A near-infrared fluorescent probe for selective detection of fluorion

Abstract In this work, we have designed and synthesized the fluorescent probe 1, which was capable to selectively detect fluoride anion (F−). More importantly, the probe 1 possessed near-infrared excitation and emission wavelengths (excitation at 650 nm and emission at 695), and the probe solution had changed dramatically from yellow to cyan with the addition of F–. In addition, the fluorescence intensity exhibited perfectly positive correlation with concentration of F− concentration from 0 to 40 μM (R2 = 0.9972), which offered the important condition for quantitative analysis. The probe 1 owned detection limit of 46 nM. Therefore, this near-infrared probe can be of great benefit for detecting F− in practical application.


INTRODUCTION
Fluoride ion (F -) is the smallest anion, which is indispensable trace element in the human body and plays critical roles in a variety of biological processes 1, 2 . Low levels of Fintake are closely connected with risk of decayed tooth and osteoporosis 3 . However, excessive Fingestion easily gave rise to many diseases that are, skeletal fl uorosis, kidney failure, or urolithiasis, even cancer 4 . In addition, Fcontamination in soil and water affect plant growth and development. Fis also abundant in many industrial workplaces, which can be used as oxidizer in rocket fuel burn, in nuclear fuel, in computer chips and microelectronic sensors, and television screens 5 . Therefore, the accurate determination of fl uoride is of growing importance in both environmental and biological systems. The conventional assay methods of fl uoride commonly include ion-selective electrode, colorimetry, capillary electrophoresis and so on 6, 7 . Among them, fl uorescence detection techniques have attracted much attention based on operational simplicity, high sensitivity, and cell bioimaging 8 .
Up to now, although a lot of fl uorescent probes have been developed to detect F -9-13 , they still have some limitations more or less in practical application. Poor selectivity and sensitivity is the common problem for some probes. Moreover, most probes have short-wavelength excitation and emission spectra, which increase diffi culty for detecting Fin the biological systems because of poor anti-interference ability. Meanwhile, light in the short-wavelength region has limited tissue penetration and it is easily scattered and absorbed by biomolecules. Therefore, in order to improve this drawback, the fl uorescent probe should be constructed with excitation and emission wavelengths in the near-infrared.

Materials and apparatus
1 H/ 13 C NMR spectra were recorded on a Bruker AVIII-400/600 MHz spectrometer. High resolution mass spectra (HRMS) were measured with Thermo (orbitrap Elite). Absorption spectra were measured using a Thermo (BioMate 3S) UV/Vis spectrophotometer. Fluorescence measurements were carried out with a F97pro fl uorospectrophotometer. Unless otherwise noted, materials were purchased from commercial suppliers and used without further purifi cation. All the solvents were purifi ed and dried according to general methods.

Bioimaging of probe 1 in MCF-7 cells
MCF-7 cells (5×10 4 /mL) were seeded in 6-well fl at microtiter plates for adherence for 24 h. The control group. Cells were incubated with probe 1 (20 μM) for 30 min, and then washed with PBS for 3 times. The experiment group. The cells were pre-incubated with probe 1 (20 μM) for 30 min and subsequently incubated by the appointed concentration of Ffor another 30 min, and then washed with PBS for 3 times. Fluorescence images of probe 1 were got by a fl uorescence microscope.

RESULTS AND DISSCUSSION
We synthesized the probe 1 and the synthetic route is shown in Scheme 1. First, we synthesized the (2-(3,5,5-trimethylcyclohex-2-en-1-ylidene)malononitrile) (Compound 1) according to relevant references 15 . Subsequently, compound 1 reacted with p-hydroxy benzaldehyde to give the compound 2 by Aldol reaction. Lastly, the compo-und 2 was further reacted with diphenyl tert-butylchlorosilane to obtain the probe 1 under basic condition in 3% all yield and the probe structure was confi rmed by NMR and HRMS (ESI).
As the probe 1 was obtained, we next investigate its optical properties. Firstly, the absorption b ands of probe 1 and probe 1 toward Fwere surveyed in the solution (THF/DMSO = 1 : 1, v/v). As shown in Fig. 1, probe 1 showed an absorption band at 425 nm. Upon the addition of F -, the peak of absorption band red shifted to 675 nm and the new absorption peak increased gradually with increasement of reaction time from 1 to 5 min at room temperature. In order to avoid the infl uence of excitation light for emission spectrum, we choose the 650 nm as excitation wavelength. More interestingly, the solution of probe 1 changed from yellow to cyan with the addition of F -. Hence, this result showed that probe 1 can identify and detect Fby ''naked-eye'' colorimetric change, which provided convenience for practical application.   (Fig. 2). The fl uorescence intensity increased markedly at 695 nm, when the probe 1 (20 μM) was incubated by F − for 2 min in solution (THF/DMSO = 1 : 1, v/v) at room temperature with excitation wavelength at 650 nm. As we all known, the excitation and emission wavelength was longer than that of most pre-To study the mechanism, we test the absorption of compound 2 in the same condition (THF/DMSO = 1:1, v/v) (Fig. 5). The result displayed compound 2 have absorption peak at 437 nm. However, with addition of TBAF, the absorption peak of compound 2 was red shifted to 675 nm, which was consistent with the absorption peak of probe 1 toward F -. Therefore, we proposed possible mechanism in the Fig. 6. The probe with tert-butyldiphenylsilyl (TBDPS) moiety identifi es the fl uoride ion by Si-O band breakage, which had been proved in many previous reports 9- 13 . The compound 2 can be obtained with reaction of probe 1 and F -. Based on the basic conditions provided by TBAF, the compound 2 further loses the hydrogen proton to lead absorption red shift and the turn on the fl uorescence 16 .
To explore the pra ctical application, we inspected the bioimaging of probe 1 in the living cells. Firstly, we measured the cytotoxicity of the probe in MCF-7 cells by MTT method (Fig. 7). The results displayed the survival rate did not signifi cantly change with the addition of 80 μM probe. Therefore, the probe can be used in the biosystem. Subsequently, we executed the cell imaging experiment in MCF-7 cells (Fig. 8). The results showed the blue fl uorescence markedly appeared when the cells was pre-incubated with 30 min and then continued to incubate with F − for 30 min. Moreover, the fl uorescence intensity was increased clearly along with increase of TBAF concentration from 100 to 500 μM.
vious reports 11-13 . More interestingly, other anions did not cause signifi cant increase of fl uorescence intensity, even change induced by hydroxyl ion also was negligible. Moreover, the relationship between fl uorescence intensity and concentration was quantitatively analyzed by titration (Fig. 3a). Results exhibited the satisfactory positive correlation between fl uorescence intensity and concentration of F − from 0 to 40 μM (R 2 = 0.9972) (Fig. 3b), which is an advantage condition for quantitative analysis of F − .
Meanwhile, w e further finished the competition experiment to survey interference of coexisting ions for detecting F − (Fig. 4a). The results exhibited the fl uorescence intensity change of probe 1 toward F − is insignifi cant in the presence of other anions. Therefore, these results displayed probe 1 can selectively detect F − . On the other hand, we investigated the fl uorescence intensity change of probe toward F − with incubation time increase (Fig. 4b). The reaction was completed after the probe was incubated by F − for 2 min and the fl uorescence intensity hardly decayed within 20 minutes. Furthermore, the probe itself did not cause a change in fl uorescence intensity with the reaction time extension. Thus, these results provided the condition for practical application. Moreover, the detection limit of the probe was determined (3σ/k) to be 46 nM.

CONCLUSIONS
In summary, we have developed the fl uorescent probe 1, which exhibited a highly selective and sensitive response to F − in the test. When the F − was added, the fl uorescence intensity enhanced obviously with the color change from yellow to cyan by ''naked-eye'' within 2 min. Therefore, the probe 1 could selectively detect F − . Furthermore, the probe 1 possessed near-infrared excitation and emission wavelengths (excitation at 650 nm and emission at 695), which offered the favorable condition for anti-interference and penetration. In addition, the fl uorescence intensity exhibited perfectly positive correlation with concentration of F − concentration from 0 to 40 μM (R 2 = 0.9972). Therefore, we believe this near-infrared probe can be of great benefi t for detecting F − in practical application.