First report on a BODIPY-based fluorescent probe for sensitive detection of oxytetracycline: application for the rapid determination of oxytetracycline in milk, honey and pork

Abstract

A new BODIPY-based fluorescent probe (S) for oxytetracycline bearing hydrophilic carboxyl has been designed and synthesized. Fluorescence is quenched by oxytetracycline (OTC), and the effect was used to develop a method for the determination of OTC. This is the first report on the determination of OTC by a BODIPY-based “turn-off” fluorescence probe, which provides a sensitive and selective method for OTC detection. S exhibited higher properties, such as acting as a “naked eye” probe, stronger anti-jamming capability, outstanding thermal stability, good stability to pH, fast-response characteristics, and better live-cells imaging with low cytotoxicity when compared with other probes. Moreover, this is the first time that one chemosensor could be successfully applied to OTC imaging in zebrafish, which demonstrated its excellent organism permeability. Furthermore, the feasibility of the developed assays in milk, honey, and pork samples was verified through recovery experiments using spiked samples. Under the optimum conditions, good linearity was obtained in the range of 0–42 µM and the recoveries ranged from 101.3% to 104.8%, 101.9% to 107.9%, and 98.3% to 109.5% for milk, honey, and pork, respectively, with relative standard deviations less than 1.74%.

---

1 Introduction

Oxytetracycline (OTC) is a member of the tetracycline family.¹ Because of its strong antibacterial activity against both Gram-positive and Gram-negative bacteria, high clinical curative effect, and less adverse reactions,^2,3 it has been applied to the treatment of a variety of diseases both systemically (furunculosis, vibriosis, pasteurellosis, and yersiniosis) and topically in nature (columnaris disease).^4,5 Today, OTC is also used in a wide variety of food-producing animals such as cattle, chicken, sheep, and pigs. However, OTC can accumulate in the bone tissue of mammals, which is a serious threat to human health.⁶ What is more, excessive abuse will also lead to bacterial drug resistance, water pollution, soil contamination side effects, and toxic effects.⁷ To safeguard consumer health in China, the maximum residue limit for OTC is 100 µg kg⁻¹ in milk and muscle, 200 µg kg⁻¹ in eggs, and 300 µg kg⁻¹ in liver. Therefore, reliable analytical methods are required for monitoring OTC in food products.

To date, conventional approaches to detect OTC are still limited. The most common method for the determination of OTC is HPLC, and it offers good selectivity with a good detection limit,^8–10 but this method needs professional operators as well as expensive and complicated instrumentation. To determinate OTC in different environments, several other methods have also been reported such as UV-vis spectrophotometry,¹¹ capillary zone electrophoresis,¹² potentiometry,¹³ colorimetry, and light scattering agglutination aptasensing.¹⁴ Although they are frequently used, they have many shortages such as being not sensitive enough to detect low concentrations, time consuming, being expensive, requiring the use of a specially packed column and is laborious.^15–18 Recently, several studies on the detection of OTC in food using fluorescence assays have been reported. For example, Gao et al.¹⁹ developed a novel fluorescence assay for (OTC) based on fluorescence quenching of water-soluble CdTe nanocrystals. The synthetic TGA-capped CdTe nanoparticle, with high stability in an aqueous system, was feasible for fluorescence detection, and the practical usefulness of the QDs achieved satisfactory results with a wide linear range, low detection limit, and good precision. However, this method is easily affected by pH, and is therefore not conducive to quantifying OTC present in food. Yuan et al.²⁰ performed a fluorescence assay for OTC detection based on an indirectly fluorescein-labelled aptamer probe, which was fabricated through the partial hybridization of an OTC long-chain aptamer with a FAM-labelled short-chain ssDNA (S1). Upon combination of the target, OTC, and its aptamer, S1, with quenched fluorescein, was released from the probe to the graphene sheet freely. But it was relatively time-consuming because the long time required for preparation of graphene prior to detection. And the designed OTC aptamer is very expensive. Xu et al.²¹ developed a versatile probe for the “turn off” sensing of Hg²⁺ based on the aggregation-induced fluorescence quenching and for the “turn-on” sensing of OTC with a high sensitivity and selectivity; the method had a good linearity (0.375 to 12.5 µM⁻¹) with good recoveries ranging from 98.80 to 103.00%.

The aim of this work was to develop a simple, cheap and sensitive assay with better anti-jamming ability to determine residues of OTC in food products compared to other fluorescent probes. BODIPY dye (boron-dipyrromethene), since it was synthesized by Treibs et al. in the 1960s, is becoming a hot spot of research in recent decades because of its unique photophysical properties and photochemistry, good light stability, high quantum yield, high photo-stability, and relative insensitivity to environmental perturbations.^22–29 Recently, through structure modifications,^30–34 it has also been used in many fields such as sensitizers for living cells, cationic and anionic chemical sensors, medical applications, and electroluminescent agents.^35–43

In this paper, a novel water-soluble daylight-fluorescent BODIPY derivative (S) with a polar group (COOH) was synthesized adopting a one-pot reaction. S shows a glow characteristic of a daylight-fluorescent material since the emitted fluorescent light is visible under daylight illumination. This green daylight fluorescence is diminished upon interaction with OTC, resulting in a highly selective and sensitive naked-eye detection of OTC over other common anions. Under optimal conditions, the method exhibits good linearity in the range 0–42 µM with a limit of detection of 0.72 µM. In addition, the method was successfully applied to the analysis of milk, honey, and pork samples with high selectivity. It is worth noting that this is the first time that BODIPY was used for antibiotics detection (OTC) in food samples with fluorometry, which has not been previously reported.

2 Experimental

2.1 Materials

Oxytetracycline (OTC), tetracycline (TET), doxycycline (DOX), and chlortetracycline (CTE) were purchased from Shanghai Chemical Reagent Factory and Shanghai Biotechnology Co. Ltd. Glutaric anhydride, 2,4-dimethylpyrrole, dichloromethane, boron trifluoride ether complex, sodium hydroxide, hydrochloric acid, rhodamine, and all other reagents were all obtained from Xilong Chemical CO., Ltd. Zebrafish were bought from China Zebrafish Resource Center. HepG2 cells, fetal bovine serum (FBS), and Dulbecco's Modified Eagle's Medium (DMEM) were all purchased from Shanghai Gefan Biotechnology Co., Ltd. Milk, honey, and pork were purchased from a local Yonghui supermarket. All reagents were of analytical reagent grade. Solutions were prepared with distilled water (prepared by Fuzhou University). LIVE SUBJECT STATEMENT: This housing facility has a barrier and maintains the national standard “Laboratory Animal-Requirements of Environment and Housing Facilities” (GB 14925-2001). The care of laboratory animals and the animal experimental operations conformed to the “Beijing Administration Rule of Laboratory Animal”. The Zebrafish experiments were approved by the China Zebrafish Resource Center (CZRC), Key deployment project of the Chinese Academy of Sciences (KSZD-EW-Z-001), and Project of Major Scientific Research Projects of the Ministry of Science and Technology (2012CB944504). All experiments with human subjects (cell experiments) were according to the Chinese Pharmacopoeia.

2.2 Instruments

All experiments were carried out at room temperature, unless stated otherwise. Reactions were magnetically stirred and monitored by thin layer chromatography (TLC). Column chromatography was conducted over silica gel (mesh 200–300). ¹H NMR and ¹³C NMR spectra were carried out on a Bruker 400 spectrometer in CDCl₃. Mass spectra were measured on a HP 1100 LC-MS spectrometer. The molecular structure and the groups of the prepared material were characterized by Fourier transform infrared spectroscopy (FT-IR) (Bruker VERTEX 70 FTIR). UV-visible spectra were recorded on a U-4100 spectrophotometer with a scan rate of 1000 nm s⁻¹. Fluorescence spectra were determined on a Hitachi F-4600 fluorospectrophotometer with a quartz cuvette (path length = 1 cm). An Olympus Zeiss 710 laser scanning confocal microscopy was used for fluorescence imaging of the cells. The Jingke pH measurements were measured by a PHS-3D digital pH meter. Solutions of metal ions were prepared from their analytical grade nitrate or chloride salts.

2.3 Synthesis of S

To a nitrogen-flushed round-bottom flask were added, consecutively, glutaric anhydride (56 mg, 0.50 mmol), dry CH₂Cl₂ (15 mL), 2,4-dimethylpyrrole (93 mg, 1.00 mmol), and BF₃·OEt₂ (0.096 g, 0.70 mmol). The mixture was heated at reflux for 4 h. After the mixture was cooled to room temperature, BF₃·OEt₂ (0.49 g, 3.5 mmol) and Et₃N (0.30 g, 3.0 mmol) were added. The reaction mixture was stirred under nitrogen at room temperature for 8 h. TLC showed that the reaction was completed. Then, the mixture was extracted with petroleum ether/CH₂Cl₂, v/v, 2 : 1. The solvent was evaporated under vacuum and the resulting dark oil was purified by chromatography on silica gel (elution with hexane/ethyl acetate, v/v, 2 : 1) to give 30 mg (18%) of BODIPY as a dark red solid. ¹H NMR δ 6.06 (s, 2H), 3.05–2.98 (m, 2H), 2.58–2.48 (m, 2H), 2.51 (s, 6H), 2.42 (s, 6H), 2.02–1.93 (m, 2H); ¹³C NMR δ 177.9, 154.3, 144.7140.4, 131.5, 121.9, 27.4, 26.5, 16.3, 14.5, 14.4; ¹⁹F NMR δ −146.3 to 146.8 (m); HRMS [EI, M⁺ of 10B species] m/z calcd for C₁₇H₂₁BF₂N₂O₂ 333.1701, found 333.1710.

2.4 Fluorescence quantum yield

The relative fluorescence quantum yields were determined with pure rhodamine B (fluorescence in 10 mM PBS; Φ = 0.95) as a standard, taking the total area under the curve, and calculated by the following equation:

Φ_X = Φ_S(F_X/F_S)(A_S/A_X)(λ_exs/λ_exx)(n_x/n_s)²

where Φ is the quantum yield, F is the integrated area under the corrected emission spectrum, A is the absorbance at excitation wavelength, λ_ex is the excitation wavelength, and n is the refractive index of the solution. The subscripts x and s refer to the unknown and the standard, respectively.

2.5 Determination of OTC

Deionized water was used throughout all experiments. The fluorescence intensity was monitored at an excitation wavelength of 358 nm and recorded at an emission wavelength of 515 nm. The slits for excitation and emission were both set at 5 nm. Solutions of Ca²⁺, Fe³⁺, Hg²⁺, Cd²⁺, Mg²⁺, Na⁺, K⁺, and Al³⁺ were prepared from their chloride salts; solutions of Sn²⁺, Co²⁺, Ni²⁺, and Cu²⁺ were prepared from their nitrate salts. All the metal ion solutions were at the concentration of 200 µM. In the metal ion effect experiment, a 3.0 mL solution of S (10 µM) was poured into a quartz optical cell of 1 cm optical path length each time, followed by an addition of 20 equiv. metal ions (200 µM). TET, DOX, and CTE were examined under the same procedure for the specificity tests. In the titration experiments, a 3.0 mL solution of S (10 µM) was poured into a quartz optical cell of 1 cm optical path length each time, and OTC solution was added into the quartz optical cell gradually using a micro-pipette.

2.6 Cell culture and fluorescence bioimaging

HepG2 cells were grown in high glucose DMEM (4.5 g of glucose per L) supplemented with 10% FBS at 37 °C and 5% CO₂. Cells (5 × 112 per L) were plated on 14 mm glass coverslips and allowed to adhere for 24 h. Experiments to assess the cytotoxicity of S were performed over 1 h in the same medium supplemented with various concentrations of S (10, 20, 30, 40, 50, and 60 µM) by MTT assays. MTT (Sigma) solution (5.0 mg mL⁻¹, PBS) was then added to each well. After 4 h, the remaining MTT solution was removed and 150 mL of DMSO was added to each well to dissolve the formazan crystals. Absorbance was measured at 570 nm in a TRITURUS microplate reader. Subsequently, immediately before the cell imaging experiments, cells were first incubated with 10 µM of S for 6 h at 37 °C. Then, they were washed with PBS buffer and then incubated with 42 µM OTC in PBS for 30 min at 37 °C. After washing the cells with PBS, confocal fluorescence imaging was performed with a Zeiss LSM 710 laser scanning microscope and a 63× oil-immersion objective lens. For zebrafish experiments, wild type zebrafish were firstly incubated with S (10 µM) for 30 min. After washing three times with PBS, it was then fed with a solution containing various concentrations of OTC (10, 20, 30, and 50 µM) at 28.5 °C for 1 h. Zebrafish imaging was recorded under UV lamp (365 nm) irradiation.

2.7 Pre-treatment of actual samples for OTC detection

Three types of actual samples (milk, honey, and pork) were chosen to evaluate the feasibility of this assay based on the corresponding pre-treatment. For milk samples, Ca²⁺ and Mg²⁺ in milk can form chelation complexes with OTC, which will affect the detection of OTC. To remove these components including protein, fat, carbohydrate, and cations, milk samples (5 mL; fat, 4.3%) were mixed with 5 mL of McIlvaine buffer containing 20 mM EDTA (pH 5.0) and 0.5% (v/v) trifluoroacetic acid to denature milk proteins. Then, the mixture was defatted and deproteinized by centrifugation at 4 °C for 10 min at 12 000 rpm. Then, 1 M NaOH was added dropwise to the supernatant to adjust the pH to 7.0 and subjected to analysis.

The honey samples (2.00 ± 0.03 g) were dissolved in 20 mL of 10 mM PBS (pH 7.0) and mixed 10 min using a vortex mixer. Then, it was ultrasonicated for 10 min. The supernatant was then subjected to analysis.

The pork samples (2.00 ± 0.04 g, muscle tissue) were weighed into 50 mL polythene centrifuge tubes. The mixture was then placed in an ultrasonic bath for about 10 min to disperse the analyte thoroughly into the matrix. Then, 8 mL of the extraction solution (trichloroacetic acid at 0.90 g L⁻¹, 19.6% methanol, and 0.1% mercaptoethanol) was added and mixed with a vortex mixer for 10 min. The mixture was then centrifuged for 10 min at 12 000 rpm. The supernatant was diluted fivefold with 10 mM PBS (pH 7.0), and was then subjected to analysis.

3 Results and discussion

A fluorescent probe 4-(4,4-difluoro-1,3,5,7-tetramethyl-4-bora-3a,4a-diaza-s-indacene-8-yl)-butyric acid, as a OTC sensor, was synthesized according to Scheme 1 using a one-pot reaction. The fluorescence quantum yield of S was very high (Φ = 0.89) in PBS buffer (10 mM, pH 7.0). Then, the fluorescence spectra of S in PBS buffer (10 mM, pH 7.0) was investigated. As shown in Fig. 1, the emission fluorescence spectrum of S appeared at the maximum emission wavelength and was 504 nm in 10 mM PBS (pH 7.0) when excited at λ_ex = 358 nm. When 20 equivalents of OTC were added to the PBS solution of sensor S, dramatic fluorescent quenching was observed immediately. The apparent fluorescence emission color changed from bright green to colorless and could be distinguished by the naked-eye through UV lamp irradiation. This indicated that sensor S shows a specific response to OTC.

Scheme 1 Synthetic procedure for S and schematic of the quenching mechanism of S by OTC.

Fig. 1 Fluorescence spectra of S (10 µM) upon an excitation at 358 nm in PBS buffer (10 mM, pH = 7.0) in the presence of OTC (20 equiv.). Inset: image from left to right shows the change in the fluorescence of S and S + OTC (20 equiv.) in PBS buffer (10 mM, pH = 7.0).

3.1 Selectivity study of BODIPY toward OTC

The selectivity of S as a fluorescent sensor for OTC towards the analyte over potentially competing species was investigated in PBS buffer (10 mM, pH 7.0) by incubating S (10 µM) with a wide range of heavy metal ions and tetracycline derivatives (200 µM). The fluorescence intensity results were recorded 5 minutes later. As can be seen in the Fig. 2a, the addition of Fe³⁺ caused small quenches of fluorescence intensities while the addition of other metal ions did not induce any observable change in the fluorescence intensity. Three structurally similar tetracycline derivatives were also used to verify the good selectivity of S, and Fig. 2b show that the addition of TET can also cause small quenches among the tetracycline derivatives. In contrast, the addition of OTC resulted in substantial quenching (approximately 92.2%) of the original fluorescence intensity. Such a variation by Fe³⁺ and TET was relatively small compared with OTC, indicating that S showed the stronger response in absorbance spectrum to OTC among these metal cations and tetracycline derivatives.

Fig. 2 (a) Fluorescence intensity of S (10 µM) in PBS buffer (10 mM, pH = 7.0) in the presence of 20 equiv. of different aqueous metal ion solutions. (b) Fluorescence intensity of S (10 µM) in the presence of 20 equiv. OTC, TET, DOX, and CTE. Experiments were carried out in PBS buffer (10 mM, pH = 7.0).

3.2 UV-vis absorption responses of S

To examine whether BODIPY has the potential be used to measure OTC, we first conducted UV-vis absorption assays. The absorption spectra of S (10 µM) was first explored in PBS buffer (10 mM, pH 7.0) solution in the presence of different concentrations of OTC (0–1.6 equiv.) and the results are shown in Fig. 3. S exhibited a very strong absorption rate at 468 nm. When increasing concentrations of OTC were added, the absorption spectra at 468 nm did not change, and a new absorption band at 353 nm gradually appeared, which may be ascribed to the characteristic peak of OTC according to the literature.⁴⁴ An obvious color change from bright green to colorless can be observed by the naked eye.

Fig. 3 UV-vis spectra of S (10 µM) in PBS buffer (10 mM, pH = 7.0) in the presence of increasing concentrations of OTC. The final ratio of OTC to S is 4.2 equiv.

3.3 Fluorescence titration studies of S

Subsequently, we carried out the fluorescence titration studies of S towards OTC under the same conditions and a series of S spectra with different concentrations of OTC were recorded (Fig. 4). S (10 µM) exhibits high fluorescence in the visible region upon treatment with increasing concentrations of OTC. The fluorescence at 504 nm decreased gradually and the fluorescence was almost quenched when 4.2 equivalents of OTC were added to the PBS solutions of sensor S. Under the present conditions, a good linear relationship between the changes of fluorescence intensity and the OTC concentration could be obtained in the 0–42 µM range (R² = 0.9990). The detection limit was calculated to be 0.72 µM based on the equation: detection limit = 3b_i/m, where b_i is the standard deviation of the blank measurements and m is the slope between fluorescence intensity and sample concentration.

Fig. 4 (a) Fluorescence changes of S (10 µM) with various concentrations of OTC (0–4.2 equiv.) in PBS buffer (10 mM, pH = 7.0) recorded after 5 min, excitation at 358 nm. (b) A plot of the fluorescence intensity changes of S (10 µM) at 504 nm in the presence of different concentrations of OTC (0–4.2 equiv.) in PBS buffer (10 mM, pH = 7.0), ΔF = F₀ − F, F₀: fluorescence intensity of S (10 µM), F: fluorescence intensity after addition of OTC. (c) Photographs of S (10 µM) with various concentrations of OTC (0–4.2 equiv.) from left to right under UV excitation (λ_ex = 365 nm) in PBS buffer (10 mM, pH = 7.0).

3.4 Effect of pH on S

The influence of pH on S both in the absence and in the presence of OTC was examined. In these studies, each experiment was replicated at least three times. The same amount of S (10 µM) was dissolved in the buffer solutions with different pH values (pH = 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, and 13), and the fluorescence intensity was recorded in the presence and absence of OTC (2.7 equiv.) after reacting for 5 min. As shown in Fig. 5, pH has an important role on the S sensing system. With the pH value increasing from 2 to 7, the intensity of S almost keeps unchanged and keeps relatively stable in acidic conditions. While fluorescence quenching becomes stronger as pH values increase from 7 to 11, which indicates that the S sensing system is pH dependent.

Fig. 5 Fluorescence response of S (10 µM) in PBS buffer (10 mM, pH = 7.0) and after the addition of 2.7 equiv. OTC as a function of different pH values. Slit widths were 5 nm.

3.5 Effect of temperature on S

The experimental temperature influenced the experiment and the results from temperature studies were analyzed. The same amount of S (10 µM) was dissolved in PBS buffer (10 mM, pH 7.0) solution in the presence and absence of OTC (2.7 equiv.) and reacted for 5 min. Then, it was heated in a water bath of different temperatures (0–100 °C) followed by cooling to room temperature. Obviously, Fig. 6 reveals that S kept a nearly constant value in the temperature range of 10–50 °C. When the temperature was below 10 °C, the fluorescence was enhanced a little. However, the fluorescence decreased sharply when the temperature was over 60 °C, which may be ascribed to the damage to the structure of S. Therefore, room temperature (about 25 °C) was chosen as the appropriate temperature for the experiments.

Fig. 6 Fluorescence responses of the S solution (10 µM) in PBS buffer (10 mM, pH = 7.0) and after the addition of 2.7 equiv. OTC as a function of temperature. Slit widths were 5 nm.

3.6 Effect of contact time on S

Next, time course studies were investigated to judge whether the fluorescence color changes could be observed immediately upon the addition of OTC. In the experiment, the changes in the fluorescence intensity depending on the reaction time were recorded from 0 to 15 min for the mixture of S (10 µM) and OTC (2.7 equiv.) in PBS buffer (10 mM, pH 7.0) at room temperature (Fig. 7). It clearly shows that the reaction is completed within 5 min after the addition of OTC and S exhibits a quick response time to OTC.

Fig. 7 Fluorescence responses of the S solution (10 µM) in PBS buffer (10 mM, pH = 7.0) after the addition of 2.7 equiv. OTC as a function of different reaction time. Slit widths were 5 nm.

3.7 Reaction mechanism of S with OTC

To investigate the recognition mechanism of S with OTC, S and S + OTC were subjected to mass spectral analyses. The ion peak was detected at m/z 334.1, which corresponds to [S + H]⁺. The ion peaks at m/z 238.0 and 100.94 demonstrated the presence of a breakdown product of S (C₈H₉BF₂N₂ and pentanoic acid). We speculated the presence of OTC leads to the destruction of S, which then causes a structural change (Scheme 2). To obtain more detailed insights into the mechanism, interaction between S and OTC was monitored by HPLC. Similar with mass spectrometry analysis, results from the HPLC monitoring test show that the BODIPY peak around 9.7 min disappears gradually with increases in OTC. Meanwhile, a new peak corresponding to structural damage of S at 6.8 min appears and increased with OTC addition (Scheme 3).

Scheme 2 The proposed mechanism of S with OTC.

Scheme 3 HPLC analysis of the reaction process.

FT-IR spectrum has also proven to be a suitable technique to give enough information to elucidate the way of the recognition mechanism of the sensor S to OTC. S was mixed with OTC in methanol solution and dried in vacuo prior to obtaining its IR spectrum. As shown in Fig. 8, FT-IR spectrum of S showed a strong resonance at 1695 cm⁻¹, which could be assigned to the C=O stretching vibration of the carboxy group of unsaturated carboxylic acid. Upon complexed with OTC, the C=O stretch of the carboxy group (1695 cm⁻¹) diminished and a new strong peak appeared at 1705 cm⁻¹, which is the C=O stretching vibration of the carboxy group of the saturated carboxylic acid. This indicated that the structure of unsaturated carboxylic acid S was destroyed and a new saturated carboxylic acid was formed. This is in agreement with results above.

Fig. 8 FT-IR spectra of compound S and S + OTC mixture.

3.8 MTT assays and HepG2 cells

Next, we evaluated the cytotoxicity of S by MTT assays with the concentration of S ranging from 10.0 µM to 60.0 µM. As shown in Fig. 9, when the S concentration is not more than 30 µM, the probe exhibits little cytotoxicity for the cells. The results demonstrate that S has a low toxicity to the cultured cell line when concentrations are not more than 30 µM.

Fig. 9 Cell viability of HepG2 cells incubated with varying concentrations of S (10–60 µM) for 1 h. Viability of the cells was assessed using the MTT assay. The reported percent cell survival values are relative to untreated control cells.

Because of the highly sensitive and selective responses for OTC, S could be considered a promising tool for imaging in living cells. Thus, the capability of S for the fluorescence quench imaging of OTC in living cells was investigated. As can be seen in Fig. 10, HepG2 cells were found to give a strong fluorescence in the green channel when the cells were incubated with S (10 µM) in culture medium for 1 h at 37 °C, indicating that S is cell-permeable. When the S-loaded cells were subsequently treated with OTC (5 equiv., excited at 488 nm) for 30 min, an obvious fluorescence quenching was observed. In order to enlarge the applied range, we explored the feasibility of S for the detection of OTC in zebrafish imaging (Fig. 11). With increasing concentrations of OTC in zebrafish, the fluorescence intensity was gradually decreased, which can be observed by the naked eye under UV lamp (365 nm) irradiation. These results suggested that S has good cell permeability and has the potential to visualize OTC in living organisms.

Fig. 10 Confocal fluorescence images of HepG2 cells. (a–c) HepG2 cells incubated with S (10 µM) for 1 h; (a₁–c₁) then incubated with OTC (4.2 equiv.) for 30 min (λ_ex = 488 nm) (a and a₁) bright-field image; (b and b₁) merged image; (c and c₁) fluorescence image.

Fig. 11 Fluorescence changes of S (10 µM) after the addition of OTC into zebrafish. (a–e) Dark-field image of zebrafish after the addition of S (10 µM) followed by incubating with various concentrations of OTC (0, 1, 2, 3, and 5 equiv.).

3.9 Detection of OTC in actual samples

In order to further investigate the feasibility of the developed sensor S for OTC, OTC in milk, honey, and pork samples were assayed. These samples were purchased from the local market and the spiked method was used to detect OTC in the spiked samples. The pretreatment of the samples was mentioned above. The accuracy was evaluated by comparing the results obtained from the analysis of OTC by the proposed fluorescence method with those obtained using the recommended HPLC method described in the US Pharmacopeia for OTC. Recovery tests were used to evaluate the accuracy of the two methods. The results are shown in Table 1. The results in Table 1 showed that the mean recoveries of spiked samples were between 101.3% and 104.8%, 101.9% and 107.9%, and 98.3% and 109.5%; the relative standard deviation ranged from 0.53% to 1.14%, 0.75% to 1.49%, and 1.09% to 1.74%. This indicated that the method developed can be used to screen OTC in a variety of food matrices with acceptable accuracy and precision.

Table 1 OTC detection in actual samples and spiked recoveries^a

Sample	Spiked concentration (µM)	Determined for OTC (µM, mean ± SD)	OTC recovery(%, mean)	RSD (%)	HPLC for OTC (µM, mean ± SD)	RSD (%)
a The results were obtained from five repetitive assays.
Milk	10	10.48 ± 0.172	104.8 ± 1.72	0.53	10.33 ± 0.117	0.73
	20	20.36 ± 0.165	101.8 ± 0.83	1.09	20.18 ± 0.132	0.87
	30	30.39 ± 0.333	101.3 ± 1.11	1.14	30.24 ± 0.226	1.12
Honey	10	10.79 ± 0.156	107.9 ± 1.56	1.49	10.66 ± 0.128	0.85
	20	21.44 ± 0.322	107.2 ± 1.61	1.46	21.31 ± 0.139	0.72
	30	30.58 ± 0.225	101.9 ± 0.75	0.75	30.64 ± 0.267	0.83
Pork	10	10.95 ± 0.214	109.5 ± 2.14	1.74	10.87 ± 0.144	0.76
	20	19.66 ± 0.218	98.3 ± 1.09	1.19	19.52 ± 0.156	0.94
	30	31.47 ± 0.225	104.9 ± 2.75	1.09	30.93 ± 0.193	0.71

---

4 Conclusions

In summary, a novel BODIPY-based fluorescent probe for OTC detection, with high selectivity and sensitivity, was successfully designed and synthesized. The bright green-to-colorless color change of S in the presence of OTC was found to be easily observed by the naked eye or measured by UV/vis spectrometry. Under optimized conditions, the developed method possesses a short analysis time (less than 5 min), outstanding thermal stability, high sensitivity, and high selectivity. Furthermore, S has good membrane permeability and low toxicity for biological imaging applications. Finally, the developed method was successfully applied to the quantitative determination of OTC in real samples with satisfactory accuracy and recoveries. It is expected that this BODIPY-based fluorescent one-shot analytical method could pave a facile route for the development of a low-cost and highly sensitive method in food and environmental applications.

Notes and references

1. J. Ghodsi, A. A. Rafati and Y. Shoja, First report on electrocatalytic oxidation of oxytetracycline by horse radish peroxidase: application in developing a biosensor to oxytetracycline determination, Sens. Actuators, B, 2016, 224, 692–699.
2. C. Kowalski, M. Pomorska, B. Lebkowska and T. Slawik, Determination of oxytetracycline in biological matrix, Acta Pol. Pharm., 2006, 63(5), 409–411.
3. M. L. Loke, J. Tjornelund and B. Halling-Sorensen, Determination of the distribution coefficient (log K_d) of oxytetracycline, tylosin A, olaquindox and metronidazole in manure, Chemosphere, 2002, 48(3), 351–361.
4. I. Chopra and M. Roberts, Tetracycline antibiotics: mode of action, applications, molecular biology, and epidemiology of bacterial resistance, Microbiol. Mol. Biol. Rev., 2001, 65(2), 232–260.
5. M. Boonsaner and D. W. Hawker, Transfer of oxytetracycline from swine manure to three different aquatic plants: implications for human exposure, Chemosphere, 2015, 122, 176–182.
6. A. Pena, N. Pelantova, C. M. Lino, M. I. N. Silveira and P. Solich, Validation of an analytical methodology for determination of oxytetracycline and tetracycline residues in honey by HPLC with fluorescence detection, J. Agric. Food Chem., 2005, 53(10), 3784–3788.
7. S. J. Jiao, S. R. Meng, D. Q. Yin, L. H. Wang and L. Y. Chen, Aqueous oxytetracycline degradation and the toxicity change of degradation compounds in photoirradiation process, J. Environ. Sci., 2008, 20(7), 806–813.
8. J. L. Martinez Vidal, M. del Mar Aguilera-Luiz, R. Romero-Gonzalez and A. Garrido Frenich, Multiclass Analysis of Antibiotic Residues in Honey by Ultraperformance Liquid Chromatography-Tandem Mass Spectrometry, J. Agric. Food Chem., 2009, 57(5), 1760–1767.
9. R. Khosrokhavar, M.-J. Hosseini, S. Shoeibi and B. Jannat, Detection of Oxytetracycline Residue in Infant Formula by High-Performance Liquid Chromatography (HPLC), Iran. J. Pharm. Res., 2011, 10(2), 221–224.
10. L. Wang, H. Yang, C. Zhang, Y. Mo and X. Lu, Determination of oxytetracycline, tetracycline and chloramphenicol antibiotics in animal feeds using subcritical water extraction and high performance liquid chromatography, Anal. Chim. Acta, 2008, 619(1), 54–58.
11. M. Jelikicstankov, D. Veselinovic, D. Malesev and Z. Radovic, Spectrophotometric Determination of Oxytetracycline in Pharmaceutical Preparations Using Sodium Molybdate as Analytical Reagent, J. Pharm. Biomed. Anal., 1989, 7(12), 1565–1570.
12. P. P. Maia, J. Amaya-Farfan, S. Rath and F. G. Reyes Reyes, Simultaneous determination of streptomycin and oxytetracycline in agricultural antimicrobials by CZE after an experimental design, J. Pharm. Biomed. Anal., 2007, 43(2), 450–456.
13. X. X. Sun, X. Zhang and H. Y. Aboul-Enein, Construction and characterization of potentiometric sensor for the determination of oxytetracycline hydrochloride, Farmaco, 2004, 59(4), 307–314.
14. Y. S. Kim, J. H. Kim, I. A. Kim, S. J. Lee, J. Jurng and M. B. Gu, A novel colorimetric aptasensor using gold nanoparticle for a highly sensitive and specific detection of oxytetracycline, Biosens. Bioelectron., 2010, 26(4), 1644–1649.
15. L. H. Hansen, B. Ferrari, A. H. Sorensen, D. Veal and S. J. Sorensen, Detection of oxytetracycline production by Streptomyces rimosus in soil microcosms by combining whole-cell biosensors and flow cytometry, Appl. Environ. Microbiol., 2001, 67(1), 239–244.
16. Y. S. Kim, J. H. Niazi and M. B. Gu, Specific detection of oxytetracycline using DNA aptamer-immobilized interdigitated array electrode chip, Anal. Chim. Acta, 2009, 634(2), 250–254.
17. D. Vega, L. Aguei, A. Gonzalez-Cortes, P. Yanez-Sedeno and J. M. Pingarron, Voltammetry and amperometric detection of tetracyclines at multi-wall carbon nanotube modified electrodes, Anal. Bioanal. Chem., 2007, 389(3), 951–958.
18. N. Link, W. Weber and M. Fussenegger, A novel generic dipstick-based technology for rapid and precise detection of tetracycline, streptogramin and macrolide antibiotics in food samples, J. Biotechnol., 2007, 128(3), 668–680.
19. C. Gao, Z. Liu, J. Q. Chen and Z. Y. Yan, A novel fluorescent assay for oxytetracycline hydrochloride based on fluorescence quenching of water-soluble CdTe nanocrystals, Luminescence, 2013, 28(3), 378–383.
20. F. Yuan, H. M. Zhao, Z. N. Zhang, L. C. Gao, J. T. Xu and X. Quan, Fluorescent biosensor for sensitive analysis of oxytetracycline based on an indirectly labelled long-chain aptamer, RSC Adv., 2015, 5(72), 58895–58901.
21. S. H. Xu, X. L. Li, Y. N. Mao, T. Gao, X. Y. Feng and X. L. Luo, Novel dual ligand co-functionalized fluorescent gold nanoclusters as a versatile probe for sensitive analysis of Hg²⁺ and oxytetracycline, Anal. Bioanal. Chem., 2016, 408(11), 2955–2962.
22. A. Treibs and F.-H. Kreuzer, Difluorboryl-Komplexe von Di- und Tripyrrylmethenen, Justus Liebigs Ann. Chem., 1968, 718(1), 208–223.
23. A. B. Nepomnyashchii, A. J. Pistner, A. J. Bard and J. Rosenthal, Synthesis, Photophysics, Electrochemistry and Electrogenerated Chemiluminescence of PEG-Modified BODIPY Dyes in Organic and Aqueous Solutions, J. Phys. Chem. C, 2013, 117(11), 5599–5609.
24. J. Isaad and A. El Achari, A water soluble fluorescent BODIPY dye with azathia-crown ether functionality for mercury chemosensing in environmental media, Analyst, 2013, 138(13), 3809–3819.
25. A. C. Benniston, G. Copley, A. Harriman, D. Howgego, R. W. Harrington and W. Clegg, Cofacial Boron Dipyrromethene (Bodipy) Dimers: Synthesis, Charge Delocalization, and Exciton Coupling, J. Org. Chem., 2010, 75(6), 2018–2027.
26. A. B. Nepomnyashchii, S. Cho, P. J. Rossky and A. J. Bard, Dependence of Electrochemical and Electrogenerated Chemiluminescence Properties on the Structure of BODIPY Dyes. Unusually Large Separation between Sequential Electron Transfers, J. Am. Chem. Soc., 2010, 132(49), 17550–17559.
27. A. Harriman, G. Izzet and R. Ziessel, Rapid energy transfer in cascade-type bodipy dyes, J. Am. Chem. Soc., 2006, 128(33), 10868–10875.
28. L. Wang, L. Li and D. Cao, Dual binding site assisted chromogenic and fluorogenic discrimination of fluoride and cyanide by boryl functionalized BODIPY, Sens. Actuators, B, 2016, 228, 347–359.
29. A. Loudet and K. Burgess, BODIPY dyes and their derivatives: syntheses and spectroscopic properties, Chem. Rev., 2007, 107(11), 4891–4932.
30. S. L. Niu, G. Ulrich, R. Ziessel, A. Kiss, P.-Y. Renard and A. Romieu, Water-Soluble BODIPY Derivatives, Org. Lett., 2009, 11(10), 2049–2052.
31. K. J. Yeum, R. M. Russell, N. I. Krinsky and G. Aldini, Biomarkers of antioxidant capacity in the hydrophilic and lipophilic compartments of human plasma, Arch. Biochem. Biophys., 2004, 430(1), 97–103.
32. S. Zhu, J. Zhang and J. Janjanam, et al. Highly water-soluble BODIPY-based fluorescent probes for sensitive fluorescent sensing of zinc(II), J. Mater. Chem. B, 2013, 1(12), 1722–1728.
33. T. Komatsu, Y. Urano and Y. Fujikawa, et al. Development of 2,6-carboxy-substituted boron dipyrromethene (BODIPY) as a novel scaffold of ratiometric fluorescent probes for live cell imaging, Chem. Commun., 2009, 45, 7015–7017.
34. G. Fan, L. Yang and Z. Chen, Water-soluble BODIPY and aza-BODIPY dyes: synthetic progress and applications, Front. Chem. Sci. Eng., 2014, 8(4), 405–417.
35. G. Wu, F. Zeng and S. Wu, A water-soluble and specific BODIPY-based fluorescent probe for hypochlorite detection and cell imaging, Anal. Methods, 2013, 5(20), 5589–5596.
36. S. Atilgan, T. Ozdemir and E. U. Akkaya, A sensitive and selective ratiometric near IR fluorescent probe for zinc ions based on the distyryl-bodipy fluorophore, Org. Lett., 2008, 10(18), 4065–4067.
37. J. Zhang, F. Peng, X. Dong and W. Zhao, Water-soluble BODIPY Derivative as a Highly Selective “Turn-on” Fluorescent Probe for Hydrogen Sulfide in Living Cells, Chem. Lett., 2015, 44(11), 1524–1526.
38. F. Song, Y. Xue, X. Wang, J. Wang, X. Xiong and X. Peng, Ratiometric Fluorescent Probe Based on Novel Red-emission BODIPY for Determination of Bovine Serum Albumin, Chem. Res. Chin. Univ., 2014, 30(5), 738–742.
39. L. Jiao, M. Liu, M. Zhang, C. Yu, Z. Wang and E. Hao, Visual and Colorimetric Detection of Cyanide Anion Based on a “Turn-off” Daylight Fluorescent Molecule, Chem. Lett., 2011, 40(6), 623–625.
40. Z. G. Li, E. Mintzer and R. Bittman, First synthesis of free cholesterol–BODIPY conjugates, J. Org. Chem., 2006, 71(4), 1718–1721.
41. D. W. Domaille, L. Zeng and C. J. Chang, Visualizing Ascorbate-Triggered Release of Labile Copper Within Living Cells Using a Ratiometric Fluorescent Sensor, J. Am. Chem. Soc., 2010, 132(4), 1194–1195.
42. E. Wang, L. Pang and Y. Zhou, et al. A high performance Schiff-base fluorescent probe for monitoring Au(3+) in zebrafish based on BODIPY, Biosens. Bioelectron., 2016, 77, 812–817.
43. L. Li, J. Han, B. Nguyen and K. Burgess, Syntheses and spectral properties of functionalized, water-soluble BODIPY derivatives, J. Org. Chem., 2008, 73(5), 1963–1970.
44. S. Orellana, C. Soto and M. Ines Toral, UV-vis, IR and ¹H NMR spectroscopic studies and characterization of ionic-pair crystal violet-oxytetracycline, Spectrochim. Acta, Part A, 2010, 75(1), 437–443.

---