Near-Infrared Turn-On Fluorescent Probe for Aqueous Fluoride Ion Detection and Cell Imaging.

Fluoride ions are one of the most essential anions in the human body and have been implicated in various pathological and physiological processes. The detection of fluoride ions in aqueous solution, as well as the imaging of fluoride ions in living cells, remains a challenge. We herein report a BODIPY-based fluorescent probe employing a pinacol borate group as the recognition moiety for the detection of fluoride ions in aqueous solutions. This probe shows high selectivity and sensitivity to fluoride ions with a significant near-infrared fluorescence turn-on response. In addition, this probe was successfully employed in fluorescence bioimaging of fluoride ions in the human cervical cancer cell and mouse mammary cancer cell, demonstrating its good cell permeability and stability under physiological conditions.

Introduction

Fluoride ion (F–), the smallest anion,[1−3] is one of the essential trace elements which plays various roles in biochemical processes. Appropriate intake of fluoride is contributed to preventing dental cavities and osteoporosis.[4,5] However, excess fluoride ingestion can cause detrimental health effects and can calcify bone and teeth, resulting in various symptoms and diseases such as dental fluorosis, skeletal fluorosis, urolithiasis, and kidney failure.[6−10] Several analytical techniques, including ion chromatography, ion-selective electrodes, ultraviolet–visible spectroscopy, mass spectroscopy, and 19F NMR, have been demonstrated for quantitative fluoride determination.[11−13] However, most of these methods have limitations, such as high experimental costs, tediousness, and time-consuming procedures, as well as inaccessibility for studying biological processes. Therefore, the development of a simple and highly sensitive method for F– detection is still urgently needed. Compared with other analytical methods, fluorescence detection technology has attracted great attention as it has been proved to have simple techniques, cost-efficient experiments, high sensitivities, and selectivity and can also be applied to live cell imaging of anions.[14−16] Recently, several organic molecules have been reported as fluorescent probes for the detection of F–.[17−22] These probes are based on the hydrogen bond mechanism where the O–H or N–H group is protonated or bonded under the action of fluorine, resulting in the change of molecular spectrum properties. However, since F– is the most electronegative anion, it can also readily form hydrogen bonds with water molecules, which severely limits the detection of F– in aqueous systems by hydrogen-bonded fluorescent probes. To overcome the issue of detecting F– in aqueous solutions, a series of probes based on fluorine–boron complexation have been developed during the last few years.[23−28] The mechanism of such probes is based on the Lewis acid–base interactions. Since the boron atom has an empty p orbital, the electron-deficient trivalent organic boron readily binds to F–, thus breaking the p-π conjugation of the boron center to the aromatic group, leading to a change in the photophysical properties of the probe molecule and thus achieving selective detection of F–. The molecular probe based on the second mechanism is not affected by the aqueous solution, which can be used to detect F– in vivo and in cells with high sensitivity and selectivity.

Near-infrared (NIR) fluorescent probes have great advantages for in vivo imaging because of their low background interference, low light source energy, high tissue penetration depth, low tissue damage, and high image sensitivity.[29,30] A series of fluorescent probes have been reported for detecting F– in the last decade (Table S1), but fewer probes can be used as a sensitive NIR fluorescent turn-on sensor to detect F–. In recent years, considerable progress has been reported for NIR fluorescent F– probes based on rhodamine,[31] coumarin,[32] and hemicyanine structures;[33,34] however, achieving high sensitivity and selective F– detection in aqueous solutions and living cells remains a challenge.[16,35−37] In this work, we have designed and synthesized a novel 4,4-difluoro-4-bora-3a,4a-diaza-s-indacene (BODIPY) derivative bearing a pinacol boronate group for imaging F– in aqueous solutions (Scheme ). Microwave technology and palladium-catalyzed boronization borylation reactions were used in the synthesis of this probe molecule, which provides the advantages of both ease of operation and high product yield. The pinacol boronate group will bond to F– and lead to borate anion, which results in a change in the absorption, emission properties, and fluorescence of the probe. Further experiments disclosed that it can be used for the real-time detection in vivo bioimaging of exogenous F–.

Scheme 1

BODIPY-Based Turn-On Fluorescent Probe for the Detection of F–

Results and Discussion

Synthesis of Probe BEB-F

A total of three steps are required for the synthesis of the target probe BEB-F, and the synthetic procedure is described in Scheme . First, 4-bromobenzaldehyde reacted with 3-ethyl-2,4-dimethyl-1H-pyrrole and subsequently complexed with Et2O·BF3 in the presence of triethylamine to produce BODIPY-1. Then, for the second step of Knoevenagel condensation reaction between BODIPY-1 and N,N-diethylaminobenzaldehyde, we used microwave technology to facilitate this reaction, which has the advantage of not having by-products with similar polarity to the probe BEB-F and facilitates the isolation of high purity products. The unreacted raw material can also be separated and then recycled. In the third reaction step, the borylation of BODIPY-2 using Pd(dppf)Cl2 as the catalyst and XPhos as the ligand afforded the probe BEB-F after careful column chromatography. All intermediate compounds and products were characterized by 1H NMR, 13C NMR, and high-resolution mass spectrometry (Figures S1–S9).

Scheme 2

Preparation of Probe BEB-F

Spectroscopic Property of Probe BEB-F and Its Response to F–

The detection of aqueous solution with high sensitivity is a challenge. In this work, after careful screening, 100 μM cetyltrimethylammonium bromide (CTAB) was found to facilitate the detection of F– by BEB-F in aqueous solution (Figure S10). Then, the UV–vis absorption spectra of BEB-F before and after the addition of F– were measured in N-(2-hydroxyethyl)piperazine-N′-ethanesulfonic acid (HEPES) buffer (10 mM, pH = 7.4) with CTAB (100 μM). As shown in Figure a, BEB-F has an intense absorption peak at 630 nm, which slightly blue shifts after the addition of F– to the solution. Next, the optimized excitation wavelength for fluorescence measurements was studied. As seen in Figure S11, without the addition of F–, the fluorescence intensity of the probe BEB-F was essentially unchanged with different excitation wavelengths. At the same time, after the addition of F– for 30 min, the fluorescence intensity of BEB-F gradually increased with the increase in excitation wavelength. To avoid scattered light interference from the excitation source, 600 nm was chosen as the excitation wavelength for the fluorescence measurements in this work. Under 600 nm excitation, an obvious enhancement of emission intensity at 677 nm was also observed after the addition of F– (Figure b). To further investigate the interaction between BEB-F and F–, the fluorescence titration experiment was carried out. As shown in Figure c, as anticipated, the fluorescence of BEB-F (10 μM) showed a large enhancement upon the addition of F– (120 μM). In the range of 0–120 μM, the fluorescence intensity of probe BEB-F increases linearly with the concentration of fluoride ions. As displayed in Figure d, the relationship between the concentration of fluorine ion (x) and the fluorescence intensity (F) was obtained as follows: F = 33.99 + 3.61x (correlation coefficient R2 = 0.9850). The detection limit (LOD) (3σ/slope, σ is the standard deviation of the blank measurement) was calculated to be 231 nM of F–. The results indicated that the probe BEB-F can effectively detect F– in aqueous solution. The photostability of probe BEB-F was studied by using laser irradiation (660 nm, 600 mW·cm–2) as the light source, and the results indicate that the probe BEB-F exhibits good photostability (Figures S12 and S13).

Figure 1

Absorption spectra (a) and fluorescence spectra (b) of BEB-F (10 μM) in pH 7.4 HEPES buffer (10 mM) with CTAB (100 μM) before and after reaction with F– (100 μM) for 0.5 h. (c) Fluorescence response of the probe (10 μM) toward F– at varied concentrations (0–120 μM) in pH 7.4 HEPES buffer (10 mM) with CTAB (100 μM). (d) Linear fitting curve of F against the concentration of F–. λex/em = 600/677 nm.

Time-Dependence of BEB-F Response to F–

The response time is an important parameter for evaluating the effectiveness of a probe. The response time of BEB-F to F– was studied by a kinetic experiment (Figure ). The result showed that the fluorescence intensity increased to its maximum in 10 min after adding F– to the probe solution, which indicated that the probe instantly responds to F–. In addition, the fluorescence intensity of the solution was stable for at least an hour after the reaction. Therefore, to obtain more accurate test results, we set the reaction time at 0.5 h in subsequent experiments.

Figure 2

Reaction time on the fluorescence of BEB-F (10 μM) with varied concentrations (0 and 100 μM) of F–. The fluorescence was measured in the pH 7.4 HEPES buffer (10 mM) with CTAB (100 μM). λex/em = 600/677 nm.

pH Effect of BEB-F Response to F–

The pH value is a crucial factor for evaluating the feasibility of fluorescent sensors for biological applications. We next examined the pH dependencies of the response of BEB-F to F–. The results are shown in Figure , the probe BEB-F is stable, and the emission is constant in a very wide range of pH 3.0–9.0. After the reaction of probe BEB-F and F– for 0.5 h, the fluorescence intensity increased, and it was stable in the pH range of 3.0–9.0. The results suggested that the probe BEB-F can be used for the biological detection of F–.

Figure 3

Effects of pH on the fluorescence of probe BEB-F (10 μM) with varied concentrations (0 and 100 μM) of F–. The fluorescence was measured in the pH 7.4 HEPES buffer (10 mM) with CTAB (100 μM). λex/em = 600/677 nm.

Selectivity of Probe BEB-F for F–

To investigate the selectivity of the BEB-F for F–, various anions such as Cl–, Br–, I–, CH3COO–, ClO4–, HSO4–, H2PO4–, NO3–, HPO42–, OH–, SO42–, ONOO–, ClO–, and H2O2 were introduced into the system, the absorption (Figure S14) and fluorescence intensity (Figure a) of BEB-F did not change as compared with the blank signal. Furthermore, various cations (K+, Mg2+, Ca2+, Zn2+, Na+, Cu2+, Ni2+, Co2+, Al3+, and Fe3+) were also tested, which also did not cause significant changes in the absorption (Figure S15) and fluorescence of BEB-F (Figure b). Furthermore, the response experiments of the probe BEB-F in the presence of F– in coexistence with other common ions were also carried out (Figures S16 and S17). The pinacol boronate group is widely reported as a recognition group for some ROS, such as ONOO–, ClO–, and H2O2. However, while examining the selectivity of probe BEB-F for F–, no fluorescence response was found with the addition of these ROS for 30 min (Figure S18). However, after 1 h, the fluorescence intensity slightly decreased (Figure S18). We presume that the ROS may react with BEB-F and result in a non-fluorescent product. Anyway, such fluorescence turn-off responses are not significant and are quite slow, resulting in limited interference for the detection of F–. All these results indicate that BEB-F was a highly selective probe for detecting F–.

Figure 4

Fluorescence responses of 10 μM BEB-F toward various anions and cations. All experiments were performed in the pH 7.4 HEPES buffer (10 mM) with CTAB (100 μM). (a): (0) probe only; (1) 100 μM F–; (2) 100 μM Cl–; (3) 100 μM Br–; (4) 100 μM I–; (5) 100 μM CH3COO–; (6) 100 μM ClO4–; (7) 100 μM HSO4–; (8) 100 μM H2PO4–; (9) 100 μM NO3–; (10) 100 μM HPO42–; (11) 100 μM OH–; (12) 100 μM SO42–; (13) 100 μM H2O2; (14) 100 μM ONOO–; and (15) 100 μM ClO–. (b) (0) probe only; (1) 100 μM F–; (2) 150 mM K+; (3) 2.5 mM Mg2+; (4) 2.5 mM Ca2+; (5) 200 μM Zn2+; (6) 100 mM Na+; (7) 100 μM Cu2+; (8) 100 μM Ni2+; (9) 100 μM Co2+; (10) 100 μM Al3+; and (11) 100 μM Fe3+. λex/em = 600/677 nm.

Cytotoxicity Assays

After confirming the good sensitivity and selectivity of BEB-F in detecting F– in vitro, we went on to explore whether BEB-F could be useful in living cells. The cytotoxicity of BEB-F was first studied using the MTT assay. We incubated two experimental cells (HeLa cells and 4T1 cells) with different concentrations of BEB-F (0–50 μM) and then stained the cells with MTT to investigate the viability of the cells treated with BEB-F. Cytotoxicity studies in Figure illustrated that cell viability was consistently above 85% even when the concentration of BEB-F was increased to 5 × 10–5 M. These results indicate that BEB-F has low cytotoxicity and good biocompatibility.

Figure 5

(a) Cell viability of HeLa treated with BEB-F. The viability of the cell without the probe is defined as 100%. The results are presented as mean ± standard deviation (n = 5). (b) Cell viability of 4T1 treated with the probe. The viability of the cell without the probe is defined as 100%. The results are presented as mean ± standard deviation (n = 5).

Imaging of Fluoride Ions in Live Cells

BEB-F was further used for imaging exogenous F– in cells to explore its potential applications in studying bioimaging application. We examined the cell imaging ability of BEB-F in the presence of KF in HeLa cells and 4T1 cells using an inverted fluorescence imaging microscope. The results are shown in Figure ; the BEB-F showed only a weak red fluorescence signal in the HeLa cells and 4T1 cells under the excitation of a green channel source. However, the fluorescence in the cells was significantly enhanced after being treated with KF in the medium, and the fluorescence rose with the increasing amount of F–. These results illustrated that the BEB-F had excellent living cell imaging ability. As we learned from the literature, F– above 1 mM results in osteoblast cytotoxicity, particularly to the nucleus and endoplasmic reticulum.[38] Neuronal cell lines exposed to ≥3 mM NaF undergo DNA damage, oxidative stress, mitochondrial agglutination, and cytoskeleton damage.[39,40] Owing to the high sensitivity and selectivity of the probe BEB-F, it remains potential for the application of BEB-F for F– detection under some pathological conditions, which is our future work to be continued.

Figure 6

Fluorescence images of BEB-F in HeLa cells (a) and 4T1 cells (b) incubated with different concentrations of F–. Fluorescence images of cells from the green channel (λex = 460–550 nm). Scale bar: 20 μm.

Detection of Fluoride Ions in Water Samples

BEB-F was also employed to determine F– content in tap water and lake water samples. As shown in Table S2, BEB-F can detect F– in the collected samples spiked with 80 and 100 μm F– concentrations with the recovery close to 100%. In addition, the fact that the compositions of tap water and lake water do not significantly interfere with F– suggested the potential of utilizing this compound in water quality monitoring applications.

Reaction Mechanism of BEB-F with Fluoride Ions

The mechanism of BEB-F with F– was proposed in Figure . The boron atom of the boronic ester group has an sp2 triangular planar geometry with an empty P orbital; after the addition of F–, the boron center of the boronic ester is converted to sp3 hybridization in the fluoroborate, thus leading to a significant perturbation of the p system and resulting in a response with a blueshift in the emission wavelength from 720 to 677 nm and an increase in fluorescence intensity.[41,42] We monitored the changes in 11B NMR spectra produced via the addition of F– to BEB-F solution. As shown in Figure , the boronate ester group exhibits a singlet at δ 21.65 ppm, and this singlet changes to 4.37 ppm, which was assigned to the boronic ester converted to sp3 hybridization in the fluoroborate. Also, a new single peak at −123.87 ppm was observed by 19F NMR after the addition of F–, which is attributed to the new B–F bonding (Figure S19).

Figure 7

Proposed reaction mechanism of BEB-F with F–.

Figure 8

11B NMR spectra of BEB-F in the presence of fluoride anion.

Conclusions

In summary, we have developed a BODIPY-based fluorescent probe, BEB-F, for the detection of fluoride ions in aqueous media. The probe synthesis was performed using microwave technology and a palladium-catalyzed efficient borylation reaction, which improved the yield of the product and facilitated the operation. Significantly, this probe BEB-F employing a pinacol borate group as the recognition moiety was capable of detecting fluoride ions in the aqueous phase, accompanied by a NIR fluorescence (at 677 nm) turn-on process. The probe BEB-F exhibited high selectivity and sensitivity for fluoride ions over other reactive cations and anions and has been demonstrated to have a linear response to fluoride ions with an LOD of 0.231 μM. Furthermore, this probe can be successfully utilized for detecting fluoride ions in real water samples and imaging those in HeLa cells and 4T1 cells. Therefore, this work provides a promising NIR probe for the rapid detection of fluoride ions found in environmental and biological samples.

Materials and Methods

Materials and Instruments

Unless otherwise noted, all reagents and materials were commercially available and used without further purification. All aqueous solutions were prepared with deionized water, which was purified using a Nex Up 1000 (Human, South Korea). The human cervical cancer cells (HeLa cells) and mouse mammary cancer cells (4T1 cells) were obtained from the Key Laboratory of Basic Pharmacology, Zunyi Medical University. 1H NMR and 13C spectra were obtained on the Agilent 400 MHz-DD2 (Agilent, USA). High-resolution mass spectrometry (HR-MS) was obtained on the Agilent QTOF 6550 mass spectrometer. The UV–vis absorption spectra were obtained on a TU-1901 spectrometer (Persee, Beijing, China). The fluorescence spectrum measurements were performed on a Vary Eclipse spectrophotometer (Varian, USA) with the excitation and emission slit widths at 5 and 10 nm, respectively. The incubation was performed in the 3131 CO2 incubator (Thermo, USA). Cytotoxicity was tested on WD-2102B (Liu Yi, Beijing, China). Fluorescence imaging experiments were performed on an Olympus fluorescent inverted microscope (IX73+DP73, Japan).

Synthesis and Characterization of BEB-F

The synthetic process for fluorescent BEB-F is displayed in Scheme .

Synthesis of BODIPY-1

Three drops of trifluoroacetic acid were added to a stirred solution of 4-bromobenzaldehyde (1.00 g, 5.40 mmol) and 3-ethyl-2,4-dimethyl-1H-pyrrole (1.82 mL, 13.51 mmol) in anhydrous CH2Cl2 (100 mL). The reaction mixture was stirred at r.t. under N2 in a darkened flask overnight. Then, 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (1.23 g, 5.40 mmol) was added in a single portion, and the mixture was stirred at r.t. for 6 h. Anhydrous triethylamine (4.51 mL, 32.43 mmol) was added to the mixture and stirred at r.t. for 2 h. After that, Et2O·BF3 (6.67 mL, 54.05 mmol) was added, and the mixture was stirred at r.t. overnight. Next, the reaction mixture was washed with brine (3 × 100 mL). The separated organic fractions were dried (Na2SO4) and filtered, and then the solvent was removed to yield a black/dark-violet residue with a green tint. The crude product was purified via chromatography over silica gel with a 20/1 (v/v) petroleum ether/ethyl acetate to provide a red powder (744.00 mg, yield 30%). 1H NMR (400 MHz, CDCl3): δ 7.63 (d, J = 8.2 Hz, 1H), 7.18 (d, J = 8.2 Hz, 1H), 2.53 (s, 3H), 2.30 (q, J = 7.5 Hz, 2H), 1.31 (s, 3H), 0.98 (t, J = 7.6 Hz, 3H). 13C NMR (101 MHz, CDCl3): δ 154.10, 138.11, 134.71, 132.98, 132.30, 130.47, 130.08, 122.97, 17.05, 14.63, 12.53, 11.95. HRMS (ESI-TOF, m/z): for C23H26BBrF2N2 [M + H]+ calcd, 459.1419; found, 459.1439.

Synthesis of BODIPY-2

BODIPY-1 (700.00 mg, 1.52 mmol), p-N,N-diethylaminobenzaldehyde (324.24 mg, 1.83 mmol), piperidine (0.76 mL), acetic acid (0.76 mL), and 20 mL of dry toluene were added in a 50 mL glass tube sealed with a Teflon cap. The sample was irradiated at 200 W and 190 °C for 20 min in a CEM-Discover monomode microwave apparatus. After completion of the reaction, the mixture was directly poured into a silica gel column and eluted with a 15/1 (v/v) dichloromethane/petroleum ether to provide a blue powder (102.19 mg, yield 10%). 1H NMR (400 MHz, CDCl3): 7.63 (d, J = 8.2 Hz, 2H), 7.58–7.47 (m, 3H), 7.24–7.17 (m, 3H), 6.66 (d, J = 8.8 Hz, 2H), 3.40 (q, J = 6.9 Hz, 4H), 2.65–2.51 (m, 5H), 2.31 (q, J = 7.5 Hz, 2H), 1.37–1.24 (m, 6H), 1.22 – 1.13 (m, 9H), 0.99 (t, J = 7.5 Hz, 3H). 13C NMR (101 MHz, CDCl3): δ 152.95, 148.21, 138.58, 136.76, 135.19, 132.81, 132.21, 130.47, 129.09, 124.56, 122.85, 119.06, 114.74, 111.43, 44.49, 31.43, 30.15, 18.45, 17.12, 14.69, 14.00, 12.64, 11.93, 11.76. HRMS (ESI-TOF, m/z): for C34H39BBrF2N3 [M + H]+ calcd, 618.2467; found, 618.2479.

Synthesis of BEB-F

BODIPY-2 (100.00 mg, 0.16 mmol), bis(pinacolato)diboron (61.59 mg, 0.24 mmol), sodium acetate anhydrous (53.06 mg, 0.24 mmol), and Pd(dppf)Cl2 (11.83 mg, 0.02 mmol) were dissolved in toluene (5 mL) and stirred for 7 h at 110 °C. After the reaction mixture was cooled to room temperature, toluene was evaporated under reduced pressure. The residue was diluted with dichloromethane (20 mL), washed with saturated NaHCO3 (10 mL × 3), and dried by Na2SO4; then, the solvent was removed and purified by chromatography over silica gel (petroleum ether/ethyl acetate = 10/1) to provide a blue powder (83.94 mg, yield 78%). 1H NMR (400 MHz, CDCl3): δ 7.89 (d, J = 7.6 Hz, 2H), 7.48 (d, J = 8.6 Hz, 3H), 7.31 (d, J = 7.5 Hz, 2H), 7.18 (d, J = 16.6 Hz, 1H), 6.66 (d, J = 8.4 Hz, 2H), 3.39 (d, J = 9.2 Hz, 4H), 2.58 (t, J = 8.1 Hz, 5H), 2.30 (q, J = 7.6 Hz, 2H), 1.40 (s, 12H), 1.27–1.25 (m, 6H), 1.19 (t, J = 7.1 Hz, 6H), 1.14 (t, J = 7.4 Hz, 3H), 0.98 (t, J = 7.5 Hz, 3H). 13C NMR (101 MHz, CDCl3): δ 148.11, 139.11, 136.34, 135.16, 128.99, 127.99, 124.68, 114.92, 111.45, 109.98, 103.62, 84.08, 44.47, 30.14, 29.69, 24.94, 24.82, 24.54, 18.42, 17.11, 14.66, 13.99, 12.63, 11.80, 11.61. HRMS (EI-TOF, m/z): for C40H51B2F2N3O2 [M + H]+ calcd 666.4214; found, 666.4210.

General Procedures for Spectral Measurements

The stock solutions (3 mM) of tetrabutylammonium fluoride, KHSO4, KH2PO4, NaClO, NaNO3, Na2HPO4, NaOH, Na2SO4, NaCl, KBr, KI, CH3COONa, KClO4, H2O2, and ONOOH were prepared in deionized water. The stock solutions (3 mM) of KCl, MgCl2, CaCl2, ZnCl2, NaCl, AlCl3, CuCl2, CoCl2, NiCl2, and FeCl3 were prepared in deionized water. A stock solution of BEB-F (3 mM) was prepared in dimethyl sulfoxide; then, 10 μL of the stock solution was diluted with a mixed solution of CTAB containing HEPES buffer (10 mM, pH = 7.4), and different volume stock solutions (3 mM) of F– were added until the total volume was 3 mL. Then, the final concentration of 10 μM BEB-F with 100 μM CTAB solution was used for F– spectral measurement. The same amount of analyte is added to the probe test solutions for subsequent selective determination. Titration experiments are performed by adding different amounts of F– to the probe solution and measuring optical changes. The reaction conditions were optimized with a certain amount of fluoride ions and probe.

Cytotoxicity Assay

The cytotoxicity of BEB-F was determined using MTT assays. HeLa cells and 4T1 cells were incubated in Dulbecco’s modified Eagle medium (DMEM) supplemented with 10% FBS (fetal bovine serum) in an atmosphere of 5% CO2 and 95% air at 37 °C. The cells were placed in a 96-well plate and incubated for 24 h upon different concentrations of BEB-F of 0.0, 10.0, 20.0, 30.0, 40.0, and 50.0 μM, respectively. Finally, the MTT assay followed. The cytotoxic effect (VR) of BEB-F was assessed using the following equationwhere A and A0 are the absorbance of the experimental group and control group using a microplate reader at 490 nm, respectively.

Living Cell Culture and Imaging

HeLa cells and 4T1 cells were incubated in DMEM supplemented with 10% FBS in an atmosphere of 5% CO2 and 95% air at 37 °C. The cells were seeded in six-well flat-bottomed plates and incubated for 24 h before cell imaging. For the control experiment, the cells were incubated with a 10 μM probe for 0.5 h. Meanwhile, the other groups were incubated with different concentrations of F– for 0.5 h and then treated with the 10 μM probe for another 0.5 h. Before imaging, all cells were washed three times with PBS buffer. All fluorescence images were acquired using an inverted fluorescence imaging microscope, and the fluorescence signals at the green field were collected.