A highly effective “naked eye” colorimetric and fluorimetric curcumin-based fluorescent sensor for specific and sensitive detection of H₂O₂in vivo and in vitro

Abstract

Hydrogen peroxide (H₂O₂) is involved in many important tasks in normal cell metabolism and signaling. However, abnormal levels of H₂O₂ are associated with the occurrence of several diseases. Therefore, it is important to develop a new method for the detection of H₂O₂in vivo and in vitro. A turn-off sensor, 2,2-difluoro-4,6-bis(3-methoxy-4-((4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)benzyl)oxy)styryl)-2H-1,3,2-dioxaborine (DFCB), based on curcumin was developed for the detection of H₂O₂. The DFCB, an orange-emitting sensor, was constructed by employing 2,2-difluoro-4,6-bis(4-hydroxy-3-methoxystyryl)-2H-1,3,2-dioxaborine (DFC) as the main carrier, and 2-(4-bromomethylphenyl)-4,4,5,5-tetramethyl-1,3,2-doxaborolane as the recognition site. The recognition group on the DFCB sensor could be completely cleaved by H₂O₂ to generate the intermediate DFC, which would lead to a colorimetric change from bright orange to light blue accompanying by a significantly quenched fluorescence, which could be seen by the naked eye. This sensor exhibited a highly specific fluorescence response to H₂O₂, in preference to other relevant species, with an excellent anti-interference performance. The sensor DFCB also possessed some advantages including a wide pH response range (6–11), a broad linear range (0–300 μM), and a low detection limit (1.31 μM). The sensing mechanism of the DFCB sensor for H₂O₂ was verified by HRMS analysis, ¹H-NMR titration and DFT calculations. In addition, the use of the DFCB sensor was compatible with the fluorescence imaging of H₂O₂ in living cells and zebrafish.

---

Introduction

Hydrogen peroxide (H₂O₂) is one of the most important reactive oxygen species (ROS), and one of its key tasks is to activate immune cells and maintain the normal metabolism.^1–4 However, abnormal H₂O₂ generation will cause irreversible damage to the organism, and may attack cellular biomolecules such as protein and DNA,^5–10 affect cell metabolism and proliferation, and cause cancer, abnormal amino acid modification, Alzheimer's disease and other diseases.^11–13 Due to the complex pathologies that arise in organisms, H₂O₂ can cause oxidative stress and eventually lead to cell damage and necrosis.^14–16 Therefore, to develop efficient methods for detecting H₂O₂ is of great importance in disease prevention and medical diagnosis.

Existing quantitative detection methods, such as titrimetry, electrochemistry, spectrophotometry and so on, are relatively mature techniques.^17–19 Fluorescent sensor analysis has the advantages of fast response, strong selectivity, adaptability and microscopic properties, and it has been widely used for detecting some analytes in biological and environmental systems.^20–23 At present, there are many reported sensors for H₂O₂ which use aryl borate or phenyl boric acid reactions to generate phenol, phenol oxidation to give quinone, oxonium or diketone to give acid, carbon–carbon double bond breaking, and so on.^24–26 In recent years, a limited number of fluorescent sensors have been widely used for monitoring H₂O₂ in living organisms and cells.^27–30 However, some reported sensors for H₂O₂ have suffered from some drawbacks, such as unsatisfactory sensitivity, narrow pH usage range, and so on. Therefore, it is necessary to design pH-compatible fluorescent sensors for detecting H₂O₂ in biological systems.

Curcumin, a natural product with a unique fluorescence nature, is found mainly in the rhizomes of the plant families Zingiberaceae and Araceae.^31,32 Curcumin has received much attention due to its wide range of health care functions and biological properties, such as antibacterial, anti-inflammatory, anti-infective, anti-cancer and liver protective activities. We propose that the curcumin-based fluorescent sensor may have low cytotoxicity.^33–35 Moreover, curcumin and its derivatives have a large conjugate structure, which could enhance the π electron delocalization. The curcumin and its derivatives are better to be emitted to generate outstanding long-wavelength fluorescence. The fluorescence sensor with a long-wavelength emission can effectively avoid interference with self-emission fluorescence, and is an ideal tool for in vivo imaging. In this case, some excellent curcumin-based fluorescent sensors have constructed.^36–38 Therefore, it is promising for the development of a new curcumin-based fluorescent sensor for H₂O₂ detection.

It has been reported that β-dicarbonyl compounds which form a rigid structure with boron trifluoride (BF₃), could enhance the fluorescence emission wavelength.^39,40 Based on this research, BF₃ was purposefully added to curcumin to amplify its conjugation effect and sequentially modulate its fluorescence properties. In the research reported in this paper, a novel curcumin-based fluorescent sensor based on DFCB was developed. The quenching mechanism of DFCB on H₂O₂ was confirmed by high resolution mass spectrometry (HRMS) analysis, ¹H-NMR titration and DFT calculations. The bilateral ether bond of the DFCB sensor would be broken in the presence of H₂O₂, which leads to a significant change of the color which can be seen by the “naked eye”, and the spectral properties of the sensor. The DFCB sensor emitted orange fluorescence at 601 nm and exhibited a significantly quenched fluorescence response to H₂O₂. The DFCB sensor showed a high sensitivity and favorable selectivity for H₂O₂ in PBS solution (containing 50% CH₃CN, pH = 7.4). More importantly, the DFCB sensor could detect H₂O₂ in HeLa cells as well as in living zebrafish, which confirmed that the DFCB sensor could be used to effectively monitor exogenous and endogenous H₂O₂ in a biological system.

Results and discussion

Design and synthesis of the DFCB sensor

In this research, a new curcumin-based fluorescent sensor was constructed for H₂O₂ detection. In this sensor, the curcumin unit is used logically as a highly effective fluorescent framework due to its lipophilic long-chain conjugate. After that, boron trifluoride diethyl etherate (BF₃·(C₂H₅)₂O) was introduced into the curcumin framework to form the DFC compound, which can produce a strong electron push–pull effect. Next, the 2-(4-bromomethylphenyl)-4,4,5,5-tetramethyl-1,3,2-doxaborolane group was introduced into the molecule and used as a recognition group for H₂O₂. As expected, the sensor exhibited a significant fluorescence emission in the PBS solution (containing 50% CH₃CN, pH = 7.4). After reaction with H₂O₂, the 2-(4-bromomethylphenyl)-4,4,5,5-tetramethyl-1,3,2-doxaborolane group of the DFCB sensor was selectively broken, allowing for the formation of the intermediate DFC, which would lead to a significantly quenched fluorescence emission. The chemical structure of the DFCB sensor was confirmed using ¹³C-NMR, ¹H-NMR and HRMS analyses (see the ESI†).

Spectral properties of DFCB in response to H₂O₂

The optical properties of the DFCB sensor in the absence and presence of H₂O₂ were determined using UV-vis absorption and fluorescence spectroscopy. As shown in Fig. 1a, the DFCB sensor had a sharp and strong absorption band at 503 nm. After adding 300 μM H₂O₂ into the solution of the DFCB sensor, the absorbance at 503 nm decreased significantly, whereas a new absorption band appeared at 601 nm. At the same time, the color of the solution changed from bright orange to light blue. The previous data indicated that the DFCB sensor can serve as a colorimetric sensor for detecting H₂O₂. As shown in Fig. 1b, the fluorescence spectra of DFCB sensor had a strong fluorescence emission at 601 nm. However, after addition of 300 μM H₂O₂, the fluorescence emission of the DFCB sensor decreased significantly, and the color of the fluorescence changed from bright orange to colorless. The fluorescence spectral results indicated that the DFCB sensor could be used as a turn off fluorescence sensor for H₂O₂ detection.

Fig. 1 (a) The UV-vis absorption, and (b) the fluorescence spectra of the DFCB (100 μM) sensor before and after addition of H₂O₂ (300 μM) in PBS solution (containing 50% CH₃CN, pH = 7.4). λ_ex = 425 nm.

Sensitivity of the DFCB sensor to H₂O₂

In order to explore the sensitivity of the DFCB sensor for detecting H₂O₂, the UV-vis absorption and fluorescence spectra of the DFCB sensor in response to different concentrations of H₂O₂ were studied. As shown in Fig. 2a, with gradual addition of H₂O₂ (0–300 μM), the absorption peak of sensor DFCB at 503 nm weakened gradually, whereas a new UV absorption peak appeared at 601 nm and was gradually enhanced. The absorption intensity (A_{601 nm}/A_{503 nm}) of the DFCB sensor showed a good linear relationship with the H₂O₂ concentration (0–300 μM), and the linear regression equation was y = 0.00469 + 0.00117x (R² = 0.99152). These experimental data indicated that the DFCB sensor could be employed for the colorimetric quantitative detection of H₂O₂.

Fig. 2 (a) The UV-vis absorption spectra of the DFCB sensor (100 μM) with the addition of H₂O₂ (0–300 μM) in PBS buffer (containing 50% CH₃CN, pH = 7.4). (b) The linear fitting relationship between the H₂O₂ concentration and the absorption intensity ratio (A_{601 nm}/A_{503 nm}) of the DFCB sensor. (c) The fluorescence spectra of the DFCB sensor (100 μM) with the addition of H₂O₂ (0–300 μM) in PBS buffer (containing 50% CH₃CN, pH = 7.4). (d) The linear fitting relationship between H₂O₂ concentration and the fluorescence intensity of the DFCB sensor at 601 nm.

The fluorescence spectra of the DFCB sensor for detecting H₂O₂ were measured. As shown in Fig. 2c, with the gradual addition of 0–300 μM H₂O₂, the fluorescence emission peak of the DFCB sensor at 601 nm gradually weakened, and the fluorescence intensity of the DFCB sensor showed a linear relationship with the concentration of H₂O₂ (0–300 μM). The linear regression equation was: y = 829.10482 − 3.08365x (R² = 0.99001). According to the literature, the formula for the limit of detection was: LOD = 3σ/κ, and the detection limit was calculated to be 1.31 μM,¹¹ which was comparable to that of a previously reported sensor. Based on the UV-vis absorption and fluorescence spectroscopy, the DFCB sensor had high sensitivity and a low detection limit for H₂O₂.

Selectivity of the DFCB sensor to H₂O₂

The selectivity and anti-interference properties of the DFCB sensor for the detection of H₂O₂ were measured. As shown in Fig. 3a, the fluorescence spectra of the DFCB sensor after the addition of different analytes (AcO⁻, Ag⁺, Ba²⁺, Br⁻, Ca²⁺, Cl⁻, ClO⁻, ClO₄⁻, Co²⁺, CrO₇²⁻, Cs²⁺, Cu²⁺, H₂O₂, H₂PO₄⁻, Hg²⁺, HSO₃⁻, HSO₄⁻, I⁻, K⁺, meta-chloroperoxybenzoic acid (MCPBA), Mg²⁺, Mn²⁺, Na⁺, Ni²⁺, NO₂⁻, NO₃⁻, ˙OH, ONOO⁻, PAA, SO₄²⁻, TBHP, Zn²⁺). After adding these analytes, the fluorescence intensity of the DFCB sensor was sharply quenched after the reaction with H₂O₂. In contrast, after the addition of the other analytes, the fluorescence intensity of the DFCB sensor exhibited no significant changes. These data demonstrated the superior selectivity of the DFCB sensor to H₂O₂. The anti-interference properties of the DFCB sensor for detecting H₂O₂ were investigated. As shown in Fig. 3b, the fluorescence intensity of the DFCB sensor with H₂O₂ was not affected by the presence of other analytes. Previous results had indicated that the DFCB sensor had excellent selectivity and anti-interference properties for H₂O₂.

Fig. 3 (a) The fluorescence spectra of DFCB after the introduction of various analytes (300 μM) in PBS buffer (containing 50% CH₃CN, pH = 7.4). (b) The fluorescence intensity of the DFCB sensor with H₂O₂ at 601 nm in the presence of other analytes (300 μM) in PBS buffer (containing 50% CH₃CN, pH = 7.4). λ_ex = 425 nm.

Effects of pH and response time

The pH detection range determines whether the sensor can be applied in a biological environment. As shown in Fig. 4a, the DFCB sensor exhibited a high fluorescence intensity at 601 nm, which indicated its good stability. When the pH was located in the range of 2–6, the DFCB sensor showed no remarkable fluorescence change because the H₂O₂ can react with acid to produce H₂O and O₂ in an acidic environment. However, when the pH value increased to above 7, the fluorescence emission of the DFCB sensor was sharply quenched, which suggested that the DFCB sensor had an excellent detection ability over a wide pH range, and could be applied to living organisms. The response time of the DFCB sensor to H₂O₂ was evaluated. As shown in Fig. 4b, after the addition of H₂O₂, the fluorescence intensity of the DFCB sensor dramatically decreased and reached saturation after 44 min, which suggested that the DFCB sensor can be employed for real-time monitoring of H₂O₂ (Table 1).

Fig. 4 (a) Fluorescence intensity of the DFCB sensor (100 μM) at different pH values in the absence and presence of H₂O₂ (300 μM) in PBS buffer (containing 50% CH₃CN, pH = 7.4). (b) Response time of the DFCB sensor (100 μM) to H₂O₂ (300 μM) in PBS buffer (containing 50% CH₃CN, pH = 7.4).

Table 1 Comparison of other fluorescence sensors reported for specific detection of H₂O₂

Sensors	Response time	Detection limit (μM)	Linear range	pH range	Application
	120 min	4.0	0–1 μM	Not mentioned	Living mice
Ref. 41
	70 min	20	0–10 μM	48	Living mice
Ref. 42
	180 min	6.0	0–100 μM	5–8	Living cells
Ref. 43
	60 min	4.6	0–1 mM	4–11	Living cells
Ref. 44
	60 min	1.35	0–200 μM	4–10	Living cells and living zebrafish
Ref. 45
	44 min	1.31	0–300 μM	6–11	Living cells and living zebrafish
This work

---

Detection mechanism

In order to demonstrate the reaction mechanism of DFCB to H₂O₂, HRMS analysis of the reaction product of the DFCB sensor with H₂O₂ was performed. As shown in Fig. 5(a), it was observed that the main molecular ion peaks were recorded at an m/z value of 439.1144, which was consistent with the stable DFC ([M + Na]⁺ = m/z 439.1135). It was hypothesized that the boronate ester group of the DFCB sensor was oxidized and then dissociated fully to form the intermediate ([M + K]⁺ = m/z 665.1566) in the presence of H₂O₂, followed by quinoid dissociation to form stable DFC. In addition, the results of the HRMS analysis confirmed that the probable peak was recorded at an m/z value of 665.1585 when about 2 eq. H₂O₂ was added. The HRMS analysis confirmed the logic of our preliminary analytical reaction mechanism.

Fig. 5 (a) The proposed sensing mechanism of DFCB with H₂O₂. (b) The HRMS spectra of the reaction of the DFCB sensor with 5 eq. H₂O₂. (c) The HRMS spectra of the reaction of the DFCB sensor with 2 eq. H₂O₂.

To demonstrate the reaction mechanism between the DFCB sensor and H₂O₂, the ¹H-NMR spectra of the DFCB sensor after reaction with different amounts of H₂O₂ were recorded. As shown in Fig. 6, after adding H₂O₂ into the DMSO-d₆ solution of the DFCB sensor, the proton signal peaks (H_a) of the phenylborate groups at 1.29 ppm disappeared with the increase of the H₂O₂ equivalent. At the same time, the proton signal peaks (H_b) of the methylene units at 5.5 gradually disappeared, which indicated that the boronate ester group of the DFCB sensor can react with H₂O₂ and was then removed. Based on HRMS analysis and ¹H-NMR titration, the sensing mechanism of DFCB on H₂O₂ was selective boron oxidation which formed the intermediate, and the elimination reactions to release DFC, which might break the original structure of the sensor and would result in a significant weakening of the fluorescence emission at 601 nm.

Fig. 6 The ¹H-NMR spectra of the DFCB sensor before and after the reaction with different amounts of H₂O₂.

The DFT calculation was carried out using the Gaussian 09 program to further reveal the optical changes of the DFCB sensor before and after the reaction with H₂O₂. Fig. 7 shows the spatial distribution and energy orbitals of the lowest unoccupied molecular orbital (LUMO) and the highest occupied molecular orbital (HOMO) of the DFCB sensor and the DFC compound. The LUMO and HOMO of the DFCB sensor were mainly distributed on the fluorine-boron core, the C=C bond, and the benzene ring, indicating the weak charge transfer effect of the DFCB sensor. However, the LUMO of the DFCB sensor was distributed on the fluorine-boron core and the C=C bond, whereas its HOMO was distributed on the phenol group, which would activate the charge transfer effect of the DFCB sensor after reaction with H₂O₂, and then it led to a red-shifted absorption spectra.

Fig. 7 Optimized geometric configurations and frontier molecular orbitals of the DFCB sensor and the DFC intermediate.

Imaging of DFCB in HeLa cells

The use of the DFCB sensor for imaging H₂O₂ was investigated in living HeLa cells. Prior to the cell fluorescence imaging experiment, the cytotoxicity of the DFCB sensor was measured. As shown in Fig. S7 (ESI†), the DFCB sensor (0–4 μM) had no significant effect on the survival rate of the HeLa cells, and the cell survival rate was over 90%. This indicated that the DFCB sensor was less cytotoxic to cells and that it could be applied in the cell imaging experiment.

Furthermore, the DFCB sensor was used for fluorescence imaging in living cells. As shown in Fig. 8, the control cells showed no observable fluorescence emission. With the addition of the DFCB sensor, the HeLa cells emitted an orange fluorescence, which demonstrated that the sensor had good cell permeability and biocompatibility. After further incubation with increasing concentrations of H₂O₂ (6 and 12 μM), the fluorescence intensity of the cells gradually decreased and was completely quenched. The 3D surface plot and the relative fluorescence intensity were calculated to support the fluorescence quenching effect in living cells, which demonstrated that the DFCB sensor could be utilized for imaging exogenous H₂O₂in vitro.

Fig. 8 (a) Imaging of H₂O₂ in HeLa cells: (A–C) only the HeLa cells were cultured for 1 h; (E–G) HeLa cells supplemented with the DFCB sensor (4 μM) were cultured for 1 h; (I–K) HeLa cells supplemented with the DFCB sensor (4 μM) and H₂O₂ (6 μM) were cultured for 30 min; and (M–O) HeLa cells were cultured with the DFCB sensor (4 μM) and H₂O₂ (12 μM) for 1 h. (D, H, L and P) are the corresponding 3D surface plots of the fluorescence images. (b) Quantitative intensity obtained from the fluorescent images of the HeLa cells.

The fluorescence labeling, with endogenous H₂O₂, was carried out in HeLa cells using the DFCB sensor. As shown in Fig. 9, in the presence of the free DFCB sensor (4 μM), there was significant fluorescence emission in the HeLa cells. Since phorbol myristate acetate (PMA, a ROS stimulant) could promote endogenous H₂O₂ production, the HeLa cells displayed quenching which was supplemented with PMA (1 μg ml⁻¹), and subsequently the cells were cultured for 30 min. If the cells were cultured with PMA for 60 min, the fluorescence intensity decreased further, which indicated that the DFCB sensor could detect endogenous H₂O₂. In addition, to verify that the endogenous H₂O₂ production could cause the fluorescence quenching in the cells, the PMA-stimulated cells were further supplemented with N-acetyl-L-cysteine (NAC, a ROS scavenger, 1 mM) and then stained with the DFCB sensor. From Fig. 9(b) it was concluded that the average fluorescence intensity was enhanced after the removal of the endogenous H₂O₂ from the HeLa cells. The previously described experimental imaging data confirmed the excellent imaging ability of the DFCB sensor for generating endogenous H₂O₂. In addition, the untreated HeLa cells were selected as the control group. Significant fluorescence emission could not be observed in the cells due to the absence of the DFCB sensor, and these experimental data confirmed that the DFCB sensor could be used to monitor endogenous H₂O₂ production in living cells. All these cellular experiments confirmed that the DFCB sensor could image endogenous H₂O₂ in living cells.

Fig. 9 (a) Imaging of H₂O₂ in HeLa cells: (A–C) HeLa cells alone were cultured for 1 h and acted as the control; (D–F) HeLa cells supplemented with the DFCB sensor (4 μM) were cultured for 1 h; (G–I) the HeLa cells supplemented with PMA (1 μg ml⁻¹) were cultured for 30 min; (J–L) subsequently, the HeLa cells supplemented with PMA (1 μg ml⁻¹) were cultured for 60 min; and (M–O) the HeLa cells were cultured with PMA (1 μg ml⁻¹) and then incubated with NAC (1 mM), followed by staining with the DFCB sensor (4 μM) for 1 h. (b) The quantitative intensity was obtained from the fluorescent images of the HeLa cells.

Bioimaging in zebrafish

The DFCB sensor was further used to detect H₂O₂in vivo. As shown in Fig. 10, zebrafish treated with the DFCB sensor (4 μM) exhibited an intense orange fluorescence, suggesting that the DFCB sensor possessed excellent cell permeability. However, the fluorescence intensity of the DFCB sensor gradually decreased until it was completely quenched by the increase of H₂O₂ concentration (6 and 12 μM). Furthermore, the fluorescence imaging phenomenon in living zebrafish was supported by the 3D surface and the relative fluorescence intensity plots. The previous experimental results showed that the DFCB sensor can detect exogenous H₂O₂ in organisms.

Fig. 10 (a) Imaging of H₂O₂ in zebrafish: (A–C) zebrafish alone were cultured for 1 h and acted as the control; (E–G) zebrafish supplemented with the DFCB sensor (4 μM) were cultured for 1 h; (I–K) zebrafish supplemented with the DFCB sensor (4 μM) and H₂O₂ (6 μM) were cultured for 30 min; and (M–O) zebrafish were cultured with the DFCB sensor (4 μM) and H₂O₂ (12 μM) for 1 h. (D, H, L and P) are the corresponding 3D surface plots of the fluorescence images. (b) Quantitative intensity was obtained from fluorescent images of the zebrafish.

As shown in Fig. 11, the zebrafish incubated with the DFCB sensor (4 μM) emitted an orange fluorescence unlike the control group, which confirmed that zebrafish cannot emit fluorescence by themselves. With the addition of PMA (1 μg ml⁻¹) for different times, the emission fluorescence of the DFCB sensor in zebrafish was gradually quenched, which indicated that the PMA supplement could stimulate the production of endogenous H₂O₂ in zebrafish. Compared with the zebrafish with the DFCB sensor, the fluorescence intensity was stronger in the zebrafish which was treated with NAC (1 mM) and then cultured with the DFCB sensor. These experimental results confirmed that the DFCB sensor had good potential for use in biological imaging and in the detection of endogenous H₂O₂. In addition, the relative fluorescence intensity chart was used to support the biological imaging experiments. These experimental data indicated that the DFCB sensor can detect exogenous and endogenous levels of H₂O₂ in biological systems.

Fig. 11 (a) The imaging of H₂O₂ in zebrafish: (A–C) zebrafish alone were cultured for 1 h; (D–F) zebrafish supplemented with the DFCB sensor (4 μM) were cultured for 1 h; (G–I) zebrafish supplemented with PMA (1 μg ml⁻¹) were cultured for 30 min; (J–L) subsequently, the zebrafish supplemented with PMA (1 μg ml⁻¹) were cultured for 60 min; and (M–O) the zebrafish were cultured with PMA (1 μg ml⁻¹) and then incubated with NAC (1 mM), followed by staining with the DFCB sensor (4 μM) for 1 h. (b) Quantitative intensity obtained from fluorescent images of zebrafish.

Experiments

Instruments and materials

The chemical reagents used in the experiments are normally available commercially and were used directly. Curcumin was purchased from the Shanghai Titan Technology Company. The BF₃·(C₂H₅)₂O was obtained from the Shanghai Lingfeng Chemical Reagent Company. The 2-(4-bromomethylphenyl)-4,4,5,5-tetramethyl-1,3,2-doxaborolane was obtained from the Shanghai Haohong Biomedical Technology Company. The ¹H-NMR and ¹³C-NMR spectra were recorded on a AV-600 NMR spectrometer (Bruker). The HRMS was carried out on a JMS-800D mass spectrometer (Jeol). The UV-vis absorption spectra were recorded on a UV-2450 spectrometer (Shimadzu). The fluorescence emission spectra were recorded using a LS55 fluorescence spectrometer (PerkinElmer).

Synthesis of compound DFC

Curcumin (1 mmol) was dissolved in 30 mL of CH₂Cl₂ at ambient temperature, and BF₃·(C₂H₅)₂O (2.4 mmol) was added dropwise. The reaction was continued for 5 h, and monitored with thin layer chromatography. After concentrating and removing half of the solvent, the reaction mixture was placed in a refrigerator for crystallization to take place. The red solid product DFC was obtained by filtering, washing with CH₂Cl₂, and drying, with a yield of 98%. ¹H-NMR (600 MHz, DMSO-d₆) δ: 10.09 (s, 2H), 7.92 (d, J = 15.5 Hz, 2H), 7.47 (s, 2H), 7.34 (d, J = 10.3 Hz, 2H), 7.01 (d, J = 15.6 Hz, 2H), 6.87 (s, 2H), 6.45 (s, 1H), 3.85 (s, 6H). ¹³C-NMR (151 MHz, DMSO-d₆) δ: 178.89, 151.52, 148.35, 147.14, 126.16, 125.44, 118.03, 116.13, 112.56, 101.29, 55.94. HRMS (m/z): [M + H]⁺ calcd for C₂₁H₁₉BO₆F₂Na + H⁺, 439.1140; found, 439.1141.

Synthesis of the DFCB sensor

The DFC (1 mmol) and 2-(4-bromomethylphenyl)-4,4,5,5-tetramethyl-1,3,2-doxaborolane (2.4 mmol), and K₂CO₃ (2.4 mmol) were dissolved in 8 mL of CH₃CN, and the reaction mixture was then refluxed for 6 h. The reaction mixture was kept in cryogenic refrigerator overnight, and the brown precipitate was filtered to obtain the crude DFCB. The crude DFCB was dissolved in ethyl acetate, and washed with deionized water until the pH became neutral. After drying with MgSO₄, followed by filtration, and distillation of the solvent, the DFCB was finally obtained, with a yield of 53% (Scheme 1). ¹H-NMR (600 MHz, DMSO-d₆) δ: 7.96 (d, J = 15.6 Hz, 2H), 7.70 (d, J = 7.6 Hz, 4H), 7.53 (s, 2H), 7.44 (dd, J = 17.0, 8.1 Hz, 6H), 7.17–7.07 (m, 4H), 6.52 (s, 1H), 5.23 (s, 4H), 3.86 (s, 6H), 1.29 (s, 24H). ¹³C-NMR (151 MHz, DMSO-d₆) δ: 179.43, 158.87, 158.65, 158.42, 158.20, 151.59, 149.70, 147.12, 144.36, 140.22, 134.91, 127.77, 127.32, 125.34, 124.51, 119.42, 117.70, 115.75, 113.67, 111.96, 84.00, 70.03, 56.12, 24.95. HRMS (m/z): [M + H]⁺ calcd for C₄₇H₅₃B₃O₁₀F₂Na + H⁺, 871.3784; found, 871.3794.

Scheme 1 The synthesis route for the DFCB sensor.

Spectral measurements

The stock solution of DFCB (1 mM) was prepared in CH₃CN. The stock solutions of different metal ions (10 mM) (Ag⁺, Ba²⁺, Ca²⁺, Co²⁺, Cs²⁺, Cu²⁺, Hg²⁺, K⁺, Mg²⁺, Mn²⁺, Na⁺, Ni²⁺, Zn²⁺), anions (AcO⁻, Br⁻, Cl⁻, ClO⁻, ClO₄⁻, CrO₇⁻, H₂PO₄⁻, HSO₃⁻, HSO₄⁻, I⁻, NO₃⁻, NO₂⁻, SO₄²⁻) and ROS (MCPBA, ˙OH, ONOO⁻, PAA, TBHP) were prepared in deionized water. The UV-vis absorption and fluorescence spectra of the samples were recorded in PBS solution (50% CH₃CN, pH = 7.4) at room temperature. For the fluorescence spectroscopy measurements, the excitation wavelength was λ_ex = 425 nm, and the slit width was λ_ex/λ_em = 5 nm/5.5 nm.

HeLa cell and zebrafish imaging

The HeLa cells were cultured in DMEM containing 10% fetal bovine serum (FBS) with 5% CO₂ at 37 °C. Next, the cells were divided into four groups. Except for group 1, the other cells from groups 2 to 4 were incubated in 4 μM DFCB solution for 1 h at 37 °C, and then incubated with different concentrations of H₂O₂ (0, 6, 12 μM) for another 30 min. Fluorescence images of the HeLa cells were recorded using a laser confocal fluorescence microscope in the orange channel (λ_ex = 405 nm, λ_em = 450–650 nm).

Four-day-old zebrafish were randomly divided into three groups. Each group was incubated with 4 μM of the DFCB sensor for 30 min, and then incubated with different concentrations of H₂O₂ (0 μM, 6 μM, and 12 μM) for 44 min. After washing the zebrafish three times with the PBS solution, the fluorescence images of the zebrafish were obtained using a laser confocal fluorescence microscope in the orange channel (λ_ex = 405 nm, λ_em = 450–650 nm).

Conclusions

In summary, a new curcumin-based fluorescent DFCB sensor was designed and synthesized for use in H₂O₂ detection. The DFCB sensor emitted an intense orange fluorescence and exhibited a significant quenched fluorescence response to H₂O₂. The color of the DFCB sensor solution changed from orange to blue in the presence of H₂O₂, when the solution was viewed with the naked eye. The DFCB sensor could selectively detect H₂O₂ and showed a good anti-interference performance. Moreover, the DFCB sensor exhibited the advantages of a low detection limit (1.31 μM), wide linear range (0–300 μM), and broad pH detection range (6–11). Furthermore, the DFCB sensor with low cytotoxicity and good biocompatibility was successfully used for imaging endogenous and exogenous H₂O₂ in living HeLa cells, as well as in living zebrafish. We predict that the results of this research will have the potential to be used in future for monitoring the physiological and pathological effects of H₂O₂in vivo.

Author contributions

Wenhao Du: investigation, methodology and writing – original draft. Zheyu Shen: formal analysis. Yueying Liang: software. Shuai Gong: investigation. Zhiyuan Meng: software. Mingxing Li: data curation. Zhonglong Wang: writing – review and editing. Shifa Wang: project administration.

Ethics declarations

Ethics approval

All the animal procedures were performed in accordance with the Guidelines for Care and Use of Laboratory Animals of Nanjing University and approved by the Animal Ethics Committee of China.

Conflicts of interest

The authors declare that they have no conflict of interest.

Acknowledgements

The authors wish to thank the National Natural Science Foundation of China (Grant No. 32071707 and 32101466) and the Natural Science Foundation of Jiangsu Province (Grant No. BK20210624) for their financial support for this research.

References

1. X. Li, N. Gao, C. Liu, M. Yu, X. Rong, Y. Zhang, M. Su, X. Wang, H. Zhu, K. Wang, Y. Liu, W. Sheng and B. Zhu, Sens. Actuators, B, 2022, 353, 131051.
2. X. Zhang, L. Zhang, X. Y. Cheng, S. D. Liu, S. J. Fang, L. W. Zhang, X. Y. Wang and L. X. Chen, Sens. Actuators, B, 2022, 366, 131982.
3. Y. Huang, L. Yu, P. P. Lu, Y. H. Wei, L. L. Fu, J. J. Hou, Y. Q. Wang, X. Y. Wang and L. X. Chen, J. Hazard. Mater., 2022, 424, 127425.
4. Y. Q. Shi, Q. C. Wu, W. T. Li, L. Lin, F. F. Qu, C. J. Shen, Y. Z. Wei, P. C. Nie, Y. He and X. P. Feng, J. Hazard. Mater., 2022, 432, 128605.
5. Q. Xu, Y. Y. Tang, P. D. Zhu, W. Y. Zhang, Y. Q. Zhang, O. S. Solis, T. S. Hu and J. C. Wang, Nanoscale, 2022, 14, 13771–13778.
6. Y. Wu, Z. Y. Li and Y. M. Shen, ACS Omega, 2019, 4, 16242–16246.
7. Z. L. Wang, Y. Zhang, J. Song, M. X. Li, Y. Q. Yang, X. Xu, H. J. Xu and S. F. Wang, Sens. Actuators, B, 2019, 284, 148–158.
8. J. Q. Xu, J. S. Guo, K. F. Xie, M. J. Gao, R. Wei, Z. H. Xin and Y. F. Kang, Dyes Pigm., 2022, 204, 110437.
9. X. L. Liu, M. D. Yan, Z. G. Chen, B. X. Zhang, N. C. Yao, S. Zhao, X. X. Zhao, T. Zhang and G. F. Hai, Spectrochim. Acta, Part A, 2023, 286, 121955.
10. M. Chen, Z. H. Liang, G. L. Zeng, Y. Wang, Z. H. Mai, X. Y. Chen, G. Wu and T. S. Chen, Dyes Pigm., 2022, 198, 109995.
11. L. Q. Li, M. H. Zheng, X. Y. Yan, H. Huang, S. X. Cao, K. M. Liu and J. B. Liu, J. Photochem. Photobiol., A, 2022, 432, 114069.
12. T. Ma, K. Fu, Z. C. Li, C. C. Yuan and W. B. Ma, Spectrochim. Acta, Part A, 2022, 276, 121218.
13. M. X. Wang, F. Zhang, C. Q. Wang, N. Yin, Y. T. Wang, G. X. Qin, Q. L. Xu, J. H. Gong, H. Z. Liu and X. R. Duan, Anal. Chem., 2022, 94, 5962–5969.
14. Y. Q. Li, Y. Zhou, J. N. Lei, Q. J. Lu, X. Qin, Q. Xu, Y. Q. Wang, C. Y. Wu, Z. Yang and B. S. He, J. Mol. Struct., 2023, 1271, 134042.
15. Y. Fang, J. Wang, H. Yu, Q. Zhang, S. J. Chen, K. P. Wang and Z. Q. Hu, Sens. Actuators, B, 2022, 371, 132514.
16. Y. Tian, S. Y. Liu, W. W. Cao, P. Wu, Z. M. Chen and H. Xiong, Anal. Chem., 2022, 94, 11321–11328.
17. M. Chen, Z. H. Liang, X. H. Fan, R. M. Qu, H. H. Wang and T. S. Chen, Spectrochim. Acta, Part A, 2022, 276, 121163.
18. L. Guo, S. Chen, Y. L. Yu and J. H. Wang, Anal. Chem., 2021, 93, 16240–16247.
19. M. R. Li, B. W. Wang, J. Y. Liu, Z. Z. Zhang, L. G. Chen, Y. Li and X. L. Yan, Anal. Chem., 2022, 94, 9732–9739.
20. K. Yin, F. B. Yu, D. Y. Liu, Z. H. Xie and L. X. Chen, Sens. Actuators, B, 2016, 223, 799–805.
21. Z. Y. Zhang, Z. P. Chen, F. B. Cheng, Y. W. Zhang and L. X. Chen, Biosens. Bioelectron., 2017, 89, 932–936.
22. X. Wang, Q. Ding, Y. Tian, W. Wu, F. D. Che, P. Li, W. Zhang, W. Zhang and B. Tang, Chem. Commun., 2022, 58, 6320–6323.
23. J. K. Liang, H. Li, J. R. Wang, H. L. Yu and Y. He, Anal. Chem., 2020, 92, 6548–6554.
24. S. Wang, Y. Zhang, T. R. Wang, Y. J. Liu, S. L. Shen and X. Q. Cao, Spectrochim. Acta, Part A, 2022, 266, 120435.
25. Y. Hua, Y. J. Shang, M. J. Gao, J. Li and Y. F. Kang, Spectrochim. Acta, Part A, 2022, 265, 120320.
26. X. D. Zeng, C. Jiang, Q. Zhang, D. K. Chai, M. S. Ma, J. Chen and Z. G. Liu, J. Lumin., 2021, 240, 118422.
27. J. Su, S. P. Zhang, C. R. Wang, M. Li, J. J. Wang, F. Su and Z. J. Wang, ACS Omega, 2021, 6, 14819–14823.
28. L. L. Xu, Y. Zhang, L. H. Zhao, H. Han, S. Q. Zhang, Y. B. Huang, X. H. Wang, D. Q. Song, P. Y. Ma, P. Ren and Y. Sun, Talanta, 2021, 233, 122578.
29. G. Q. Yang, T. Zhu, D. Wang, Z. J. Liu, R. L. Zhang, G. M. Han, X. H. Tian, B. H. Liu, M. Y. Han and Z. P. Zhang, Chem. Commun., 2021, 57, 6628–6631.
30. R. R. Zhou, Q. Y. Peng, D. Wan, C. Yu, Y. Zhang, Y. Hou, Q. Luo, X. Li, S. H. Zhang, L. Xie, P. H. Ou and Y. B. Peng, RSC Adv., 2021, 11, 24032–24037.
31. Y. M. Hao, H. P. Wang, Z. H. Wang, W. J. Dong, Q. Hu, S. M. Shuang, C. Dong and X. J. Gong, Mikrochim. Acta, 2021, 188, 16.
32. X. H. Pan, Y. H. Zhao, T. T. Cheng, A. S. Zheng, A. B. Ge, L. X. Zang, K. H. Xu and B. Tang, Chem. Sci., 2019, 10, 8179–8186.
33. Z. W. Gan, T. Zhang, X. X. An, Q. Tan, S. J. Zhen, Y. M. Hu and X. L. Hu, Microchem. J., 2022, 182, 107939.
34. H. Yang, C. Yu, Z. Yin, P. Guan, S. Jin, Y. Wang and X. Feng, J. Sci. Food Agric., 2022, 103, 1550–1560.
35. H. L. Dong, P. Wang, Z. Y. Yang, R. Li, X. L. Xu and J. Shen, Ultrason. Sonochem., 2022, 90, 106188.
36. H. Li, T. Wang, J. Q. Su and P. V. Meeren, Food Hydrocolloids, 2022, 133, 108020.
37. J. P. Hu, Y. L. Wang, T. Y. Shao, G. C. Lian, K. B. Hu, Y. Liu, M. Zhou, X. P. Wang, L. Z. Huang, X. L. Meng and G. F. Jin, Arabian J. Chem., 2022, 15, 104087.
38. S. Kang, B. Y. Park, S. Lee, N. Lee and M. S. Han, Analyst, 2021, 146, 463–470.
39. Q. Jiang, Z. L. Wang, M. X. Li, J. Song, Y. Q. Yang, X. Xu, H. J. Xu and S. F. Wang, Tetrahedron Lett., 2020, 61, 152103.
40. Y. Gao, M. X. Li, X. C. Tian, S. Gong, Y. Zhang, Y. Q. Yang, Z. L. Wang and S. F. Wang, Microchem. J., 2022, 169, 106631.
41. C. S. Lim, M. Y. Cho, M. Y. Park and H. M. Kim, ChemistryOpen, 2017, 7, 53–56.
42. R. F. Xu, Y. Wang, H. Y. You, L. W. Zhang, Y. Q. Wang and L. X. Chen, Analyst, 2019, 144, 2556–2564.
43. Y. Liu, J. Nie, J. Niu, F. F. Meng and W. Y. Lin, Sci. Rep., 2017, 7, 7293.
44. G. Masanta, C. H. Heo, C. S. Lim, S. K. Bae, B. R. Cho and H. M. Kim, Chem. Commun., 2012, 48, 3518–3520.
45. K. Xu, L. He, X. Yang, Y. Yang and W. Lin, Analyst, 2018, 43, 3555–3559.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d3an00340j

---