A dual-channel chemosensor based on rhodamine and BODIPY conjugated dyad for ratiometric detection of Hg²⁺ and fluorescence on–off recognition of Cu²⁺ in aqueous solution and living cells

Abstract

A novel probe RB for discriminative detection of Hg²⁺ and Cu²⁺ constructed by rhodamine and BODIPY was designed and synthesized. Probe RB consists of three parts: BODIPY as the energy donor, rhodamine as the energy receptor, and thiosemicarbazide as the reacting site. Probe RB selectively binds with Hg²⁺ and Cu²⁺ ions in in EtOH/H₂O (V/V = 1:1) among tested metal ions. Probe RB exhibited a fluorescence red shift from 513 to 593 nm for Hg²⁺ and a fluorescence quenching for Cu²⁺. The mechanism study showed that the fluorescence changes of probe RB after the addition of Hg²⁺ was due to oxadiazole formation when the thiosemicarbazide moiety was liberated by Hg²⁺-facilitated ring opening of the spirocyclic group, while Cu²⁺ not only opened the spirolactam ring, but also bonded to the opening product through 1:2 mode. It is worth noting that all these types of probes could only detect Hg²⁺ or Cu²⁺ through metal ions-facilitated ring opening of the spirocyclic group mechanism. To the best of our knowledge, probe RB was firstly reported to detect Hg²⁺ and Cu²⁺ discriminatively. Moreover, probe RB demonstrated excellent sensitivity toward Hg²⁺ (LOD 8.36 nM) and Cu²⁺ (LOD 0.16 μM), fast responsive time (within 30 s) and wide pH range (6.0–10.0 for Hg²⁺, 6.0–11 for Cu²⁺). This probe was further successfully applied to Hela cells and could easily discriminate Hg²⁺ and Cu²⁺ through double-channel imaging.

Introduction

Heavy metal pollutants have attracted much attention because of their harm to environments and human health. Therefore, selective detection of heavy metal pollutants, especially using fluorescent probes, is currently under in-depth research (Lian et al.2020; Nagarajan et al. 2021; Duong and Kim 2010). Among them, Cu²⁺ and Hg²⁺ ions as two common metal pollutants have been fully researched due to their wide application and high toxicity. Cu²⁺ is closely related to the hematopoietic function of human body. However, when a large number of Cu²⁺ ions are retained in the human body, it is easy to cause a burden to various organs in the body, especially the liver and gallbladder. When problems occur in these two organs, the maintenance of the metabolism in the human body will appear disorder (Hosseini et al. 2014; Zhao et al.2018). While its wide application in infrastructure constructions, power generation, as well as in the production of electronic products and equipment, industrial machinery and transport vehicles, the level of Cu²⁺ pollution in the environment is constantly increasing. Hence, it is of great significance to develop fluorescent probe for Cu²⁺ detection.

Unlike Cu²⁺, Hg²⁺ is considered to be one of the most toxic heavy metal ions to human health because it can coordinate with many negatively charged groups in enzymes or proteins, such as sulfhydryl groups, which affect many metabolic pathways in cells, such as energy generation, protein and nucleic acid synthesis, thus further affecting cell function and growth of cells (Tetsuro et al. 2005). Moreover, Hg²⁺ ions could easily pass through membranes and induce massive damage to central nervous system and organs, such as brain and kidney. Therefore, it is important to develop probes which could detect trace amounts of Hg²⁺ ions in living cells.

Currently, most reported fluorescent probes could only detect Cu²⁺ or Hg²⁺ (Zhu et al. 2021; Slassi et al. 2021; Du et al. 2021; Zhang et al. 2021; Yang et al. 2013, 2021; Culzoni et al. 2012; Chen et al. 2020, 2019; Li et al. 2020; Petdum et al. 2020; Liu et al. 2020a, b; Hu et al. 2020; Lin et al.2020), the fluorescent probes which could discriminately detect two metal ions through different spectra changes were rarely developed, especially in living cells (Huang et al.2013, 2019; Divya and Thennarasu 2020; Shi et al. 2019). Some of reported dual-signal fluorescence sensors have shortcomings in practical applications, such as the indistinguishability of Hg²⁺ and Cu²⁺ ions, long response time, low sensitivity and interference from other metal ions. Therefore, it is still needed to develop dual-signal fluorescence sensors which could discriminately detect Hg²⁺ and Cu²⁺ ions with large spectral differences.

Rhodamine and BODIPY dyes are widely used in the construction of fluorescent probes due to their excellent spectral properties, such as long absorption and emission wavelength, high fluorescence quantum yield, high extinction coefficient and excellent photostability (Zhang et al. 2020, 2016; Zhang and Wong 2020; Beija et al. 2009; Nguyen et al. 2021; Kaur and Singh 2019; Kowada et al. 2015). Most reported rhodamine probes for Hg²⁺ detection undergo a transformation from the spirocyclic type to the open-loop amide type, whose fluorescence properties are completely different. The spirocyclic (closed loop) form is basically non-fluorescent, while the open-loop form produces strong fluorescence emission. Based on this mechanism, some chemical sensors for Hg²⁺ and Cu²⁺ detection have been developed (Zhang and Zhang 2014; Saleem and Lee 2014; Chen et al. 2020). Among them, a series of ratiometric fluorescent probes by fluorescence resonance energy transfer (FRET) mechanism have been reported. Previously, our group has synthesized a FRET probe based on BODIPY- rhodamine system (Compound 1, Scheme 1), which shows high selectivity toward Hg²⁺ [Wen et al. 2021]. While all these types of probes could only detect Hg²⁺ or Cu²⁺ (Table S1). Recently, our group found a novel FRET probe system with thiosemicarbazide moiety (probe RB, Scheme 1) could discriminatively detect Hg²⁺ and Cu²⁺ with excellent sensitivity. Most importantly, probe RB was successfully applied to Hela cells and could easily discriminate Hg²⁺ and Cu²⁺ through double-channel imaging. To the best of our knowledge, this type of probe was firstly reported to discriminatively detect Hg²⁺ and Cu²⁺ through different mechanisms.

Scheme 1

Synthesis route of probe RB

Experimental section

Materials and apparatus

All the materials used for the synthesis of probe RB and analytical experiments were purchased from Sigma-Aldrich without further purification. EtOH in HPLC grade purity and redistilled water were used in all analytical experiments. PerkinElmer Lambda 25 and HITACHI F-4600 Fluorescence spectrophotometer were used for absorption and fluorescence spectra measuring. Bruker Advance 400 MHz spectrometer was used for ¹H-NMR and ¹³C-NMR recording. HRMS data were obtained from an SCIEX TripleTOF 5600 + high resolution spectrometer (American). Leica TCS SP8 Confocal Laser Scanning Microscope was used to obtain the fluorescence images of living cells.

Synthesis of probe RB

Compound 1 (0.3 g, 0.4 mmol) and phenyl isothiocyanate (0.2 mL, 1.3 mmol) were dissolved in dry DMF (5 mL), the mixture was stirred overnight at room temperature. After completion of the reaction, the solvent was removed by a rotary evaporator under reduced pressure to give a red residue, which was purified directly by gel column chromatography (CH₂Cl₂/MeOH = 50/1) to give a red solid (0.33 g, 0.36 mmol) in 90% yield. ¹H NMR (400 MHz, CDCl₃) δ 8.04 (d, J = 7.9 Hz, 1H), 7.69–7.60 (m, 2H), 7.52 (s, 1H), 7.18 (t, J = 7.5 Hz, 2H), 7.08 (dd, J = 19.0, 7.4 Hz, 3H), 6.90 (s, 1H), 6.68 (s, 1H), 6.53–6.46 (m, 4H), 6.33–6.31 (m, 1H), 6.06 (s, 2H), 3.79–3.75 (m, 2H), 3.62–3.58 (m, 2H), 3.44–3.26 (m, 4H), 3.21–3.20 (m, 4H), 3.13–3.00 (m, 2H), 2.51 (s, 6H), 2.46 (s, 6H), 2.15–1.91 (m, 2H), 1.66 (s, 3H), 1.16 (t, J = 7.0 Hz, 6H). ¹³C NMR (100 MHz, CDCl₃) δ 170.28, 167.02, 154.10, 153.99, 125.12, 145.52, 140.46, 137.58, 134.46, 131.61, 129.36, 129.08, 128.34, 127.73, 126.12, 124.88, 124.70, 124.05, 121.81, 112.30, 103.23, 66.83, 48.27, 48.11, 45.19, 44.46, 41.25, 32.68, 27.57, 26.99, 16.51, 14.49, 12.51. HRMS (m/z): calculated for C₅₂H₅₆BF₂N₈O₃S [M + H]⁺, 921.4257; found, 921.4333.

Absorption and fluorescence spectra measurement procedure

The stock solution of probe RB (1 mM) was prepared with EtOH in HPLC grade purity and stored in refrigerator at 4 °C. All the tested metal ions (Ag⁺, Ba²⁺, Ca²⁺, Cd²⁺, Co²⁺, Cu²⁺, Fe³⁺, K⁺, Hg²⁺, Mg²⁺, Mn²⁺, Na⁺, Ni⁺ and Zn²⁺, 10 mM) were prepared in ultrapure water. The UV–vis and fluorescence spectra were recorded in the mixture solution EtOH/H₂O (V/V = 1:1) at room temperature.

Cell imaging

Hela cells were cultured in Dulbecco's modified Eagle medium (DMEM), which contains 10% Fetal Bovine Serum (FBS) in two days. Then Hela cells were firstly treated with probe RB (5 μM, PBS containing 1% DMSO), which were incubated over 30 min at 37 °C. Then the Hela cells were washed with phosphate buffer saline (PBS) (1 mL × 3). The treated cells were further incubated with Cu²⁺ or Hg²⁺ (5 μM, PBS containing 1% DMSO) over 30 min. After removing the culture solvent, the cells were washed with PBS for three times. The fluorescence images of living cells were captured in PBS with Leica TCS SP8 Confocal Laser Scanning Microscope, upon excitation at 488 nm. Emissions were collected with green channel (513 ± 15 nm) and red channel (593 ± 15 nm).

Results and discussion

Probe RB synthesis

The procedures of probe RB synthesis are depicted in Scheme 1. Probe RB was obtained by the reaction of phenyl isothiocyanate with compound 1 in 90% yield. Probe RB was fully characterized by ¹H NMR, ¹³C NMR and HRMS data, which were offered in in the Experimental Section. The spectra of ¹H NMR, ¹³C NMR and HRMS were provided in the supporting information.

Studies of spectral response of probe RB to different metal ions

With probe RB in hand, we firstly investigated its sensing ability toward several metal ions (Ag⁺, Ba²⁺, Ca²⁺, Cd²⁺, Co²⁺, Cu²⁺, Fe³⁺, K⁺, Hg²⁺, Mg²⁺, Mn²⁺, Na⁺, Ni⁺ and Zn²⁺) in the mixture solution EtOH/H₂O (V/V = 1:1) by absorption and fluorescence spectra. As shown in Fig. 1a, probe RB displays an absorption band around 499 nm, which is the characteristic absorption band of BODIPY. Addition of Cu²⁺ or Ag⁺ led to partial decrease in absorption band around 499 nm, accompanied by the appearance of a new band around 564 nm, which was attributed to the typical absorption spectrum of rhodamine group. While with the addition of Hg²⁺, the absorption spectrum of open-loop rhodamine appeared immediately. The above phenomenon indicated that probe RB interacted with Cu²⁺/Hg²⁺/Ag⁺, further enhancing the conjugation degree of RB-Cu²⁺/Hg²⁺/Ag⁺ system, resulting in the red shift of RB absorption spectrum. In addition, probe RB is slight pink in the mixed solution of EtOH/H₂O (V/V = 1:1). When Hg²⁺ was added, its color changed from pink to purple, while the addition of Cu²⁺ or Ag⁺ induced color changed from pink to lavender, indicating that probe RB can detect Hg²⁺ by naked eye (Fig. 2a). Moreover, the fluorescence spectra of probe RB in the presence of different metal ions were also measured. As shown in Fig. 1b, probe RB exhibited an emission band around 513 nm upon excitation at 480 nm. With addition of Cu²⁺, its fluorescence was almost quenched completely; while addition of Hg²⁺ induced the disappearance of emission band at 513 nm, along with the appearance of a new emission band centered at 593 nm. As shown in Fig. 2b, emission color changed from green to blank or red were easily observed by addition of Cu²⁺ or Hg²⁺ through naked eye under the illumination with a 365 nm UV-lamp. In addition, Ag⁺ ions had a slight effect, because it could also induce the fluorescence quenching of BODIPY, with the appearance of a new weak emission band centered at 593 nm. However, it has much longer equilibrium time (10 min, Fig. S1) than that of Hg²⁺ ions (30 s).

Fig. 1

a Absorption spectra of probe RB (10 μM) with different metal ions (100 μM) in the mixture solution EtOH/H₂O (V/V = 1:1). b fluorescence spectra of probe RB (10 μM) with different metal ions (100 μM) in the mixture solution EtOH/H₂O (V/V = 1:1)

Fig. 2

a The color changes of probe RB (10 μM) with different metal ions (100 μM) in the mixture solution EtOH/H₂O (V/V = 1:1) under day light. b The color changes of probe RB (10 μM) with different metal ions (100 μM) in the mixture solution EtOH/H₂O (V/V = 1:1) under 365 nm UV-lamp

Fluorescence titrations of probe RB with different amounts of Cu²⁺ or Hg²⁺ were carried out in the mixture solution EtOH/H₂O (V/V = 1:1). As depicted in Fig. 3a, probe RB showed a fluorescence emission band centered at 513 nm, which was decreased gradually and remained unchanged after addition of 8 Mμ Cu²⁺. Plot of fluorescence intensity changes at 513 nm as a function of concentrations of Cu²⁺ (0–2.5 Mμ) exhibited an excellent linearity (R² = 0.99277). While upon progressive addition of Hg²⁺ (0–20 Mμ), fluorescence intensity at 513 nm decreased, along with a new fluorescence emission band around 593 nm appeared and increased gradually (Fig. 3b). In addition, the fluorescence titration curve showed that the ratio of fluorescence intensity of different emission band (I_{593 nm}/I_{513 nm}) increased dramatically with the increasing concentrations of Hg²⁺ and reached a plateau with addition of 20 μM Hg²⁺. Plot of I_{593 nm}/I_{513 nm} changes of probe RB as a function of concentrations of Hg²⁺ (5–15 μM) exhibited an excellent linearity (R² = 0.99549). The detection limits of probe RB to Cu²⁺ and Hg²⁺ were calculated to be 0.16 μM (0.01 ppm) and 8.36 nM (1.68 ppb) (SI), respectively, which were lower than the maximum permissible amount of Cu²⁺ (1.3 ppm) and Hg²⁺ (2.0 ppb) in drinking water that proposed by USA Environmental Protection Agency (EPA).

Fig. 3

a The fluorescence spectra changes of probe RB (10 μM) with increasing concentration of Cu²⁺ (0–8 μM) in EtOH/H₂O (V/V = 1:1). Insert: plot of I_{513 nm} changes of probe RB (10 μM) as a function of Cu²⁺ concentration (0–2.5 μM). b The fluorescence spectra changes of probe RB (10 μM) with increasing concentration of Hg²⁺ (0–20 μM) in EtOH/H₂O (V/V = 1:1). Insert: plot of I_{593 nm}/I_{513 nm} changes of probe RB (10 μM) as a function of Hg²⁺ concentration (5–15 μM)

Moreover, in order to further confirm the sensing ability of probe RB to Cu²⁺ and Hg²⁺ in complicated environments, competition experiments of probe RB for Cu²⁺ and Hg²⁺ detection were measured through adding the above-mentioned tested metal ions, respectively. As shown in Fig. 4a, the fluorescence intensity of probe RB with tested metal ions (Ba²⁺, Ca²⁺, Cd²⁺, Co²⁺, Fe³⁺, K⁺, Mg²⁺, Mn²⁺, Na⁺, Ni⁺ and Zn²⁺) remained basically unchanged, while Cu²⁺ addition resulted a ca. 96% quenching of fluorescence at 513 nm. In addition, Ag⁺ could lead to a circa (ca.) 91% quenching of fluorescence at 513 nm, while the fluorescence intensity was further decreased to ca. 96%, following addition of Cu²⁺. Moreover, I_{593 nm}/I_{513 nm} was significantly increased with addition of Hg²⁺ in presence of other tested ions (Ag⁺, Ba²⁺, Ca²⁺, Cd²⁺, Co²⁺, Fe³⁺, K⁺, Mg²⁺, Mn²⁺, Na⁺, Ni⁺ and Zn²⁺), indicating excellent anti-interference ability of probe RB for the detection of Cu²⁺ and Hg²⁺. In addition, we also investigated whether Cu²⁺ and Hg²⁺ interfered with each other (Figs. S2, S3). The results showed that when Cu²⁺ was firstly added into the solution of probe RB in EtOH/H₂O (V/V = 1:1), it induced a ca. 90% quenching of fluorescence at 513 nm, and the fluorescence intensity at 513 nm was further decreased, along with the appearance of a new emission band centered at 593 nm after another addition of Hg²⁺. While when Hg²⁺ was firstly added into the solution of probe RB in EtOH/H₂O (V/V = 1:1), it led to the appearance of a new emission band centered at 593 nm, and then another addition of Cu²⁺had no effect on the new emission band. The above results indicated that the probe RB for detecting Hg²⁺ was not disturbed by other measured ions, while for detecting Cu²⁺ was interfered by Hg²⁺.

Fig. 4

a I_{513 nm} of probe RB (10 μM) toward Cu²⁺ in presence of other metal ions (100 μM) in EtOH/H₂O (V/V = 1:1). b I_{593 nm}/I_{513 nm} of probe RB (10 μM) toward Hg²⁺ in presence of different metal ions (100 μM) in EtOH/H₂O (V/V = 1:1)

Sensing mechanism study

According to the literature (Yang et al. 2005; Ko et al. 2006; Wang et al. 2015; Li et al. 2019; Zhu et al. 2020; Ji et al. 2017), rhodamine fluorophore containing thiosemicarbazide group usually shows no fluorescence, and it could be easily transformed into 1,3,4-oxadiazole by Hg²⁺-promoted cyclization reaction. Therefore, we assume that probe RB with thiosemicarbazide group shows green fluorescence of BODIPY upon excitation at 480 nm, which is due to no FRET process. When Hg²⁺ facilitated the cyclization reaction of thiosemicarbazide to form oxadiazole, the FRET was switched on, accompanied by fluorescence emission of rhodamine appearance upon excitation at 480 nm. To verify this hypothesis, we carried out ¹H NMR titration and HRMS experiments. The ¹H NMR titration experiment was carried out in (CD₃)₂SO, due to its high solubility for probe RB. Moreover, the response of probe RB to Hg²⁺ in (CD₃)₂SO was the same as that in EtOH/H₂O (V/V = 1:1). As shown in Fig. 5, the peaks at 8.86 (s, 1H) and 9.40 (s, 1H) are the protons of amine 1 and amine 2, addition of Hg²⁺ resulted the disappearance of proton 1 and downfield shift to 10.63 of proton 2. Moreover, the protons of rhodamine group showed a downfield shift and the protons of the pyrrole group on BODIPY kept the same at 6.23 (s, 2H) upon addition of Hg²⁺. These results clearly showed that thiosemicarbazide group could be transformed into oxadiazole by Hg²⁺ addition. In addition, the HRMS spectrum of probe RB with Hg²⁺ was recorded. The peak at m/z 887.4457 corresponded to the calculated m/z at 887.4375 for rhodamine containing oxadiazole group (Fig. S4). ¹H NMR and HRMS data supported the mechanism which is depicted in Scheme 2. Moreover, The FRET energy transfer efficiency (E) was calculated as 97.8% (E = 1 − I_DA/I_D) through testing the fluorescence intensity of BODIPY and probe RB (Fig. S5).

Fig.5

The ¹H NMR of probe RB and probe RB with Hg²⁺ in (CD₃)₂SO solvent

Scheme 2

The proposed mechanism of probe RB for Cu²⁺ and Hg²⁺ detection

Most reported rhodamine containing thiosemicarbazide group showed no response toward Cu²⁺, in order to understand the sensing mechanism of probe RB to Cu²⁺, we investigated reversibility, job plot, HRMS experiments. As shown in Fig. 6a, Cu²⁺ addition induced a ca. 96% quenching of fluorescence at 513 nm, and the fluorescence intensity was almost recovered after addition of EDTA, indicating that the coordination of probe RB with Cu²⁺ is chemically reversible. While the fluorescence changes induced by Hg²⁺ could not recover through addition of EDTA, due to the formation of oxadiazole (Fig. 6b). Job plot based on fluorescence change was applied to study the binding stoichiometry of probe RB and Cu²⁺. The fitting result is depicted in Fig S6 and 2:1 stoichiometry was calculated for probe RB bonding with Cu²⁺. In addition, the HRMS spectrum of probe RB with Cu²⁺ showed peaks at 1926.46, which was corresponding to the calculated m/z for 2(RB) + Cu²⁺  + Na⁺ (Fig. S7). All above results indicated that the response of probe RB to Cu²⁺ was due to the coordination of probe RB with Cu²⁺ through 2:1 binding mode (Scheme 2).

Fig. 6

a The fluorescence spectra changes of probe RB after the addition of Cu²⁺ and EDTA; b The fluorescence spectra changes of probe RB after the addition of Hg²⁺ and EDTA

Kinetics study

Time dependent fluorescence change of probe RB in the presence of different concentrations of Cu²⁺ or Hg²⁺ were studied. As shown in Fig. 7a, the fluorescence intensity at 513 nm decreased sharply with the increasing concentrations of Cu²⁺. It was noted that I_{513 nm} significantly decreased after 30 s and reached equilibrium within 1 min, which was much faster than most reported probes (5–30 min, Table S1). Moreover, Hg²⁺ addition also induced the value of I_{593 nm}/I_{513 nm} augmented with the increasing concentration of Hg²⁺, and reached equilibrium within 1 min (Fig. 7b). Because of its fast Kinetics, probe RB has a broad application prospect in real-time detection of Cu²⁺ and Hg²⁺.

Fig. 7

Time dependent I_513nm or I_{593 nm}/I_{513 nm} of probe RB (10 μM) with different concentrations of Cu²⁺ or Hg²⁺ (0–20 μM) in EtOH/H₂O (V/V = 1:1)

pH influence

To evaluate the practical applicability of probe RB, the suitable operating pH ranges for Cu²⁺ and Hg²⁺ detection were measured through fluorescence spectra, respectively. As shown in Fig. 8, the fluorescence intensity at 513 nm or I_{593 nm}/I_{513 nm} of probe RB was basically stable in the pH range of 4–12. In the presence of Cu²⁺, the fluorescence intensity at 513 nm was decreased 45–75% of its initial level in the range of pH = 4–5. Interestingly, the fluorescence intensity decreased significantly (~ 95% of the initial level) at pH 6 to 12, especially at pH = 6–11 (Fig. 8a). The result showed that probe RB was adequate for Cu²⁺ detection in biological surroundings. On the other hand, upon addition of Hg²⁺, I_{593 nm}/I_{513 nm} of probe RB showed no obvious changes at pH below 5 or above 11, which indicated that Hg²⁺ could not promote the ring-closed reaction of thiosemicarbazides under strong acidic condition or basic condition. In the range of pH 6–10, the presence of Hg²⁺ significantly improved I_{593 nm}/I_{513 nm} of probe RB (Fig. 8b), indicating that probe RB is suitable for Hg²⁺ detection under physiological conditions.

Fig. 8

I_{513 nm} and I_{593 nm}/I_{513 nm} of probe RB (10 μM) without and with Cu²⁺ or Hg²⁺ (100 μM) as a function of pH

Living cellular imaging

In order to further study the practical applicability of the probe RB in biological systems, fluorescence microscopy was used to conduct cell imaging experiments using Leica TCS SP8 Confocal Laser Scanning Microscope. Green channel (513 nm ± 15 nm) and red channel (593 nm ± 15 nm) were used to capture fluorescence emission images. As shown in Fig. 9, obvious green fluorescence in the cells was observed, after incubating living Hela cells with probe RB (5 μM) for 30 min at room temperature, indicating the cell permeability of probe RB. Moreover, in red channel no fluorescence can be observed, which demonstrated that probe RB is very steady and there is no interference for Hg²⁺ detection in Hela cells. Incubated the Hela cells stained with the probe RB with Cu²⁺ and Hg²⁺ (5 μM) in PBS for 30 min, respectively, and then washed with PBS three times. As shown in Fig. 9d, e, the green fluorescence intensity in the green channel was partially quenched (Fig. 9d), and the red color increased significantly, which indicated Hg²⁺ could promote the cyclization reaction of thiosemicarbazide group to form oxadiazole in living cells. Compared to Hg²⁺, Cu²⁺ only induced the green fluorescence intensity partially quenched, which was in accordance with the fluorescence spectrum changes in EtOH/H₂O (V/V = 1:1) solution. All above results showed that the probe RB could be used as a fluorescent probe to discriminate Hg²⁺ and Cu²⁺ through double-channel imaging in living cells.

Fig. 9

The confocal fluorescence images of Hela cells incubated with probe BR (5 μM) over 30 min a green channel, b red channel, c overlay image of a, b; the fluorescence images of Hela cells stained with the probe RB incubated with Hg²⁺ (5 μM) over 30 min d green channel, e red channel, f overlay image of d, e; the fluorescence images of Hela cells stained with the probe RB incubated with Cu²⁺ (5 μM) over 30 min g green channel, h red channel, i overlay image of g, h

Conclusion

In summary, we have synthesized a dual-function probe RB containing thiosemicarbazide group, which showed high sensitivity and selectivity toward Cu²⁺ and Hg²⁺. This probe could detect Cu²⁺ and Hg²⁺ through significant fluorescence on–off and ratiometric fluorescence changes in EtOH/H₂O (V/V = 1:1) solution and living cells. The detection limits of probe RB to Cu²⁺ and Hg²⁺ were calculated to be 0.01 ppm and 1.68 ppb, respectively, which was lower than the maximum permissible amount of Cu²⁺ (1.3 ppm) and Hg²⁺ (2.0 ppb) in drinking water proposed by USA Environmental Protection Agency (EPA). Probe RB containing thiosemicarbazide group was firstly reported to detect Cu²⁺ through 2:1 binding mechanism. Most importantly, probe RB has been successfully applied for discriminating Hg²⁺ and Cu²⁺ through double-channel imaging in living cells.