Coumarin-Based Fluorescence Probe for Differentiated Detection of Biothiols and Its Bioimaging in Cells

In this work, a coumarin derivative, SWJT-14, was synthesized as a fluorescence probe to distinguish cysteine (Cys), homocysteine (Hcy) and glutathione (GSH) in aqueous solutions. The detection limit of Cys, Hcy and GSH for the probe was 0.02 μM, 0.42 μM and 0.92 μM, respectively, which was lower than biothiols in cells. The probe reacted with biothiols to generate different products with different conjugated structures. Additionally, it could distinguish Cys, Hcy and GSH using fluorescence and UV-Vis spectra. The detection mechanism was confirmed by MS. SWJT-14 was successfully used in cellular experiments and detected both endogenous and exogenous biothiols.

A variety of methods have been developed for the detection of biothiols [18][19][20][21], most of which have the disadvantages of being inconvenient, time-consuming and nonspecific. Fluorescence detection technology based on fluorescence probes can take images of analytes in situ and has the advantages of simple operation and low cost; thus, it developed rapidly. In order to target the identification of Cys, Hcy and GSH, the general approach is to exploit the nucleophilic nature of biothiols. Therefore, some organic reactions have been used to design chemodosimeters as fluorescent probes, such as the cyclization of aldehydes [22,23], Michael addition [24,25], conjugate addition cyclization with acrylates [26,27], aromatic nucleophilic substitution-rearrangement reactions [28], cleavage of sulphonamides and sulphonates [29,30], as well as metal complexation [31,32] and dithiol breakage reactions [33,34]. However, most of the reported probes can only identify one biothiol or do not discriminate between the three biothiols. Thus, chemodosimeters with multisite recognition have been developed as fluorescence probes to discriminate biothiols [35][36][37][38][39][40] (Table S1, Supplementary Materials). Although these probes were good differentiators for the three biothiols, differentiating them in the high water phase remained a great challenge.
In connection with our continuing research on biothiols [41][42][43], herein we report a coumarin derivative as a fluorescence probe to distinguish three biothiols in an aqueous solution. Sodium 2-mercaptoethanesulfonate was introduced as a leaving group on coumarin, which can increase the aqueous solubility of the probe. It also synergistically provided multiple reaction sites with double bonds. The reaction of the probe with three biothiols could produce different products with different conjugated structures (Scheme 1), resulting in different UV-Vis changes and fluorescence changes at different excitation wavelengths for the purpose of the identification of the three biothiols. The reaction mechanism was confirmed by mass spectrometry and TD-DFT calculations. Furthermore, the probe could be used to image biothiols in HeLa cells.
differentiators for the three biothiols, differentiating them in the high water phase remained a great challenge.
In connection with our continuing research on biothiols [41][42][43], herein we report a coumarin derivative as a fluorescence probe to distinguish three biothiols in an aqueous solution. Sodium 2-mercaptoethanesulfonate was introduced as a leaving group on coumarin, which can increase the aqueous solubility of the probe. It also synergistically provided multiple reaction sites with double bonds. The reaction of the probe with three biothiols could produce different products with different conjugated structures (Scheme 1), resulting in different UV-Vis changes and fluorescence changes at different excitation wavelengths for the purpose of the identification of the three biothiols. The reaction mechanism was confirmed by mass spectrometry and TD-DFT calculations. Furthermore, the probe could be used to image biothiols in HeLa cells.

Synthesis of Probe SWJT-14
The synthetic route was shown in Scheme 2. Compounds 2-4 were obtained based on the published literature [36,44,45]. A solution of sodium 2-mercaptoethanesulfonate (52 mg, 0.317 mmol) in 1 mL water was added to a solution of 4 (100 mg, 0.267 mmol) in 4 mL acetonitrile and stirred for 5 min, then two drops of triethylamine were added to the mixture. After stirring at room temperature for 15 min, the solvents were removed by evaporation and the residue was purified by column chromatography (DCM:MeOH = 15:1) to obtain red solids (87 mg, yield: 65%). 1
Biosensors 2023, 13, x FOR PEER REVIEW 2 of 10 differentiators for the three biothiols, differentiating them in the high water phase remained a great challenge. In connection with our continuing research on biothiols [41][42][43], herein we report a coumarin derivative as a fluorescence probe to distinguish three biothiols in an aqueous solution. Sodium 2-mercaptoethanesulfonate was introduced as a leaving group on coumarin, which can increase the aqueous solubility of the probe. It also synergistically provided multiple reaction sites with double bonds. The reaction of the probe with three biothiols could produce different products with different conjugated structures (Scheme 1), resulting in different UV-Vis changes and fluorescence changes at different excitation wavelengths for the purpose of the identification of the three biothiols. The reaction mechanism was confirmed by mass spectrometry and TD-DFT calculations. Furthermore, the probe could be used to image biothiols in HeLa cells.

Synthesis of Probe SWJT-14
The synthetic route was shown in Scheme 2. Compounds 2-4 were obtained based on the published literature [36,44,45]. A solution of sodium 2-mercaptoethanesulfonate (52 mg, 0.317 mmol) in 1 mL water was added to a solution of 4 (100 mg, 0.267 mmol) in 4 mL acetonitrile and stirred for 5 min, then two drops of triethylamine were added to the mixture. After stirring at room temperature for 15 min, the solvents were removed by evaporation and the residue was purified by column chromatography (DCM:MeOH = 15:1) to obtain red solids (87 mg, yield: 65%). 1  A solution of sodium 2-mercaptoethanesulfonate (52 mg, 0.317 mmol) in 1 mL water was added to a solution of 4 (100 mg, 0.267 mmol) in 4 mL acetonitrile and stirred for 5 min, then two drops of triethylamine were added to the mixture. After stirring at room temperature for 15 min, the solvents were removed by evaporation and the residue was purified by column chromatography (DCM:MeOH = 15:1) to obtain red solids (87 mg, yield: 65%). 1 13

The Spectral Responses of SWJT-14 for Biothiols in Aqueous Solution
The absorption and fluorescence spectra of SWJT-14 in different solvents were used for the research (Table S2, Figures S4 and S5, Supplementary Materials). The absorption of SWJT-14 in most solutions did not show recognition for biothiols. However, there was a clear distinction between the probe and the three biothiols in HEPES, Tris and PBS buffer solutions. In HEPES and Tris buffer solutions, the absorption peaks of the probe and GSH were almost the same. However, in the PBS buffer solution, SWJT-14 exhibited a maximum absorption peak at 504 nm. Upon the addition of Cys, Hcy or GSH to the solution, the maximum absorption peak blue shifted to 380 nm, 460 nm and 490 nm, respectively. For fluorescence spectra, when the excitation wavelength was 380 nm, the fluorescence of the probe increased after the addition of three biothiols in HEPES, Tris and PBS buffer solutions. Then, the pH effect on the probe was measured. As shown in Figure S6 (Supplementary Materials), the SWJT-14 and SWJT-14 + biothiols system had good fluorescence in aqueous solutions and exhibited good stability from pH 2.0 to 11.0. Considering the application in biological systems, the PBS solution with pH 7.40 was selected for the subsequent experiment.
In order to study the SWJT-14 for three biothiols in detail, UV-Vis spectra ( Figure S7, Supplementary Materials) and color changes ( Figure 1) in PBS buffer solution were used. UV-Vis spectra of the probe + Cys system showed that the absorption peak at 504 nm decreased, accompanied by a new peak at 380 nm, enhanced within 20 min ( Figure S7a a clear distinction between the probe and the three biothiols in HEP solutions. In HEPES and Tris buffer solutions, the absorption peaks were almost the same. However, in the PBS buffer solution, SWJT mum absorption peak at 504 nm. Upon the addition of Cys, Hcy o the maximum absorption peak blue shifted to 380 nm, 460 nm and For fluorescence spectra, when the excitation wavelength was 380 n the probe increased after the addition of three biothiols in HEPE solutions. Then, the pH effect on the probe was measured. As sho plementary Materials), the SWJT-14 and SWJT-14 + biothiols syst cence in aqueous solutions and exhibited good stability from pH 2 the application in biological systems, the PBS solution with pH 7.4 subsequent experiment.
In order to study the SWJT-14 for three biothiols in detail, UV-Supplementary Materials) and color changes ( Figure 1) in PBS buff UV-Vis spectra of the probe + Cys system showed that the absor decreased, accompanied by a new peak at 380 nm, enhanced with Supplementary Materials). For the probe + Hcy system, the absor decreased, accompanied by new peaks at 460 nm and 375 nm, en ( Figure S7b   Then, the fluorescence of SWJT-14 for the three biothiols was studied in detail. The emission peak of SWJT-14 itself at different excitation wavelengths showed almost no difference, and the photostability of the probe was good ( Figure S8, Supplementary Materials). SWJT-14 showed weak fluorescence at 584 nm, the quantum yield φ [46,47] was 0.32% (λ ex = 490 nm), 0.26% (λ ex = 460 nm) and 0.06% (λ ex = 380 nm). As shown in Figure 2a, compared with other analytes, upon the addition of Cys to the solution, the fluorescence enhanced at 470 nm (λ ex = 380 nm, φ = 2.8%) and Hcy and GSH made weak emission peaks at 467 nm and 481 nm, respectively. Notably, when the excitation wavelength was changed to 460 nm, the SWJT-14 + Hcy system had a distinct emission wavelength at  (Figure 2b). When the excitation wavelength was 490 nm, GSH exhibited a maximum emission peak at 553 nm (φ = 1.67%) (Figure 2c). As shown in Figure 2d, the presence of other analytes had no effect on the recognition of Cys by SWJT-14. Due to different reaction rates, there was, to a certain extent, an effect on the recognition of Hcy or GSH by SWJT-14 (Figure 2e,f). These results showed that SWJT-14 could detect the three biothiols in fluorescence mode, and combined with UV and naked-eye recognition, the three biothiols could be distinguished.
Biosensors 2023, 13, x FOR PEER REVIEW peaks at 467 nm and 481 nm, respectively. Notably, when the excitation wavelen changed to 460 nm, the SWJT-14 + Hcy system had a distinct emission wavelengt nm (Ф = 2.3%) (Figure 2b). When the excitation wavelength was 490 nm, GSH exh maximum emission peak at 553 nm (Ф = 1.67%) (Figure 2c). As shown in Figure presence of other analytes had no effect on the recognition of Cys by SWJT-14 different reaction rates, there was, to a certain extent, an effect on the recognition or GSH by SWJT-14 (Figure 2e,f). These results showed that SWJT-14 could de three biothiols in fluorescence mode, and combined with UV and naked-eye reco the three biothiols could be distinguished. To further investigate the change in fluorescence intensity with increasing co tions of biothiols, titration experiments were made. As shown in Figure S9   To further investigate the change in fluorescence intensity with increasing concentrations of biothiols, titration experiments were made. As shown in Figure S9 (Supplementary Materials), with the addition of biothiols to the solution, the fluorescence intensity of SWJT-14 was enhanced 311-fold for Cys at 470 nm (λ ex = 380 nm), 13.7-fold for Hcy at 550 nm (λ ex = 460 nm) and 2-fold for GSH at 553 nm (λ ex = 490 nm). As shown in Figure S10 (Supplementary Materials), SWJT-14 had a good linear relationship with biothiols at low concentrations, thus the detection limits were calculated to be 0.0287 µM for Cys, 0.422 µM for Hcy and 0.924 µM for GSH [43]. The kinetic studies of SWJT-14 with the three biothiols were performed at 37 • C using the corresponding excitation wavelengths. As shown in Figure S11 (Supplementary Materials), the reaction process conformed to the pseudo first-order kinetic equation. The slope of the linear fit was the rate constant, which was calculated to be 10.98 min −1 for Cys, 0.326 min −1 for Hcy, and 0.442 min −1 for GSH [48].

Proposed Mechanism
According to the proposed mechanism [40,45], SWJT-14 would react with biothiols to obtain the thiol derivatives, which were formed by replacing the mercaptosulfonic acid from the probe with sulfhydryl groups of biothiols, which quickly underwent intramolecular rearrangement to form a Schiff base (Figure 3). The exposed sulfhydryl groups further underwent a Michael addition reaction and formed structurally similar polycyclic products with Cys and Hcy. The difference in reaction rate was attributed to different potential space resistances. For GSH, the product underwent thiol coumarin and formed Schiff bases due to the large size and strong spatial site resistance of GSH. As shown in Figure S12 (Supplementary Materials), the peak of m/z 458.08 in MS was confirmed as a heptacyclic coumarin derivative was obtained after the reaction of SWJT-14 with Cys. The peaks of m/z 472.13 and m/z 533. 26 were the corresponding products of Hcy and GSH with SWJT-14, respectively (Figures S13 and S14, Supplementary Materials). These results indicated that different products were obtained, which was the origin of different spectral changes. The precise mechanism is still under study.
PEER REVIEW 5 of 10 biothiols were performed at 37 °C using the corresponding excitation wavelengths. As shown in Figure S11 (Supplementary Materials), the reaction process conformed to the pseudo first-order kinetic equation. The slope of the linear fit was the rate constant, which was calculated to be 10.98 min -1 for Cys, 0.326 min -1 for Hcy, and 0.442 min -1 for GSH [48].

Proposed Mechanism
According to the proposed mechanism [40,45], SWJT-14 would react with biothiols to obtain the thiol derivatives, which were formed by replacing the mercaptosulfonic acid from the probe with sulfhydryl groups of biothiols, which quickly underwent intramolecular rearrangement to form a Schiff base (Figure 3). The exposed sulfhydryl groups further underwent a Michael addition reaction and formed structurally similar polycyclic products with Cys and Hcy. The difference in reaction rate was attributed to different potential space resistances. For GSH, the product underwent thiol coumarin and formed Schiff bases due to the large size and strong spatial site resistance of GSH. As shown in Figure S12 (Supplementary Materials), the peak of m/z 458.08 in MS was confirmed as a heptacyclic coumarin derivative was obtained after the reaction of SWJT-14 with Cys. The peaks of m/z 472.13 and m/z 533. 26 were the corresponding products of Hcy and GSH with SWJT-14, respectively ( Figures S13 and S14, Supplementary Materials). These results indicated that different products were obtained, which was the origin of different spectral changes. The precise mechanism is still under study. To investigate the relationship between spectra changes and structure, Gaussian calculations were performed. The optimized structures used the B3LYP/6-31G (d, p) basis set [49]. As shown in Figure 4, the electron density of LUMO was mainly distributed over coumarin and the adjacent ethyl cyanoacetate group. When the probe reacted with Cys or Hcy to obtain 5 or 6, the decrease in electron density of the coumarin fraction indicated that the conjugated system was reduced and therefore higher energy excitation wave- To investigate the relationship between spectra changes and structure, Gaussian calculations were performed. The optimized structures used the B3LYP/6-31G (d, p) basis set [49]. As shown in Figure 4, the electron density of LUMO was mainly distributed over coumarin and the adjacent ethyl cyanoacetate group. When the probe reacted with Cys or Hcy to obtain 5 or 6, the decrease in electron density of the coumarin fraction indicated that the conjugated system was reduced and therefore higher energy excitation wavelengths were required. Compared with 5 or 6, the electron density of 7 was similar to that of the probe, and therefore required a similar excitation wavelength to the probe. For 5, 6 and 7, the larger energy gap indicated a blue shift in absorption wavelength, which was consistent with the observed experimental results.

Bioimaging in Living Cells
To further investigate the imaging of the SWJT-14 inside living cells, HeLa cells were selected for the experiments. As shown in Figure 5(a1-a4), faint fluorescence in the blue, green and red channels was observed after 40 min of co-incubation of the SWJT-14 with HeLa cells. After adding 200.0 μM of NEM (N-Ethylmaleimide, thiol blocking reagent) to block endogenous biothiols, the fluorescence of SWJT-14 itself in the red channel was shown ( Figure 5(b1-b4)). When Cys was added to the cells that were shielded from endogenous biothiols, and incubated for 30 min, followed by incubation with SWJT-14 for 40 min, weak fluorescence in the blue channel and almost no fluorescence in the red or green channels was observed ( Figure 5(c1-c4)). Hcy ( Figure 5(d1-d4)) or GSH ( Figure  5(e1-e4)) were added in the same way and showed fluorescence in the blue and green or green channels, while there was almost no fluorescence in the red channel. These results indicated that SWJT-14 could detect exogenous biothiols in living cells, which was consistent with the fluorescence detection results.

Bioimaging in Living Cells
To further investigate the imaging of the SWJT-14 inside living cells, HeLa cells were selected for the experiments. As shown in Figure 5(a1-a4), faint fluorescence in the blue, green and red channels was observed after 40 min of co-incubation of the SWJT-14 with HeLa cells. After adding 200.0 µM of NEM (N-Ethylmaleimide, thiol blocking reagent) to block endogenous biothiols, the fluorescence of SWJT-14 itself in the red channel was shown ( Figure 5(b1-b4)). When Cys was added to the cells that were shielded from endogenous biothiols, and incubated for 30 min, followed by incubation with SWJT-14 for 40 min, weak fluorescence in the blue channel and almost no fluorescence in the red or green channels was observed ( Figure 5(c1-c4)). Hcy ( Figure 5(d1-d4)) or GSH ( Figure 5(e1-e4)) were added in the same way and showed fluorescence in the blue and green or green channels, while there was almost no fluorescence in the red channel. These results indicated that SWJT-14 could detect exogenous biothiols in living cells, which was consistent with the fluorescence detection results.

Conclusions
A coumarin derivative as a fluorescent probe was designed and synthesized. It can react with biothiols in aqueous solutions to generate products with different conjugated structures, resulting in different emission peaks at different excitation wavelengths, which achieved the purpose of differential detection of Cys, Hcy and GSH. The fluorescence imaging of SWJT-14 had also been demonstrated in living cells.

Conclusions
A coumarin derivative as a fluorescent probe was designed and synthesized. It can react with biothiols in aqueous solutions to generate products with different conjugated structures, resulting in different emission peaks at different excitation wavelengths, which achieved the purpose of differential detection of Cys, Hcy and GSH. The fluorescence imaging of SWJT-14 had also been demonstrated in living cells.