The synthesis, crystal, hydrogen sulfide detection and cell assement of novel chemsensors based on coumarin derivatives

A series of chemsensors (1–4) containing fluorobenzene group based on coumarin derivatives have been developed for the selective and sensitive detection of H2S. The advantages of the synthesized fluorescent probe (compound 1) were the low detection limit (4 × 10−6 mol·L−1), good selectivity and high sensitivity which had been demonstrated through UV-vis, fluorescent titration experiments. Besides cytotoxicity test of compounds (1 and 2) was studied and the results indicated that compounds (1 and 2) showed almost no cytotoxicityat at a concentration of 150 μg·mL−1. The interacted mechanism was the thiolysis reaction of dinitrophenyl ether which had been confirmed by fluorescence and HRMS titration experiment. In addition, probe 1 can also detect HS− selectively by naked eye in pure DMSO solvent.

In the past decade, we has seen a boost of research interest in hydrogen sulfide (H 2 S), a colorless, flammable, toxic gas with unpleasant smell, which is recognized as a signal gasotransmitter in the body as same as nitric oxide (NO) 1-10 and carbon monoxide (CO) 11 . Endogenous concentration of H 2 S is related to some diseases such as Alzheimer's disease, Down syndrome, liver cirrhosis and diabetes 2,9,[12][13][14][15][16][17] . What's more, the regulation of H 2 S levels is also a potential drug development strategy 18,19 and the importance of accurate detection of H 2 S cannot be over-emphasized. Therefore, it presents significant research related to track and quantify H 2 S inside living cells being crucial in order to understand the biological and pathological roles of H 2 S. Recently, some methods to determine H 2 S concentration in biological sample have been developed including the methylene blue, the monobromobimane (MBB), gas chromatography (GC), the sulphide ion selective electrodes (ISE) and fluorescent analysis [20][21][22][23] . Among these methods, fluorescent analysis has attracted great attention due to the high sensitivity and selectivity for the detection of H 2 S in many fields such as environment area, pharmacy area and so on 24-31 . In the present work, a series of "OFF-ON" probes based on coumarin derivatives to detect HS − (Fig. 1). The results of UV-vis titration experiments indicated that the synthesized compounds showed high binding ability for HS − among the tested anions (NaHS (HS − ), (n-C 4 H 9 ) 4 NAcO (AcO − ), (n-C 4 H 9 ) 4 NH 2 PO 4 (H 2 PO 4 − ), (n-C 4 H 9 ) 4 NF (F − ), (n-C 4 H 9 ) 4 NCl (Cl − ), (n-C 4 H 9 ) 4 NBr (Br − ), (n-C 4 H 9 ) 4 NI (I − )) and amino acids (Glutathione (GSH), Cysteine (Cys), Homocysteine (Hcy)). Besides, four compounds designed and synthesized could exhibit the probes with strong electron-withdrawing groups located in 2,4-positions of fluorobenzene have strong binding ability for HS − detection which provides a good idea for the design of probe in future.

Results and Discussion
X-ray crystallography. Compound 2 was synthesized according to the route shown in Fig. 1. Fortunately, the crystallographic of compound 2 was obtained by the standing method. The suitable single light yellow crystal was obtained by volatilizing ethyl acetate containing compound 2 at room temperature. The details of the crystallographic determination, selected bond lengths and angles were given in Table 1  The crystal of compound 2 suitable for X-ray crystal analysis was obtained and the structure was also confirmed (Fig. 2a). The fluoride atom in benzene cycle forms hydrogen bonds with hydrogen atom (H3) (supplementary material). The overall crystal structure features a chain type joining in through the hydrogen bonds (H….F) along the b axis (Fig. 2b). In the crystal packing of compound 2 (Fig. 2c), there are two stacked forms: (1) π-π stacking of one fluorobenzene ring with another; (2) π-π stacking of one coumarin ring with another, which connected into "chair" conformation along a axis.   The data of UV-vis titration experiments manifested only compounds (1, 2) displayed different binding abilities with the above anions and amino acids. The free 1 showed a main absorption at 320 nm, as the HS − increases in pure DMSO solution of probe 1, the absorbance at 320 nm was decreased gradually, along with the simultaneous emergence of a new absorption at 470 nm. In this process, two isosbestic points noted at 330 nm and 348 nm suggesting a clear chemical reaction. Based on the well-establish thiolysis reaction of dinitrophenyl ether, the new absorption at 470 nm could be attributed to coumarin derivative, which was also supported by fluorescence and HRMS titration experiment. Furthermore, the UV-vis spectra of probe 1 with HS − in aqueous solution were also performed (shown in Fig. 3b), however, comparison with DMSO solvent, the probe 1 showed a weak response of UV-vis spectra. The additions of amino acids (Cys, GSH, Hcy) and other anions (AcO − , H 2 PO 4 − , F − , Br − , Cl − and I − ) to pure DMSO solution of probe 1, only Cys induced similar changes in the UV-vis spectra compared with HS − , which exhibited Cys also interacted with compound 1 (supplementary material). However, the additions of the above amino acids and anions to aqueous solution induced almost no spectra changes of compound 1. The result indicated that compound 1 showed different binding abilities for HS − and Cys in pure DMSO solution and almost no binding abilities with above amino acids and anions in aqueous solution. Therefore, compound 1 could be used as a sensor to detect HS − in aqueous solution.

UV-vis
Subsequently, the colorimetric sensing capabilities of compound 1 were carried out with different anions (Fig. 3c). Obvious color changes from colorless to bright yellow was observed in the presence of HS − , while faint or no color changes happened in the presence of other anions (AcO − , H 2 PO 4 − , F − , Cl − , Br − , I − and Cys), which indicated that compound 1 can be used for the detection of HS − as a colorimetric sensor.
Next   Binding constant. The job-pot curves suggested two compounds (1 and 2) interacted with amino acids and various anions as the ratio of 1:1 or 1:2. The UV-vis spectral data was used to calculate the binding constants by non-linear least square method 34,35 , and the binding constants were listed in the Table 2. Obviously, the binding ability of two compounds (1 and 2) with amino acids and various anions followed the order of HS − ≫ H 2 PO 4 − , Cys, AcO − , F − , Br − ,Cl − , I − , Hcy and GSH. In general, both compound 1 and compound 2 showed the strongest binding ability for HS − among amino acids and anions. Besides, a theoretical basis and these binding constants were necessary for the optimization of sensor.
The anion binding abilities of compounds (1 and 2) with two electron-withdrawing groups on the fluorobenzene ring were stronger than that of compounds (3 and 4) which had one electron-withdrawing group on the fluorobenzene ring. In addition, the anion binding ability of compound 1 was stronger than that of compound 2 due to the electron-withdrawing ability of nitro group was greater than the trifluoromethyl group 36 . The above results indicated that strong electron-withdrawing groups located in 2, 4-positions of fluorobenzene provides a Mechanism. The interacted mechanism was measured by fluorescence and HRMS titration experiment. The broad emission peak of compound 1 appeared at about 396 nm in the absence of HS − . The emission peak shifted to the short wavelength from 396 nm to 380 nm after HS − was added. The emission peak of the single 7-hydroxy-4-methylcoumarin appeared at about 380 nm. The same emission peak observed in the presence of HS − which suggested free 7-hydroxy-4-methylcoumarin was released after compound 1 interacted with HS − (Fig. 5a). Furthermore, the interacted mechanism of host-guest (1-HS − ) was also carried by performing MS-HRMS after the addition of 2 equiv. NaHS. The probe 1 itself exhibited a dominant peak at m/z = 341.0551(M-H) − (supplementary material). However, the above peak vanished and a new peak at m/z = 199.0364 (M + Na) + appeared which was the ion-peak of 7-hydroxy-4-methylcoumarin (Fig. 5b) after the addition of HS − . The MS-HRMS titration experiment suggested the binding ratio of host-guest (1-HS − ) was 1:2. The above results indicated the possible interacted mechanism was thiolysis reaction of dinitrophenyl ether (Fig. 5c) 37-44 . Cytotoxicity Assay. For further biological application point of view, a quantitative cytotoxicity study of two compounds (1 and 2) in vitro was conducted using MTT assay. Owning to Glutathione peroxidase (GPx1) being an important selenoprotein and not found in human breast cancer cells (MCF-7) according to scientific research, therefore, the special cell lines (MCF-7 cell) were selected by us 45,46 . The MTT assay results indicated that compound 1 and 2 showed very low cototoxicity over a concentration range of 0-150 μg·mL −1 especial probe 2 (Fig. 6). Cellular viability was minimally affected (80%, cellular viability) with the compounds (1 and 2) ( <150 μg·mL −1 ). In agreement with the binding constants, compounds (1 and 2) showed a high binding capacity and low cytotoxicity, especial probe 1, which indicated that compounds (1 and 2) have a potential application to H 2 S detection in cells.

Material and Methods
Most staring materials were obtained commercially, all reagents and solvents were of analytical grade and used as received without further purification unless otherwise stated. Sodium hydrolfide and all anions in the form of tetrabutylammonium salts (such as (n-C 4 H 9 ) 4 NCl, (n-C 4 H 9 ) 4 NBr, (n-C 4 H 9 ) 4 NI, (n-C 4 H 9 ) 4 NAcO and (n-C 4 H 9 ) 4 NH 2 PO 4 ) and amino acides (Cys, GSH and Hcy) were purchased from Aladdin Chemistry Co. Ltd (Shanghai, China). Dimethyl sulfoxide (DMSO) was distilled in vacuum after being dried with CaH 2 . 1H NMR spectra were recorded using an Unity Plus-400-MHz spectrometer. HRMS was obtaioned with a Mariner apparatus. Absorption spectra were obtained on UV-vis spectrophotometer (Shimadzu, UV-2600, Japan). Fluorescence emmission spectra were taken on a Cary Eclipse Fluorescence Spectrophotometer (Agilent, USA). The binding constant (K s ) was obtained by non-linear least squares calculation method for data fitting. The cells that were at logarithmic growth phase were seeded in a 96-well plate at a density of 2.0 × 10 4 cell peer well for 24 h, followed by treatment different concentration of compound at 37 °C for additional 24 h. Next, cells were collected and washed with PBS three times then 100 µL of culture medium and 20 µL of MTT solution were added to each well respectively, and the cells were incubated for 4 h. The absorbance were detected by microplate reader (Thermo Multiscan MK3, Thermo Fisher Scientific, MA, USA) at 490 nm wavelength measurement. Cell viability (expressed in%) was calculated considering 100% growth at the absence of fluorescence probe, and the viability of other groups were calculated by comparing the optical density reading with the control. The IC50 was defined as the compound concentrations required for 80% inhibition of cell growth.

Synthesis of 7-hydroxy-4-methylcoumarin
Synthesis of compound 1. 7-Hydroxy-4-methyl coumarin (2.16 g, 12 mmol) and 2,4-dinitrofluorobenzene (1.5 g, 8 mmol), K 2 CO 3 (3.39 g, 24 mmol) were dissolved in DMF(40 mL) solution with stirring. The above mixture were heated for 4 h at 90 °C at N 2 atmosphere and then cooled to room temperature. The reaction was poured into     A light yellow crystal of compound 2 with dimensions of 0.45 nm × 0.32 nm × 0.23 nm was mounted on a glass fiber. X-ray single-crystal diffraction data was collected on a Rigaku saturn CCD area detector at 293 K with Mo-Ka radiation (λ = 0.71073 Å). The structure was solved by direct methods and refined on F 2 by full-matrix least squares methods with SHELXL-97 48 .

Conclusion
In conclusion, we developed a series of fluorescence probes based on coumarin derivatives for the detection of H 2 S successively with OFF-ON" fluorescence response. The fluorescence probes, especially compound 1, exhibited remarkable response to H 2 S against other anions and amino acids in pure DMSO solvent. Otherwise, the probe 1 also showed strong binding ability for HS − in HEPES buffer solution. The interacted mechanism of host-guest was the thiolysis reaction of dinitrophenyl ether. In addition, compound (1 and 2) showed highly sensitivity and low cytotoxicity to MCF-7 cells and probe 1 can also detect HS − selectively by naked eye in pure DMSO solvent. The results of our efforts highlight that the probes, especially compound 1, hold a potential chemical tool for the detection of H 2 S.