Shedding Light on Heavy Metal Contamination: Fluorescein-Based Chemosensor for Selective Detection of Hg2+ in Water

This article discusses the design and analysis of a new chemical chemosensor for detecting mercury(II) ions. The chemosensor is a hydrazone made from 4-methylthiazole-5-carbaldehyde and fluorescein hydrazide. The structure of the chemosensor was confirmed using various methods, including nuclear magnetic resonance spectroscopy, infrared spectroscopy with Fourier transformation, mass spectroscopy, and quantum chemical calculations. The sensor’s ability in the highly selective and sensitive discovery of Hg2+ ions in water was demonstrated. The detection limit for mercury(II) ions was determined to be 0.23 µM. The new chemosensor was also used to detect Hg2+ ions in real samples and living cells using fluorescence spectroscopy. Chemosensor 1 and its complex with Hg2+ demonstrate a significant tendency to enter and accumulate in cells even at very low concentrations.


Introduction
Heavy metals are significant pollutants due to their ubiquitous presence in almost all natural ecosystems and their resistance to degradation processes, unlike organic pollutants [1].Metal ions accumulate in soil easily, but their removal is difficult and slow, leading to their intensive accumulation in the tissues and organs of humans, living organisms, and hydrobionts [2,3].Mercury occupies an unique position among heavy metals as its compounds are among the most toxic and hazardous substances [4].The areas with the highest levels of mercury pollution in Russia are in proximity to metallurgical plants located in the Kola Peninsula, Urals, and Norilsk, where concentrations are several times higher than background levels [5,6].The World Health Organization (WHO) defines the maximum allowable mercury concentration in drinking water as 0.006 mg/L [7].Regular consumption of seafood increases the risk of methylmercury poisoning, as various studies worldwide have discovered [8].Once in the body, the bloodstream easily transports mercury ions, leading to serious damage to the liver, kidneys, and brain [8][9][10].Mercury salts commonly exist as divalent cations in an aqueous solution.Therefore, monitoring the Hg 2+ ions in surface waters is a significant environmental safety task for numerous countries.
In the present article, we describe the synthesis and characteristics of a new sensor derived from 4-methylthiazole-5-carbaldehyde and hydrazide fluorescein (chemosensor 1) to recognize mercury(II) ions.The ability of the compound to sense cations (Na + , K + , Ag + , Ca 2+ , Mg 2+ , Ba 2+ , Co 2+ , Ni 2+ , Cu 2+ , Zn 2+ , Cd 2+ , Pb 2+ , Hg 2+ , Cr 3+ , Fe 3+ , Ce 3+ , Al 3+ , In 3+ ) was evaluated using UV-Vis and fluorescence experimental methods in an H 2 O/DMSO mixture.To supplement the experimental data, density functional theory (DFT) calculations were performed to determine the structure of 1 in solution.We also detected a strong enough fluorescence signal from the sensor with mercury ions in living cells, so it makes possible further work on the improvement of this sensor for application on living models.

Selection of the Optimal H 2 O/DMSO Ratio for the Determination of Hg 2+ Ions
Chemosensor 1 is insoluble in water, but soluble in many organic solvents (acetonitrile, alcohols, DMSO, THF, and DMF).Of these, DMSO is the most suitable for the determination of analytes because it has low toxicity to living organisms and is a high-boiling solvent, which makes it easier to prepare solutions of known concentrations.For these reasons, to select the optimal conditions for the determination of Hg 2+ ions, we studied the fluorescence of the mixture 1-Hg 2+ in the H 2 O-DMSO system.Up to a 30 vol.% of water, chemosensor 1 exhibits no reaction with Hg 2+ ions, likely due to the strong solvation of the cations by DMSO molecules [42].Additionally, the opening reaction of the spirolactam ring of fluorescein derivatives occurs most efficiently in aqueous solutions [43].Optimal fluorimetric conditions for the determination of mercury ions are observed in H 2 O-DMSO (8:2 v/v) (Figure 1).A sharp decrease in fluorescence intensity at 90 vol.%H 2 O is caused by aggregation of chemosensor 1.

Selectivity of Chemosensor 1 towards Hg 2+ and other Cations in H2O-DMSO (8:2 v/v)
UV-Vis and fluorescence studies were performed using a 50 µM solution of chemosensor 1 in H2O-DMSO (8:2 v/v) with 5 eq. of common metal ions (Na + , K + , Ag + , Ca 2+ , Mg 2+ , Ba 2+ , Co 2+ , Ni 2+ , Cu 2+ , Zn 2+ , Cd 2+ , Pb 2+ , Hg 2+ , Cr 3+ , Fe 3+ , Al 3+ , Ga 3+ , In 3+ , and Ce 3+ ) (Figure 2).Chemosensor 1 has two maxima in the UV region (276 and 336 nm).There are no absorption bands in the visible spectral region because of the closed spirolactam cycle, which makes the solution of 1 colorless (Figure 2a,c) [44].However, when Hg 2+ ions are added, the solution changes from transparent to brown, and a broad absorption band with a maximum of 448 nm appears.Additionally, fluorescence intensity at 520 nm significantly increases with the addition of 5 equiv. of Hg 2+ .The quantum yield also increases to 3.9% after the probe reacts with Hg 2+ ions only, which demonstrates the excellent selectivity of chemosensor 1 for Hg 2+ over other metal ions (Figure 2d).This selectivity can be utilized for the fluorescent determination of Hg 2+ ions in solution.Fe 3+ ions also induce the opening of the spirolactam cycle.However, their absorption is observed in the UV region and no fluorescence enhancement is observed (Figure 2c,d).
Chemosensor 1 has a substantial affinity towards mercury ions, which renders hydrazone a fitting alternative for detecting Hg 2+ .Accordingly, the fluorescence response of 1 (50 µM) to Hg 2+ ions (1 equiv.)mixed with some other metal ions (1 equiv.) in H 2 O-DMSO (8:2 v/v) is shown in Figure 3a.As can be seen from Figure 3, fluorescence intensity does not change significantly when 1 equiv. of various metal ions apart from Fe 3+ is added to the chemosensor 1 solution.However, Fe 3+ ions only interfere with the quantitative recognition of Hg 2+ ions.Probably, Hg 2+ and Fe 3+ ions are competitors for opening the spirolactam ring of chemosensor 1.Therefore, to determine Hg 2+ in the presence of Fe 3+ ions, it is necessary to mask the iron(III) ions.Fluorides are a suitable masking agent as they form a stable colorless complex [FeF 5 ] 2− .Chemosensor 1 exhibits high selectivity for Hg 2+ ions, which is not hindered by most of the other metal ions in the H 2 O-DMSO mixture.Anions have a negligible impact on the fluorimetric detection of Hg 2+ ions in solution, except halides such as Cl − , Br − , and I − (Figure 3b).According to the literature, Hg 2+ ions can form stable complexes of the composition [HgHal n ] 2−n with halides [45][46][47].Such anions as Cl − , Br − , and I − are competing with chemosensor 1 for Hg 2+ ions in solution.However, qualitative determination of mercury(II) ions is possible, when Hg 2+ ions and halides (Cl − , Br − , or I − ) are equimolar in solution.
Chemosensor 1 has two maxima in the UV region (276 and 336 nm).There are no absorption bands in the visible spectral region because of the closed spirolactam cycle, which makes the solution of 1 colorless (Figure 2a,c) [44].However, when Hg 2+ ions are added, the solution changes from transparent to brown, and a broad absorption band with a maximum of 448 nm appears.Additionally, fluorescence intensity at 520 nm significantly increases with the addition of 5 equiv. of Hg 2+ .The quantum yield also increases to 3.9% after the probe reacts with Hg 2+ ions only, which demonstrates the excellent selectivity of chemosensor 1 for Hg 2+ over other metal ions (Figure 2d).This selectivity can be utilized for the fluorescent determination of Hg 2+ ions in solution.Fe 3+ ions also induce the opening of the spirolactam cycle.However, their absorption is observed in the UV region and no fluorescence enhancement is observed (Figure 2c,d).
Chemosensor 1 has a substantial affinity towards mercury ions, which renders hydrazone a fitting alternative for detecting Hg 2+ .Accordingly, the fluorescence response of 1 (50 µM) to Hg 2+ ions (1 equiv.)mixed with some other metal ions (1 equiv.) in H2O-DMSO (8:2 v/v) is shown in Figure 3a.As can be seen from Figure 3, fluorescence intensity does not change significantly when 1 equiv. of various metal ions apart from Fe 3+ is added to the chemosensor 1 solution.However, Fe 3+ ions only interfere with the quantitative recognition of Hg 2+ ions.Probably, Hg 2+ and Fe 3+ ions are competitors for opening the spirolactam ring of chemosensor 1.Therefore, to determine Hg 2+ in the presence of Fe 3+ ions, it is necessary to mask the iron(III) ions.Fluorides are a suitable masking agent as they form a stable colorless complex [FeF5] 2− .Chemosensor 1 exhibits high selectivity for Hg 2+ ions, which is not hindered by most of the other metal ions in the H2O-DMSO mixture.Anions have a negligible impact on the fluorimetric detection of Hg 2+ ions in solution, except halides such as Cl − , Br − , and I − (Figure 3b).According to the literature, Hg 2+ ions can form stable complexes of the composition [HgHaln] 2−n with halides [45][46][47].Such anions as Cl − , Br − , and I − are competing with chemosensor 1 for Hg 2+ ions in solution.However, qualitative determination of mercury(II) ions is possible, when Hg 2+ ions and halides (Cl − , Br − , or I − ) are equimolar in solution.Adding increasing amounts (up to 180 µM) of Hg 2+ ions (Figure 4a) leads to an increase in the intensity of the emission band, with a maximum at 520 nm in the fluorescent spectra of chemosensor 1.This increase in fluorescence is attributed to the formation of a coordination compound with a higher quantum yield in comparison to chemosensor 1.We attribute such changes to opening the spirolactam ring of the chemosensor by complexation.The mechanism of spirolactam opening in xanthene dyes, leading to fluorescence turn-on, is well known [48,49].The stability constant of the 1-Hg 2+ complex was computed from the spectrofluorimetric data (Figure 4) using KEV software, version 0.7 [50], while the stoichiometry of the 1-Hg 2+ complex was determined via mass spectrometry Adding increasing amounts (up to 180 µM) of Hg 2+ ions (Figure 4a) leads to an increase in the intensity of the emission band, with a maximum at 520 nm in the fluorescent spectra of chemosensor 1.This increase in fluorescence is attributed to the formation of a coordination compound with a higher quantum yield in comparison to chemosensor 1.We attribute such changes to opening the spirolactam ring of the chemosensor by complexation.The mechanism of spirolactam opening in xanthene dyes, leading to fluorescence turn-on, is well known [48,49].The stability constant of the 1-Hg 2+ complex was computed from the spectrofluorimetric data (Figure 4) using KEV software, version 0.7 [50], while the stoichiometry of the 1-Hg 2+ complex was determined via mass spectrometry (Figure S2).Furthermore, the best description of the experimental fluorimetric data is observed for the 1:1 stoichiometric model (R 2 = 0.958) (Figure S3).The stoichiometric model 1:2 is suboptimal, leading to the conclusion that the formation of a 1:2 complex is improbable (Figure S4).The conditional stability constant of 1-Hg 2+ lg β = 3.54 ± 0.24 was determined using KEV software, version 0.7 [50] (Figure S3).A possible reaction scheme between chemosensor 1 and the Hg 2+ ion is shown in Figure S5.The detection limit of chemosensor 1 for Hg 2+ was also calculated from the titration to be 0.23 µM (Figure S6), which was lower than the majority of the reported LOD values for Hg 2+ chemosensors (Table S1).
(Figure S2).Furthermore, the best description of the experimental fluorimetric data is observed for the 1:1 stoichiometric model (R 2 = 0.958) (Figure S3).The stoichiometric model 1:2 is suboptimal, leading to the conclusion that the formation of a 1:2 complex is improbable (Figure S4).The conditional stability constant of 1-Hg 2+ lg β = 3.54±0.24was determined using KEV software, version 0.7 [50] (Figure S3).A possible reaction scheme between chemosensor 1 and the Hg 2+ ion is shown in Figure S5.The detection limit of chemosensor 1 for Hg 2+ was also calculated from the titration to be 0.23 µM (Figure S6), which was lower than the majority of the reported LOD values for Hg 2+ chemosensors (Table S1).  Figure 5a shows that the absorption properties of chemosensor 1 remain stable for 8 days, proving its practicality for Hg 2+ recognition purposes.Upon the addition of Hg 2+ to the chemosensor 1 solution, the fluorescence intensity quickly approached its peak after 100 min and remained stable thereafter (indicated by red dots).This demonstrates the chemosensor's rapid detection capability for mercury(II) ions.

Practical Application
We tested the practical usefulness of chemosensor 1 by examining water samples from local rivers for Hg 2+ ions.The water was collected from the Uvod (57°00′09.0″N Figure 5a shows that the absorption properties of chemosensor 1 remain stable for 8 days, proving its practicality for Hg 2+ recognition purposes.Upon the addition of Hg 2+ to the chemosensor 1 solution, the fluorescence intensity quickly approached its peak after 100 min and remained stable thereafter (indicated by red dots).This demonstrates the chemosensor's rapid detection capability for mercury(II) ions.
(Figure S2).Furthermore, the best description of the experimental fluorimetric data is observed for the 1:1 stoichiometric model (R 2 = 0.958) (Figure S3).The stoichiometric model 1:2 is suboptimal, leading to the conclusion that the formation of a 1:2 complex is improbable (Figure S4).The conditional stability constant of 1-Hg 2+ lg β = 3.54±0.24was determined using KEV software, version 0.7 [50] (Figure S3).A possible reaction scheme between chemosensor 1 and the Hg 2+ ion is shown in Figure S5.The detection limit of chemosensor 1 for Hg 2+ was also calculated from the titration to be 0.23 µM (Figure S6), which was lower than the majority of the reported LOD values for Hg 2+ chemosensors (Table S1). Figure 5a shows that the absorption properties of chemosensor 1 remain stable for 8 days, proving its practicality for Hg 2+ recognition purposes.Upon the addition of Hg 2+ to the chemosensor 1 solution, the fluorescence intensity quickly approached its peak after 100 min and remained stable thereafter (indicated by red dots).This demonstrates the chemosensor's rapid detection capability for mercury(II) ions.

Practical Application
We tested the practical usefulness of chemosensor 1 by examining water samples from local rivers for Hg 2+ ions.The water was collected from the Uvod (57°00′09.0″N

Practical Application
We tested the practical usefulness of chemosensor 1 by examining water samples from local rivers for Hg 2+ ions.The water was collected from the Uvod (57  It is also possible to qualitatively determine mercury(II) ions using a UV lamp.Yellowgreen fluorescence is only observed when Hg 2+ ions are in solution (Figure 2b).

Molecular Structure and Electronic Spectra of Chemosensor 1
Electronic absorption spectra play an important role in the design of chemosensors.Quantum chemical (QC) calculations are a powerful tool for simulating and then understanding the nature of electronic absorption spectra.It is also important to take into account that compounds can exist as several forms (conformers), which affect reactivity and spectral characteristics.
Quantum chemical calculations showed the coexistence of two conformers (1a and 1b) of 1 at room temperature (RT, T = 298 K), differing by an orientation of methylthiazole fragment (Figure 6).Both conformers possess a C s symmetry point group, i.e., there is a plane of symmetry including the atoms of thiazole and isoindoline rings.The contributions of 1a and 1b are directly proportional to e −∆G i RT and were evaluated as: where ∆G i is the relative Gibbs energy of the conformer.So, 1a dominates in equilibrium at RT; its amount is 96 mol.%.The transition from 1a to 1b can be achieved by rotating the thiazole ring around the hydrazone bridge with a barrier of ~7.7 kcal/mol (numbers of atoms correspond to those in the scheme for synthesizing).The potential energy surface profile of such a rotation is depicted in Figure 6; enlarged images of molecular models of 1a, 1b, and transition state (TS), along with their relative energies and contributions, can be found in ESI (Figure S8).Since the simulated electronic absorption spectra of 1a and 1b are close to each other (maximal difference in band position is 4 nm, Figure 7), we describe below only the spectrum of 1a, which is favorable according to QC calculations.The highest occupied molecular orbital (HOMO, Figure 8) with a" symmetry is localized on the bonding πorbitals of the thiazole and isoindoline rings and the C-N-bridge between them-however, without orbitals of the sulfur atom.The most intensive peak in the spectra is at 319 nm and corresponds to electronic transitions from HOMO to the lowest unoccupied molecular orbital (LUMO).The latter includes π-orbitals centered on the same rings as HOMO but with sulfur atomic orbitals.So, the band at 319 nm is caused by π → π* transitions in these fragments.The calculated composition of the lowest excited states and corresponding oscillator strengths are listed in Table S2.Shapes of orbitals involved in electronic transitions are represented in Table S3.Since the simulated electronic absorption spectra of 1a and 1b are close to each other (maximal difference in band position is 4 nm, Figure 7), we describe below only the spectrum of 1a, which is favorable according to QC calculations.The highest occupied molecular orbital (HOMO, Figure 8) with a" symmetry is localized on the bonding π-orbitals of the thiazole and isoindoline rings and the C-N-bridge between them-however, without orbitals of the sulfur atom.The most intensive peak in the spectra is at 319 nm and corresponds to electronic transitions from HOMO to the lowest unoccupied molecular orbital (LUMO).The latter includes π-orbitals centered on the same rings as HOMO but with sulfur atomic orbitals.So, the band at 319 nm is caused by π → π* transitions in these fragments.The calculated composition of the lowest excited states and corresponding oscillator strengths are listed in Table S2.Shapes of orbitals involved in electronic transitions are represented in Table S3 Figure 7. Calculated (solid lines) and experimental (dashed line) electronic absorption spectra of 1.Since the simulated electronic absorption spectra of 1a and 1b are close to each other (maximal difference in band position is 4 nm, Figure 7), we describe below only the spectrum of 1a, which is favorable according to QC calculations.The highest occupied molecular orbital (HOMO, Figure 8) with a" symmetry is localized on the bonding π-orbitals of the thiazole and isoindoline rings and the C-N-bridge between them-however, without orbitals of the sulfur atom.The most intensive peak in the spectra is at 319 nm and corresponds to electronic transitions from HOMO to the lowest unoccupied molecular orbital (LUMO).The latter includes π-orbitals centered on the same rings as HOMO but with sulfur atomic orbitals.So, the band at 319 nm is caused by π → π* transitions in these fragments.The calculated composition of the lowest excited states and corresponding oscillator strengths are listed in Table S2.Shapes of orbitals involved in electronic transitions are represented in Table S3

Cytotoxicity of 1 and 1:Hg 2+ Complex
The detection of mercury and its ions in living organisms may be very useful for various medical and research tasks due to their danger to the metabolism.Potential chemosensors must be both neutral to biological systems and highly sensitive, to detect very small amounts of target ions.Thus, to investigate possible harmful effects, 1 and its complex with Hg 2+ were tested using the MTT assay (Figure 9, Table 2).Compound 1 is found toxic enough to both tumor (HCT116) and non-tumor (HEK293T) cell lines when incubated for 3 days at a concentration of 20 µM or more.Half-maximal inhibitory concentrations (IC 50 ) in these cases are about 21 and 20 µM, respectively.In turn, the cytotoxicity of the 1:Hg 2+ complex is a little higher: the corresponding IC 50 values are about 15 and 16 µM.These results partly limit the application of the 1 compound as a chemosensor for living systems but pose the task of further modifying it in order to reduce such negative effects.

Cytotoxicity of 1 and 1:Hg 2+ Complex
The detection of mercury and its ions in living organisms may be very useful for various medical and research tasks due to their danger to the metabolism.Potential chemosensors must be both neutral to biological systems and highly sensitive, to detect very small amounts of target ions.Thus, to investigate possible harmful effects, 1 and its complex with Hg 2+ were tested using the MTT assay (Figure 9, Table 2).Compound 1 is found toxic enough to both tumor (HCT116) and non-tumor (HEK293T) cell lines when incubated for 3 days at a concentration of 20 µM or more.Half-maximal inhibitory concentrations (IC50) in these cases are about 21 and 20 µM, respectively.In turn, the cytotoxicity of the 1:Hg 2+ complex is a little higher: the corresponding IC50 values are about 15 and 16 µM.These results partly limit the application of the 1 compound as a chemosensor for living systems but pose the task of further modifying it in order to reduce such negative effects.
Figure 9. Measurements of cytotoxicity of 1 compound (green) and Hg 2+ salt (black), as well as their complex (blue), to human colon carcinoma cells (HCT116) and non-tumor human embryonal kidney cells (HEK293T) after 72 h of incubation.Doxorubicin (red) was taken as a standard for the comparison.

Cytotoxicity of 1 and 1:Hg 2+ Complex
The detection of mercury and its ions in living organisms may be very useful for various medical and research tasks due to their danger to the metabolism.Potential chemosensors must be both neutral to biological systems and highly sensitive, to detect very small amounts of target ions.Thus, to investigate possible harmful effects, 1 and its complex with Hg 2+ were tested using the MTT assay (Figure 9, Table 2).Compound 1 is found toxic enough to both tumor (HCT116) and non-tumor (HEK293T) cell lines when incubated for 3 days at a concentration of 20 µM or more.Half-maximal inhibitory concentrations (IC50) in these cases are about 21 and 20 µM, respectively.In turn, the cytotoxicity of the 1:Hg 2+ complex is a little higher: the corresponding IC50 values are about 15 and 16 µM.These results partly limit the application of the 1 compound as a chemosensor for living systems but pose the task of further modifying it in order to reduce such negative effects.Molecules of the 1:Hg 2+ complex are successfully accumulated by HCT116 cells in a quantity that is sufficient for fine fluorescence detection using flow cytometry.Although concentrations of 25 and 50 µM are lethal for 50% of the cell population after 3 days of exposition, a short incubation (1.5 or 24 h) allows us to visualize cellular uptake and the accumulation of the complex with a strong signal (Figure 10A).In non-toxic concentrations up to 2 µM, the signal is much weaker even after 24 h; however, the complex is still detectable (see green and pink lines in Figure 10A).
quantity that is sufficient for fine fluorescence detection using flow cytometry.A concentrations of 25 and 50 µM are lethal for 50% of the cell population after 3 exposition, a short incubation (1.5 or 24 h) allows us to visualize cellular uptake accumulation of the complex with a strong signal (Figure 10A).In non-toxic co tions up to 2 µM, the signal is much weaker even after 24 h; however, the comple detectable (see green and pink lines in Figure 10A).
The difficulty is to estimate the intracellular uptake and long accumulation o (Figure 10B, pink and green lines) because of its weak fluorescence and solubility ties.At the same time, we have shown that 4 h preincubation with 25 µM of 1, f by the addition of 50 µM Hg 2+ , almost does not increase the cellular accumulatio newly formed complex after 24 h of total incubation (see orange line in Figure 10 fact may also be explained by a poor reaction of 1 and Hg 2+ in the cultural mediu We also investigated the intracellular fluorescence of 1 and the 1:Hg 2+ com confocal microscopy.As shown in Figure 11, the 1:Hg 2+ complex is well visua HCT116 cells both at chemosensor concentrations of 10 and 25 µM.The fluoresce nal of compound 1 at a concentration of 10 µM is not registered in cells.At a conce of 25 µM, compound 1 shows extremely weak, almost invisible fluorescence.Ac tion of the 1:Hg 2+ complex occurred mainly in the cytoplasm (there is also the po of accumulation in the mitochondria, lysosomes, and endoplasmic reticulum) a much lesser extent, in the cell nucleus.Summarizing the data of in vitro expe chemosensor 1 and its complex with Hg 2+ have the fundamental ability to perm accumulate in cells even in a nanomolar concentration; however, the cells acc compound 1 alone at a much lower level than its complex with mercury.Thus relatively high cytotoxic effects for tumor and non-tumor cells, the tested compou great interest for future investigation as a fluorescence probe of mercury ions.The difficulty is to estimate the intracellular uptake and long accumulation of 1 alone (Figure 10B, pink and green lines) because of its weak fluorescence and solubility properties.At the same time, we have shown that 4 h preincubation with 25 µM of 1, followed by the addition of 50 µM Hg 2+ , almost does not increase the cellular accumulation of the newly formed complex after 24 h of total incubation (see orange line in Figure 10B).This fact may also be explained by a poor reaction of 1 and Hg 2+ in the cultural medium.
We also investigated the intracellular fluorescence of 1 and the 1:Hg 2+ complex by confocal microscopy.As shown in Figure 11, the 1:Hg 2+ complex is well visualized in HCT116 cells both at chemosensor concentrations of 10 and 25 µM.The fluorescence signal of compound 1 at a concentration of 10 µM is not registered in cells.At a concentration of 25 µM, compound 1 shows extremely weak, almost invisible fluorescence.Accumulation of the 1:Hg 2+ complex occurred mainly in the cytoplasm (there is also the possibility of accumulation in the mitochondria, lysosomes, and endoplasmic reticulum) and, to a much lesser extent, in the cell nucleus.Summarizing the data of in vitro experiments, chemosensor 1 and its complex with Hg 2+ have the fundamental ability to permeate and accumulate in cells even in a nanomolar concentration; however, the cells accumulate compound 1 alone at a much lower level than its complex with mercury.Thus, despite relatively high cytotoxic effects for tumor and non-tumor cells, the tested compound is of great interest for future investigation as a fluorescence probe of mercury ions.

Synthesis of Chemosensor 1
Solutions of 4-methylthiazole-5-carbaldehyde (0.1281 g, 1.0 mmol) and fluorescein hydrazide (0.3466 g, 1.0 mmol) in ethanol were mixed in a flask.The resultant mixture was refluxed and stirred for 6 h.Upon cooling to room temperature, a finely dispersed pale grey precipitate formed.The obtained crystalline product was filtered, washed with ice-cold ethanol and acetone, and dried at 40 °C to a constant weight.The yield was 0.3097 g (68%).The synthesis pathway for chemosensor 1 is depicted in Figure 12.

Synthesis of Chemosensor 1
Solutions of 4-methylthiazole-5-carbaldehyde (0.1281 g, 1.0 mmol) and fluorescein hydrazide (0.3466 g, 1.0 mmol) in ethanol were mixed in a flask.The resultant mixture was refluxed and stirred for 6 h.Upon cooling to room temperature, a finely dispersed pale grey precipitate formed.The obtained crystalline product was filtered, washed with ice-cold ethanol and acetone, and dried at 40 • C to a constant weight.The yield was 0.3097 g (68%).The synthesis pathway for chemosensor 1 is depicted in Figure 12.
Solutions of 4-methylthiazole-5-carbaldehyde (0.1281 g, 1.0 mmol) and fluorescein hydrazide (0.3466 g, 1.0 mmol) in ethanol were mixed in a flask.The resultant mixture was refluxed and stirred for 6 h.Upon cooling to room temperature, a finely dispersed pale grey precipitate formed.The obtained crystalline product was filtered, washed with ice-cold ethanol and acetone, and dried at 40 °C to a constant weight.The yield was 0.3097 g (68%).The synthesis pathway for chemosensor 1 is depicted in Figure 12.IR, MS, 1 H, and 13 C NMR spectra are given in the supplementary information (Figures S9-S12).

Spectral Measurements
The UV-Vis spectra were measured using a double-beamed Shimadzu UV1800 spectrophotometer (Shimadzu, Somerset, NJ, USA).The wavelength range was 260-700 nm and the absorbance range was 0-1 in H 2 O-DMSO (8:2 v/v).H 2 O-DMSO (8:2 v/v) was used as a blank solution.The temperature was kept constant at 298.2 ± 0.1 K with an external thermostat.
Fluorescence spectra were recorded using the RF6000 setup (Shimadzu, Somerset, NJ, USA) with an excitation wavelength of λ ex = 446 nm and an emission wavelength range of 470-750 nm.The excitation and emission slit widths were set to 5 nm.The temperature was maintained at 298.2 ± 0.1 K using an external thermostat.

Spectral Measurements
The UV-Vis spectra were measured using a double-beamed Shimadzu UV1800 spectrophotometer (Shimadzu, Somerset, NJ, USA).The wavelength range was 260-700 nm and the absorbance range was 0-1 in H2O-DMSO (8:2 v/v).H2O-DMSO (8:2 v/v) was used as a blank solution.The temperature was kept constant at 298.2 ± 0.1 K with an external thermostat.
Fluorescence spectra were recorded using the RF6000 setup (Shimadzu, Somerset, NJ, USA) with an excitation wavelength of λex = 446 nm and an emission wavelength range of 470-750 nm.The excitation and emission slit widths were set to 5 nm.The temperature was maintained at 298.2 ± 0.1 K using an external thermostat.
Three-dimensional fluorescence spectra were obtained by recording the spectra within the range of λex = 300-600 nm and λem = 320-800 nm (Figure 13).The optimal excitation wavelength for detecting Hg 2+ with chemosensor 1, which results in strong emission intensity, was found to be λex = 446 nm.The NMR experiments for chemosensor 1 were conducted on a Bruker Avance III 500 NMR spectrometer (Bruker, Billerica, MA, USA) with 500.17MHz and 125.77MHz frequencies for 1 H and 13 C, respectively, in DMSO-d6.Temperature control was maintained using a Bruker variable temperature unit (BVT-2000), and the experiments were carried out at 298 K without sample spinning.The accuracy of the chemical shift measurement was determined to be ±0.01 ppm for 1 H NMR spectra and ±0.1 ppm for 13 C NMR, accord- The NMR experiments for chemosensor 1 were conducted on a Bruker Avance III 500 NMR spectrometer (Bruker, Billerica, MA, USA) with 500.17MHz and 125.77MHz frequencies for 1 H and 13 C, respectively, in DMSO-d6.Temperature control was maintained using a Bruker variable temperature unit (BVT-2000), and the experiments were carried out at 298 K without sample spinning.The accuracy of the chemical shift measurement was determined to be ±0.01 ppm for 1 H NMR spectra and ±0.1 ppm for 13 C NMR, according to the external standard, HMDSO.
The MS (MALDI TOF) spectra for chemosensor 1 and 1+Hg 2+ were obtained using the Shimadzu Biotech Axima Confidence system, which was produced by Shimadzu in the NJ, United States.The samples were dissolved in an EtOH-H 2 O mixture and applied to the plate.The samples were then allowed to air-dry before the experiment.
The Avatar 360 FTIR spectrometer, manufactured by Thermo Nicolet in the MA, United States, was used to record the IR spectra for chemosensor 1.The sample was dispersed in KBr, and a range of 400-4000 cm −1 was scanned.

Computational Details
The Gaussian 09 [51] program was used for geometry optimization, followed by harmonic frequencies calculation.The calculations were carried out using density functional theory (B97D functional [52]) with a triple-zeta def2-TZVP basis set [53] taken from the EMSL BSE library [54][55][56].Basis sets for S, O, and N atoms were also augmented by diffuse functions (def2-TZVPD [53,57]).Electronic absorption spectra were calculated using the time-dependent density functional theory approach.The number of excited states was 30.The polarizable continuum model (PCM, the solvent is water) was applied to take into account solvation effects in electron absorption spectra calculations.Cartesian coordinates of quantum chemical structures are presented in Figure S13.

MTT Assay
The cytotoxicity of compounds was measured by using the MTT assay (MTT is an abbreviation for the dye compound 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide).Doxorubicin (Veropharm, Moscow, Russia) was taken as a control.Cells were seeded in 96-well plates in a density of 5000 cells per well.At 24 h after attachment, the compounds were added at a different concentration for the next 72 h.After incubation, 5 mg/mL of MTT reagent (Dia-M, Moscow, Russia) was added to each well; then, the plates were incubated for another 3 h under the same conditions.The precipitate of formazan crystals was dissolved in 100 µL DMSO.Colorimetric measurement was performed at λ = 570 nm using a CLARIOstar Plus microplate reader (BMG LABTECH, Ortenberg, Germany).The experiment was repeated at least 3 times.

Flow Cytometry
The accumulation of 1 and its complex with Hg 2+ in cells was investigated using flow cytometry.HCT116 cells were seeded on 35 mm Petri dishes in a density of 300,000 cells per dish.At 24 h after attachment, compounds were added to the culture; then, the cells were incubated for a different time.Before measuring, the cells were rinsed with PBS, detached by Versene solution (PanEco, Moscow, Russia), and centrifuged at 500 rcf for 5 min.Cell pellets were resuspended and rinsed twice in fresh PBS and analyzed using a CytoFLEX flow cytometer (Beckman Coulter, Brea, CA, USA) in FITC channel (λ ex = 488 nm, λ em = 525/40 nm).

Figure 9 .
Figure9.Measurements of cytotoxicity of 1 compound (green) and Hg 2+ salt (black), as well as their complex (blue), to human colon carcinoma cells (HCT116) and non-tumor human embryonal kidney cells (HEK293T) after 72 h of incubation.Doxorubicin (red) was taken as a standard for the comparison.

Figure 9 .
Figure9.Measurements of cytotoxicity of 1 compound (green) and Hg 2+ salt (black), as well as their complex (blue), to human colon carcinoma cells (HCT116) and non-tumor human embryonal kidney cells (HEK293T) after 72 h of incubation.Doxorubicin (red) was taken as a standard for the comparison.

Figure 10 .
Figure 10.Accumulation of 1:Hg 2+ complex in HCT116 cells at various concentrations an tion times (A), as well as of compound 1 alone followed by the addition of Hg 2+ ions (B).

Figure 10 .
Figure 10.Accumulation of 1:Hg 2+ complex in HCT116 cells at various concentrations and incubation times (A), as well as of compound 1 alone followed by the addition of Hg 2+ ions (B).

Table 1 .
Determination of Hg 2+ ions in various water samples.