A Fast-Response Red Shifted Fluorescent Probe for Detection of H2S in Living Cells

Near-infrared (NIR) fluorescent probes are attractive tools for bioimaging applications because of their low auto-fluorescence interference, minimal damage to living samples, and deep tissue penetration. H2S is a gaseous signaling molecule that is involved in redox homeostasis and numerous biological processes in vivo. To this end, we have developed a new red shifted fluorescent probe 1 to detect physiological H2S in live cells. The probe 1 is based on a rhodamine derivative as the red shifted fluorophore and the thiolysis of 7-nitro 1,2,3-benzoxadiazole (NBD) amine as the H2S receptor. The probe 1 displays fast fluorescent enhancement at 660 nm (about 10-fold turn-ons, k2 = 29.8 M−1s−1) after reacting with H2S in buffer (pH 7.4), and the fluorescence quantum yield of the activated red shifted product can reach 0.29. The probe 1 also exhibits high selectivity and sensitivity towards H2S. Moreover, 1 is cell-membrane-permeable and mitochondria-targeting, and can be used for imaging of endogenous H2S in living cells. We believe that this red shifted fluorescent probe can be a useful tool for studies of H2S biology.


Introduction
Recently biological reports have demonstrated the important role of H 2 S, and the results suggested that endogenously-produced hydrogen sulfide (H 2 S) has been marked as gasotransmitter, allowing the regulation of numerous important physiological functions including; cardiovascular, gastrointestinal, endocrine, nervous, and immune systems [1][2][3]. Generally, endogenous H 2 S can be produced enzymatically from L-cysteine (Cys) by means of three distinctive enzymatic pathways; cystathionine γ-lyase (CSE), cystathionine β-synthetase (CBS), and 3-mercaptopyruvate sulfurtransferase (3-MST) [4]. The interest in the molecular mechanisms of H 2 S associated with physiology and pathology was sparked out of its recognition as a vital signaling molecule. However, the abnormal levels of H 2 S production lead to number of different human diseases including; diabetes [5], Alzheimer's disease [6] liver cirrhosis [7], and the symptoms of Down's syndrome [8][9][10]. As an important role played by H 2 S in tumor biology, it is proposed that the both production and inhibition of H 2 S concentration beyond a NIR dyes, including cyanine (Cy), are considered the classic NIR fluorescent dyes [60]. However, due to their flexible molecular structure, some of NIR fluorescent dyes accompanied some short comings for example, small Stokes shift, limited fluorescence quantum yield, low photo-stability lying, and high occupied molecular orbital (HOMO) energy levels [61,62]. Such photo-physical properties strongly affect the fluorescence signals due to the high background signal, which in turn result in low contrast for bioimaging [63][64][65][66]. In 2017, we developed a Cy-NBD probe (Scheme 1) which has the limitation of low quantum yield after H2S activation [58]. On the other side, classical rhodamine dyes contributed much in the field of biomolecular detection and biomedical imaging because of its magnificent photophysical and chemical properties [30,[67][68][69][70][71]. Due to limited πconjugated system of xanthene core derivatives, such as rhodamine B, rhodamine 6G, and rhodamine 123 have their emission wavelengths in the visible region (<600 nm). Recently, significant advancements have been made in the improvement of rhodamines-based fluorescent dyes with extended the π-conjugated system possessing long emission wavelength, high fluorescence quantum yield, and outstanding photostability [70][71][72][73][74][75]. Herein, we report the development of an extended π- NIR dyes, including cyanine (Cy), are considered the classic NIR fluorescent dyes [60]. However, due to their flexible molecular structure, some of NIR fluorescent dyes accompanied some short comings for example, small Stokes shift, limited fluorescence quantum yield, low photo-stability lying, and high occupied molecular orbital (HOMO) energy levels [61,62]. Such photo-physical properties strongly affect the fluorescence signals due to the high background signal, which in turn result in low contrast for bioimaging [63][64][65][66]. In 2017, we developed a Cy-NBD probe (Scheme 1) which has the limitation of low quantum yield after H 2 S activation [58]. On the other side, classical rhodamine dyes contributed much in the field of biomolecular detection and biomedical imaging because of its magnificent photophysical and chemical properties [30,[67][68][69][70][71]. Due to limited π-conjugated system of xanthene core derivatives, such as rhodamine B, rhodamine 6G, and rhodamine 123 have their emission wavelengths in the visible region (<600 nm). Recently, significant advancements have been made in the improvement of rhodamines-based fluorescent dyes with extended the π-conjugated system possessing long emission wavelength, high fluorescence quantum yield, and outstanding photostability [70][71][72][73][74][75]. Herein, we report the development of an extended π-conjugated rhodamine-NBD based probe 1 for the highly selective imaging of endogenous H 2 S in a red shifted region [70,71]. The probe is high selectivity towards H 2 S among other biothiols, with fluorescence emission in the red shifted region (>660 nm) and high fluorescence quantum yield (0.29) after H 2 S activation. The probe is successfully used for bioimaging of endogenous H 2 S in living cells.

Synthesis of 1
Probe 1 was constructed using a three-step route with a good yield (Scheme 2). By using the procedure described in the literature [76], 6-(dimethylamino)-3,4-dihydronaphthalen-1 (2H)-one 2 was first synthesized from commercially available 6-amino-3,4-dihydronaphthalen-1(2H)-one, which was then transformed to 3. Finally, the probe 1 in 79% yield, was prepared by the coupling of compound 3 with NBD-piperazine. The facile and economic synthesis is important for the wide use of the probe. The structure of compound 1 was confirmed by 1 H NMR, 13 C NMR, and high resolution mass spectrum (HRMS). The spectra ( Figure S8) are included in the Supplemental Information.
Molecules 2020, 25, 437 3 of 13 conjugated rhodamine-NBD based probe 1 for the highly selective imaging of endogenous H2S in a red shifted region [70,71]. The probe is high selectivity towards H2S among other biothiols, with fluorescence emission in the red shifted region (>660 nm) and high fluorescence quantum yield (0.29) after H2S activation. The probe is successfully used for bioimaging of endogenous H2S in living cells.

Synthesis of 1
Probe 1 was constructed using a three-step route with a good yield (Scheme 2). By using the procedure described in the literature [76], 6-(dimethylamino)-3,4-dihydronaphthalen-1 (2H)-one 2 was first synthesized from commercially available 6-amino-3,4-dihydronaphthalen-1(2H)-one, which was then transformed to 3. Finally, the probe 1 in 79% yield, was prepared by the coupling of compound 3 with NBD-piperazine. The facile and economic synthesis is important for the wide use of the probe. The structure of compound 1 was confirmed by 1 H NMR, 13 C NMR, and high resolution mass spectrum (HRMS). The spectra ( Figure S8) are included in the Supplemental Information.

UV-Vis and Fluorescence Response of 1 towards H2S
With the probe in hand, we first tested the solubility of 1 in buffer solution. The linearity of 1 verified its good solubility up to 20 µM ( Figure S1). Further, we tested the optical properties of 1, and the absorbance and emission profiles are illustrated in Figure 1. As shown in Figure 1A, 1 displayed UV absorbance maxima at 620 nm and 500 nm, which are assigned to the rhodamine and NBD absorbance respectively. Since Na2S is a well-known inorganic H2S donor that is widely employed in the study of H2S effects on physiology, we used it as a H2S equivalent [77]. When reacted with H2S, the increase in intensity of absorbance peaks appeared between 600 nm and 520 nm, which could be assigned to the yielding of 4 and NBD-SH. The reaction between 1 and 500 µM H2S in PBS buffer (50 mM, pH 7.4) finished within 5 min. Furthermore, such thiolysis reaction was characterized by NMR with the formation of NBD-SH peaks ( Figure S7) and HRMS with the production of peak at 535.3070 (calculated value for [4] + : 535.3068) ( Figure S2). Scheme 2. Synthetic route for probe 1 and its reaction with H 2 S. Reagents and conditions: a) 6-aminotetralone, CH 3 I, K 2 CO 3 , DMF, 0 • C, 24 h, 53%. b) 2-(4-diethylamino-2-hydroxybenzoyl)-benzoic acid, N-substituent ketone, H 2 SO 4 0-100 • C, 2 h, 80%. c) NBD-piperazine, HATU, DIPEA, DMF, 79%.

UV-Vis and Fluorescence Response of 1 towards H 2 S
With the probe in hand, we first tested the solubility of 1 in buffer solution. The linearity of 1 verified its good solubility up to 20 µM ( Figure S1). Further, we tested the optical properties of 1, and the absorbance and emission profiles are illustrated in Figure 1. As shown in Figure 1A, 1 displayed UV absorbance maxima at 620 nm and 500 nm, which are assigned to the rhodamine and NBD absorbance respectively. Since Na 2 S is a well-known inorganic H 2 S donor that is widely employed in the study of H 2 S effects on physiology, we used it as a H 2 S equivalent [77]. When reacted with H 2 S, the increase in intensity of absorbance peaks appeared between 600 nm and 520 nm, which could be assigned to the yielding of 4 and NBD-SH. The reaction between 1 and 500 µM H 2 S in PBS buffer (50 mM, pH 7.4) finished within 5 min. Furthermore, such thiolysis reaction was characterized by NMR with the formation of NBD-SH peaks ( Figure S7) and HRMS with the production of peak at 535.3070 (calculated value for [4] + : 535.3068) ( Figure S2). The probe 1 showed weak fluorescence (quantum yield , 0.021) upon excitation at 620 nm, indicating that fluorescence in 1 could be mainly quenched by the photoinduced electron transfer process (PET) effect from the NBD moiety [58]. When 1 reacted with H2S, an excellent fluorescence change in a red shifted range with high brightness was observed ( Figure 1B) with 10-fold turn-ons at 660 nm, and the quantum yield of the red shifted product was 0.29. The absorbance and emission data suggested the stokes shift up to 40 nm in PBS. A large Stokes shift could reduce the risk of background fluorescence and thus avoid self-quenching and backscattering effect upon excitation. These preliminary studies suggest the extended π-conjugated system of rhodamines provides excellent red shifted fluorescence probes for detection in long range with high brightness.

Kinetics Studies
Reaction kinetics, as an important parameter, was investigated for the probe 1 with H2S on account of its biological applicability under physiological conditions. To this end, the time-dependent fluorescence at 660 nm was recorded for data analysis (Figure 2A). The pseudo-first-order rate, kobs, was found by fitting the data with a single exponential function. Plotting log[H2S] versus log[kobs] confirmed a first-order dependence in H2S ( Figure 2B). The reaction rate k2 (29.8 M −1 s −1 ) was obtained by linear fitting of the kobs versus H2S concentration ( Figure 2C). The H2S-reaction rate of 1 is faster than our previous Rh-NBD-based probe [30], implying that such NBD-based probes can be employed for fast detection of H2S. On the other hand, HPLC was further employed to identify the fast reaction of 1 with H2S ( Figure S4). Furthermore, fluorescent titration ( Figure 3) was performed to determine the limit of detection (LOD) of 1 for H2S as 0.27 µM by using the 3σ/k method [58]. The probe 1 showed weak fluorescence (quantum yield φ, 0.021) upon excitation at 620 nm, indicating that fluorescence in 1 could be mainly quenched by the photoinduced electron transfer process (PET) effect from the NBD moiety [58]. When 1 reacted with H 2 S, an excellent fluorescence change in a red shifted range with high brightness was observed ( Figure 1B) with 10-fold turn-ons at 660 nm, and the quantum yield of the red shifted product was 0.29. The absorbance and emission data suggested the stokes shift up to 40 nm in PBS. A large Stokes shift could reduce the risk of background fluorescence and thus avoid self-quenching and backscattering effect upon excitation. These preliminary studies suggest the extended π-conjugated system of rhodamines provides excellent red shifted fluorescence probes for detection in long range with high brightness.

Kinetics Studies
Reaction kinetics, as an important parameter, was investigated for the probe 1 with H 2 S on account of its biological applicability under physiological conditions. To this end, the time-dependent fluorescence at 660 nm was recorded for data analysis (Figure 2A). The pseudo-first-order rate, k obs , was found by fitting the data with a single exponential function. Plotting log[H 2 S] versus log[k obs ] confirmed a first-order dependence in H 2 S ( Figure 2B). The reaction rate k 2 (29.8 M −1 s −1 ) was obtained by linear fitting of the k obs versus H 2 S concentration ( Figure 2C). The H 2 S-reaction rate of 1 is faster than our previous Rh-NBD-based probe [30], implying that such NBD-based probes can be employed for fast detection of H 2 S. On the other hand, HPLC was further employed to identify the fast reaction of 1 with H 2 S ( Figure S4). Furthermore, fluorescent titration ( Figure 3) was performed to determine the limit of detection (LOD) of 1 for H 2 S as 0.27 µM by using the 3σ/k method [58].

Selectivity and Co-Interference Studies
With above promising outcomes, we further investigated the selectivity and sensitivity of probe 1. The fluorescent ''off-on'' response of 1 towards biothiols was measured. Probe 1 (2 µM) was treated with Cys, Hcy, and GSH individually (each 1 mM). As shown in Figure 4, the results showed that fluorescence intensity enhancement for analytes was nearly negligible except H2S, suggesting that 1

Selectivity and Co-Interference Studies
With above promising outcomes, we further investigated the selectivity and sensitivity of probe 1. The fluorescent ''off-on'' response of 1 towards biothiols was measured. Probe 1 (2 µM) was treated with Cys, Hcy, and GSH individually (each 1 mM). As shown in Figure 4, the results showed that fluorescence intensity enhancement for analytes was nearly negligible except H2S, suggesting that 1

Selectivity and Co-Interference Studies
With above promising outcomes, we further investigated the selectivity and sensitivity of probe 1.
The fluorescent "off-on" response of 1 towards biothiols was measured. Probe 1 (2 µM) was treated with Cys, Hcy, and GSH individually (each 1 mM). As shown in Figure 4, the results showed that fluorescence intensity enhancement for analytes was nearly negligible except H 2 S, suggesting that 1 can selectively sense H 2 S. In order to check the interference of biothiols with coexistent H 2 S, we also tested 1 with these analytes in the presence of H 2 S (Figure 4). These findings suggested that all analytes did not interfere the H 2 S-specific thiolysis reaction. Furthermore, pH-dependent experiments were carried out to check whether 1 could sense at physiological pH ( Figure S5). Obviously, the fluorescence enhancement occurred at pH 7.0-9.0, implying that 1 could work efficiently at physiological conditions. can selectively sense H2S. In order to check the interference of biothiols with coexistent H2S, we also tested 1 with these analytes in the presence of H2S (Figure 4). These findings suggested that all analytes did not interfere the H2S-specific thiolysis reaction. Furthermore, pH-dependent experiments were carried out to check whether 1 could sense at physiological pH ( Figure S5). Obviously, the fluorescence enhancement occurred at pH 7.0-9.0, implying that 1 could work efficiently at physiological conditions.

Imaging of Probe 1 in Living Cells
Encouraged by the above results, we moved forward to study the biological applications of 1. The cytotoxicity of the 1 was evaluated firstly by using the normal human umbilical vein endothelial cell (HUVEC) line via a standard MTT assay ( Figure S6). After 24 h incubation with a varied concentration range of 1 from 5 µM, over 85% of the cells still remained viable, implying the relatively good biocompatibility of 1.
To examine the application potential of 1 for H2S detection in living cells, HeLa cells were chosen as the model biological system. Briefly, the cells were incubated with 1 alone or co-incubated with 1 and Na2S/D-Cys for 30 min. Then, all cells were examined via the confocal microscopy. Cells with probe 1 treatment displayed faint fluorescence ( Figure 5E), while cells displayed remarkable red fluorescence in the presence of 1 and Na2S ( Figure 5F). These results demonstrated that 1 could be used for selective imaging of exogenous H2S. For detection of endogenous production of H2S, cells were co-incubated with D-Cys and 1, as D-Cys can induce H2S biosynthesis via the 3-MST pathway [4]. Strong fluorescence was observed in cells ( Figure 5G), which revealed that the endogenous production of H2S from D-Cys could be detected by 1. To further confirm the detection of endogenous production of H2S from D-Cys by 1, an inhibitor (aminooxyacetic acid, AOAA) was introduced to block the pathway for H2S production from D-Cys [4]. No obvious fluorescence was detected in the AOAA-treated cells ( Figure 5H). These preliminary studies suggested that probe 1 could be used for visualization of H2S in cells efficiently and selectively.

Imaging of Probe 1 in Living Cells
Encouraged by the above results, we moved forward to study the biological applications of 1. The cytotoxicity of the 1 was evaluated firstly by using the normal human umbilical vein endothelial cell (HUVEC) line via a standard MTT assay ( Figure S6). After 24 h incubation with a varied concentration range of 1 from 5 µM, over 85% of the cells still remained viable, implying the relatively good biocompatibility of 1.
To examine the application potential of 1 for H 2 S detection in living cells, HeLa cells were chosen as the model biological system. Briefly, the cells were incubated with 1 alone or co-incubated with 1 and Na 2 S/D-Cys for 30 min. Then, all cells were examined via the confocal microscopy. Cells with probe 1 treatment displayed faint fluorescence ( Figure 5E), while cells displayed remarkable red fluorescence in the presence of 1 and Na 2 S ( Figure 5F). These results demonstrated that 1 could be used for selective imaging of exogenous H 2 S. For detection of endogenous production of H 2 S, cells were co-incubated with D-Cys and 1, as D-Cys can induce H 2 S biosynthesis via the 3-MST pathway [4]. Strong fluorescence was observed in cells ( Figure 5G), which revealed that the endogenous production of H 2 S from D-Cys could be detected by 1. To further confirm the detection of endogenous production of H 2 S from D-Cys by 1, an inhibitor (aminooxyacetic acid, AOAA) was introduced to block the pathway for H 2 S production from D-Cys [4]. No obvious fluorescence was detected in the AOAA-treated cells ( Figure 5H). These preliminary studies suggested that probe 1 could be used for visualization of H 2 S in cells efficiently and selectively. The probe 1 contains a positive charge, which might be mitochondria-targeting [33]. To this end, a fluorescent co-localization assay with Mito-Tracker Green FM (a well-known mitochondria specific dye) and probe 1 with D-Cys was carried out. As shown in Figure 6, the green fluorescence signal produced by Mito-Tracker Green FM and the red fluorescence signal from probe 1 merged well in the cells ( Figure 6C). The Pearson's coefficient is 0.946. These data implied that the probe 1 is a promising tool for imaging of mitochondria H2S.  The probe 1 contains a positive charge, which might be mitochondria-targeting [33]. To this end, a fluorescent co-localization assay with Mito-Tracker Green FM (a well-known mitochondria specific dye) and probe 1 with D-Cys was carried out. As shown in Figure 6, the green fluorescence signal produced by Mito-Tracker Green FM and the red fluorescence signal from probe 1 merged well in the cells ( Figure 6C). The Pearson's coefficient is 0.946. These data implied that the probe 1 is a promising tool for imaging of mitochondria H 2 S. The probe 1 contains a positive charge, which might be mitochondria-targeting [33]. To this end, a fluorescent co-localization assay with Mito-Tracker Green FM (a well-known mitochondria specific dye) and probe 1 with D-Cys was carried out. As shown in Figure 6, the green fluorescence signal produced by Mito-Tracker Green FM and the red fluorescence signal from probe 1 merged well in the cells ( Figure 6C). The Pearson's coefficient is 0.946. These data implied that the probe 1 is a promising tool for imaging of mitochondria H2S.

Materials and Methods
All chemicals and solvents used for the synthesis were purchased from commercial suppliers and applied directly in the experiments without further purification. The progress of the reaction was monitored by TLC on pre-coated silica plates (60F-254, 250 µm) in thickness (Merck, Darmstadt, Germany), and spots were visualized by basic KMnO 4 , UV light or iodine. Merck silica gel 60 (100-200 mesh) was used for general column chromatography purification. 1

Synthesis of Probe 1
Dissolved compound 2 (0.121 g, 0.2 mmol) in 5 mL DMF, followed by the addition of HATU (0.122 g, 0.32 mmol) and DIPEA (102 µL, 0.75 mmol). Stirred the solution for 5 min, NBD-piperazine (0.064 g, 0.2 mmol) was added to the solution and continue the stirring for 12 h at room temperature. After completion of reaction DMF was removed in vacuo. The residue was purified by silica gel column chromatography to give dark-red solid 1 (0.12 g, 79%). 1

Procedure for Spectroscopic Studies
All spectroscopic measurements were performed in phosphate-buffered saline buffer (PBS, 50 mM, pH 7.4, containing 10% DMSO) at room temperature. Compounds were dissolved into DMSO to prepare the stock solutions with a concentration of 5 mM. 1-500 mM Stock solutions of Na 2 S in degassed (by bubbling N 2 for 30 min) PBS buffer were used as H 2 S source. Probes were diluted in PBS buffer (50 mM, pH 7.4, containing 10% DMSO) to afford the final concentration of 2-5 µM. For the selectivity experiment, different biologically relevant molecules (100 mM) were prepared as stock solutions in degassed PBS buffer. Appropriate amount of biologically relevant species was added to separate portions of the probe solution and mixed thoroughly. All measurements were performed in a 3 mL corvette with 2 mL solution. The reaction mixture was shaken uniformly before emission spectra were measured.

Cell Culture and Cytotoxicity Assay
The HUVEC and HeLa cell lines were purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). And the cells were cultured in RPMI 1640 medium with 10% fetal bovine serum and 1% penicillin/streptomycin under standard cell culture conditions at 37 • C in a humidified CO 2 incubator. Before the cytotoxicity assay, the HUVEC cells were transferred to the 96-well plate and cultured for one night. After that, the culture medium was replaced with a fresh one and the HUVEC cells were pre-incubated with probe 1 with a concentration range of 5-25 µM for 24 h. The cell viability was then measured by the standard MTT assay.

Cell Imaging
Glass bottom dishes were added into a 24-well plate for cell imaging before cells were seeded. Then, the HeLa cells were transferred to the 24-well plate and cultured for one night before the experiments. After that, the culture medium was replaced with the fresh one and the cells were treated with the desired reagents. After incubation, the HeLa cells were quickly washed with PBS three times, and then fixed with 4% paraformaldehyde solution for 10 min. Finally, the HeLa cells were washed with PBS and imaged using a confocal microscope (Olympus FV1000) with a 40× objective lens. Emission was collected at the green channel (500-530 nm, excitation at 488 nm) and the red channel (620-660 nm, excitation at 594 nm).

Conclusions
In summary, we have developed a new, extended π-conjugation rhodamine-NBD a red shifted fluorescence probe 1 capable of detection H 2 S in live cells. The probe shows a relatively large Stokes shift (40 nm), fast response (k = 29.8 M −1 s −1 ), and good quantum yield (ø = 0.29) after H 2 S activation. Moreover, 1 was water-soluble, cell-membrane-permeable, and had high selectivity and sensitivity for H 2 S. We believe that this red shifted range probe 1 could be a useful tool for studies of H 2 S biology in the future.