assays indicated that Cys did not show obvious cytotoxicity at the experimental conditions (Fig. S20, ESI†). The cells were then treated with NIR-DNP (50 μM). As shown in Fig. 4B, Cys induced a production enhancement of endogenous H₂S,^39,40 and a clear fluorescence was observed, implying that NIR-DNP could also be used for endogenous H₂S imaging in U-251 cells. Moreover, the probe NIR-DNP can also be used for H₂S imaging in murine hepatocytes (Fig. S21, ESI†). All of the results above suggested that NIR-DNP was cell-permeable and can react with intracellular H₂S efficiently, which can be used for H₂S imaging in cells with low cytotoxicity.

Conclusions

In this study, a dual functional probe NIR-DNP was synthesized by introducing 2,4-dinitrobenzene to a hemicyanine skeleton. The probe NIR-DNP exhibited a colorimetric response to SO₃²⁻ along with a color change from purple to colorless and a NIR fluorescence enhancement response to H₂S along with a color change from purple to cyan. The probe could be used to sense H₂S and SO₃²⁻ with a high selectivity, even in the presence of other biologically relevant species. The detection limit of the probe was 36.53 nM for H₂S and 33.33 nM for SO₃²⁻. ¹H NMR and ESI-MS analyses as well as DFT and TDFT theoretical calculations demonstrated that the sensing mechanisms of the probe toward H₂S and SO₃²⁻ were based on H₂S-induced thiolysis of dinitrophenyl ether and SO₃²⁻-induced nucleophilic addition, respectively. Further investigation showed that the probe NIR-DNP could be used to develop an easy-to-prepare and easy-to-detect paper-based test strip for cheap and effective detection of SO₃²⁻. Also, the probe NIR-DNP has the potential to image exogenous and endogenous H₂S in living cells. Based on these results, we are actively pursuing more sensitive and wide-ranging analogues for simultaneous detection of SO₃²⁻ and H₂S as well as fluorescence imaging of H₂S in living cells, tissues and animals in our future work.

Acknowledgements

This study was supported by the National Natural Science Foundation of China (No. 21375106 and 21202131), the Fund of Youth Science and Technology Stars by Shaanxi Province (2015KJXX-15), and the Fundamental Research Funds for the Central Universities (2014YB026).

Notes and references

1. J. W. Calvert, S. Jha, S. Gundewar, J. W. Elrod, A. Ramachandran, C. B. Pattillo, C. G. Kevil and D. J. Lefer, Circ. Res., 2009, 105, 365.
2. S. W. Benson, Chem. Rev., 1978, 78, 23.
3. M. Noji, M. Saito, M. Nakamura, M. Aono, H. Saji and K. Saito, Plant Physiol., 2001, 126, 973.
4. L. Li, P. Rose and P. K. Moore, Annu. Rev. Pharmacol. Toxicol., 2011, 51, 169.
5. L. Li, M. Bhatia, Y. Z. Zhu, Y. C. Zhu, R. D. Ramnath, Z. J. Wang, F. B. M. Anuar, M. Whiteman, M. Salto-Tellez and P. K. Moore, FASEB J., 2005, 19, 1196.
6. K. Abe and H. Kimura, J. Neurosci., 1996, 16, 1066.
7. G. Yang, L. Wu and R. Wang, FASEB J., 2006, 20, 553.
8. S. Fiorucci, E. Antonelli, A. Mencarelli, S. Orlandi, B. Renga, G. Rizzo, E. Distrutti, V. Shah and A. Morelli, Hepatology, 2005, 42, 539.
9. K. Eto, T. Asada, K. Arima, T. Makifuchi and H. Kimura, Biochem. Biophys. Res. Commun., 2002, 293, 1485.
10. C. Winkler, B. Frick, K. Schroecksnadel, H. Schennach and D. Fuchs, Food Chem. Toxicol., 2006, 44, 2003.
11. H. Vally, N. Misso and V. Madan, Clin. Exp. Allergy, 2009, 39, 1643.
12. N. S. Lawrence, J. Davis, L. Jiang, T. G. Jones, S. N. Davies and R. G. Compton, Electroanalysis, 2000, 12, 1453.
13. J. Radford-Knoery and G. A. Cutter, Anal. Chem., 1993, 65, 976.
14. M. Ishigami, K. Hiraki, K. Umemura, Y. Ogasawara, K. Ishii and H. Kimura, Antioxid. Redox Signaling, 2009, 11, 205.
15. A. R. Lippert, E. J. New and C. J. Chang, J. Am. Chem. Soc., 2011, 133, 10078.
16. L. A. Montoya and M. D. Pluth, Chem. Commun., 2012, 48, 4767.
17. K. Sasakura, K. Hanaoka, N. Shibuya, Y. Mikami, Y. Kimura, T. Komatsu, T. Ueno, T. Terai, H. Kimura and T. Nagano, J. Am. Chem. Soc., 2011, 133, 18003.
18. C. Liu, J. Pan, S. Li, Y. Zhao, L. Y. Wu, C. E. Berkman, A. R. Whorton and M. Xian, Angew. Chem., 2011, 123, 10511.
19. X. Cao, W. Lin, K. Zheng and L. He, Chem. Commun., 2012, 48, 10529.
20. M. G. Choi, J. Hwang, S. Eor and S.-K. Chang, Org. Lett., 2010, 12, 5624.
21. H. Xie, F. Zeng, C. Yu and S. Wu, Polym. Chem., 2013, 4, 5416.
22. Y.-Q. Sun, J. Liu, J. Zhang, T. Yang and W. Guo, Chem. Commun., 2013, 49, 2637.
23. Y. Sun, C. Zhong, R. Gong, H. Mu and E. Fu, J. Org. Chem., 2009, 74, 7943.
24. H. Tian, J. Qian, Q. Sun, C. Jiang, R. Zhang and W. Zhang, Analyst, 2014, 139, 3373.
25. Z. Guo, S. Park, J. Yoon and I. Shin, Chem. Soc. Rev., 2014, 43, 16.
26. L. Yuan, W. Lin, S. Zhao, W. Gao, B. Chen, L. He and S. Zhu, J. Am. Chem. Soc., 2012, 134, 13510.
27. N. Karton-Lifshin, E. Segal, L. Omer, M. Portnoy, R. Satchi-Fainaro and D. Shabat, J. Am. Chem. Soc., 2011, 133, 10960.
28. E. Sasaki, H. Kojima, H. Nishimatsu, Y. Urano, K. Kikuchi, Y. Hirata and T. Nagano, J. Am. Chem. Soc., 2005, 127, 3684.
29. S. K. Bae, C. H. Heo, D. J. Choi, D. Sen, E.-H. Joe, B. R. Cho and H. M. Kim, J. Am. Chem. Soc., 2013, 135, 9915.
30. S. J. Mason, J. L. Hake, J. Nairne, W. J. Cummins and S. Balasubramanian, J. Org. Chem., 2005, 70, 2939.
31. G. A. Reynolds and K. H. Drexhage, J. Org. Chem., 1977, 42, 885.
32. J. Furne, A. Saeed and M. D. Levitt, Am. J. Physiol.: Regul., Integr. Comp. Physiol., 2008, 295, 1479.
33. J. C. Mathai, A. Missner, P. Kügler, S. M. Saparov, M. L. Zeidel, J. K. Lee and P. Pohl, Proc. Natl. Acad. Sci. U. S. A., 2009, 106, 16633.
34. M. P. Lutolf, N. Tirelli, S. Cerritelli, L. Cavalli and J. A. Hubbell, Bioconjugate Chem., 2001, 12, 1051.
35. M. Sun, H. Yu, K. Zhang, Y. Zhang, Y. Yan, D. Huang and S. Wang, Anal. Chem., 2014, 86, 9381.
36. Q. Wan, S. Chen, W. Shi, L. Li and H. Ma, Angew. Chem., Int. Ed., 2014, 53, 10916.
37. Y. Li, Y. Wang, S. Yang, Y. Zhao, L. Yuan, J. Zheng and R. Yang, Anal. Chem., 2015, 87, 2495.
38. U. Schatzschneider, J. Niesel, I. Ott, R. Gust, H. Alborzinia and S. Wölfl, ChemMedChem, 2008, 3, 1104.
39. C. Zhang, L. Wei, C. Wei, J. Zhang, R. Wang, Z. Xi and L. Yi, Chem. Commun., 2015, 51, 7505.
40. Y. Qian, J. Karpus, O. Kabil, S.-Y. Zhang, H.-L. Zhu, R. Banerjee, J. Zhao and C. He, Nat. Commun., 2011, 2, 495.

---

Footnote

† Electronic supplementary information (ESI) available: Synthesis, additional absorption and fluorescence spectral details, calculation of the detection limit, data for investigation of sensing mechanisms, effect of pH, cytotoxicity of various concentrations of the probe to cell, and ¹H-NMR, ¹³C-NMR, HRMS, ESI-MS, FT-IR spectra. See DOI: 10.1039/c6ra15065a

---