A new intramolecular proton transfer (ESIPT)‐based fluorescent probe for selective visualization of cyanide ion

Fluorescent probes for detection of CN− still have many limitations, such as small Stokes shift, irreversible, and background interference, which hamper their applications for on‐site detection and bioimaging of CN−. In this work, we design a new CN−‐activatable fluorescent probe (named AHMM) containing an ESIPT (excited‐state intramolecular proton transfer) and hydrogen bond features, which show a large Stokes shift (225 nm) and molecular structural reversible detection. The probe AHMM exhibits an excellent selectivity for CN− without any interference from other anions in aqueous DMSO system. Furthermore, the mechanism of the interaction of AHMM with CN− is concluded by various experiments. The limit of detection of AHMM for CN− is calculated as low as 4.47 × 10−8 M, lower than the concentration of CN− deemed acceptable by WHO (World Health Organization). AHMM can recognize CN− in tap water quantitatively and on‐site by a smartphone APP. Moreover, food samples such as almond and cassava including CN− are visualized by fluorescence imaging. In addition, the probe shows practical applications of CN− imaging in cells and mice. This concept can be applied for designing multifunctional fluorescent probes with ESIPT and reversible characteristics for detection of CN−.


INTRODUCTION
Water and food pollution arising from anions continues to be one of the major serious problems of environmental concern, which has a threat to human health and safety. [1][2][3] Among the anions, cyanide ion (CN − ) has been widely used in the chemical industry and agriculture and existing in common food, such as almonds and cassava. [4][5][6] In addition, CN − can cause serious water pollution through industrial production. And some food like bitter almond and cassava are considered toxic if CN − in the form of cyanogenic glycosides outpaced 200 ppm. [7][8][9] Moreover, CN − can bind to cytochrome oxidase to inhibit the mitochondrial electron transport chain and cause death. [10][11][12] According to the data of the World Health Organization (WHO), the safety level of CN − in drinking water is set below 1.9 μM. [13][14][15] Considering the extreme toxicity of CN − , an effective and reliable method enabling monitoring CN − is necessary. Analytical methods including inductively coupled plasma mass spectrometry (ICP-MS), atomic absorption spectrometry (ABS), and electrochemistry have been developed for the detection of CN − . [16][17][18][19] Compared to these analytical methods, fluorescent probes exhibit many advantages including high sensitivity, fast and easy detection, and excellent biocompatibility. [20][21][22][23] Most recently reported CN − probe are based on the photonic mechanisms, such as PET (photoinduced electron transfer), FRET (fluorescence resonance energy transfer), ICT (internal charge transfer), and ESIPT (excited-state intramolecular proton transfer). [24][25][26][27] Compared to other mechanisms, the probes based on the ESIPT process have attracted wide attention because of their ultra-fast electronic transfer rate and large Stokes shift, which can improve the detection sensitivity in vitro and in vivo by avoiding self-reabsorption effects. 28-30 CN − has good nucleophilicity; thus, many CN − probes are designed by using nucleophilic addition. [31][32][33] However, this mechanism usually presents a "turn-off" effection that is undesirable for bioimaging, and the property of irreversible cycling of the probes greatly restricted their practical applications. 34,35 Thus, an ESIPT-based probe owing large Stokes shift and reversible properties for detecting CN − is highly desirable.
According to the previous work, the compound 2-(2′hydroxyphenyl) benzothiazole (HBT) with ESIPT readily forms hydrogen bond. 36,37 We propose to design a probe that shows ESIPT-based for large Stokes shift and reversible dual-path for detecting CN − (Figure 1). In addition, the introduction of an electron-withdrawing group (diaminoacetonitrile) and an electron-donating group (phenol group) to form a push-pull electron system can effectively red-shift the emission wavelength of the system, which is conducive to bioimaging. Herein, we synthesize a new probe (AHMM) for CN − . The probe AHMM shows four advantageous characteristics: (a) a large Stokes shift in responding to CN − (225 nm); (b) excellent selectivity and anti-interference ability; (c) reversible, qualitative, and quantitative detection of CN − ; and (d) near-infrared emission and good biocompatibility enabling detection of CN − in cancer cells and mice.

Materials and instruments
The equipment and materials required for the experiments are given in the Supporting Information.

Synthesis of HBT-CHO and AHMM
The compound HBT-CHO was prepared according to the previous work. 38,39 The synthesis and the characterization of the compound (HBT-CHO) is shown in Figure (S1). The compound HBT-CHO (1 mmol, 0.269 g) and diaminomaleonitrile A (1.5 mmol, 0.162 g) were added to 40 ml dry EtOH, then the mixture was reacted at 50 • C, HCl (20 μl, 37%) was added while stirring. When the reaction was completed, the deposit was filtered while hot, and the mixture was washed with EtOH and CH 2 Cl 2 , respectively, and dried under vacuum to gain probe AHMM (Scheme 1). 1 Figure S4).

Preparation of solutions and spectral measurements
The probe AHMM (100 μM) stock solutions were prepared in CH 3

Crystal structures of AHMM
At room temperature, the probe AHMM was dissolved in CH 3 CN. After slow evaporation, the single crystals were obtained for x-ray single-crystal diffraction analysis. The CCDC number of the crystal was 2085453.

Detection of CN − by smartphone (RGB)
The different concentrations of CN − were added to the AHMM concentration (20 μM). Then, the RGB experiment was operated on a smartphone APP, while the UV spectra of corresponding solutions were gained by spectrophotometer.

Food sample imaging
According to the literature, 28,40 food samples like almonds and cassava were considered to contain CN − in the form of cyanogenic glycosides. The almonds, cooked almonds, and cassava were prepared and then soaked in a solution of AHMM (100 μM) for 2 h before imaging.

CCK-8 assay, cell culture, and fluorescence imaging
The detailed experiment of CCK-8 could be found in Supporting Information. The HeLa cells were cultured in DMEM medium containing 0.25% trypsin at 37 • C in a 5% CO 2 incubator. The cells were paved on a 14 mm glass cover and allowed to adhere for 24 h. Then the cells were incubated with AHMM (10 μM) for 30 min before imaging. For the control experiments, confocal fluorescent imaging was performed immediately after adding the different concentrations of CN − (5, 10, and 15 μM) for another 2 h.

Mice and mice tissues imaging
The probe AHMM and analyte (CN − ) were injected subcutaneously into mice. In accordance with ethical considerations involving animal research, all procedures performed in this study were approved by the Institutional (Fuyang Normal University), Approval ID (IACUC-NO.202204081).
The study was conducted in compliance with the guidelines set forth by the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals and the Animal Welfare Act. All efforts were made to minimize animal suffering and to reduce the number of animals used in the study.

Solvent selection and the ESIPT characteristics
Given the previous reports, 37,41 ESIPT-based probe has unique dual-emission wavelength, the short or large wave-length emission is attributed to its enol-form emission or keto-form emission, respectively. In addition, ESIPT phenomenon could also be affected by different solvents. First, the various analytical solvents are prepared for the AHMM, such as CH 3 CN, THF, DMSO, DMF, PhMe, CH 2 Cl 2 , MeOH, EtOAc, acetone, and EtOH. When applied excitation at 405 nm, the double emission of enol and keto tautomers of AHMM were all observed in the various solutions (Figure 2A-C). Among the solutions, AHMM in DMSO showed the largest Stokes shift with ESIPT process (λ ex /λ em = 405/620). Because of trace amount of DMSO is often used as a cosolvent for biological reagents compared to other organic solvents. Thus, the aqueous DMSO system was selected for the next experiments.
To further demonstrate the ESIPT phenomenon, singlecrystal structure of AHMM was obtained for analyzation ( Figure 2D,E). The result showed a strong intramolecular OH . . . N hydrogen bond in the structure of AHMM with a corresponding distance of 1.86 Å. The intramolecular hydrogen bond will promote intramolecular proton transfer and generate the photoinduced structural transformation of enol-form to keto-form. 42 Moreover, we also observed that the hyperconjugated planar construction of rhodol-benzothiazole created the formation of intermolecular π-π stacking of AHMM with the interaction distance of 3.715 Å. These results limited the rotation of the C-C single bond between rhodol and benzothiazole, thereby promoting the ESIPT process. 43

3.2
The selectivity of AHMM for CN − The selectivity of probe AHMM for various anions was detected by visualizing color and optical spectra changes in solution. From Figure (2F,G), the solution mixed with  Figure 2H,I). The results indicated that AHMM had high selectivity and specificity for CN − . And then, the effect of pH interference on the probe was studied by fluorescence spectrum (see Figure S5). It was obvious that pH between 6.8 and 8.0 could not affect the selectivity of AHMM to CN − . This phenomenon was helpful to fluorescence imaging studies. In addition, the limit of detection (LOD) of AHMM to CN − was 4.47 × 10 −8 M on the ground of 3δ/k (see Figure S6), much lower than [CN − ] deemed acceptable by WHO. 44,45

Response mechanism
The preceding results indicated that AHMM could effectively detect CN − from different anions with ESIPT process. To understand the interaction between AHMM and CN − , this interaction was studied using Job's plots, HRMS, 1 H NMR titrations, and theoretical computations. The Job's plots indicated a stoichiometric ratio of 1:1 between  Figure S10). The phenomenon showed that the removal of hydrogen could promote the ESIPT process. To shed light on the nature of the sensing mechanism between AHMM and CN − , theoretical studies were carried out. As shown in Figure (3C), for AHMM, the HOMO was solely localized on benzothiazole moiety and phenolic units, while LUMO was localized on the diaminomaleonitrile. However, after binding with CN − , the HOMO and LUMO of AHMM-CN were changed merely. Theoretical computations indicated that the energy gap between HOMO and LUMO was 2.3465 eV for AHMM-CN, lower than that for AHMM and indicating the formation of hydrogen bonding upon inclusion of CN − . At the same time, the occurrence of the ESIPT phenomenon was further confirmed. According to the above data, the reaction mechanism was demonstrated as we proposed ( Figure 3D).

Reversibility applications of AHMM
To reveal the reversibility of AHMM, the reversible switching cycle experiments were operated by trifluoroacetic acid (TFA) as the reagent. 46 From Figure (S11), when a solution of TFA (3.0 equiv. for AHMM) was introduced into the solution containing AHMM and CN − , the color change from dark red to light red was observed due to the

Detection of CN − by smartphone (RGB)
To further achieve the goal of monitoring CN − on-site, the smartphone fascinated our attention. 47 Based on the excellent "naked-eye" identification capability, RGB (red, green, blue) values of the solutions of AHMM-CN − could be readout by a smartphone. R/B (red/blue) ratio showed a good linear relationship with the concentration of CN − (R 2 = .96736). Then, to demonstrate reliability of the method, the above various CN − solutions were tested to obtain a standard curve by a UV-Vis spectrophotometer to quantify the concentration of CN − . Then, we prepared a CN − solution with a random concentration of 13.78 μM to obtain a smartphone app image with an R/B ratio of 2.49. According to the standard curve of R/B to [CN − ], the concentration of CN − could be calculated as 13.58 μM ( Figure 3E,F). The results only showed a relative difference of 1.45%.

Food samples and cell imaging
We also examined the applicability of AHMM to our life. For this purpose, the food samples, such as cassava, almond, and cooked almond, are selected. As shown in Figure (4A,C), cassava and raw almond barely showed fluorescence. After incubation with AHMM for 3 h, the strong green and red fluorescence were observed due to the CN − in the samples and the two-color imaging was attributed to the ESIPT feature. Subsequently, the food sample of cooked almonds showed dim fluorescence, because CN − was destroyed after the treatment of high temperature ( Figure 4B). The above results indicated that the probe AHMM could detect endogenous CN − in the food.
The CCK-8 assay on HeLa cells illustrated that probe AHMM and AHMM-CN could be applied in biological imaging ( Figure S12). So, the probe AHMM was applied to the detection of CN − in cells. As shown in Figure ( fluorescence increased with ESIPT characteristics. These results indicated that the probe AHMM could be applied to highly selective detection of intracellular CN − with good biocompatibility.

Mice and Mice tissues imaging
Based on the excellent performance of probe AHMM in cell imaging, we examined its capability for visualization of CN − in mice model. At the beginning, keeping the concentration of probe AHMM to CN − at 1:2, and the concentration of AHMM (0, 5, 10, 15, 20, 25 μM) was studied ( Figure 5A,C). Finally, we conducted AHMM with concentration of 20 μM for mice imaging. As depicted in Figure (5B), no fluorescent signal was observed in mice before the AHMM injection. Then, a weak fluorescent signal was observed after injection of the probe AHMM (20 μM) into the mice, whereas strong fluorescence signals were observed after injection of solution of CN − in the same injection point. The time-dependent fluorescent signals were recorded after subcutaneous injection of AHMM and CN − . The intensity of the injection point remained higher than that before injection of CN − within 2 h. Results showed that CN − could be detected by the probe AHMM in vivo.
To further understand the biodistribution of the probe, fluorescence imaging of the main tissues of mice was performed. The probe AHMM was injected into the thigh muscles of the two mice. Subsequently, CN − was injected into the same injection point in one of the mice. After 2 h, the liver, kidney, and heart tissues of the mice were collected for fluorescence imaging. From Figure (5F-H), through comparison of tissue cells of main organs, the tissue cells involved green and red fluorescence significantly because of the ESIPT characteristic when AHMM interacted with CN − in vivo.

CONCLUSIONS
In summary, we have rationally designed an ESIPT-based probe for detecting CN − with large Stokes shift, reversible transformed, rapid off-on response, high selectivity, and extremely low LOD. The interaction mechanism of AHMM to CN − is revealed by spectrum experiments, mass spectrometry, 1 H NMR titrations, and theoretical computations under an ESIPT-on process. Furthermore, a smartphone APP is used to achieve qualitative detection of tap water contaminated with CN − on-site. Also, AHMM is successfully applied to the imaging of CN − in food, cancer cells, and mice in a physiological environment.