A novel highly sensitive and selective fluorescent sensor for imaging mercury(II) in living cells

Abstract

A novel phenothiazine-based derivative was designed and synthesized in this work, this molecule exhibited an obvious fluorescence quenching upon addition of Hg²⁺ due to the formation of a 1 : 1 metal–ligand complex. In addition to the high sensitivity toward Hg(II), it also displayed a high selectivity for Hg²⁺ over other transition metal ions in a rapid response time (<10 s). Simultaneously, the cell imaging experiment demonstrated the value of the sensor in fluorescent visualization of Hg²⁺ in biological systems.

---

1. Introduction

Mercury is one of the most hazardous elements in ecosystems which is released through natural events or human activities.¹ Mercury accumulation in the human body will affect a wide variety of diseases like digestive, kidney, and especially neurological diseases, even in a low concentration.² Unfortunately, mercury contamination is widespread and occurs in the environment through a variety of natural and anthropogenic sources,³ and a high percentage of mercury contamination can be attributed to the industrial processes such as oceanic and volcanic emissions, coal-burning power plants, gold mining, solid waste incineration and the combustion of fossil fuels.⁴ Therefore, developing the real-time Hg²⁺ sensors with high sensitivity and selectivity are in great demand.

The development of fluorescent chemosensors for sensing and reporting heavy transition-metal ions has been receiving considerable attention in recent years.⁵ Until now, many fluorescent sensors for Hg²⁺ have been reported, most of them are based on common fluorophore such as rhodamine or naphthylamine derivatives,⁶ which are often hampered by interference from chemically related cations like Cu²⁺ and exhibit relatively long response times, and the phenothiazine based fluorescent sensor is rarely reported in early literatures.⁷ In these regards, our synthesis of the novel phenothiazine based fluorescent sensor which displays extremely selective and sensitive for Hg²⁺ is of great advantage. Besides, the rapid response time make it as a perfect tool for illustrating and dynamically mapping the intracellular fluctuations of Hg²⁺ by utilizing microscopy techniques to allow real-time local imaging.⁸

Herein we report a new fluorescent sensor combine quinoline with phenothiazine, which serving as a Hg²⁺ chelator based on ICT (intramolecular charge transfer) mechanism, presented very excellent affinity for Hg²⁺ among other metal ions and the living cells image experiments illustrated its favourable biocompatibility, in addition, its simple structure and rapid response will expand its application prospects.

2. Experimental

2.1 Materials and instruments

All the materials for synthesis were purchased from commercial suppliers and used without further purification. Solutions for spectra detection was HPLC reagent without fluorescent impurity. ¹H NMR spectra were taken on a Varian mercury-400 spectrometer with TMS as an internal standard and DMSO as solvent. HRMS spectra was analysed on an Agilent 1290-micro TOF QII. Fluorescence spectra measurements were performed on a Hitachi F-4500 spectrofluorimeter. The pH measurements were made with Mettler-Toledo Instruments DELTE 320 pH. Cell experiment were applied on an inverted fluorescence microscope (Olympus IX-70) connected with a digital camera (Olympus, c-5050).

2.2 UV-vis and fluorometric analysis

The solution of sensor 1 was prepared in C₂H₅OH and the metal ion buffer solutions were using various cations dissolved in Tris–HCl buffer solution at pH 7.2. In titration and selectivity experiments, the test samples were prepared by placing appropriate amounts of ions into corresponding solution of sensor 1. For fluorescence measurements, excitation was provided at 368 nm, and emission was collected from 330 to 550 nm, both the excitation and emission slit widths were 5 and 5 nm, respectively.

2.3 Synthesis

Preparation of intermediate 4. Phosphoryl chloride of 3.0 g (19 mmol) was added slowly to 2 mL of dry N,N-dimethylformamide (DMF) at 0 °C, and the mixture was stirred for 2 h at room temperature. Then, a solution of 10-ethylphenothiazine (2.4 g, 9.5 mmol) in 1,2-dichloroethane (25 mL) was added to it. After another 1 h, the reaction mixture was heated to 90 °C for 15 h. Finally, the solution was cooled and poured into 300 mL of water. After the solution was extracted with CH₂Cl₂ (70 mL × 3) and dried with anhydrous magnesium sulfate, the solvent was removed and the residue was purified by column chromatography with petroleum ether–ethyl acetate(15 : 1, volume ratio) as eluent to afford 1.75 g of compound 4 as a yellow solid, yield 32%.δ 1.31 (m, J = 6.0, 5.9 Hz, 3H), 3.96 (d, J = 3.1 Hz, 2H), 7.03 (m, J = 9.2, 5.7, 6.2 Hz, 1H), 7.15 (m, J = 6.5, 8.9 Hz, 1H), 7.55 (s, 1H), 7.70 (d, J = 6.3 Hz, 1H), 9.78 (s, 1H). ¹³C NMR (75 MHz DMSO, 25 °C, TMS): δ 12.91, 42.22, 115.49, 116.52, 122.26, 123.30, 123.98, 127.58, 128.01, 128.42, 130.70, 131.22, 143.02, 149.95, 190.94.

Preparation of intermediate 2 (ref. 9). (2.0 g, 13.78 mmol) 8-hydroxyquinoline and (3.8 g, 27.56 mmol) K₂CO₃ were mixed in 50 mL acetonitrile with stirring for 30 min, added (2.5 g, 15.15 mmol) ethyl bromoacetate and kept stirring for 6 h at room temperature, extracted the product with CH₂Cl₂ and H₂O for 3 times, retained organic layer and dried on MgSO₄ for 12 h before distilling the solvent. The crude product compound 3 (red oil) was purified by column chromatography (silica gel, EtOAc : petroleum ether = 1 : 2) with a yield of 80% [18]. (0.5 g, 2.16 mmol) compound 3 in 5 mL MeOH was added dropwise into 1 mL hydrazine for stirring 1 h, filtered the generated precipitates and washed with H₂O and CH₂Cl₂ for 3 times respectively, purified the residue through silica gel column chromatography (silica gel, dichloromethane : ethanol = 1 : 1). The product intermediate 2 was collected as white solid with a yield of 69% (Scheme 1). ¹H NMR (300 MHz DMSO, 25 °C, TMS): δ 4.39 (s, 2H), 4.76 (s, 2H), 7.26 (d, J = 6.0 Hz, 1H), 7.53 (d, J = 3.0 Hz, 2H), 7.59 (s, H), 8.36 (d, J = 3.9 Hz, 1H), 8.91 (d, J = 1.6 Hz, 1H), 9.46 (s, 1H). ¹³C NMR (75 MHz DMSO, 25 °C, TMS): δ 68.77, 112.19, 121.42, 122.48, 127.23, 129.6, 136.51, 140.33, 149.88, 154.42, 167.32. LC-MS. Calc. for: C₁₁H₁₁N₃O₂ m/z = 217.09, found [(M + 1)]⁺ m/z = 218.1.

Scheme 1 Synthetic route of sensor 1.

Synthesis of sensor 1. To 20 mL of anhydrous DMF containing intermediate 4 (0.22 g, 1.0 mmol) was added intermediate 2 (0.25 g, 1.0 mmol) and the mixture was vigorously at 155 °C for 8 h. The reaction progress was monitored by thin-layer chromatography. After completion of the reaction, removed the solvent and the residue was purified by column chromatography (silica gel, dichloromethane : methanol = 2 : 1) to afford 0.11 g of sensor 1 as yellow solid, yield 23%. ¹H NMR (300 M DMSO, 25 °C, TMS): δ 1.44 (m, J = 5.8, 5.9 Hz, 3H), 3.95 (d, J = 3.0 Hz, 2H), 4.99 (s, 2H), 6.89 (m, J = 6.2, 2.9, 3.2 Hz, 3H), 7.16 (m, J = 5.9, 8.8 Hz, 2H), 7.31 (m, J = 6.4, 2.7 Hz, 2H), 7.57 (m, J = 3.0, 3.3 Hz, 5H), 8.25 (m, J = 6.3, 5.7 Hz, 2H), 8.96 (d, J = 3.6 Hz, 1H). ¹³C NMR (75 MHz DMSO, 25 °C, TMS): δ 13.00, 41.83, 69.59, 110.38, 115.72, 115.78, 116.11, 120.37, 121.82, 123.29, 127.11, 127.32, 127.53, 128.24, 128.76, 136.23, 136.69, 143.36, 143.99, 146.09, 147.23, 148.39, 149.39, 149.88, 154.53, 169.18. LC-MS. Calc. for: C₂₆H₂₂N₄O₂S m/z = 454.15.13, found [(M + 1)⁺] m/z = 455.1.

3. Results and discussion

3.1 Hg²⁺-titration and spectral responses

With the synthesis complete, the optical property of sensor 1 in the presence of Hg²⁺ was investigated. Evidence for ion interaction with the chromophore was first sought using UV-visible spectroscopy, titration experiments were conducted under the condition of 0.01 mol L⁻¹ Tris–HCl buffer solution at pH = 7.2. As illustrated in Fig. 1, sensor 1 exhibited an absorption maximum at 368 nm, with the addition of a Hg²⁺ buffer solution ([Hg²⁺] = 0–0.01 mmol L⁻¹), the absorption band at 368 nm increased until the first stoichiometry. In the fluorescence emission (Fig. 2), sensor 1 (5 × 10⁻⁶ mol L⁻¹) alone exhibited strong fluorescence at 526 nm (V(C₂H₅OH)–V(buffer) = 7 : 3, pH = 7.2, excited at 368 nm). Upon the addition of Hg²⁺ from 0 to 1 × 10⁻⁵ mol L⁻¹, the fluorescence intensity gradually weakened until quenching and the reaction responsible for these changes reached completion well within the time frame (<10 s), this “switch-off” process could be observed under the irradiation of ultraviolet lamp (Scheme 2). The decreasing fluorescence intensity of sensor 1 depend on the concentration of Hg²⁺ was in a linear manner as illustrated in Fig. 3 (R = 0.99012), which indicated sensor 1 had potential application for quantitative determination of Hg²⁺, and the detecting limit could reach to 3.139 × 10⁻⁷ mol L⁻¹ by calculation from this linear relationship (based on DL = KSb₁/S). Simultaneously, this 10⁻⁷ mol L⁻¹ degree detecting limit is much lower than the TLV (10 ppb) set by the EPA. The Job's plot (Fig. 4) exhibited the maximum absorbance appeared at w = 50%, demonstrated that sensor 1 and Hg²⁺ formed a 1 : 1 stoichiometry complex. Furthermore, the association constant of complex of sensor 1 with Hg²⁺ was found to be 8.667 × 10⁶ mol⁻¹ (R = 0.99611) according to the linear Benesi–Hildebrand (Fig. 5) expression [10–12], which based on the relationship between absorbance [1/(A − A₀)] and 1/[Hg²⁺] at 368 nm in absorption spectrum.¹⁰

Fig. 1 UV-Vis absorption response of sensor 1 (1 μM L⁻¹) upon addition of different concentrations of Hg²⁺ (0.1 μM L⁻¹) in Tris–HCl solution [V(C₂H₅OH)–V(H₂O) = 7 : 3, pH = 7.2].

Fig. 2 Fluorescence spectra of sensor 1 (1 μM L⁻¹) upon addition of different concentrations of Hg²⁺ (0.1 μM L⁻¹) in Tris–HCl solution [V(C₂H₅OH)–V(H₂O) = 7 : 3, pH = 7.2] (λ_ex = 368 nm).

Scheme 2

Fig. 3 Normalized response of the fluorescent signal to changing Hg²⁺ concentrations.

Fig. 4 Job's plot for determining the stoichiometry of sensor 1 and Hg²⁺ in Tris–HCl solution [V(C₂H₅OH)–V(H₂O) = 7 : 3, pH = 7.2] [Hg²⁺] + [sensor 1], the total concentration of sensor 1 and Hg²⁺ was 1 × 10⁻⁵ mol L⁻¹.

Fig. 5 Benesi–Hildebrand plot of sensor 1 with Hg²⁺.

To gain the insight into the observed fluorescence quenching, we may consider that the specific binding between sensor 1 and Hg²⁺ change the stereochemical structure of sensor 1, and then interrupted the π-conjugation and affected the intramolecular electron density distribution, resulting in the fluorescence quenching to Hg²⁺ in aqueous solution.¹¹ The observation was in good agreement with the afore-mentioned design concept.

3.2 Selective and competitive experiments

To gain insight into the selectivity of sensor 1 for Hg²⁺, various common metal ions in environmental and biological interest were introduced to investigate their impact on the fluorescence response of sensor 1. In selectivity experiments the fluorometric behavior of sensor 1 (5.0 × 10⁻⁶ mol L⁻¹) was investigated upon addition of several metal ions such as Na⁺, K⁺, Ca²⁺, Mg²⁺, Cu²⁺, Ba²⁺, Fe³⁺, Mn²⁺, Fe²⁺, Pb²⁺, Cd²⁺, Hg²⁺, Zn²⁺, Al³⁺ (5.0 × 10⁻⁶ mol L⁻¹) in C₂H₅OH–Tris buffer solution [V(C₂H₅OH)–V(H₂O) = 7 : 3, pH 7.2, excited at 368 nm]. As shown in Fig. 6 (black bar), only Hg²⁺ to sensor 1 caused an obvious fluorescence quenching, the introduction of other metal ions of Na⁺, K⁺, Ca²⁺, Mg²⁺, Ba²⁺, Pb²⁺, Al³⁺, Fe³⁺, Zn²⁺, Cd²⁺, Cu²⁺ and Mn²⁺ had negligible effect to fluorescence intensity, whereas some metal ions like Ni²⁺ and Cr³⁺ slightly weaken the fluorescence. In general, all these ions would not induce any remarkable disturbance for detecting Hg²⁺ in the fluorescence change at 526 nm. In order to further test the interference of other common cations in recognition of Hg²⁺, the competition experiments were performed: sensor 1 was conducted with the addition of 1.0 equiv. Hg²⁺ to induced fluorescence quenching before mixed 10 equiv. Na⁺, K⁺, Ca²⁺, Mg²⁺, Cd²⁺, Ba²⁺, Fe³⁺, Pb²⁺, Al³⁺, Ni²⁺, Cr³⁺, Mn²⁺ and Zn²⁺ respectively. The fluorescence intensity of the mixed system at 526 nm was shown in Fig. 6 (red bar), all of them exhibited still quenching, so the experimental results indicated that these ions had no interference for Hg²⁺ detection. Thus the fluorometric analysis above had proven that sensor 1 could serve as an outstanding sensitive and selective fluorescent sensor for Hg²⁺ in our prospective.

Fig. 6 Fluorescence intensities of sensor 1 (1 μM L⁻¹) in the presence of various metal ions (black bar) and competition experiment (red bar) in Tris–HCl solution [V(C₂H₅OH)–V(H₂O) = 7 : 3, pH = 7.2] (λ_ex = 368 nm, λ_em = 526 nm).

We selected the filter paper as the supporter of the sensor 1 to make a dipstick for the Hg²⁺ detecting. After dropping a solution of sensor 1 on the neutral filter paper and drying it, a fluorescent dipstick was formed. When exposed the modified filter-paper to the aqueous solution of various metal ions under UV lamp, only Hg²⁺ induced the fluorescence quenching as shown in Fig. 7.

Fig. 7 Fluorescence changes of filter paper containing sensor 1 treated with various metal ions under UV lamp.

3.3 Detection of Hg²⁺ in living cells

To explore the potential biological application of this sensor, we researched the capability of sensor 1 to track Hg²⁺ in living GES-1 (human stomach lining cells) cell. The images were obtained upon irradiation at 405 nm with a band path from 395 to 425 nm under identical exposure conditions. The living cells were first incubated with 0.1 mmol L⁻¹ sensor 1 in DMF for 30 min at 37 °C in 5% CO₂ atmosphere, then washed the cells with phosphate buffered saline (PBS, pH = 7.4) 3 times, the strong intracellular fluorescence illustrate the sufficient accumulation of sensor 1 in the living cells (Fig. 8a). However, after adding 0.2 mmol L⁻¹ Hg²⁺ in to the cells which supplemented with sensor 1, the intracellular fluorescence quenched as shown in Fig. 8b. So Fig. 8 displays the fluorescence distinction of sensor 1 before and after treated with Hg²⁺ in living GES-1 cells, illustrating that it can be a valuable molecular sensor for studying biological processes involving Hg²⁺ within living cells.

Fig. 8 Fluorescence microscope imaging of GES-1 cells stained with 1 × 10⁻⁴ mol L⁻¹ sensor 1 before and after treating with 2 × 10⁻⁴ mol L⁻¹ of Hg²⁺.

4. Conclusions

We have developed a fluorescent chemosensor for recognizing Hg²⁺ by the conjugation of quinoline and phenothiazine. It displayed excellent selectivity and sensitivity to Hg²⁺ with a 1 : 1 binding mode. Upon addition common metal ions, only Hg²⁺ caused a striking fluorescence quenching. Moreover, the cell-permeable experiment indicated this sensor can indeed visualize the change of intracellular Hg²⁺ in living cells. We expect that such Hg²⁺ sensor will have broad application in biomedical and environmental detections.

References

1. (a) A. Jana, J. S. H. Kim, S. Jung and P. K. Bharadwaj, Chem. Commun., 2009, 4417; (b) D. W. Boening, Chemosphere, 2000, 40, 1335; (c) H. H. Harris, I. J. Pickering and G. N. George, Struct. Sci., 2003, 301, 1203.
2. (a) M. Harada, Crit. Rev. Toxicol., 1995, 25, 1; (b) N. K. Mottet, M. E. Vahter, J. S. Charleston and L. T. Friberg, Met. Ions Biol. Syst., 1997, 34, 371; (c) T. W. Clarkson, L. Magos and G. J. Myers, N. Engl. J. Med., 2003, 349, 1731.
3. (a) Q. Wang, D. Kim, D. D. Dionysiou, G. A. Sorial and D. Timberlake, Environ. Pollut., 2004, 131, 323; (b) P. B. Tchounwou, W. K. Ayensu, N. Ninashvili and D. Sutton, Environ. Toxicol., 2003, 18, 149; (c) R. Eisler, Environ. Geochem. Health, 2003, 25, 325.
4. (a) E. M. Nolan and S. J. Lippard, Chem. Rev., 2008, 108, 3443–3480; (b) Q. He, E. W. Miller, A. P. Wong and C. J. Chang, J. Am. Chem. Soc., 2006, 128, 9316–9317; (c) K. Harano, S. Hiraoka and M. Shionoya, J. Am. Chem. Soc., 2007, 129, 5300; (d) M. Leermakers, W. Baeyens, P. Quevauviller and M. Horvat, Trends Anal. Chem., 2005, 24, 383.
5. (a) L. Zeng, E. W. Miller, A. Pralle, E. Y. Isacoff and C. J. Chang, J. Am. Chem. Soc., 2006, 128, 10; (b) G. Farruggia, S. Iotti, L. Prodi, N. Zaccheroni, M. Montalti, P. B. Savage, G. Andreani, V. Trapani and F. I. Wolf, J. Fluoresc., 2008, 19, 11.
6. (a) H. Komatsu, T. Miki, D. Citterio, T. Kubota, Y. Schindo, Y. Kitamura, K. Oka and K. Suzuki, J. Am. Chem. Soc., 2005, 127, 10798; (b) M. Taki, M. Desaki, A. Ojida, S. Iyoshi, T. Hirayama, I. Hamachi and Y. Yamamoto, J. Am. Chem. Soc., 2008, 130, 12564; (c) Q. He, E. W. Miller, A. P. Wong and C. J. Chang, J. Am. Chem. Soc., 2006, 128, 9316; (d) B. K. Kanungo, M. Baral, R. K. Bera and S. K. Sahoo, Monatsh. Chem., 2010, 141, 157; (e) R. Akbar, M. Baral and B. K. Kanungo, Spectrochim. Acta, Part A, 2014, 129, 365.
7. H. Yang, Z. Zho, K. Huang, M. Yu, F. Li, T. Yi and C. Huang, Org. Lett., 2007, 9, 4729.
8. (a) H. S. Jung, P. S. Kwon, J. W. Lee, J. I. Kim, C. S. Hong, J. W. Kim, S. Yan, J. Y. Lee, J. H. Lee, T. Joo and J. S. Kim, J. Am. Chem. Soc., 2009, 131, 2008; (b) E. M. Nolan and S. J. Lippard, Acc. Chem. Res., 2009, 42, 193.
9. S. Goswami, K. Aich, A. K. Das, A. Manna and S. Das, RSC Adv., 2013, 3, 2412.
10. (a) H. A. Benesi and J. H. Hildebrand, J. Am. Chem. Soc., 1949, 71, 2703; (b) M. Barra, C. Bohne and J. C. Scaiano, J. Am. Chem. Soc., 1990, 112, 8075; (c) M. Zhu, M. J. Yuan, X. F. Liu, J. L. Xu, J. Lv, C. S. Huang, H. B. Liu, Y. L. Li, S. Wang and D. B. Zhu, Org. Lett., 2008, 10, 1481.
11. M. D. Sun, S. D. Wang, Q. B. Yang, X. L. Fei, Y. X. Li and Y. P. Li, RSC Adv., 2014, 4, 8295.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra03908d

---