A novel near-infrared fluorescent probe for highly selective recognition of hydrogen sulfide and imaging in living cells

A novel near-infrared fluorescent probe (L) based on a 1,4-diethyl-1,2,3,4-tetrahydro-7H-pyrano[2,3-g]quinoxalin-7-one scaffold has been synthesized and characterized. Probe L displays highly selective and sensitive recognition to H2S over various anions and biological thiols with a large Stokes shift (125 nm) in THF/H2O (6/4, v/v, Tris–HCl 10 mM, pH = 7.4). This probe exhibits turn-on fluorescence for H2S through HS− induced thiolysis of dinitrophenyl ether. Confocal laser scanning micrographs of MCF-7 cells incubated with L confirm that L is cell-permeable and can successfully detect H2S in living cells.


Introduction
Hydrogen sulde (H 2 S), a newly identied gaseous signalling molecule, has recently become a research focus in biological elds due to its multiple functions in physiological and pathological processes. 1 Previous studies have demonstrated that endogenous H 2 S can affect the functions of neuronal, cardiovascular, immune, endocrine, and gastrointestinal systems and its antioxidant and anti-apoptotic signaling effects also show therapeutic benet for the treatment of ischemia-induced heart failure. 2 Excessive H 2 S can irritate the eyes and respiratory tract, causing severe loss of consciousness, respiratory failure, and even death. 3 The concentration-dependent effects of H 2 S on health and disease demand precise methods to track the production of this important signaling molecule in living organisms. 4 In view of the biological importance of H 2 S, several methods such as colorimetry, 5-8 electrochemistry, [9][10][11] and gas chromatography 12 have been reported for H 2 S detection. Compared with these traditional methods, uorescence analysis is widely used because of its advantages such as good selectivity, high sensitivity, simplicity, rapidness, and nondestructive bioimaging. 13 Much attention has been given by chemists and biologists to develop various uorescent chemosensors that would allow real-time tracking of a small molecule of interest in living cells and animals. 14 In recent years, a number of uorescent H 2 S probes based on different design strategies have been developed, including using nucleophilic substitution reaction, 15-17 chemoselective reduction of azide to amine, 18-20 Cu 2+ -uorophore complex 21-23 through indicator displacement assays. However, most of the reported H 2 S uorescent probes still have some limitations, such as short emission wavelength, working in pure organic solvent, poor biological applications, etc. In fact, uorescent probes with near-infrared (NIR) emission are more suitable for bioimaging applications due to their deep tissue penetration, minimum photo damage and background autouorescence interference. [24][25][26][27][28][29][30][31][32] Therefore, it is important to design and synthesize NIR uorescent probe for selective and sensitive detection of H 2 S. [33][34][35] In view of this, we judiciously designed an "off-on" uorescence probe (L) for H 2 S detection with NIR emission. Coumarinyl chalcone (1) was selected as the NIR emitting uorophore owing to its high molar absorption coefficient and long-wavelength emission. 36 The incorporated 2,4dinitrophenyl ether moiety in L was anticipated to function as both H 2 S recognition site and uorescence quencher. [37][38][39][40][41] Aer reaction with H 2 S, a uorescence turn-on response would occur through HS À -triggered thiolysis of dinitrophenyl ether to release 1. Further investigations demonstrate that the proposed uorescent probe L displays high selectivity and sensitivity to H 2 S with fast response and NIR emission, and L is applicable to image H 2 S in living cells. To the best of our knowledge, NIR uorescence turnon H 2 S probe based on thiolysis of the dinitrophenyl ether were rarely reported.

Experimental
Instruments and materials 1 H NMR and 13 C NMR spectra were measured on an Agilent 400 MR spectrometer, and the chemical shis were expressed in ppm and coupling constants (J) in hertz. High-resolution mass spectra (HRMS) were measured using a Bruker micrOTOF-Q mass spectrometer (Bruker Daltonik, Bremen, Germany). Fluorescence measurements were carried out on a Sancho 970-CRT spectrouorometer (Shanghai, China). UVvis absorption spectra were measured with a SP-1900 spectrophotometer (Shanghai Spectrum Instruments Co., Ltd., China). pH measurements were made with a Model PHS-25 Bmeter (Shanghai, China). Cell imaging was observed under a confocal laser scanning microscope (LEICA TCS SP5 II, Germany) with excitation at 405 nm. Fluorescence quantum yields were measured with an absolute uorescence quantum yield spectrometer (Quantaurus-QY C11347, Hamamatsu Photonics).
Unless otherwise noted, reagents were purchased from commercial suppliers and used without further purication. 3 was prepared according to the reported method 42 (ESI †). MCF-7 (human breast carcinoma) cells were obtained from Institute of Basic Medical Sciences (IBMS) of Chinese Academy of Medical Sciences (CAMS).

General methods
Stock solutions (50 mM) of anions (sodium or potassium salts), NaHS used as H 2 S source, and biological thiols (glutathione (GSH), homocysteine (Hcy) and cysteine (Cys)) were prepared in double-distilled water. Stock solution of L (10 mM) was prepared in DMSO and was further diluted with a mixed solution of THF/H 2 O (v/v, 6 : 4 Tris-HCl 10 mM, pH ¼ 7.4) to make a nal concentration at 10 mM. Fluorescence spectra were measured using 10 mM solution of L aer 30 minutes upon addition of anions by excitation at 500 nm. The excitation and emission slit widths were 10 and 10 nm, respectively. All titration experiments were carried out at room temperature. Double-distilled water was used throughout the experiments.

Synthesis of compound 2
Compound 3 (468.2 mg, 2.0 mmol), acetylacetic ether (260.4 mg, 2.0 mmol), and piperidine (0.1 mL) were dissolved in dry ethanol, the mixture was heated to reuxed for 4 h, then it was poured into hydrochloric acid (4 M, 10 mL) and stirred at 60 C for 30 minutes. Aer that, the mixture was further diluted with H 2 O and extracted with ethyl acetate, and organic layer was dried with anhydrous Na 2 SO 4 and removed under reduced pressure. The residue was puried by column chromatography with petroleum ether/ethyl acetate (5 : 1, v/v) as eluent to give compound 2 (420.5 mg, 70%). 1

Synthesis of compound 1
Compound 2 (300.4 mg, 1.0 mmol), p-hydroxybenzaldehyde (134.2 mg, 1.1 mmol), and piperidine (0.1 mL) were dissolved by toluene, the mixture was heated to reuxed for 4 h, then it was poured into hydrochloric acid (4 M, 10 mL) and stirred for 50 min at 60 C. Aer that, water were added into the mixture, then mixture was extracted with ethyl acetate (3 Â 100 mL), and combined organic layer was dried with anhydrous Na 2 SO 4 and removed under reduced pressure. The crude products were puried by column chromatography with petroleum ether/ethyl acetate (1 : 1, v/v) as eluent to give compound 1 (199. 8

Synthesis of probe L
Compound 1 (404.5 mg, 0.5 mmol), 2,4-dinitrouorobenzene (186.0 mg, 0.6 mmol), potassium carbonate (187 mg, 1.0 mmol) were dissolved into DMF (5.0 mL), the mixture was stirred for 3 h at room temperature. Aer that, water was added and then the mixture was extracted with ethyl acetate, organic layer was dried with anhydrous Na 2 SO 4 . The crude products was puried by column chromatography with petroleum ether/ dichloromethane (1 : 4, v/v) as eluent to give compound L

Cell viability assays
The cytotoxicity of the probe was examined by MTT assay. MCF-7 cells with 90% conuence was digested by 0.25% of trypsin, and transferred into 96-well plates with a cell suspension of 5000 cells per hole. The cells were incubated for 24 h at 37 C, and then different concentrations of L (0, 1.0, 5.0, 10, 20, 40 mM) were added to the 96-well plates. Aer 24 h incubation at the same condition, 10 mL MTT was added and incubated for another 4 h. The cells were washed twice with PBS (1 mL) and dissolved with DMSO (1 mL). The absorbance was recorded at 490 nm using the microplate spectrophotometer system, and each experiment was run in triplicate. Cell viability was calculated based on the equation: Cell viability (%) ¼ (A with probe À A blank )/(A control À A blank ) Â 100%.

Cell culture and confocal image
Similar to the cytotoxicity assay and incubation conditions, MCF-7 cells of proper density were transferred in confocal dishes. Aer 24 h incubating at 37 C, the cell layer were washed three times with PBS buffer, then the same solution with a concentration of 2 mM L was added into each well and incubated for 30 min. Subsequently, the L treated MCF-7 cells were washed three times with PBS, and then further incubated with different concentrations of NaHS (10, 20, and 50 mM) for 30 min. Aer washing the culture dishes three times with PBS, uorescence imaging experiments were performed using a LEICA TCS SP5 II confocal laser scanning microscope.

Optical responses of L to H 2 S
The specic selectivity of a probe determines its basic performance. Therefore, the optical responses of L toward various anions including F À , Cl À , Br À , I À , NO 2 When L was treated with 100 equiv. of HS À , a strong emission band centered at 650 nm occurred (F ¼ 0.011) and the uorescence color changed from non-uorescence to vivid red ( Fig. 1, inset or S4 †). However, the uorescence emission intensity of L underwent negligible or very slight changes upon addition of other anions and biological thiols to solution of L. These results demonstrate that L possesses high selectivity to H 2 S. In addition, UV-vis absorption experiments were carried out, and the absorption band centered at 525 nm of L (3 ¼ 428 144 M À1 cm À1 ) underwent a blue shi of 8 nm upon addition of HS À (3 ¼ 391 138 M À1 cm À1 ). There was no significant change in the absorption and color of L solution upon addition of other anions and biological thiols (Fig. S5 †). The results indicate that L is hardly to recognize H 2 S via UV-vis measurements. The Stokes shi was found to be 125 nm, such a large Stokes shi and emission wavelength are benecial to biological imaging because they could efficiently minimize self-absorption and reduce the interference from auto-uorescence (Fig. 2). 44 Fluorescence titration experiments of L with HS À were performed to further explore the sensing behaviors of L. With Scheme 1 Synthetic route to probe L.  increasing the added amount of HS À to L solution, the emission intensity of L at 650 nm concomitantly increased and reached a plateau when 100 equiv. of HS À was employed. The intensity of L towards HS À exhibited a good linear correlation with the concentration of HS À ranging from 200 to 550 mM (R 2 ¼ 0.9938). Based on the signal to noise ratio, the detection limit (LOD ¼ 3s/k, s is the standard deviation of the blank solution; k is the slope of the calibration curve) of L to HS À was calculated to be 7.3 Â 10 À7 M (Fig. S6 †), indicating that L was highly sensitive to H 2 S and had a potential applicability for bioimaging of HS À . To assess the specic nature of L toward HS À , competitive experiments were then performed to estimate the availability of L in complicated systems. As illustrated in Fig. 3, HS À could still produce a signicant uorescence enhancement in the presence of co-existing anions or biothiols, indicating the excellent anti-interference ability of L for HS À recognition. At the same time, we also prove that S 2 2À and p-toluenethiol have no interference for recognition of H 2 S (Fig. S7 †), which is benecial to its potential applications in complicated biological systems. In addition, the time-dependent uorescence variations of the probe were also monitored. As shown in Fig. 4, the uorescence intensity of L increased along with the time and reached a maximum within $18 min, which is faster than that of some reported probes (Table S1 †), indicative of a fast response of L for H 2 S detection. For biological applications, pH dependence of L was examined in THF-Tris (6/4, v/v) solution (Fig. 5). Probe L is pH insensitive in a wide pH range from 2 to 12. However, there is signicant uorescence enhancement with addition of HS À when L is within pH range from 3 to 12, which includes the biologically relevant range of pH 4-8, indicating that L is applicable to detect H 2 S in the biological system.
The sensing mechanism of L to H 2 S Preliminary investigations reveal that the uorescence spectrum of L + HS À exhibits an almost identical emission pattern as that of free 1 (Fig. S8 †), suggesting that HS À can completely cleave the dinitrophenyl ether moiety to release 1. [45][46][47] To further corroborate this reaction process, HRMS of the reaction mixture L + HS À was analyzed (Fig. S9 †). The prominent peak observed at m/z ¼ 405.1816 can be ascribed to reaction released compound 1 (Calcd m/z ¼ 405.1815). These results reveal that HS À -triggered thiolysis of dinitrophenyl ether to release uorescent dye 1 indeed happened. The sensing mechanism of L toward H 2 S was proposed in Scheme 2.

Live cell imaging of L to H 2 S
To validate the biological applicability of L, live cell imaging experiments were also performed using MCF-7 cells. Firstly, the cytotoxicity of the probe was evaluated by MTT assay (Fig. S10 †). The results showed that the cell viability was estimated to be >88% at 24 h when L was used less than 10 mM. Thus 2 mM of the probe was selected as non-cytotoxic dose for further studies. Subsequently, MCF-7 cells were incubated with L (2 mM) for 30 min at 37 C and then washed three times with PBS buffer, there was negligible uorescence signal in the red channel (Fig. 6b). When L-pretreated MCF-7 cells were further incubated with different concentrations of H 2 S (5, 10, and 20 mM) for 30 min, the brightness of the observed red uorescence from the red channel gradually increased with increase of H 2 S concentration (Fig. 6e, h and k). The integrated optical density analysis manifests that the uorescence intensity was   signicantly increased aer addition 20 mM of HS À , which has prominent statistically signicance compared with the control group (P > 0.01, Fig. 6m). These results indicate that L possesses good cell permeability and is capable of imaging HS À in MCF-7 cells.

Conclusion
In summary, we reported herein a unique NIR uorescent turnon H 2 S probe based on thiolysis of dinitrophenyl ether. Probe L displays highly selective and sensitive recognition to H 2 S over various anions and biological thiols with a large Stokes shi in THF/H 2 O (6/4, v/v, Tris-HCl 10 mM, pH ¼ 7.4) solution. Moreover, probe L is suitable for uorescence imaging of H 2 S in living MCF-7 cells. Based on the unique uorescence feature, the coumarinyl chalcone conjugate (dye 1) will be a promising platform for the development of various NIR uorescent probes.

Conflicts of interest
There are no conicts to declare.