A Sensitive Water-Soluble Reversible Optical Probe for Hg2+ Detection

4 hours ago - Synopsis. We report a highly water-soluble, reversible, optical sensor for detecting low ppb levels of inorganic mercury in contaminated...
31 downloads 9 Views 4MB Size
Download PDF
Article pubs.acs.org/IC

Cite This: Inorg. Chem. XXXX, XXX, XXX−XXX

A Sensitive Water-Soluble Reversible Optical Probe for Hg2+ Detection Sayani Das,† Anindita Sarkar,† Ananya Rakshit,† and Ankona Datta*,† †

Department of Chemical Sciences, Tata Institute of Fundamental Research (TIFR), 1 Homi Bhabha Road, Colaba, Mumbai 400005, India S Supporting Information *

ABSTRACT: We report the serendipitous discovery of an optical mercury sensor while trying to develop a water-soluble manganese probe. The sensor is based on a pentaaza macrocycle conjugated to a hemicyanine dye. The pentaaza macrocycle earlier designed in our group was used to develop photoinduced electron transfer (PET)-based “turn-on” fluorescent sensors for manganese.1 In an attempt to increase the water-solubility of the manganese sensors we changed the dye from BODIPY to hemicyanine. The resultant molecule qHCM afforded a distinct reversible change in the absorption features and a concomitant visible color change upon binding to Hg2+ ions, leading to a highly water-soluble mercury sensor with a 10 ppb detection limit. The molecule acts as a reversible “ON−OFF” fluorescent sensor for Hg2+ with a 35 times decrease in the emission intensity in the presence of 1 equiv of Hg2+ ions. We have demonstrated the applicability of the probe for detecting Hg2+ ions in living cells and in live zebrafish larvae using confocal fluorescence microscopy with visible excitation. High selectivity and sensitivity toward Hg2+ detection make qHCM an attractive probe for detecting Hg2+ in contaminated water sources, which is a major environmental toxicity concern. We have scrutinized the altered metal-ion selectivity of the probe using density functional theory (DFT) and time-dependent DFT calculations, which show that a PET-based metal-sensing scheme is not operational in qHCM. 1H NMR studies and DFT calculations indicate that Hg2+ ions coordinate to oxygen-donor atoms from both the chromophore and macrocycle, leading to sensitive mercury detection.

■

INTRODUCTION

soluble in water up to millimolar concentrations, making it suitable for Hg2+ detection in contaminated potable water. Additionally, qHCM affords both fluorescence excitation (λex = 535 nm) and emission (λmax = 589 nm) in the visible region, allowing applications in in vivo optical imaging. The sensor exhibits a highly selective 35 times decrease in the fluorescence emission intensity at 589 nm in the presence of 1 equiv of Hg2+ ions. The uptake of inorganic Hg in fish is a major concern for Hg toxicity through fish intake.39 Inorganic Hg is converted to alkyl Hg by bacteria living in the gut of fish.40 Alkyl Hg can be readily converted back to inorganic Hg and vice versa within living systems.41−43 The detection of inorganic Hg in living marine systems can therefore afford information on food contamination and also provide tools for elucidating Hg2+ uptake and removal pathways in biological systems. In this context, we have used confocal fluorescence microscopy to demonstrate that the sensor is readily cell-permeable and can be used to detect inorganic Hg both in live cells and in zebrafish larvae.

Heavy-metal ions like mercury (Hg) are toxic to most living organisms.2−5 Exposure to Hg from environmental, industrial, and food sources leads to renal dysfunction and neurological disorders.2−5 One of the major sources of Hg toxicity is contaminated potable water. The relevant speciation for Hg toxicity via drinking water is inorganic Hg in the form of labile Hg2+ ions.2−5 The World Health Organization guideline for the tolerable limit of inorganic Hg in drinking water is 6 parts per billion (ppb).2 Therefore, a sensitive optical probe6−9 that can detect the ppb levels of inorganic Hg in drinking water will be extremely useful for establishing the drinkability of water, especially in remote areas where sophisticated instrumentation for analysis of the metal content in water is not available. While there are numerous reports on optical sensors for detecting Hg2+,6,10−35 major shortcomings are related to limited aqueous solubility and irreversibility (Table S1).6,9,36,37 Very few sensors have a combination of high aqueous solubility and reversibility along with the requisite limit of detection.26,38 We report the serendipitous discovery of a highly sensitive, water-soluble, reversible, colorimetric, and fluorescent sensor, qHCM, for Hg2+ detection (Scheme 1a). The sensor has a limit of detection of 10 ppb (50 nM) for inorganic Hg and is readily © XXXX American Chemical Society

Received: February 4, 2018

A

DOI: 10.1021/acs.inorgchem.8b00310 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Scheme 1. (a) Colorimetric and Fluorescent Sensor qHCM for Detecting Hg2+ Ions and (b) Synthetic Scheme for qHCM

■

RESULTS AND DISCUSSION qHCM was synthesized via condensation of a pentaaza macrocycle aldehyde 2 with a water-soluble hemicyanine dye 3 (Scheme 1b). We have previously reported a Mn2+-selective “turn-on” sensor M1 based on the pentaaza macrocycle aldehyde 2 condensed to a boron dipyrromethene (BODIPY) dye (Figure S1).1 M1 was applied successfully for Mn2+ detection in live cells.1 However, because of the lipophilic nature of both the receptor unit 1 and the BODIPY unit the probe accumulated in lipid-rich regions within live cells.1 qHCM was designed as a second-generation water-soluble counterpart of M1 and synthesized in three steps starting from the synthesis of macrocycle 1 (Scheme 1b and Figure S2). qHCM was readily soluble in water. The absorbance values at 430 nm of aqueous buffered solutions of qHCM scaled linearly up to 2.5 mM concentration (Figure S5). Hence, all experiments with qHCM were performed in aqueous buffered solutions without the addition of any organic solvent. The absorption spectrum of qHCM showed a peak at 535 nm (Figure 1a). Upon excitation at 535 nm, an emission peak at 589 nm was observed in the fluorescence spectrum (Figure 2a). In order to test the in vitro response of qHCM as a potential divalent metal-ion sensor, absorption and emission spectra were recorded in the presence of different metal ions (Figures 1a, 2a, and S6, S10, and S11). The absorption spectrum of qHCM (2.5 μM) showed a distinct decrease in the intensity of the peak at 535 nm upon the addition of Hg2+ ions (Figures 1a and S6a). Simultaneously, a new peak appeared at 390 nm, which increased in the intensity with increasing Hg2+ levels. We observed a visible color change of the sensor solution from pink to faint yellow upon Hg2+ addition (Figure 1a, inset), which correlated with the blue shift from the green to blue region recorded in the absorption spectrum. Furthermore, the Hg2+ dose-dependent absorption intensity decrease at 535 nm was linear within the range of 50 nM to 1.6 μM and was fitted to a straight line to obtain the limit of detection for Hg2+ (Figure 1b). The limit of detection for Hg2+ with qHCM was determined to be 50 nM (10 ppb).

Figure 1. (a) Absorbance response of qHCM (2.5 μM) to Hg2+ (0− 2000 nM, starting from the 50 nM Hg2+ concentration) in 20 mM HEPES and 100 mM KNO3 buffer at pH 7.1. Inset: Image depicting glass vials containing qHCM (100 μM, left) and qHCM (100 μM) in the presence of Hg2+ (100 μM, right). (b) Linear correlation between the absorption of qHCM at 535 nm and the concentration of Hg2+ (50−1600 nM) used to calculate the limit of detection for Hg2+.

Biologically relevant metal ions as well as other heavy-metal ions like Pb2+ did not afford any change in the absorption spectrum of qHCM upon metal ion addition (Figures S6 and S7). The only other heavy-metal ion that afforded a similar change in the absorption spectrum of qHCM was Cd2+ (Figure S6k). However, when the absorption intensity decrease was compared at the same concentrations of Hg2+ and Cd2+ (2 μM) B

DOI: 10.1021/acs.inorgchem.8b00310 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 2. (a) Fluorescence response of qHCM (2.5 μM) to Hg2+ (0−2000 nM, starting from the 50 nM Hg2+ concentration). (b) Fluorescence intensity at 589 nm in the absence of any metal ion (F0) over the observed fluorescence intensity (F) for qHCM (2.5 μM) in the presence of various metal ions ([Mn+] = 2 μM). (c) Red bars represent F0/F values at 589 nm for qHCM (2.5 μM) upon the addition of [Mn+]: [Zn2+, Ni2+, Fe3+, Mn2+] = 10 μM, [Cu2+, Cd2+, Pb2+] = 2 μM, and [Mg2+, Na+, Ca2+] = 1 mM. Gray bars represent F0/F values at 589 nm upon the subsequent addition of Hg2+, [Hg2+] = 2 μM. (d) Fluorescence response of qHCM (2.5 μM; black line), qHCM + Hg2+ (2.5 μM each; red line), and qHCM + Hg2+ + DMSA (2.5 μM each, blue line). Inset: Image depicting glass vials containing qHCM (100 μM; left), qHCM + Hg2+ (100 μM each; middle), and qHCM + Hg2+ + DMSA (100 μM each, right). Measurements were performed in 20 mM HEPES and 100 mM KNO3 at pH 7.1. λex = 535 nm.

solutions. The fluorescence response of qHCM remained unaffected within the physiologically relevant pH range of 6−9 (Figure S12). The reversibility of the fluorescence response of a sensor is an important criterion for the detection of changes in metal-ion concentrations in live biological samples.9 Colorimetric reversibility will also be useful in the detection of Hg2+ in potable water sources. First, the probe can be used to test whether water is contaminated with Hg2+. Next, contaminated water can be passed through a purification system, and a color change will indicate whether heavy-metal contamination has been removed. We therefore checked whether the colorimetric and fluorescence responses of qHCM toward Hg2+ were reversible. Dimercaptosuccinic acid (DMSA) was used as a nonfluorescent Hg2+ chelator to check the reversibility of the sensor.44 Both the fluorescence and colorimetric response of the sensor completely recovered upon the addition of DMSA, indicating that qHCM is a reversible Hg2+ sensor (Figure 2d). LC−electrospray ionization MS (ESI-MS) data also corroborated reversible Hg2+ complexation (Figures S3 and S4). Several reported Hg2+ sensors are based on either metal-ioninduced spirolactam ring opening in rhodamine-based dyes or metal-induced reactions and do not afford a reversible response (Table S1).21,24,25,27,32,45−49 Hence, qHCM has a distinct advantage over these irreversible sensors. Encouraged by the highly selective fluorescence response of qHCM in the presence of Hg2+, we next checked the in-cell and in vivo applicability of the probe. In order to ensure that the sensor would remain intact in cellular milieu, the stability of qHCM was monitored in a HEK293T cell extract using LC− ESI-MS. qHCM was found to be highly stable, and no degradation products were observed up to the monitored time

the response to Hg2+ was 14.5 times higher (Figure S7). qHCM also showed an absorption spectral shift in the presence of Cu2+ (Figure S6f). When the absorption intensities were compared at the same concentrations of Hg2+ and Cu2+, the response to Hg2+ was 14.9 times higher (Figure S7). The absorption spectra in the presence of different metal ions (Figures S6 and S7) reveal that qHCM is a highly selective, sensitive, and water-soluble sensor for Hg2+. The Hg2+-dependent decrease in the absorption intensity for qHCM translated to a significant intensity decrease in the fluorescence emission of qHCM with visible excitation, upon Hg2+ titration (Figures 2a and S10). The fluorescence response (λex= 535 nm) was highly selective toward Hg2+ over biologically relevant as well as other heavy-metal ions. When the fluorescence response was plotted for different metal ions (2 μM), a selective 35 times decrease in the emission intensity was observed for Hg2+ (Figure 2b). Further, fluorescence competition experiments were performed in which Hg2+ ions were added to qHCM after the addition of physiologically and environmentally relevant concentrations of different metal ions. The emission intensity barely changed in the presence of competing metal ions (Figure 2c and S11). Upon the subsequent Hg2+ addition to qHCM in the presence of competing metal ions, we observed an identical fluorescence intensity decrease with an average F0/F value of 33 ± 2 (Figures 2c and S11). The absorption, fluorescence, and robust competition titration data taken together demonstrate that qHCM will be suitable for Hg2+ detection in both environmental and biological contexts. Because the sensor might encounter different pH environments in vivo, we also monitored the emission of qHCM in the absence and presence of Hg2+ under varying pH conditions using aqueous buffered C

DOI: 10.1021/acs.inorgchem.8b00310 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry of 9 h (Figures S14 and S15). Hence, we went ahead and tested whether qHCM could detect Hg2+ within live cells. HEK293T cells were incubated with qHCM in a phosphate buffer. Confocal images showed that the sensor readily entered cells within 30 min of incubation and showed bright punctated spots in the cytoplasm (Figure 3a(i)). Bright punctated spots might indicate sensor aggregation in cells.50 In order to check whether the sensor aggregated in cells, we performed dynamic light scattering (DLS) experiments with qHCM under concentrations and buffer conditions identical with those of the cell studies (Figure S13). The DLS count rates for the qHCM solution were indistinguishable from that of buffer alone, ruling out any possibility of aggregation. Colocalization studies with different organelle markers like Endosomal-tracker, Lysotracker, and Mito-tracker dyes indicated partial colocalization with endosomes and lysosomes with a slightly higher Pearson’s coefficient value for endosomes, which might explain the punctated staining (Figure S16). The cells incubated with the sensor were then treated with increasing concentrations of Hg2+ ions ranging from 2 to 10 μM in separate batches in order to evaluate the probe response in live cells (Figure 3). Hg2+-treated cells showed a significant dose-dependent decrease in the fluorescence intensity compared to control untreated cells, as shown in the intensity analysis data (Figure 3b). The live cell Hg2+ titration response clearly indicates that qHCM is suitable for detecting Hg2+ in a complex biological environment. Importantly, the dose-dependent intensity decrease confirms that the probe selectivity is translated effectively into a biological medium without any interference from other intracellular biological species and metal ions. Fish intake is a source of Hg toxicity in humans.41,43 We therefore checked whether qHCM can be used to detect Hg2+ contamination in fish larvae. Three-day-old zebrafish larvae were incubated with qHCM. Confocal images of the larvae indicated sensor uptake in the yolk sak and eyes of the larvae (Figure 4). The larvae were then washed with a phosphate buffer to remove excess sensor that did not enter the larvae and finally treated with Hg2+. Fluorescence confocal images of these larvae showed a clear decrease in the fluorescence intensity in the Hg2+-treated larvae in comparison to the untreated ones (Figure 4). The Hg2+ dose-dependent fluorescence intensity decrease in live mammalian cells, taken together with imaging in live larvae, indicates that qHCM can be used to detect Hg2+ accumulation in biological systems. Control uncontaminated samples will, however, be required to ascertain the presence of Hg2+ in biological samples using qHCM because the probe affords a turn-off fluorescence response. Finally, we wanted to investigate the mechanism of Hg2+ sensing by qHCM. We performed a 1H NMR titration of Hg2+ on qHCM to check whether Hg2+ binding induced any structural change in the sensor. Distinct peak shifts were observed in the aromatic linker and hemicyanine dye regions (Figure 5a). The peak shifts observed have been color-coded. Maximum peak shifts (0.39−0.58 ppm) were obtained in the aromatic linker region, as shown in Figure 5a. The 1H NMR peaks for the macrocycle protons broadened but did not afford distinct peak shifts (Figure S17). However, a considerable peak shift (0.26 ppm) was also observed for the methylene protons of the ester arms of the macrocycle, and a shift (0.14 ppm) was observed for the methylene protons of the carboxylate arm on the hemicyanine dye (Figure S17). Hence, the 1H NMR data indicated that the Hg2+ ion might coordinate to the oxygen-

Figure 3. (a) Confocal images of live HEK293T cells: (i) incubated with qHCM (10 μM); (ii) incubated with qHCM (10 μM), followed by treatment with HgCl2 (2 μM); (iii) incubated with qHCM (10 μM), followed by treatment with HgCl2 (5 μM); (iv) incubated with qHCM (10 μM), followed by treatment with HgCl2 (10 μM). λex = 543 nm. Scale bar: 30 μm. (b) Bar plots representing the average intensities obtained from intensity analysis of confocal images of cells shown in part a. Data are presented as mean ± standard deviation, where n = 3 for each set. Intensity analysis of cells incubated with qHCM and HgCl2 (10 μM) could not be achieved because of complete attenuation of the fluorescence signal in the images. Statistical analysis using the Student’s t test indicated a significant difference in the intensity between untreated and Hg2+-treated cells in a dose-dependent manner.

donor atoms from the dye along with the oxygen-donor atoms from the macrocycle arms, thereby placing it close to the aromatic linker region. Coordination of the oxygen-donor atoms from both the dye and macrocycle would cause a structural change in the chromophore, leading to a reversible change in the absorption spectrum of the probe. D

DOI: 10.1021/acs.inorgchem.8b00310 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

the oxygen-donor atoms from both the dye and macrocycle and induces a conformation change in the chromophore, leading to selective Hg2+ sensing. Our previously reported Mn2+-selective sensor M1 had a macrocyclic scaffold that was identical with that of qHCM (Figure S1).1 However, qHCM senses Hg2+ over Mn2+. The sensor M1 afforded a “turn-on” response upon Mn2+ binding to the scaffold because Mn2+ coordination to the macrocycle blocked photoinduced electron transfer (PET)-based quenching of BODIPY from the dimethylaniline (DMA)-linked macrocycle, as shown in Figure 6, top left. We show, using the electronic structure calculations described below, that in qHCM PET is not operational. In order to investigate the difference in the sensing mechanism between qHCM and M1, we performed DFT and time-dependent DFT (TD-DFT) calculations to determine the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) energy levels of BODIPY−DMA and hemicyanine−DMA with respect to the macrocyclic scaffold. When we scrutinized the HOMO orbital of BODIPY−DMA, we found that the electron density was localized on the BODIPY moiety (Figure 6). The DMA moiety was oriented at an angle of 90° with respect to the BODIPY moiety in the energy-optimized structure (Figure S18a). This result indicated a weak electronic coupling between the macrocyclic scaffold and BODIPY, which enables a nonadiabatic PET process between the two moieties. Additionally, ground-state DFT calculations show that the energy-level alignment of the BODIPY-centered HOMO with respect to that of the macrocyclic scaffold HOMO provides a favorable driving force for PET (Figure 6). Indeed, TD-DFT calculations on BODIPY−DMA indicated a localized HOMO-to-LUMO excitation on the BODIPY moiety, validating the conclusions drawn from the ground-state DFT calculations (Table S2). Metal-ion binding to the macrocycle would allow lowering of the macrocycle HOMO, leading to “turn-on” sensing of the metal ion in M1. In contrast, the DFT-optimized geometry of hemicyanine− DMA indicated a planar orientation of hemicyanine with respect to DMA (Figure S18b). The TD-DFT calculations on hemicyanine−DMA indicated a delocalized HOMO−3-toLUMO transition over the entire hemicyanine−DMA moiety (Figure 6 and Table S2). These results indicate a strong electronic coupling between the hemicyanine dye and the macrocyclic scaffold via the DMA moiety, which precludes a long-range nonadiabatic PET process. Rather, photoexcitation would lead to a vertical transition between states that are delocalized over the entire hemicyanine−DMA moiety. As shown by our 1H NMR data (Figure 5a) and DFT calculations on the Hg2+−qHCM complex (Figure 5b,c), Hg2+ coordination to qHCM via oxygen-donor atoms from both the dye and macrocyclic scaffold induces structural changes that manifest as an absorption shift. Thus, the reversible dual colorimetric and fluorescence response of the sensor toward Hg2+ can be explained in terms of an absorption shift, which translates to a fluorescence response. DFT and TD-DFT calculations on the BODIPY fluorophore indicate that tuning the energy levels of the dye with respect to the metal-binding scaffold should be considered for the development of PET-based sensors for metal ions. Importantly, results on the hemicyanine-based sensor qHCM highlight that, in addition to the right placement of the dye and metal-binding scaffold energy levels, the linker moiety

Figure 4. (a) Confocal images of 3-day-old zebrafish larvae: incubated with qHCM (10 μM) (top); incubated with qHCM (10 μM), followed by HgCl2 (10 μM) (bottom). λex = 543 nm. Scale bar: 300 μm. (b) Bar plots representing the average intensities obtained from intensity analysis of confocal images of zebrafish larvae shown in part a. Data are presented as mean ± standard deviation, where n = 6 for each set. Statistical analysis using the Student’s t test indicated a significant difference in the intensity between untreated and Hg2+-treated larvae.

In order to further investigate the mode of Hg2+ binding to qHCM, we performed density functional theory (DFT) calculations. However, before performing the DFT calculations, we needed information on the binding stoichiometry of Hg2+ to qHCM. The LC−ESI-MS data of the metal complex showed the presence of 1:1, 1:2, and 1:3 Hg2+−qHCM complexes with several multiply charged peaks for 1:1 and 1:2 species (Figure S3). The peak in Job’s plot was closest to the 1:2 Hg2+−qHCM binding stoichiometry (Figure S8). Further, the binding isotherm obtained from the absorption titration of Hg2+ ions to the probe could be best simulated with a 1:2 Hg2+−qHCM binding model by considering a β12 (cumulative stability constant) value of 1014 (Figure S9). Taken together, the binding isotherm and Job’s plot analysis indicated that the major species formed between Hg2+ and qHCM was a 1:2 complex. The data further indicated tight binding of qHCM to Hg2+, which implied the participation of strong donor atoms. Hence, we performed DFT calculations on a 1:2 Hg2+−qHCM complex. In order to assess the binding interactions, we also performed calculations on a 1:1 complex for comparison. The optimized structure of both 1:1 and 1:2 Hg2+−qHCM complexes indicated that the Hg2+ ion preferentially coordinated to both the carboxylate and ester arm oxygen atoms from the dye and macrocycle, respectively (Figure 5b,c and Videos S1 and S2). Attempts to place the Hg2+ ion in the center of the macrocycle failed to provide optimized structures. The 1H NMR titration data in conjunction with the structure optimization calculations via DFT indicate that Hg2+ binds to E

DOI: 10.1021/acs.inorgchem.8b00310 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 5. (a) Stacked 1H NMR spectra showing the aromatic region (6.8−8.4 ppm) of qHCM (1.18 mM) in D2O with increasing concentrations of Hg2+ ions. The color code reflects the extent of peak shift (Δδ): 0.08−0.10 ppm (blue), 0.11−0.14 ppm (green), 0.19−0.22 ppm (yellow), and 0.39−0.58 ppm (red). The corresponding protons are indicated in the molecular structure of qHCM. (b) Energy-optimized geometry of the Hg2+− qHCM (1:1) complex. (c) Energy-optimized geometry of the Hg2+−qHCM (1:2) complex. Representation: white balls, hydrogen atoms; golden balls, carbon atoms; red balls, oxygen atoms; blue balls, nitrogen atoms; purple balls, Hg2+ ions.

Figure 6. Top: Representative diagram to explain the mechanism of PET, where the blue ball represents a metal ion that can coordinate to the scaffold. Bottom: Molecular orbital energies and most probable electronic transitions of BODIPY−DMA, hemicyanine−DMA, and macrocyclic scaffold−DMA as predicted by DFT and TD-DFT calculations.

F

DOI: 10.1021/acs.inorgchem.8b00310 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Other biologically and environmentally relevant metal ions were used to test the selectivity of qHCM toward Hg2+. Hg(ClO4)2·H2O was used as the source of Hg2+. Fe3+, Zn2+, Ca2+, Mg2+, Mn2+, Na+, Ni2+, Cd2+, Pb2+, and Cu2+ were delivered in the form of their chlorides or nitrates as FeCl3·6H2O, ZnCl2, CaCl2·2H2O, MgCl2·6H2O, MnCl2· 4H2O, NaCl, NiCl2, Cd(NO3)2·4H2O, Pb(NO3)2, and Cu(NO3)2· 3H2O from stock solutions prepared in 20 mM HEPES and a 100 mM KNO3 buffer at pH 7.1. The buffers were Chelex-treated overnight prior to all measurements. The concentration of qHCM for all of the absorbance and fluorescence experiments was 2.5 μM. To test the reversibility of qHCM, 2,3-meso-DMSA was used as a known nonfluorescent Hg2+ chelator.44 A total of 1 equiv of DMSA in a buffer was added to the qHCM−Hg2+ complex in 20 mM HEPES and a 100 mM KNO3 buffer solution. The limit of detection was calculated based on absorbance titrations. Increasing amounts of M2+ were added to the sensor qHCM (2.5 μM) in 20 mM HEPES and a 100 mM KNO3 buffer. A plot of the measured absorbance at the absorbance peak wavelength (535 nm) versus concentration of M2+ added allowed calculation of the limit of detection from eq 1

responsible for electronically coupling the PET donor and acceptor should be designed carefully.

■

CONCLUSIONS In conclusion, we report the serendipitous discovery of a watersoluble, reversible “ON−OFF” colorimetric and fluorescent sensor for detecting Hg2+ ions in water. The sensor affords a highly selective 35 times decrease in the fluorescence emission intensity in the presence of Hg2+ over other metal ions. The sensor is cell-permeable and can be used to detect Hg2+ accumulation in living biological systems. An investigation of the mechanism of Hg2+ sensing indicates that the coordination of Hg2+ ions to oxygen-donor atoms from both the dye and macrocycle leads to a distinct change in the absorption spectrum, allowing the highly sensitive detection of low ppb levels of this toxic metal ion.

■

EXPERIMENTAL SECTION LOD =

Synthesis of qHCM. All compounds until compound 2 were synthesized according to previously reported procedures.1 qHCM was prepared via a condensation reaction of compound 2 with Nquaternized hemicyanine dye 3. Dye 3 was synthesized by refluxing 2,3,3-trimethylbenzoindolenine with 3-bromopropionic acid in 1,2dichlorobenzene under inert conditions for 20 h according to a literature procedure.51 Compound 2 (0.02 g, 0.045 mmol) and dye 3 (0.02 g, 0.09 mmol) were dissolved in anhydrous acetonitrile (10 mL) under an argon atmosphere. The reaction mixture was refluxed for 12 h and evaporated in vacuo. The residue was purified by reverse-phase high-performance liquid chromatography. Elution solvents were solvent B [0.1% trifluoroacetic acid (TFA) in acetonitrile] and solvent A (0.1% TFA in water). A gradient of 15% solvent B (0−15 min), 15− 20% solvent B (15−20 min), 20% solvent B (20−35 min), 20−30% solvent B (35−40 min), 30% solvent B (40−60 min), 30−40% solvent B (60−65 min), 40% solvent B (65−80 min), 40−100% solvent B (80−85 min), and 100% solvent B (85−95 min) was run through a C18 column (10 μm, 250 × 10.0 mm, Phenomenex) to afford compound qHCM as a pink solid (0.0083 g, 26.4%). 1 H NMR (600 MHz, CD3OD, 298 K): δ (ppm) 8.33 (H8, 1H, d, J = 15.5 Hz), 7.94 (H10, 2H, d, J = 8.4 Hz ), 7.66 (H1+4, 2H, m), 7.56 (H2, 1H, app t, J = 7.7 Hz), 7.51 (H3, 1H, app t, J = 7.5 Hz), 7.37 (H9, 1H, d, J = 15.5 Hz), 6.91 (H11, 2H, d, J = 8.7 Hz), 4.77 (H6, 2H, t, J = 6.7 Hz), 3.81 (H12, 4H, t, J = 6.24 Hz), 3.72 (H19, 6H, s), 3.55 (H18, 4H, s), 3.42 (H17, 4H, br t), 3.17 (H13, 4H, br t), 3.03 (H14+15, 8H, m), 2.96 (H7, 2H, t, J = 6.7 Hz), 2.80 (H16, 6H, s), 1.80 (H5, 6H, s). 13 C NMR (150 MHz, CD3OD, 298 K): δ 182.25 (carboxylate carbon atoms), 173.49 (ester carbonyl carbon atoms), 156.61, 155, 144.12, 142.26, 135.59, 130.28, 129.33, 124.74, 123.86, 114.69, 113.71, 106.69 (aromatic + alkene carbon atoms), 56.65, 56.15, 53.40, 52.95, 52.79, 52.57, 52.23, 51.50, 49.85, 42.72, 42.43, 33.16, 27.38 (aliphatic carbon atoms). All of the NMR assignments are according to the numbers assigned in the spectra given in the Supporting Information. HRMS. Calcd for C39H57N6O6 ([MH+]): m/z 705.4334. Found: m/ z 705.4333. Absorbance and Fluorescence Measurements. All spectroscopic measurements were performed in 20 mM 2-[4-(2hydroxyethyl)piperazin-1-yl]ethanesulfonic acid (HEPES) and a 100 mM KNO3 buffer at pH 7.1 and 25 °C. UV−visible spectrophotometric measurements for the sensor were performed on a SPECORD 205 (Analytik Jena AG, Germany) using a quartz cuvette having a path length of 1 cm. Fluorescence spectra were recorded on a FluoroLog-3 (Horiba Jobin Yvon Inc.) spectrofluorometer using quartz cuvettes with 10 mm × 4 mm (Hellma Analytics) or 10 mm × 2 mm (Hellma Analytics) inner dimensions. Fluorescence spectra were obtained by excitation at 535 nm with a slit width of 5 nm for both excitation and emission.

3σ k

(1)

where σ is the standard deviation of the absorbance of a blank solution, which was measured three times, and k is the slope of the calibration curve.52−54 Cell Studies and Confocal Imaging. HEK 293T cells were cultured in Dulbecco’s modified Eagle’s medium (Sigma-Aldrich) supplemented with fetal bovine serum (10%, Gibco), penicillin (50 units/mL, Gibco) and streptomycin (50 μg/mL, Gibco) in T25 culture flasks at 37 °C under humidified air containing 5% CO2. A day before the imaging, the cells were plated on glass-coverslip-bottomed Petri plates (35 mm diameter, Tarsons) coated with polylysine (0.1 mg/mL) and fibronectin (100 μg/mL). Fluorescence images of the cells were recorded on a confocal microscope (LSM 710, Carl Zeiss, Germany) using a 40× water immersion objective. A 543 nm laser (argon source) was used for qHCM excitation. A buffer containing phosphate-buffered saline (PBS; 20 mM) and glucose (5 mM), with the pH adjusted to 7.4, was used during the confocal studies. A stock solution of qHCM (1986 μM) was prepared in Milli-Q water. The cells were washed with PBS and incubated with qHCM (10 μM in PBS) for 30 min. After incubation, the cells were washed three times with PBS and imaged. For mercury treatment, the same cells earlier stained with qHCM were incubated with HgCl2 (2, 5, and 10 μM in PBS) in different batches for 30 min each at room temperature. The Hg2+-treated cells were washed with PBS multiple times to remove excess HgCl2 and imaged. In Vivo Imaging of Hg2+ in Zebrafish Larvae. All experiments were performed on 3 dpf live zebrafish larvae obtained from wild-type stock. Embryos were incubated at 28 °C in an E3 medium containing NaCl (5 mM), KCl (170 μM), CaCl2 (330 μM), MgCl2 (330 μM), and methylene blue (0.6 μM) at pH 7.5. Zebrafish were maintained and bred following protocols approved by the institutional animal ethics committee (TIFR, India). The 3 dpf zebrafish larvae (n = 12) were incubated with qHCM (10 μM) in E3 media for 30 min at 28 °C. After washing with PBS to remove the excess sensor, the larvae were randomly divided into two groups of which one batch (n = 6) of the larvae were imaged and the rest (n = 6) were further incubated with a HgCl2 (10 μM) solution in E3 media for 30 min. The larvae were then washed with PBS to remove the excess HgCl2 and imaged using a 10× air objective on a confocal microscope. Quantification of the fluorescence intensity was performed using Fiji (ImageJ, NIH, USA). The fluorescence intensity was quantified by defining identical regions of interest using a Concentric Circles plugin and measuring the integrated intensity within this region. Statistical analysis was performed using the Student’s t test on intensities obtained for Hg2+-treated and untreated larvae. G

DOI: 10.1021/acs.inorgchem.8b00310 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

■

(7) Carter, K. P.; Young, A. M.; Palmer, A. E. Fluorescent Sensors for Measuring Metal Ions in Living Systems. Chem. Rev. 2014, 114, 4564− 4601. (8) Kim, H. N.; Ren, W. X.; Kim, J. S.; Yoon, J. Fluorescent and colorimetric sensors for detection of lead, cadmium, and mercury ions. Chem. Soc. Rev. 2012, 41, 3210−3244. (9) Nolan, E. M.; Lippard, S. J. Tools and Tactics for the Optical Detection of Mercuric Ion. Chem. Rev. 2008, 108, 3443−3480. (10) Nandre, J. P.; Patil, S. R.; Sahoo, S. K.; Pradeep, C. P.; Churakov, A.; Yu, F.; Chen, L.; Redshaw, C.; Patil, A. A.; Patil, U. D. A chemosensor for micro- to nano-molar detection of Ag+ and Hg2+ ions in pure aqueous media and its applications in cell imaging. Dalton Trans. 2017, 46, 14201−14209. (11) Long, Y.; Yang, M. P.; Yang, B. Q. Development and applications of two colorimetric and fluorescent indicators for Hg2+ detection. J. Inorg. Biochem. 2017, 172, 23−33. (12) Li, Y.; Liu, N.; Liu, H.; Wang, Y.; Hao, Y.; Ma, X.; Li, X.; Huo, Y.; Lu, J.; Tang, S.; Wang, C.; Zhang, Y.; Gao, Z. A novel label-free fluorescence assay for one-step sensitive detection of Hg2+ in environmental drinking water samples. Sci. Rep. 2017, 7, 45974. (13) Knight, A. S.; Kulkarni, R. U.; Zhou, E. Y.; Franke, J. M.; Miller, E. W.; Francis, M. B. A modular platform to develop peptoid-based selective fluorescent metal sensors. Chem. Commun. 2017, 53, 3477− 3480. (14) Bhatta, S. R.; Mondal, B.; Vijaykumar, G.; Thakur, A. ICT− Isomerization-Induced Turn-On Fluorescence Probe with a Large Emission Shift for Mercury Ion: Application in Combinational Molecular Logic. Inorg. Chem. 2017, 56, 11577−11590. (15) Sarkar, P. K.; Polley, N.; Chakrabarti, S.; Lemmens, P.; Pal, S. K. Nanosurface Energy Transfer Based Highly Selective and Ultrasensitive “Turn on” Fluorescence Mercury Sensor. ACS Sens. 2016, 1, 789−797. (16) Li, Y.; Zhou, H.; Chen, W.; Sun, G.; Sun, L.; Su, J. A simple AIEbased chemosensor for highly sensitive and selective detection of Hg2+ and CN−. Tetrahedron 2016, 72, 5620−5625. (17) Minami, T.; Sasaki, Y.; Minamiki, T.; Koutnik, P.; Anzenbacher, P.; Tokito, S. A mercury(II) ion sensor device based on an organic field effect transistor with an extended-gate modified by dipicolylamine. Chem. Commun. 2015, 51, 17666−17668. (18) Diaz de Grenu, B.; Garcia-Calvo, J.; Cuevas, J.; Garcia-Herbosa, G.; Garcia, B.; Busto, N.; Ibeas, S.; Torroba, T.; Torroba, B.; Herrera, A.; Pons, S. Chemical speciation of MeHg+ and Hg2+ in aqueous solution and HEK cells nuclei by means of DNA interacting fluorogenic probes. Chem. Sci. 2015, 6, 3757−3764. (19) Chen, Y.; Yang, C.; Yu, Z.; Chen, B.; Han, Y. A highly sensitive hemicyanine-based fluorescent chemodosimeter for mercury ions in aqueous solution and living cells. RSC Adv. 2015, 5, 82531−82534. (20) Zheng, J.; Nie, Y.; Hu, Y.; Li, J.; Li, Y.; Jiang, Y.; Yang, R. Timeresolved fluorescent detection of Hg2+ in a complex environment by conjugating magnetic nanoparticles with a triple-helix molecular switch. Chem. Commun. 2013, 49, 6915−6917. (21) Lin, W.; Cao, X.; Ding, Y.; Yuan, L.; Long, L. A highly selective and sensitive fluorescent probe for Hg2+ imaging in live cells based on a rhodamine-thioamide-alkyne scaffold. Chem. Commun. 2010, 46, 3529−3531. (22) Hu, Z.-Q.; Lin, C.-s.; Wang, X.-M.; Ding, L.; Cui, C.-L.; Liu, S.F.; Lu, H. Y. Highly sensitive and selective turn-on fluorescent chemosensor for Pb2+ and Hg2+ based on a rhodamine-phenylurea conjugate. Chem. Commun. 2010, 46, 3765−3767. (23) Choi, M. G.; Kim, Y. H.; Namgoong, J. E.; Chang, S.-K. Hg2+selective chromogenic and fluorogenic chemodosimeter based on thiocoumarins. Chem. Commun. 2009, 3560−3562. (24) Zhang, X.; Xiao, Y.; Qian, X. A ratiometric fluorescent probe based on FRET for imaging Hg2+ ions in living cells. Angew. Chem., Int. Ed. 2008, 47, 8025−9. (25) Shi, W.; Ma, H. Rhodamine B thiolactone: a simple chemosensor for Hg2+ in aqueous media. Chem. Commun. 2008, 1856−1858.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.8b00310. LC−MS spectra of qHCM, a qHCM−Hg2+ complex, and qHCM−Hg2+ in the presence of DMSA, purity, solubility, and absorption titration spectra with all metal ions, Job’s plot, fluorescence competition titration spectra, binding isotherms, stability constant value determination plot, stability of qHCM in cell extract, pH response of qHCM in the presence and absence of Hg2+, 1H and 13C NMR, colocalization studies with organelle trackers, and details of DFT and TD-DFT calculations (PDF) Video S1 (AVI) Video S2 (AVI)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Ananya Rakshit: 0000-0001-8250-0399 Ankona Datta: 0000-0003-0821-6044 Author Contributions

All authors have given approval to the final version of the manuscript. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS A.D. acknowledges the TIFR and Department of Atomic Energy, India, for research funding. The authors acknowledge Dr. Ravindra Venkatramani for extensive discussions on the TD-DFT data, the Biophotonics group, TIFR, for access to confocal microscope and Department of Chemical Sciences, TIFR, for access to cell culture facility, Dr. Antara Banerjee for providing cells, Triveni Menon for setting up zebrafish crosses and supplying larvae, National NMR facility, TIFR, and Mass Laboratory, Chemistry Department, Indian Institute of Technology, Bombay, India.

■

REFERENCES

(1) Bakthavatsalam, S.; Sarkar, A.; Rakshit, A.; Jain, S.; Kumar, A.; Datta, A. Tuning macrocycles to design ’turn-on’ fluorescence probes for manganese(II) sensing in live cells. Chem. Commun. 2015, 51, 2605−2608. (2) Mercury in Drinking-water, Background document for development of WHO Guidelines for Drinking-water Quality; WHO: Geneva, Switzerland, 2005. (3) Sahu, R.; Saxena, P.; Johnson, S.; Ramanathan, S.; Juneja, S. CSE Study: Mercury Pollution in Sonbhadra District of Uttar Pradesh and Its Health Impacts; Centre for Science and Environment: New Delhi, India, 2012. (4) Regulatory Impact Analysis of the Clean Air Mercury Rule; U.S. EPA: Washington, DC, 2005; EPA-452/R-05-003. (5) Indian Standard Drinking Water Specification, 2nd revision; Bureau of Indian Standards: New Delhi, India, 2012; IS 10500. (6) Yan, Z.; Yuen, M.-F.; Hu, L.; Sun, P.; Lee, C.-S. Advances for the colorimetric detection of Hg2+ in aqueous solution. RSC Adv. 2014, 4, 48373−48388. H

DOI: 10.1021/acs.inorgchem.8b00310 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry (26) Yoon, S.; Miller, E. W.; He, Q.; Do, P. H.; Chang, C. J. A Bright and Specific Fluorescent Sensor for Mercury in Water, Cells, and Tissue. Angew. Chem. 2007, 119, 6778−6781. (27) Wu, J.-S.; Hwang, I.-C.; Kim, K. S.; Kim, J. S. Rhodamine-Based Hg2+-Selective Chemodosimeter in Aqueous Solution: Fluorescent OFF−ON. Org. Lett. 2007, 9, 907−910. (28) Nolan, E. M.; Lippard, S. J. Turn-On and Ratiometric Mercury Sensing in Water with a Red-Emitting Probe. J. Am. Chem. Soc. 2007, 129, 5910−5918. (29) Wang, J.; Qian, X. Two regioisomeric and exclusively selective Hg(II) sensor molecules composed of a naphthalimide fluorophore and an o-phenylenediamine derived triamide receptor. Chem. Commun. 2006, 109−111. (30) Tatay, S.; Gaviña, P.; Coronado, E.; Palomares, E. Optical Mercury Sensing Using a Benzothiazolium Hemicyanine Dye. Org. Lett. 2006, 8, 3857−3860. (31) Yoon, S.; Albers, A. E.; Wong, A. P.; Chang, C. J. Screening Mercury Levels in Fish with a Selective Fluorescent Chemosensor. J. Am. Chem. Soc. 2005, 127, 16030−16031. (32) Yang, Y.-K.; Yook, K.-J.; Tae, J. A Rhodamine-Based Fluorescent and Colorimetric Chemodosimeter for the Rapid Detection of Hg2+ Ions in Aqueous Media. J. Am. Chem. Soc. 2005, 127, 16760−16761. (33) Nolan, E. M.; Lippard, S. J. MS4, a seminaphthofluoresceinbased chemosensor for the ratiometric detection of Hg(II). J. Mater. Chem. 2005, 15, 2778−2783. (34) Caballero, A.; Martínez, R.; Lloveras, V.; Ratera, I.; VidalGancedo, J.; Wurst, K.; Tárraga, A.; Molina, P.; Veciana, J. Highly Selective Chromogenic and Redox or Fluorescent Sensors of Hg2+ in Aqueous Environment Based on 1,4-Disubstituted Azines. J. Am. Chem. Soc. 2005, 127, 15666−15667. (35) Palomares, E.; Vilar, R.; Durrant, J. R. Heterogeneous colorimetric sensor for mercuric salts. Chem. Commun. 2004, 362− 363. (36) Hatai, J.; Pal, S.; Jose, G. P.; Bandyopadhyay, S. Histidine Based Fluorescence Sensor Detects Hg2+ in Solution, Paper Strips, and in Cells. Inorg. Chem. 2012, 51, 10129−10135. (37) Bhalla, V.; Roopa; Kumar, M.; Sharma, P. R.; Kaur, T. New Fluorogenic Sensors for Hg2+ Ions: Through-Bond Energy Transfer from Pentaquinone to Rhodamine. Inorg. Chem. 2012, 51, 2150−2156. (38) Nolan, E. M.; Lippard, S. J. A “Turn-On” Fluorescent Sensor for the Selective Detection of Mercuric Ion in Aqueous Media. J. Am. Chem. Soc. 2003, 125, 14270−14271. (39) Harris, H. H.; Pickering, I. J.; George, G. N. The chemical form of mercury in fish. Science (Washington, DC, U. S.) 2003, 301, 1203. (40) Schaefer, J. K.; Rocks, S. S.; Zheng, W.; Liang, L.; Gu, B.; Morel, F. M. M. Active transport, substrate specificity, and methylation of Hg(II) in anaerobic bacteria. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 8714−8719. (41) Wang, X.; Wu, F.; Wang, W. X. In Vivo Mercury Demethylation in a Marine Fish (Acanthopagrus schlegeli). Environ. Sci. Technol. 2017, 51, 6441−6451. (42) Karri, R.; Banerjee, M.; Chalana, A.; Jha, K. K.; Roy, G. Activation of the Hg−C Bond of Methylmercury by [S2]-Donor Ligands. Inorg. Chem. 2017, 56, 12102−12115. (43) Björnberg, K. A.; Vahter, M.; Englund, G. S. Methylmercury, Amalgams, and Children’s Health: Björnberg et al. Respond. Environ. Health Perspect. 2006, 114, A149−A150. (44) George, G. N.; Prince, R. C.; Gailer, J.; Buttigieg, G. A.; Denton, M. B.; Harris, H. H.; Pickering, I. J. Mercury Binding to the Chelation Therapy Agents DMSA and DMPS and the Rational Design of Custom Chelators for Mercury. Chem. Res. Toxicol. 2004, 17, 999− 1006. (45) Wang, J. L.; Li, W. L.; Long, L. P. Development of a nearinfrared fluorescence turn-on probe for imaging Hg2+ in living cells and animals. Sens. Actuators, B 2017, 245, 462−469. (46) Ji, R. X.; Liu, A. K.; Shen, S. L.; Cao, X. Q.; Li, F.; Ge, Y. Q. An indolizine-rhodamine based FRET fluorescence sensor for highly sensitive and selective detection of Hg2+ in living cells. RSC Adv. 2017, 7, 40829−40833.

(47) Bera, K.; Das, A. K.; Nag, M.; Basak, S. Development of a Rhodamine−Rhodanine-Based Fluorescent Mercury Sensor and Its Use to Monitor Real-Time Uptake and Distribution of Inorganic Mercury in Live Zebrafish Larvae. Anal. Chem. 2014, 86, 2740−2746. (48) Ma, X.; Wang, J.; Shan, Q.; Tan, Z.; Wei, G.; Wei, D.; Du, Y. A “Turn-on” Fluorescent Hg2+ Chemosensor Based on Ferrier Carbocyclization. Org. Lett. 2012, 14, 820−823. (49) Jiang, J.; Liu, W.; Cheng, J.; Yang, L.; Jiang, H.; Bai, D.; Liu, W. A sensitive colorimetric and ratiometric fluorescent probe for mercury species in aqueous solution and living cells. Chem. Commun. 2012, 48, 8371−3. (50) Morgan, M. T.; Bagchi, P.; Fahrni, C. J. Designed To Dissolve: Suppression of Colloidal Aggregation of Cu(I)-Selective Fluorescent Probes in Aqueous Buffer and In-Gel Detection of a Metallochaperone. J. Am. Chem. Soc. 2011, 133, 15906−15909. (51) Hirayama, T.; Van de Bittner, G. C.; Gray, L. W.; Lutsenko, S.; Chang, C. J. Near-infrared fluorescent sensor for in vivo copper imaging in a murine Wilson disease model. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 2228−2233. (52) Goswami, S.; Das, S.; Aich, K.; Sarkar, D.; Mondal, T. K.; Quah, C. K.; Fun, H.-K. CHEF induced highly selective and sensitive turn-on fluorogenic and colorimetric sensor for Fe3+. Dalton Trans. 2013, 42, 15113−15119. (53) Hakonen, A. Plasmon Enhancement and Surface Wave Quenching for Phase Ratiometry in Coextraction-Based Fluorosensors. Anal. Chem. 2009, 81, 4555−4559. (54) Thomsen, V.; Schatzlein, D.; Mercuro, D. Limits of Detection in Spectroscopy. Spectroscopy 2003, 18, 112−114.

I

DOI: 10.1021/acs.inorgchem.8b00310 Inorg. Chem. XXXX, XXX, XXX−XXX