High-Affinity Ratiometric Fluorescence Probe Based on 6-Amino-2,2′-Bipyridine Scaffold for Endogenous Zn2+ and Its Application to Living Cells

Zinc is an essential trace element involved in many biological activities; however, its functions are not fully understood. To elucidate the role of endogenous labile Zn2+, we developed a novel ratiometric fluorescence probe, 5-(4-methoxyphenyl)-4-(methylsulfanyl)-[2,2′-bipyridin]-6-amine (6 (rBpyZ)) based on the 6-amino-2,2′-bipyridine scaffold, which acts as both the chelating agent for Zn2+ and the fluorescent moiety. The methoxy group acted as an electron donor, enabling the intramolecular charge transfer state of 6 (rBpyZ), and a ratiometric fluorescence response consisting of a decrease at the emission wavelength of 438 nm and a corresponding increase at the emission wavelength of 465 nm was observed. The ratiometric probe 6 (rBpyZ) exhibited a nanomolar-level dissociation constant (Kd = 0.77 nM), a large Stokes shift (139 nm), and an excellent detection limit (0.10 nM) under physiological conditions. Moreover, fluorescence imaging using A549 human lung adenocarcinoma cells revealed that 6 (rBpyZ) had good cell membrane permeability and could clearly visualize endogenous labile Zn2+. These results suggest that the ratiometric fluorescence probe 6 (rBpyZ) has considerable potential as a valuable tool for understanding the role of Zn2+ in living systems.


Introduction
Zinc is an essential trace element for living organisms and plays pivotal roles in many biological processes, such as gene expression, apoptosis, enzyme regulation, and neurotransmission [1][2][3][4]. Most zinc is tightly bound to proteins and is involved in structural and catalytic functions [2,3]. In contrast, labile (free) zinc exits at low concentrations (picomolar to nanomolar) and is reported to play important roles in signaling, cell proliferation, and differentiation [4][5][6]. The amount of labile zinc present is regulated by zinc homeostasis; however, it has been suggested that zinc deficiency induced by insufficient dietary intake, poor absorption, and heredity disorders is related to hypogeusia, growth retardation, and carcinogenesis [7][8][9]. Therefore, to elucidate the details of the function and behavior of zinc in cells and tissues, a reliable method is urgently required.
Biological imaging using fluorescence probes is a robust method that enables the sensitive, real-time, and noninvasive detection of target molecules. Organic molecule-based fluorescence probes are gaining importance as simple cost-effective probes with good biocompatibility [10,11]. Their properties, including target binding ability and fluorescence response, are relatively easy to optimize by modifying their molecular structure and substituents [10,11]. As zinc has a closed shell structure (3d 10 4s 0 ) and cannot be directly analyzed using conventional methods, such as ultraviolet/visible (UV/vis) and nuclear magnetic resonance (NMR) spectroscopies, the development of organic molecule-based fluorescence probes for zinc has attracted attention [12,13]. A variety of fluorescence Zn 2+ probes based on fluorophore platforms, such as quinoline, fluorescein, cyanine, and biheteroaryl have been reported, some of which have contributed to the understanding of the biological function and behavior of zinc [12][13][14][15][16]. However, even with these probes, the detailed biological functions of zinc are yet to be fully understood. Therefore, the development of novel fluorescence Zn 2+ probes that demonstrate higher affinity toward Zn 2+ as well as increased solubility and cell membrane permeability than the currently available probes is essential. Previously, we reported a 6-amino-2,2 -bipyridine-based fluorescence Zn 2+ probe 3, in which the core structure acted as both the chelating agent for Zn 2+ and the fluorescent moiety. The resulting probe had a small molecular weight, making it highly soluble and cell membrane-permeant [17]. This probe also achieved a nanomolar-range dissociation constant toward Zn 2+ by controlling intramolecular amino-imino tautomerization; however, the high background fluorescence of the probe was a significant drawback for detecting endogenous Zn 2+ [17]. The use of ratiometric probes enables highly sensitive measurements through self-calibration at different emission wavelengths in the presence and absence of analyte [18,19]. Generally, the ratiometric property is based on forming an intramolecular charge transfer (ICT) state. This is generally achieved by complex electron interaction between an electron donor and an electron acceptor within molecules [18]. In this manner, the ratiometric probe can minimize the effect of its background fluorescence, which allows for the accurate detection of analytes, including Zn 2+ .

Spectroscopic Studies
We analyzed the UV/Vis absorption and fluorescence properties of 4-6 (rBpyZ) in HEPES buffer (100 mM, 50% EtOH, pH = 7.4) and compared them with those of compound 3. As shown in Figure 1, the absorption peaks at wavelengths 260 and 345 nm of free 3 gradually increased with the addition of Zn 2+ , accompanied by a small bathochromic shift of 5-7 nm (Figure 1a). Compounds 4 and 5 also exhibited absorption maxima at 260 and 345 nm, and similar changes to 3 were observed after adding Zn 2+ (Figure 1b,c). In compound 6 (rBpyZ), the absorption peak at the longer wavelength exhibited a hypsochromic shift of 20 nm compared to that of 3, and a new peak appeared at 349 nm with the addition of Zn 2+ (Figure 1d). Figure 2 shows the fluorescence titration spectra with Zn 2+ . Similar to the spectrum of 3, the fluorescence intensities of 4 and 5 gradually increased in a Zn 2+ concentration-dependent manner, and the maximum emission wavelength of 4 and 5 exhibited small bathochromic shifts of 14 and 5 nm, respectively. The fluorescence intensities of 4 and 5 were stronger than that of 3, being enhanced 10-and 18-fold, respectively.
In contrast, a significant ratiometric response to Zn 2+ was observed in compound 6 (rBpyZ) (Figure 2d). In the absence of Zn 2+ , 6 (rBpyZ) exhibited an emission peak at 438 nm, which was blue-shifted by 30 nm compared to those of compounds 4 and 5. Upon adding Zn 2+ , the emission peak at 438 nm gradually decreased, and a new emission peak at 465 nm concomitantly appeared ( Figure 3). In addition, 6 (rBpyZ) showed a significantly larger Stokes shift (139 nm), which enabled sensitive detection without interference from the excitation light wavelength.
These results indicated that a chelation-enhanced fluorescence (CHEF) effect occurred with the addition of Zn 2+ , and the simple replacement of substituent influenced the fluorescence emission profile. The analysis of all compounds revealed that the complex formed between the compound and Zn 2+ possessed a 1:1 stoichiometry ( Figure S1, Supplementary Materials). From the fluorescence titration data, the dissociation constants (Kd) of 4-6 (rBpyZ) were calculated to be at a nanomolar level, comparable to that of compound 3 (2.2 nM) (Table 1; Figure S2). Scheme 1. Preparation of 6-amino-2,2 -bipyridine-based compounds 3-6 (rBpyZ).

Spectroscopic Studies
We analyzed the UV/Vis absorption and fluorescence properties of 4-6 (rBpyZ) in HEPES buffer (100 mM, 50% EtOH, pH = 7.4) and compared them with those of compound 3. As shown in Figure 1, the absorption peaks at wavelengths 260 and 345 nm of free 3 gradually increased with the addition of Zn 2+ , accompanied by a small bathochromic shift of 5-7 nm ( Figure 1a). Compounds 4 and 5 also exhibited absorption maxima at 260 and 345 nm, and similar changes to 3 were observed after adding Zn 2+ (Figure 1b,c). In compound 6 (rBpyZ), the absorption peak at the longer wavelength exhibited a hypsochromic shift of 20 nm compared to that of 3, and a new peak appeared at 349 nm with the addition of Zn 2+ (Figure 1d).      In contrast, a significant ratiometric response to Zn 2+ was observed in compound 6 (rBpyZ) (Figure 2d). In the absence of Zn 2+ , 6 (rBpyZ) exhibited an emission peak at 438 nm, which was blue-shifted by 30 nm compared to those of compounds 4 and 5. Upon adding Zn 2+ , the emission peak at 438 nm gradually decreased, and a new emission peak at 465 nm concomitantly appeared ( Figure 3). In addition, 6 (rBpyZ) showed a significantly larger Stokes shift (139 nm), which enabled sensitive detection without interference from the excitation light wavelength.    These results indicated that a chelation-enhanced fluorescence (CHEF) effect occurred with the addition of Zn 2+ , and the simple replacement of substituent influenced the fluorescence emission profile. The analysis of all compounds revealed that the complex formed between the compound and Zn 2+ possessed a 1:1 stoichiometry ( Figure S1, Supplementary Materials). From the fluorescence titration data, the dissociation constants (K d ) of 4-6 (rBpyZ) were calculated to be at a nanomolar level, comparable to that of compound 3 (2.2 nM) (Table 1; Figure S2). We previously reported on the properties of the Zn 2+ -coordinated bipyridine moiety of 3 and its structural analogs using 1 H-NMR [17]. As 6 (rBpyZ) exhibited a fluorescence response distinct from 3-5, we investigated the Zn 2+ coordination site of 6 (rBpyZ) using 1 H-NMR spectroscopy. As shown in Figure 4, the protons of the 3-and 3 -positions of 6 (rBpyZ) shifted 0.04 or 0.9 ppm downfield when 1.0 equiv. of Zn 2+ was added. The NH 2 proton peak at 5.26 ppm moved to 6.29 ppm without any change in integral values. These results were similar to the NMR results of 3, indicating that the Zn 2+ coordinated with the 2,2 -bipyridine moiety of 6 (rBpyZ) and that amino-imino tautomerism did not occur. These results also suggested that the ratiometric property of 6 (rBpyZ) was not due to a change of the Zn 2+ coordination site but was rather caused by the stabilization of the highest occupied molecular orbital (HOMO) energy via the introduction of an electron-donating group, similar to that observed in other ratiometric probes [20]. Ratiometric probes that emit at two distinct wavelengths are beneficial for the sensitive and selective detection of targets. Subsequently, evaluations including a metal selectivity experiment, a test of the influence of pH on fluorescence, and the analysis of the fluorescence imaging of cells, were performed on the ratiometric compound 6 (rBpyZ).  To investigate the selectivity of 6 (rBpyZ) toward cations, the fluorescence spectra of 6 (rBpyZ) in the presence of various cations (Al 3+ , Ca 2+ , Cd 2+ , Co 2+ , Cu 2+ , Fe 2+ , Fe 3+ , K + , Mg 2+ , Mn 2+ , Na + , Ni 2+ , and Zn 2+ ) were recorded, and are displayed in Figure 5. We note that the addition of Cd 2+ to 6 yields a similarly enhanced fluorescence spectrum to that produced with the addition of Zn 2+ . It has been reported that various Zn 2+ fluorescence probes also detect Cd 2+ owing to both ions being in the same column of the periodic table [12]. However, the concentration of free Cd 2+ is very low in living systems and can therefore be assumed to have a negligible influence on cellular Zn 2+ imaging. The addition of alkali ions (Na + ; K + ) and alkaline-earth ions (Ca 2+ ; Mg 2+ ), which exist ubiquitously at millimolar concentrations in living systems, cause no change to the spectrum of 6 (rBpyZ). In the transition metal cations (Co 2+ , Cu 2+ , Fe 2+ , Fe 3+ , Mn 2+ , Ni 2+ , and Al 3+ ), fluorescence quench- 1 H-NMR spectra of (a) 6 (rBpyZ), (b) 6 (rBpyZ) + 0.5 equiv. Zn 2+ , and (c) 6 (rBpyZ) + 1.0 equiv. Zn 2+ in DMSO-d 6 .
To investigate the selectivity of 6 (rBpyZ) toward cations, the fluorescence spectra of 6 (rBpyZ) in the presence of various cations (Al 3+ , Ca 2+ , Cd 2+ , Co 2+ , Cu 2+ , Fe 2+ , Fe 3+ , K + , Mg 2+ , Mn 2+ , Na + , Ni 2+ , and Zn 2+ ) were recorded, and are displayed in Figure 5. We note that the addition of Cd 2+ to 6 yields a similarly enhanced fluorescence spectrum to that produced with the addition of Zn 2+ . It has been reported that various Zn 2+ fluorescence probes also detect Cd 2+ owing to both ions being in the same column of the periodic table [12]. However, the concentration of free Cd 2+ is very low in living systems and can therefore be assumed to have a negligible influence on cellular Zn 2+ imaging. The addition of alkali ions (Na + ; K + ) and alkaline-earth ions (Ca 2+ ; Mg 2+ ), which exist ubiquitously at millimolar concentrations in living systems, cause no change to the spectrum of 6 (rBpyZ). In the transition metal cations (Co 2+ , Cu 2+ , Fe 2+ , Fe 3+ , Mn 2+ , Ni 2+ , and Al 3+ ), fluorescence quenching and small bathochromic shifts in the emission wavelength were observed, indicating that these cations also form a complex with 6 (rBpyZ). However, binary competition experiments of 6 (rBpyZ) between Zn 2+ and other cations, using the ratio of fluorescence intensity at 465 nm to that at 438 nm (F 465 /F 438 ) (Figure 6), showed that in the presence of Co 2+ , Fe 2+ , Fe 3+ , Mn 2+ , and Al 3+ , 6 (rBpyZ) selectively detects Zn 2+ . 6 (rBpyZ) is not selective for Zn 2+ when in the presence of either Cu 2+ or Ni 2+ . Still, this lack of preference may have little impact on visualizing cellular Zn 2+ as the cellular concentrations of free Cu 2+ and Ni 2+ are also very low.
addition of Cd 2+ to 6 yields a similarly enhanced fluorescence spectrum to that produced with the addition of Zn 2+ . It has been reported that various Zn 2+ fluorescence probes also detect Cd 2+ owing to both ions being in the same column of the periodic table [12]. However, the concentration of free Cd 2+ is very low in living systems and can therefore be assumed to have a negligible influence on cellular Zn 2+ imaging. The addition of alkali ions (Na + ; K + ) and alkaline-earth ions (Ca 2+ ; Mg 2+ ), which exist ubiquitously at millimolar concentrations in living systems, cause no change to the spectrum of 6 (rBpyZ). In the transition metal cations (Co 2+ , Cu 2+ , Fe 2+ , Fe 3+ , Mn 2+ , Ni 2+ , and Al 3+ ), fluorescence quenching and small bathochromic shifts in the emission wavelength were observed, indicating that these cations also form a complex with 6 (rBpyZ). However, binary competition experiments of 6 (rBpyZ) between Zn 2+ and other cations, using the ratio of fluorescence intensity at 465 nm to that at 438 nm (F465/F438) (Figure 6), showed that in the presence of Co 2+ , Fe 2+ , Fe 3+ , Mn 2+ , and Al 3+ , 6 (rBpyZ) selectively detects Zn 2+ . 6 (rBpyZ) is not selective for Zn 2+ when in the presence of either Cu 2+ or Ni 2+ . Still, this lack of preference may have little impact on visualizing cellular Zn 2+ as the cellular concentrations of free Cu 2+ and Ni 2+ are also very low.   Figure 7 shows the effects of pH on the Zn 2+ detection ability of 6 (rBpyZ). F46 6 (rBpyZ) was measured in buffer solutions at different pH values in the absen presence of Zn 2+ . Under both conditions, F465/F438 was approximately constant b pH 5.0-8.0. In addition, a limit of detection (LOD) of 0.10 nM at pH 7.0 buffer s was calculated using the equation LOD = 3σ/slope. These results indicated that 6 ( could function as a sensitive Zn 2+ probe under physiological conditions. We subse  Figure 7 shows the effects of pH on the Zn 2+ detection ability of 6 (rBpyZ). F 465 /F 438 of 6 (rBpyZ) was measured in buffer solutions at different pH values in the absence and presence of Zn 2+ . Under both conditions, F 465 /F 438 was approximately constant between pH 5.0-8.0. In addition, a limit of detection (LOD) of 0.10 nM at pH 7.0 buffer solution was calculated using the equation LOD = 3σ/slope. These results indicated that 6 (rBpyZ) could function as a sensitive Zn 2+ probe under physiological conditions. We subsequently explored its usefulness in cellular applications, which we report in the next section. Figure 7 shows the effects of pH on the Zn 2+ detection ability of 6 (rBpyZ). F46 6 (rBpyZ) was measured in buffer solutions at different pH values in the absen presence of Zn 2+ . Under both conditions, F465/F438 was approximately constant b pH 5.0-8.0. In addition, a limit of detection (LOD) of 0.10 nM at pH 7.0 buffer s was calculated using the equation LOD = 3σ/slope. These results indicated that 6 ( could function as a sensitive Zn 2+ probe under physiological conditions. We subse explored its usefulness in cellular applications, which we report in the next sectio  Figure 8 shows fluorescence microscopy images of human lung adenocar cells (A549). The cells were exposed to 100 µ M 6 (rBpyZ) for 30 min at 37 °C produced very weak fluorescence (Figure 8a). In contrast, the cells incubated w µ M 6 (rBpyZ) for 30 min and 100 µ M Zn 2+ for another 30 min exhibited a bright cence signal in the cells (Figure 8b). The fluorescence signal was reduced by tre with a cell membrane permeable Zn 2+ c N,N,N',N'-tetrakis(2-pyridylmethyl)ethylenediamine (TPEN) [21], indicating (rBpyZ) exhibited fluorescence in response to intracellular Zn 2+ (Figure 8c). In ad no significant cytotoxicity was observed during this study.

Zn 2+ Cellular Imaging
Therefore, we further investigated the possibility of detecting endogenous the apoptotic cells. Zn 2+ is known as a regulator of apoptosis and is released from cellular stores during apoptosis [22]. As shown in Figure 9, in the H2O2-induced totic cells, a bright fluorescence signal was observed after incubation with 100  Figure 8 shows fluorescence microscopy images of human lung adenocarcinoma cells (A549). The cells were exposed to 100 µM 6 (rBpyZ) for 30 min at 37 • C, which produced very weak fluorescence (Figure 8a). In contrast, the cells incubated with 100 µM 6 (rBpyZ) for 30 min and 100 µM Zn 2+ for another 30 min exhibited a bright fluorescence signal in the cells (Figure 8b). The fluorescence signal was reduced by treatment with a cell membrane permeable Zn 2+ chelator, N,N,N',N'-tetrakis(2-pyridylmethyl)ethylenediamine (TPEN) [21], indicating that 6 (rBpyZ) exhibited fluorescence in response to intracellular Zn 2+ (Figure 8c). In addition, no significant cytotoxicity was observed during this study. (rBpyZ). It was confirmed that this observed fluorescence signal does not originate from the direct effect of H2O2 on 6 (rBpyZ) ( Figure S3). These data indicated that ratiometric probe 6 (rBpyZ) could detect endogenous Zn 2+ .   Therefore, we further investigated the possibility of detecting endogenous Zn 2+ in the apoptotic cells. Zn 2+ is known as a regulator of apoptosis and is released from intracellular stores during apoptosis [22]. As shown in Figure 9, in the H 2 O 2 -induced apoptotic cells, a bright fluorescence signal was observed after incubation with 100 µM 6 (rBpyZ). It was confirmed that this observed fluorescence signal does not originate from the direct effect of H 2 O 2 on 6 (rBpyZ) ( Figure S3). These data indicated that ratiometric probe 6 (rBpyZ) could detect endogenous Zn 2+ .

Materials and Instruments
All reagents and chemicals were of the highest grade available. 1 H and 13 C-NMR spectra were recorded on JEOL ECP-400 NMR and Varian Gemini 300 systems at room temperature. The chemical shifts are reported in parts per million (ppm) relative to tetramethylsilane (TMS) as the reference. Mass spectra (MS) and high-resolution mass spectrometry (HRMS) were performed using a JMS-700 spectrometer (JEOL, Tokyo, Japan). UV/Vis absorption spectra were acquired with a UV/Vis spectrophotometer (UV-2450, Shimadzu, Kyoto, Japan). Fluorescence spectra were acquired using a fluorescence spectrophotometer (FP-8300ST, Jasco, Tokyo, Japan).

Conclusions
To reduce the influence of background fluorescence and achieve the sensitive detection of endogenous Zn 2+ in cells, we designed and synthesized 6-amino-2,2 -bipyridinebased small molecular weight Zn 2+ fluorescence compounds with electron-withdrawing or electron-donating groups on the allyl group at position 5 of the central pyridine ring. Among the synthesized compounds, compound 6 (rBpyZ) bearing a methoxy group exhibited ratiometric fluorescence profiles at 438 nm and 465 nm in the absence or presence of Zn 2+ . In addition, 6 (rBpyZ) showed a high Zn 2+ binding affinity (K d = 0.77 nM), a large Stokes shift (over 139 nm), high Zn 2+ selectivity, and high stability under physiological pH conditions, making it useful as a fluorescence Zn 2+ probe. Fluorescence microscopy studies indicated that 6 (rBpyZ) possessed cell membrane permeability in living human lung adenocarcinoma cells and enabled the visualization of endogenous labile Zn 2+ in the cells during apoptosis. We expect that 6 (rBpyZ) will be a valuable and efficient probe for elucidating the biological functions of Zn 2+ .