A novel colorimetric and “turn-on” fluorescent sensor for selective detection of Cu²⁺

2.1 Materials and instrumentations

The chemical reagents and solvents used in this experiment were all commercially available analytical grade, without further purification, and solvents for spectra detection were HPLC reagent without fluorescent impurity. ¹H and ¹³C NMR spectra were taken on a Varian mercury-300 spectrometer with TMS as an internal standard. UV–visible absorption spectrum was performed on a UV-1900 spectrometer. Fluorescence spectrum was performed on a Hitachi F-4600 spectrofluorometer. The pH measurement was made with Mettler-Toledo Instruments DELTE 320 pH. Cell experiments were applied on an inverted fluorescence microscope (Olympus IX-70) connected to a digital camera (Olympus, c-5050).

2.2 UV–vis and fluorometric analysis

The metal ions (Na⁺, K⁺, Mg²⁺, Ca²⁺, Ba²⁺, Zn²⁺, Cd²⁺, Hg²⁺, Ni²⁺, Pb²⁺, Cu²⁺, Hg²⁺,Co²⁺ Fe³⁺, Cu⁺, Au²⁺, Ag⁺) required for the test in deionized water were purchased from Aladdin reagent official website and stored in the vacuum storage boxes. The above metal ions used throughout the experiments were prepared as buffer solution (HEPES-NaOH, pH = 7.4) and diluted to 5*10⁻⁴ M to obtain the corresponding ionic solution, the test samples were prepared by placing appropriate amounts of ions stock into matching concentration solutions of sensor 1[v(acetonitrile)/v(H₂O) = 1:1, pH = 7.4]. The fluorometric analysis of sensor 1 was measured with 370 nm as excitation wavelength, and emission was collected from 525 to 650 nm, both the excitation and emission slit widths were 5 nm and 5 nm, respectively.

2.3 Synthesis of sensor 1

Rhodamine 6 G (2.39 g, 5 mmol) was dissolved in 50 mL ethanol, and excess hydrazine hydrate (85 %, 5 g, 85 mmol) was dropped into the system during stirring for 30 min. After the dripping, the solution refluxed for 12 h. After cooling, a large amount of light pink solid appeared. Filtered the solid and washed with 15 mL ethanol for three times. After vacuum drying, the intermediate (0.43 g, 1.0 mmol) and trifluoroacetic anhydride (0.42 g, 2.0 mmol) were dissolved in 10 mL ethanol, several drops of acetic acid were added to the system, and refluxed for 12 h under intense agitation. After the reaction, the solvent was evaporated and collected the resulting solid. The crude product was separated and purified by column chromatography (developing agent: methylene chloride/ethyl acetate = 10:1). Finally sensor 1 was obtained about 0.31 g as pale pink solid, yield 35 %. ¹H NMR (300 MHz DMSO, 25^◦C, TMS): δ 1.01 (d, J = 3.0 Hz, 6H), 3.46(s, 2H), 3.47(s, 2H), 3.50(s, 2H), 3.68(s, H), 3.69(s, H), 3.72(s, H), 3.74(s, H), 6.15(s, 2H), 7.18(s, 2H), 7.20(s, H), 7.24(s, H), 7.26(s, H), 7.64(s, H), 7.66(s, H), 8.21(s, H), 11.30(s, H). ¹³C NMR (75 MHz DMSO, 25^◦C, TMS): 15.75, 15.80, 22.39, 22.43, 35.64, 35.98, 69.59, 102.76, 102.79, 115.72, 115.78, 122.82, 123.07, 123.23, 123.29, 125.11, 125.3, 125.53, 128.24, 128.76, 143.59, 144.09, 146.23, 149.33, 149.39, 149.88, 154.53, 169.18. LC-MS. Calc. [(M+1)] ⁺ m/z for: C₂₈H₂₇F₃N₄O₃ = 524.20, found [(M+1)] ⁺ m/z = 525.4 (see Scheme 1).

Close

Scheme 1

Synthesis of sensor 1.

Export to PPT

3.1 Absorption spectral response

The changes of the UV–vis spectra for sensor 1 in the absence and presence of Cu²⁺ were investigated. 5 × 10⁻⁴ M sensor 1 solution was prepared with acetonitrile: HEPES-NaOH buffer solution (V: V = 1:1, pH = 7.4). Then added different metal ion (Cu²⁺, Na⁺, K⁺, Mg²⁺, Ca²⁺, Ba²⁺, Zn²⁺, Cd²⁺, Hg²⁺, Ni²⁺, Pb²⁺, Co²⁺, Cu⁺, Au²⁺, Ag⁺ and Fe³⁺⁾ buffer solutions of equal volume and concentration into sensor 1 solutions, respectively. The UV–vis absorption spectrum of the different metal ions in sensor 1 solution was illustrated in Fig. 1a. As the amide spiral ring structure of sensor 1 was in the closed state, the absorption band was not notable in absorption spectrum, the introduction of Na⁺, K⁺, Mg²⁺, Ca²⁺, Ba²⁺, Zn²⁺, Cd²⁺, Hg²⁺, Ni²⁺, Pb²⁺, Cu⁺, Au²⁺, Ag⁺ and Fe³⁺did not lead to any distinct change in absorption spectrum, whereas the addition of Cu²⁺ induced a new absorption peak located at 491 nm, and this change was also intuitively reflected in the color of the solution, upon the addition of Cu²⁺, the color of sensor 1 solution changed gradually from colorless to green, so this performance allowed the colorimetric detection of Cu²⁺ by naked eye. Furthermore, it was worth noting that Co²⁺ also caused a certain amount of change at 523 nm, but the absorbance was weaker than Cu²⁺, so Co²⁺ would not affect the detection of Cu²⁺. Then the influence of different concentrations of Cu²⁺ions (1 × 10⁻³ M ∼ 5 × 10⁻³ M) on UV–vis absorption spectrum of sensor 1 was measured, as shown in Fig. 1(b), with the gradual addition of Cu²⁺, the absorption intensity at 491 nm increased evidently, so in the subsequent fluorescence response experiments, the maximum excitation wavelength was set as 491 nm. Furthermore, such strong enhancement in the absorption behavior changed the color of the solution from colorless to green, the reason for this phenomenon was that sensor 1 existed in the form of amide screw ring closure (OFF) without the addition of Cu²⁺, showing no UV absorption and fluorescence. When Cu²⁺ was added, the N atom in the imine, the N atom and O atom in amide coordinated with Cu²⁺ to form a stable five-membered ring structure, which forced the amide spiral ring structure to open (ON), showing the enhancement of long-wavelength UV absorption and fluorescence. On account of the strong electron-absorbing effect of trifluoroacetyl group, the UV–vis absorption spectrum of sensor 1 had a blue shift compared with Rhodamine 6G, so the solution of sensor 1- Cu²⁺ complex appeared green, and it was because of the ability to change colors, sensor 1 was allowed to detect Cu²⁺ with naked eye colorimetry.

Close

Fig. 1

(a) UV–vis absorption response of sensor 1 in different metal ions solutions; (b) UV–vis absorption response of sensor 1 in different in different concentration of Cu²⁺ [v(acetonitrile)/v(H₂O) = 1:1, pH = 7.4].

Export to PPT

3.2 Fluorescence spectral response

Selectivity is an important basis for measuring fluorescence sensors, in order to verify the fluorescence response of the sensor 1 towards common metal ions, Cu²⁺, Na⁺, K⁺, Mg²⁺, Ca²⁺, Ba²⁺, Zn²⁺, Cd²⁺, Hg²⁺, Ni²⁺, Pb²⁺, Co²⁺, Cu⁺, Au²⁺, Ag⁺ and Fe³⁺ were selected to conduct selective experiments of sensor 1. The experiments took advantage of sensor 1 solution with the concentration of 5 × 10⁻⁴ M, and added equal volume of 5 × 10⁻⁴ M HEPES-NaOH buffer solution containing different kinds of metal ions (pH = 7.4), respectively. As illustrated in Fig. 2(a), sensor 1 itself had no obvious fluorescence emission, meanwhile the addition of Na⁺, K⁺, Mg²⁺, Ca²⁺, Ba²⁺, Zn²⁺, Cd²⁺, Hg²⁺, Ni²⁺, Pb²⁺, Cu⁺, Au²⁺, Ag⁺ and Fe³⁺ did not change fluorescence emission significantly, whereas Cu²⁺ distinctly enhanced fluorescence intensity at 550 nm, compared with other metal ions, the fluorescence enhancement is very prominent, and it was worth noting that Co²⁺ also induced a slight fluorescence enhancement of sensor 1, and compared with Cu²⁺, the fluorescence enhancement caused by Co²⁺ at 550 nm in the fluorescence spectrum was much weaker, which was about 1/18 of the effect of Cu²⁺, so Co²⁺ was not enough to affect the detection of Cu²⁺ by sensor 1. In addition, the unique fluorescence response of sensor 1 to Cu²⁺ reflect clear color changes under 365 nm UV lamps, only Cu²⁺ induced sensor 1 solution to emit noticeable yellow fluorescence among common metal ions, this also verified the s ability of sensor 1 to recognize Cu²⁺ under UV light. Similar to many reported rodaminespirolactam-based fluorescent chemosensors, the fluorescence enhancement response of sensor 1 toward Cu²⁺ is most likely the result of the spiro ring-opening mechanism rather than an ion-catalyzed hydrolysis reaction, and resulted in a remarkable fluorescence boost (Jung et al., 2009; Zhang et al., 2007).

Close

Fig. 2

(a) Fluorescence spectrum of sensor 1 in different metal ions solutions (λ_ex = 491 nm, slits: 5 nm/5 nm) and Fluorescence response of sensor 1 in different metal ions under 365 nm UV lamp; (b) Histograms of fluorescence selective experiments (green) and competitive experiments (red) of sensor 1 in different metal ions [v(acetonitrile)/v(H₂O) = 1:1, pH = 7.4].

Export to PPT

For further evaluate the selectivity of sensor 1, the above-mentioned metal cations were adopted to conduct competitive experiments. Under the same solvent conditions [v(acetonitrile)/v(H₂O) = 1:1, pH = 7.4], sensor 1 solution was first induced to fluorescence enhancement by Cu²⁺ solution at the same concentration (5 × 10⁻⁴ M) then 10 equivalent (5 × 10⁻³ M) other kinds of metals ion buffered solutions (Na⁺, K⁺, Mg²⁺, Ca²⁺, Ba²⁺, Zn²⁺, Cd²⁺, Hg²⁺, Ni²⁺, Pb²⁺, Co²⁺, Cu⁺, Au²⁺, Ag⁺ and Fe³⁺) were added into the fluorescent-enhanced solution, the fluorescence intensities of mixed system in emission spectrum at 550 nm were selected as the data of the competitive experiments, meanwhile the selective experiments with the competitive experiments were contrast to explain experimental results. In Fig. 2(b), the black histograms represented the data of the selective experiments, and the red histograms represented the competitive experiments, as shown in red histograms, the fluorescence intensities of the solutions mixed with competing metal ions did not change visibly and was always in enhanced states. So we could conclude that the specific bonding between Cu²⁺ and sensor 1 was stronger than other metal ions, which also proved that sensor 1 had good selectivity and anti-interference to Cu²⁺, and provided a certain guarantee for sensor 1 in detecting Cu²⁺ under actual complex conditions.

The sensitive experiments were utilized to further explore the recognition performance of sensor 1 to Cu²⁺. The effect of fluorescence emission spectra of sensor 1 solution [v(acetonitrile)/v(H₂O) = 1:1, 5 × 10⁻⁴ M] was studied under different concentrations of Cu²⁺ (5 × 10⁻⁵ M ∼ 5 × 10⁻⁴ M, HEPES-NaOH buffer solution, pH = 7.4), the Cu²⁺ concentrations dependent changes in fluorescence spectra of sensor 1 was illustrated in Fig. 3. Sensor 1 alone exhibited negligible fluorescence (excitated at 491 nm), with the concentration of Cu²⁺ increased, the fluorescence intensity at 550 nm gradually heightened. Besides, in the range of 50–500 μM Cu²⁺, the fluorescence enhancement of sensor 1 toward concentration of Cu²⁺ revealed good linear relationship, with a coefficient R = 0.99047. According to the detection limit formula: LODs = 3σ/m (Hu et al., 2021), the detection limit of sensor 1 was calculated to be 1.85 × 10⁻⁷ M, so the minimum detection concentration of sensor for Cu²⁺ could reach the level of 10⁻⁷ M. And it confirmed that sensor 1 had potential application for the quantitative determination of Cu²⁺.

Close

Fig. 3

Normalized response of fluorescence signal to changing Cu²⁺ concentrations [v(acetonitrile)/v(H₂O) = 1:1, pH = 7.4].

Export to PPT

The fluorescence quantum yields of sensor 1 and sensor 1-Cu²⁺ were determined according to the equation: $Y_u=Y_S\times \frac{F_U}{F_S}\times \frac{A_S}{A_U}$ , where F_U and F_S denoted the integral fluorescence intensity of determinand and standard substance, A_U and A_S denoted the relevant maximum absorbance of determinand and standard substance, respectively (Rhodamine 6 G was used as reference quantum yield standard, λ_ex = 488 nm, quantum yield = 0.65 in water).

Job’s plot curve was adopted to account for the association principle between sensor 1 and Cu²⁺ (Zhang et al., 2021). This curve method was evaluated from the fluorescence spectra of sensor 1 and Cu²⁺ with a total concentration of 0.2 mM. It could be seen from Fig. 4 that the strongest fluorescence intensity occurred at 0.5, this proved a 1:1 binding stoichiometry between sensor 1 and Cu²⁺. According to the form of 1:1 binding mode, and referred to previous relevant literature reports, the possible binding model between sensor 1 and Cu²⁺ was further inferred as revealed in Fig. 4. In the absence of Cu²⁺, sensor 1 existed in the form of amide screw ring closure (OFF), displayed no UV absorption and fluorescence. When Cu²⁺ was added, it bonded with one oxygen atom and two nitrogen atoms on the amide spiral ring, resulted in amide screw ring opened, which characterized by ultraviolet absorption and fluorescence emission enhancement.

Close

Fig. 4

The Job's plot curve of sensor 1 and Cu²⁺.

Export to PPT

To study the practical applicability, the effects of pH on the fluorescence response of sensor 1 in the absence and presence of Cu²⁺ were investigated. The pH range of experiments were collected from pH 3.0 to pH 9.0. Fig. 5 illustrated the fluorescence responses of sensor 1 without and with Cu²⁺ in different pH. Experimental results revealed that alone sensor 1 emitted a certain intensity of fluorescence at acidic conditions (pH < 5) on account of the formation of the open-ring state due to the strong protonation. But there was almost no fluorescence emission (λ_ex = 491 nm) from alone sensor 1 in the range from pH 4.0 to pH 9.0. But there was almost no fluorescence emission (λ_ex = 491 nm) from alone sensor 1, because of the amide helix ring was in the closed state. Upon the addition of Cu²⁺, the fluorescence enhancement was very obvious in different pH range. So the pH-control emission experiments proved that sensor 1 could effectively respond to Cu²⁺ ions in the pH range from 5 to 9, attesting that sensor 1 was capable of recognizing Cu²⁺ in a wide pH range. We could also assume that the 1:1 complexes formed between sensor 1 and Cu²⁺ was stable in the pH 5–9. In addition, considering that most samples for Cu²⁺ analysis were neutral, so the media for Cu²⁺ quantification was buffered at pH 7.4.

Close

Fig. 5

Fluorescence intensity of sensor 1 (0.5 mM) measured with and without Cu²⁺ ions (0.5 mM) as a function of pH.

Export to PPT

Fast response time and reversible response are important factors of sensor detection efficiency. The time dependence of the response of sensor 1 to Cu²⁺ was shown in Fig. 6, the experimental results proved that the recognizing process finished in less than 1 min, therefore, sensor 1 could be used as a rapid Cu²⁺ monitor. Because EDTA is a good metal coordination agent and has good coordination ability with common metal ions, we selected EDTA to research the chemical reversibility behavior of sensor 1 and Cu²⁺ in acetonitrile–water buffered solution. As shown in Fig. 7, according to the complexing ability of EDTA toward Cu²⁺, EDTA was added into sensor 1-Cu²⁺ complex solution to try to displace Cu²⁺ from the metal–ligand complex. Upon addition of excess EDTA, the color of the metal–ligand mixture did not change from green to colorless and fluorescent emission intensity of the system was stable, indicating that EDTA did not separate Cu²⁺ from sensor 1-Cu²⁺ complex, this was probably due to the better coordination ability of sensor 1 with Cu²⁺ compared with EDTA.

Close

Fig. 6

Response time course of sensor 1 (10 μM) upon addition of Cu²⁺ (2 equiv).

Export to PPT

Close

Fig. 7

Fluorescence spectra of 0.5 mM sensor 1 in the presence of 0.5 mM Cu²⁺ and 3.0 mM EDTA.

Export to PPT

3.3 Theoretical calculation research

To understand the recognition ability of sensor 1 towards Cu²⁺, the HOMO (highest occupied molecular orbitals) and LUMO (lowest unoccupied molecular orbitals) distributions of sensor 1 were determined by DFT calculations with PBE0, using the Gaussian 09 package to calculate the optimal geometry configuration, spatial electron cloud arrangement and spatial structure of molecules. The 3-21G basis set was employed for the H, C, N, and O atoms, for the Cu atom, the LANL2DZ effective core potential was employed. The HOMO-LUMO electron transition of sensor 1 was illustrated in Fig. 8, the theoretical calculation of absorption peak at 471 nm was from the ground state to the excited state (oscillator strength f = 0.01), meanwhile the maximum absorption peak came from the benzene ring unconjugated system, and the spironolactone structure did not reveal noticeable electronic transition. After sensor 1 bonded with Cu²⁺, the first electron transition became HOMO → LUMO (λ_em = 542 nm, f = 0.06). HOMO and LUMO were mainly distributed on oxanthracene units, indicated that the maximum absorption peak came from rhodamine 6G fluorophore, which generated a strong fluorescence signal. The results of theoretical calculation were highly consistent with the actual experimental data, which also verified the scientific nature of our experimental results from the theoretical perspective.

Close

Fig. 8

Molecular orbitals involved in the vertical emission of (a) sensors 1 and (b) sensor 1-Cu²⁺.

Export to PPT

3.4 Biological investigation

In order to ulteriorly explore the potential biological application, the MTT assay was performed in Hi-7702 cells (normal liver cells) to confirm the cytotoxicity in different concentrations (100.0, 50.0, 10.0, 5.0, 1.0, 0.5 and 0.1 μM) of sensor 1. As illustrated in Fig. 9, the cells were healthy and did not display any remarkable reduction in cell viability in the concentration range of 50.0–0.1 µM. At the highest concentration of sensor 1 (100 µM), the cells revealed mild decrease in the viability (63 %). These experimental results proved that sensor 1 has extraordinary low cytotoxicity under the experimental conditions. Furthermore, the IC₅₀ value was ascertained as 217.6 μM. Accordingly, the sensor 1 was suggested for monitoring Cu²⁺ status in vivo. Whereafter, Hi-7702 cells were used to measure the capability of the sensor 1 to track the changes of Cu²⁺. The live Hi-7702 cells were first incubated with 5 × 10⁻⁵ mol/L sensor 1 in the solution [v(acetonitrile)/v(H₂O) = 1:1] for 30 min at room temperature in 5 % CO₂ atmosphere, washed with phosphate buffered saline (PBS, PH = 7.4) for three times, and induced 5 × 10⁻⁵ mol/L Cu²⁺ into the solution [v(acetonitrile)/v(H₂O) = 1:1] for 30 min. As shown in Fig. 10(a) and Fig. 10(b), cells treated by sensor 1 revealed no obvious fluorescence phenomenon and still maintained clear contours, which indicated sensor 1 smoothly penetrate the cell membrane into HI-7702 cell. After the inducing of Cu²⁺, we found bright green fluorescence in Hi-7702 cells clearly from Fig. 10(c). According to the pictures we could identify that sensor 1 had successfully immersed into the cells and could track the intracellular Cu²⁺ in live Hi-7702 cells. The live cell imaging also demonstrated that sensor 1 was membrane-permeable and could be a useful molecular sensor for studying the biological processes involving Cu²⁺ in live cells. Therefore, it was concluded that sensor 1 had good biocompatibility and potential application in the field of biological tracer.

Close

Fig. 9

Concentration-dependent cell viability assay.

Export to PPT

Close

Fig. 10

(a) Bright field transmission image of Hi-7702 cells treated with sensor 1; (b) Fluorescence imaging of Hi-7702 cells treated with sensor 1;(c) Fluorescence imaging of Hi-7702 cells treated with sensor 1-Cu²⁺.

Export to PPT

Zebra fish is a common tropical freshwater fish with strong vitality, the average body length of it is about 2–4 cm and its genetic similarity to humans is 87 %. For the sake of expand the biological application value of the sensor 1, zebra fish was selected to find out whether sensor 1 could be employed for visual detecting Cu²⁺ in live animals. In this experiment, zebra fish were bred in sterile purified water for two days at room temperature, and added DMF solution dissolved with sensor 1(10 mM) into the feeding system for cultivation 1 h. Then removed the zebra fish and washed it with HEPES-NaOH buffer solution for three times. After finishing the above operations, as illustrated in Fig. 11(a), it could be seen that the zebra fish absorbed sensor 1 displayed almost no apparent fluorescence emission, the shape outline of it could only be vaguely seen under 365 nm UV light. But after further breeding of zebra fish with Cu²⁺ (20 mM) for another 1 h, the image of zebra fish under UV lamp was illustrated in Fig. 11(b). The morphology of zebra fish could be clearly observed with obvious fluorescence emission. The experimental results revealed that sensor 1 was successfully absorbed into the body of zebra fish, and the detection of Cu²⁺ by sensor 1 had occurred in zebra fish successfully. This indirectly proved that sensor 1 could be utilized in the field of biological imaging and had certain practical application value.

Close

Fig. 11

Images of zebra fish under 365 nm UV light at 365 nm: (a) Zebra fish bred with only sensor 1 for 1 h; (b) Fish bred with sensor 1 and then bred with Cu²⁺ for 1 h.

Export to PPT