Biosensors and Bioelectronics 72 (2015) 275–281
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
A colorimetric and ratiometric fluorescent probe for selective detection and cellular imaging of glutathione Chang Xu a, Hongda Li b, Bingzhu Yin a,n a Key Laboratory of Natural Resources of Changbai Mountain & Functional Molecules of the Ministry of Education, Department of Chemistry, Yanbian University, Yanji 133002, PR China b Department of Forensic Chemistry, National Police University of China, Shenyang 110854, PR China
art ic l e i nf o
a b s t r a c t
Article history: Received 12 March 2015 Received in revised form 10 May 2015 Accepted 11 May 2015 Available online 13 May 2015
A new colorimetric and ratiometric fluorescent probe 1 based on a chlorinated coumarinyl aldehyde was developed for selective detection and cellular imaging of glutathione (GSH) over cysteine (Cys) and homocysteine (Hcy). Probe 1 exhibits a dramatic colorimetric and ratiometric fluorescence responses toward biothiols Cys, Hcy and GSH with high selectivity over other amino acids. Cys (or Hcy) induces a tandem SNAr-rearrangement reaction to form the corresponding amino-coumarins (2a or 2b), which result in about 75 nm and 35 nm blue-shifts in absorption and emission, respectively. By comparison, the thio-coumarin (3′) derived from the SNAr reaction with GSH, which does not occur rearrangement because of steric hindrance, undergoes an intramolecular aldimine condensation lead to a cyclic iminium cation (3) with 47 nm and 39 nm red-shifts in absorption and emission, respectively. The significantly difference of photophysical properties enable excellent selectivity towards GSH over Cys and Hcy. Further application to cellular imaging indicates that the probe has appreciable cell permeability and is highly responsive to the changes of GSH level. As a result, it is applicable to monitor GSH level in living cells. & 2015 Elsevier B.V. All rights reserved.
Keywords: Ratiometric fluorescent probe Glutathione detection Cysteine Homocysteine chlorinated coumarinyl aldehyde Cell imaging
1. Introduction Glutathione (GSH; γ-glutamylcysteinylglycine) is ubiquitous in mammalian and many prokaryotic cells and is the most abundant intracellular thiol (1–10 mM) (Meister, 1988). It is an essential endogenous antioxidant that plays a central role in cellular and defenses against toxins and free radicals (Miller et al., 2007). Aberrant levels of GSH have been associated with a number of diseases, including cancer, AIDS, Alzheimer's, cardiovascular disease and others (Townsend et al., 2003). Therefore, the accurate concentration measurement of GSH in physiological media has been considered as an essential factor in these diseases and their therapy because of its biological and clinical significance. Because of this, the considerable contemporary effort devoted to the development of an efficient method for the detection and quantification of GSH in living systems (Chen et al., 2010). Among the various analytical methods that are available, molecular imaging based on fluorescent probes is considered to be the most sensitive approach owing to its sensitivity and simplicity. Most of the existing probes utilize the strong nucleophilicity of the thiol group, operating by Michael addition (Guy et al., 2007), cleavage of n
Corresponding author. Fax: þ 86 433 2732456. E-mail address: [email protected] (B. Yin).
http://dx.doi.org/10.1016/j.bios.2015.05.030 0956-5663/& 2015 Elsevier B.V. All rights reserved.
disulfide (Pires and Chmielewski, 2008; Lee et al., 2012) and sulfonamide (Zhang et al., 2011; Shao et al., 2012), etc (Tang et al., 2007; Zhang et al., 2007). However, although these probes can highly selectively distinguish these biothiols from other amino acids, most of them cannot distinguish Cys/Hcy/GSH from each other due to their similar structure and reactivity. Up to now, the discrimination between them has been a focal point and also a tough challenge for researchers, albeit some advances have been obtained. By means of the cyclization of Cys/Hcy with aldehydes or acrylates, selective detection of Cys/Hcy over GSH was firstly achieved by the Strongin group (Rusin et al., 2004; Yang et al., 2011). Since then, as the extended version of the two strategies, some more specific probes for Cys or Hcy were also developed (Chen. et al. 2007; Li et al., 2014; Yuan et al., 2011; Yang et al., 2012b; Guo et al., 2012a; Yang et al., 2012a; Guo et al., 2012c). Recently, discrimination of Cys from Hcy/GSH was also achieved by taking advantage of either the Cys-induced substitution-rearrangement cascade reaction (Ma et al., 2012; Niu et al., 2013) or Michael addition combined with steric and electrostatic interactions (Jung et al., 2012a; Jung et al., 2012b; Zhou et al., 2012; Zhang et al., 2012). However, probes capable of discrimination of GSH from Cys/Hcy are still rare. As far as we know, only several strategies have been reported, including supramolecular interaction between a bis-spiropyran receptor and GSH (Shao et al., 2010), a micelle-catalyzed large ring formation of GSH with acrylate (Guo
276
C. Xu et al. / Biosensors and Bioelectronics 72 (2015) 275–281
standard procedures. Fluorescence spectra were carried out on a Shimadzu RF-5301PC fluorescence spectrophotometer. UV/vis spectra were recorded with a Shimadzu UV-2550 spectrophotometer. NMR spectra were recorded on a Bruker AV-300 spectrometer (300 MHz for 1H and 75 MHz for 13C), and chemical shifts were referenced relative to tetramethylsilane. Mass data were obtained by a Shimadzu AXIMA-CFR™ plus mass spectrometry using a 1, 8, 9-anthracenetriol (DITH) matrix. PBS buffer (10 mM, pH 7.4, 10% DMSO) prepared with deionized water and spectroscopic grade of DMSO were used as the solvent for all spectroscopic experiments. Stock solution of the probe (1 10 3 M) was prepared in DMSO, which was diluted to 1 10 5 M with the PBS buffer. Stock solution of the various amino acids were prepared with the PBS buffer for future use. For all fluorescence measurements, the excitations were at 476 nm for GSH and 385 nm for Cys and Hcy, and the excitation and emission slit widths were 3 and 5 nm, respectively.
et al., 2012b), a specific reaction of GSH with o-phthalaldehyde (Xu et al., 2013), and those GSH-induced tandem reactions, such as SNAr-rearrangement (Niu et al., 2012; Liu et al., 2014; Jia et al., 2015), SNAr-rearrangement-cyclization (Liu et al., 2014; Liu et al., 2015), and nucleophilic substitution (Lim et al., 2014; Yin et al., 2014) as well as a native chemical ligation (NCL) reaction (Yang et al., 2014). Very recently, a novel strategy for selectively sensing GSH by a dual-response mechanism was reported (Wang et al., 2015). This strategy integrates two independent reaction sites with a disulfide linker and a thioether function into a fluorescent BODIPY-based probe to guarantee the synergetic dual-response in an elegant fashion to address the discrimination of GSH. Even so, there still are some limitations when considering the practical applications in biological systems, such as use of organic solvent or surfactant, long response time and require high equivalents of GSH to reach a maximal fluorescent signal, relatively poor selectivity, undesired spectra overlap, and so on. Furthermore, almost all of them are based on fluorescence measurement at a single wavelength, which means that GSH detection only depends on the changes of emission intensity, and could be significantly influenced by the excitation power and the detector sensitivity. Moreover, fluorescent intensity changes are not easy to observe by the naked eye directly (Hu et al., 2010). By contrast, ratiometric fluorescent probes allow the measurement of emission intensities at two wavelengths, which should provide a built-in correction for environmental effects and the fluorescence color change, which can be measured directly with a colorimeter or even distinguished easily by eye (Fu et al., 2011). Therefore, it is of high interest to develop new ratiometric fluorescent probes for GSH, in particular, with rapid response and high sensitivity. With these considerations in mind, we developed a new colorimetric and ratiometric fluorescent probe 1 for discrimination of GSH over Cys and Hcy based on a chlorinated coumaryl aldehyde (Scheme 1). As we expected, probe 1 displayed a high selectivity for detection of GSH over Cys and Hcy based on a novel GSH-induced SNAr-aldimine condensation strategy and that it can be used to monitor GSH in cells.
HeLa cells were obtained from American Type Culture collection and grown in Dulbecco's modification of Eagle's medium Dulbecco (Free DMEM/high: with 4500 mg/L Glucose, 4.0 mM LGlutamine, and 110 mg/L Sodium Pyruvate). The cells were incubated in a 5% CO2 humidified incubator at 37 °C and typically passaged with sub-cultivation ratio of 1:4 for two days. The HeLa cells were seeded in 6-well culture plate overnight. Stock solutions of probe (3 mM) and NEM (0.15 M) in DMSO, GSH, Cys (0.15 M) in PBS buffer (10 mM, pH 7.4) were prepared at the same day of experiment. Then, the cells were treated without or with GSH (0.1, 0.2, 0.5, 1.0, and 2.0 mM) or Cys (0.5 mM) for 30 min and NEM (0.1 mM) for 20 min at 37 °C, respectively, in culture media. After washing with phosphate buffered saline to remove the remaining GSH or Cys, the cells were further incubated with 10 mM of probe in the culture media for 30 min at 37 °C. The cells were imaged using an inverted microscope (OLYMPUS IX73) and Leica SP5II confocal microscope.
2. Experimental
3. Results and discussion
2.1. Materials and apparatus
3.1. Design concept of the probe
Commercially available compounds were used without further purification. The reaction solvents were dried according to
The design rationale is depicted in Scheme 1, and illustrated as follows. Chlorine lies in 4-position of probe 1, where is doubly
2.2. Methods for cell culture and fluorescent imaging
Scheme 1. Proposed reaction mechanisms of probe 1 with Cys, Hcy and GSH.
C. Xu et al. / Biosensors and Bioelectronics 72 (2015) 275–281
277
activated and has been proved to be subject to the thiol-halogen SNAr reaction (Liu et al., 2014). Thus, it was expected that the chlorine atom of 1 could initially be replaced by thiol group of Cys (or Hcy) to produce corresponding thio-coumarins 2a′ (or 2b′), and the subsequent intramolecular rearrangement would lead to amino-coumarins 2 (Scheme 1). We speculated that the thiol group of 2 is hard to further attack the adjacent carbonyl group due to the unstable macrocyclic hemi-thioacetal product and the HSAB effect. In the case of GSH, the initial thiol-halogen SNAr reaction would similarly lead to thio-coumarin 3′ (Scheme 1), which does not occur rearrangement due to the bulkiness of its tripeptide and the unstable 10-membered macrocyclic transition state. 3′ would undergo a intramolecular aldimine condensation leading to a cyclic iminium cation 3 that enhances the ICT and modulates both the absorbance and fluorescence. Given the distinct photophysical properties of the amino-coumarin and the cyclic iminium cation, a ratiometric colorimetric and fluorescent dual response to GSH with high selectivity is expected using this probe. 3.2. Colorimetric and ratiometric response towards GSH, Cys and Hcy To realize the above design concept, we synthesized probe 1 according to the previously reported method by formylation of 7-diethylamino-4-hydroxycoumarin under Vilsmeier reaction condition (Supporting information). With probe 1 in hand, we explored the water solubility initially. Fortunately, probe 1 was observed to display satisfactory water solubility. However, the sensing processes of the probe toward thiols were more time consuming when the water content was more than 90%. Thus, we selected a pH 7.4 PBS buffer solution containing 10% (v/v) DMSO for further experiments. Next, we examined the reactivity of 1 towards Cys, Hcy and GSH through UV–vis spectra. As shown in Fig. 1a, the UV–vis spectrum of free 1 shows a main absorption at 457 nm (Klockow et al., 2013). Upon addition of 10 equiv. of Cys, the initial absorption peak decreased, along with the emergence of a blue-shifted peak at 380 nm. According to the aforementioned speculation and the well-established tandem SNAr-rearrangement reaction mechanism (Niu et al., 2012), the absorption at 380 nm should be assigned to the amino-coumarin (2a), which was further supported by both the data of MALDI-TOF mass experiment (Fig. S1) and control compound A and B (Scheme S1). The absorption maximums of A and B were found at around 376 (Fig. S2) and 377 nm, respectively, which very close to that of 2a. Interestingly, in the titration process of Cys we have observed the corresponding absorption band of the intermediate (2a′). As seen in Fig. 1b, when increasing amount of Cys was added to the solution of probe 1, the maximum absorption peak centered at 457 nm decreased and red-shifted gradually to 475 nm and disappeared finally, which should be assigned to the thio-coumarin (2a′). The assign could be supported by control compound C (Absmax: 473 nm, Fig. S3) and D (Absmax: 468 nm) (Scheme S1). A similar observation was made for Hcy under same experimental condition. By contrast, addition of GSH to the solution of 1 led to the decrease of the initial absorption band at 457 nm followed by the emergence of two red-shifted peaks at 490 and 517 nm (Fig. 1a). This represents an unprecedented red-shift of 60 nm compared to the 35 nm shift that is typical for the coumarin aldehyde class of sensors (Feuster et al., 2003; Secor and Glass, 2004; Secor et al., 2005). Note that the absorptions at 490 and 517 nm was initially assigned to thio-coumarin 3′. However, the assign did not match the absorption spectra of the control compound C and D, which is much shorter than that of our reaction product. Fortunately, this unprecedented red-shift makes us to mind the hydrogen-bonding iminium ion 3, which would enhance the ICT and modulate both
Fig. 1. (a) Absorbance changes of probe 1 (10 μM) 30 min after addition of 5 equiv. of various biothiols (inset: color changes of the solution of 1 in the presence of 5 equiv. of different biothiols and (b) upon addition of increasing concentrations of Cys (0–2 equiv.).
the absorbance and fluorescence. In fact, the two new absorption peaks are very similar to those of the previously reported the hydrogen-bonding iminium ions E and F (Absmax: 525 nm) (Scheme S1), and can thus be assigned to 3. Importantly, the formation of the iminium ion 3 was evidenced by mass spectrum analysis of the product generated from the incubation of 1 with GSH in CH3CN–H2O. Two prominent peaks at m/z 533.90 and 489.9 corresponding to [3 þ H] þ (calcd. 532.16 for C24H28N4O8S) and [3 – CO2 þ H] þ (calcd. 488.2 for C23H28N4O6S) are clearly observed in the MALDI-TOF mass spectrum (Fig. S4), which provides strong evidence for aforementioned assign. Notably, the reaction also likely led to a spot of the by-product amino-coumarin as a result of the subsequent SNAr reaction between 3′ and excess GSH, which could be indicated by the newly emerged absorption peak around 379 nm in the titration process (Fig. S5). In a word, these results are in good accordance with our proposed reaction mechanism. Encouraged by the above results, subsequently, we examined the emission behavior of 1 upon addition of Cys, Hcy and GSH by use of two different excitations at 385 nm and 476 nm, respectively. Firstly, we selected the excitation wavelength at 385 nm to probe Cys. Upon 385 nm excitation, probe 1 is almost non-emissive. Addition of 10 equiv. of Cys to a solution of 1 elicited a big fluorescence enhancement (18-fold) at 477 nm due to the
278
C. Xu et al. / Biosensors and Bioelectronics 72 (2015) 275–281
Fig. 2. Fluorescence response of probe 1 (10 μM) to 5 equiv. of various biothiols with λex ¼476 nm (a) and λex ¼385 nm (b). Inset: the fluorescence photographs of 1 in the presence of GSH, Cys and Hcy exposed to a UV lamp at 365 nm.
production of amino-coumarin 2a (Fig. 2b). Despondently, under this excitation wavelength, addition of 10 equiv. of GSH similarly emerges a minor emission around 477 nm, which might be attributed to the amino-coumarin derivative. This phenomenon is in accordance with the results of UV spectral titration experiment. Next, we selected the excitation wavelength at 476 nm to probe GSH, which near the isosbestic point (Fig. 1a). As shown in Fig. 2a, probe 1 emits a green fluorescence with an emission maximum at 507 nm. Upon addition of 10 equiv. of GSH, the initial emission at 507 nm disappeared along with the emergence of a red-shifted emission at 546 nm, indicating a clear ratiometric fluorescence response. As mentioned before, this should be attributed to the production of the iminium ion 3. In contrast, addition of Cys/Hcy caused hardly any significant fluorescence changes in probe 1, indicating the strong ability of probe 1 to discriminate GSH from Cys/Hcy at the excitation wavelength at 476 nm, and thus it can serve as a selective ratiometric fluorescence probe for GSH in living cells. In addition, GSH can be easily distinguished from Cys/Hcy by the naked eye when illuminated by a hand-held UV lamp at 365 nm. Introduction of GSH to the solution of 1 results in a color change from yellow to red–orange and a fluorescence change from green to orange. By contrast, Cys (or Hcy) induces a color change from yellow to colorless and a fluorescence change from green to blue. These can clearly be perceived by naked eyes (Fig. 1a, inset and Fig. 2a, inset). Therefore, the colorimetric probe 1 can serve as
Fig. 3. (a) Fluorescence changes of probe 1 (10 μM) on the incremental addition of GSH, and (b) Job's plot between 1 and GSH based on data of absorption spectra. [1] þ [GSH] ¼ 40 μM. Inset: ratio of the fluorescence intensities at 546 and 507 nm (I546/I507) as a function of the GSH concentration.
a micromolar level “naked-eye” indicator for GSH. Subsequently, we performed the spectroscopic titration experiments under the same conditions. Upon treatment with the increasing concentrations of GSH, the emissions at 507 nm decreased gradually along with the simultaneous emergence and increase of the red-shifted emission at 546 nm, indicating a clear ratiometric fluorescence response as seen in Fig. 3a. Similarly, this should be attributed to the production of the iminium 3. Notably, the maximal spectral signals in both absorption and emission spectra were achieved in the presence of only 1 equiv. of GSH (Figs. S5 and S6). Moreover, the ratio of emission intensities (I546/I507) varies from 0.29 to 2.09 with a 7.2-fold ratiometric enhancement (Fig. 3a, inset). Time-dependent ratiometric changes in the fluorescence spectra showed that 10 equiv. of GSH is sufficient to complete the sensing reaction in about 20 min under the experimental condition (Fig. S7), and the observed rate constant (Kobs) (Dale et al., 2006) for sensing reaction was determined to be 0.1447 0.004 min 1 under pseudo-first-order kinetic conditions (Fig. S8). In addition, a Job's plot exhibits a maximum at 0.5 molecular fraction, which indicates that the reaction between 1 and GSH has a 1:1 stoichiometry (Fig. 3b). The result is in accordance with the data of mass experiment. More importantly, the fluorescence intensity ratio (I546/I507) was linearly proportional to the
C. Xu et al. / Biosensors and Bioelectronics 72 (2015) 275–281
279
virtually no influence on the fluorescence detection of GSH except for Cys and Hcy. Cys or Hcy might form stable amino-substituted covalent complexes with probe 1 through a sequential thiol substitution and rearrangement reaction as observed in Yang's BODIPY-based probe (Niu et al., 2012). These inspiring results suggest that probe 1 can be established as promising prototypes of ratiometric fluorescence GSH probe in actual living cells. 3.4. The effect of pH The fluorescent response of probe 1 toward GSH in PBS buffer (10 mM, 10%, DMSO) at different pH conditions were further investigated to apply probe 1 in complicated systems, such as environmental or biological systems. The results indicate that the response of probe 1 towards GSH was pH dependent. The fluorescence spectra of probe 1 had a relatively weak emission band at 546 nm and remained unaffected between pH levels 2 and 10, as shown in Fig. S10. Upon addition of GSH, the maximal fluorescence signals were observed in the pH range of 6–9 without the interference by protons, and the stable fluorescence of the probe at a pH level of approximately 7.4 was favorable for the sensing assays of GSH in environmental and biological samples. 3.5. Ratiometric cell imaging
Fig. 4. (a) Fluorescence response (λex ¼476 nm) of probe 1 (10 μM) to 5 equiv. of various amino acids and (b) changes in fluorescence intensity of 1 (10 μM) toward GSH (2 equiv.) in the presence of the competing analytes.
GSH concentration in the 0–10 μM range (R2 ¼0.998), indicating the suitability of 1 for quantitative detection of GSH. The detection limit (Sun et al., 2009 and 1.2. detection limit in Supporting information) is determined to be 0.38 μM (Fig. 3a, inset), which is significantly below the physiological levels of GSH in live cells (1– 10 mM) (Hwang et al., 1992). This result is indeed important in view of the very few fluorescent probes that could selectively and quantificationally detect GSH over Cys/Hcy in the literature so far. 3.3. Selective response of probe 1 to GSH To evaluate the specific nature of probe 1 for GSH, the spectral changes in probe 1 toward other amino acids and the biologically related analytes were explored. Compared with GSH, no obvious changes in absorption and emission spectra were observed upon addition of other amino acids and the related analytes, indicating probe 1 is highly selective to GSH over the other biologically relevant analytes tested (Figs. 4a and S9). On the other hand, whether the probe 1 could still retain the selective response to GSH in the presence of the competing analytes is very important for a fluorescent probe. Thus, competition experiments of probe 1 were conducted. As shown in Fig. 4b, probe 1 was treated with GSH in the presence of the competing analytes (10 equiv). All of the competing analytes tested have
For biological applications, the cytotoxicity in HeLa cells was firstly determined by the MTT assay with different concentrations of the probe (Supporting information). Fig. S11 shows the cell viability after incubation with probe 1 at the concentrations of 0 μM, 1 μM, 2.5 μM, 5 μM and 10 μM for 24 h, respectively. HeLa cell viabilities of more than 95% were observed during the tested time, indicative of low cytotoxicity of probe 1. Finally, we studied the capacity of probe 1 for ratiometric imaging of GSH in living cells. When HeLa cells were incubated with 1 (10 μM) for 20 min and then washed with PBS buffer, a bright fluorescence in the green channel, and weak fluorescence in both red channel and blue channel were observed simultaneously, indicating that the probe is cell-permeable (Fig. 5a–c) and is responsive to intracellular Cys and GSH. However, when the cells were pretreated with 0.1 mM N-ethylmaleimide (NEM), which is a trapping reagent for thiol species, and then treated with probe 1 (10 μM), the fluorescence in the red and blue channels are both decreased slightly while the fluorescence in green channel did not caused any detectable fluorescence suppression (Fig. 5d–f), suggesting that ratiometric imaging of 1 is responsive to changing intracellular GSH and Cys levels in the blue and red channels, respectively. Furthermore, when HeLa cells were pre-treated with 0.5 mM GSH, a significant increase in red emission and a minor decrease in green emission were observed (Fig. 5g and h). When higher concentrations of GSH (1.0 mM) were employed to image the HeLa cells, an obvious increase in red emission was found as seen in Fig. 5i, which implies that 1 is highly sensitive to concentration levels of GSH even in living cells. Indeed, concentrationdependent ratiometric intensity changes in confocal fluorescent images was linearly proportional to the GSH concentration (Figs. S12 and S13), indicating the suitability of the probe for quantitative detection of GSH in biological samples. To explore the specific nature of probe 1 toward GSH in living cells, HeLa cells were pretreated with Cys (0.5 mM), and then treated with probe 1 (10 μM). As expected, a marked enhancement in blue emission and a minor reduction in green emission were observed, respectively, compared with that of HeLa cells incubated with probe 1 only (Fig. 5j and k). More meaningfully, the fluorescence in red channel decreased slightly rather than increased (Fig. 5l), conforming that probe 1 is specific to GSH detection over Cys/Hcy in actual living cells. These results showed that the probe 1 not only
280
C. Xu et al. / Biosensors and Bioelectronics 72 (2015) 275–281
Fig. 5. Fluorescence images of living HeLa cells. (a–c) 1 (10 μM) for 30 min; (d–f) NEM (0.1 mM) for 20 min, and then 1 (10 μM) for 30 min; (g–i) GSH (0.5 mM and 1.0 mM, respectively) for 30 min, and then 1 (10 μM) for 30 min; (j–l) Cys (0.5 mM) for 30 min, and then 1 (10 μM) for 30 min. (a, d, g and j: green channel; b, e, h, i and l: red channel; c, f and k: blue channel).
discriminates GSH from Cys and Hys but also displays high sensitivity towards concentration levels of GSH, and thus, 1 can be employed to monitor GSH level in living cells.
4. Conclusions In summary, a structurally simple colorimetric and ratiometric fluorescent probe 1 was developed for selective detection and cellular imaging of GSH. The probe exhibits unique selectivity and sensitivity for ratiometric detection of GSH over Cys and Hcy in buffer environments. The probe operates by undergoing a displacement of chloride with thiolate. The discrimination of GSH from Cys/Hcy is attributed to subsequent aldimine condensation of the carbonyl of the probe with the amino group of GSH to form a
macrocyclic iminium cation, which exhibits dramatically different photophysical properties compared with the amino-coumarin derivative produced by the reaction with Cys (Hcy). In addition, the probe is cell-membrane-permeable and has low cytotoxicity. Thus, the probe is suitable for selective ratiometric fluorescent imaging of GSH in living cells. We hope that this new strategy could inspire the exploration of new systems with improved sensitivity and greater ratiometric response for probing biothiols function in biological systems.
Acknowledgments We acknowledge the National Natural Science Foundation of China (No. 21262039).
C. Xu et al. / Biosensors and Bioelectronics 72 (2015) 275–281
Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.bios.2015.05.030.
References Chen., H., Zhao, Q., Wu, Y., Li, F., Yang, H., Yi, T., Huang, C., 2007. Inorg. Chem. 46, 11075–11081. Chen, X., Zhou, Y., Peng, X., Yoon, J., 2010. Chem. Soc. Rev. 39, 2120–2135. Dale, T.J., Rebek Jr., J., 2006. J. Am. Chem. Soc. 128, 4500–4501. Feuster, E.K., Glass, T.E., 2003. J. Am. Chem. Soc. 125, 16174–16175. Fu, L., Jiang, F.-L., Fortin, D., Harvey, P.D., Liu, Y., 2011. Chem. Commun. 47, 5503–5505. Guo, Y., Wang, H., Sun, Y., Qu, B., 2012a. Chem. Commun. 48, 3221–3223. Guo, Y., Yang, X., Hakuna, L., Barve, A., Escobedo, J.O., Lowry, M., Strongin, R.M., 2012b. Sensors 12, 5940–5950. Guo, Z., Nam, S.W., Park, S., Yoon, J., 2012c. Chem. Sci. 3, 2760–2765. Guy, J., Caron, K., Dufresne, S., Michnick, S.W., Skene, W.G., Keillor, J.W., 2007. J. Am. Chem. Soc. 129, 11969–11977. Hu, R., Feng, J., Hu, D., Wang, S., Li, S., Li, Y., Yang, G., 2010. Angew. Chem. Int. Ed. 49, 4915–4918. Hwang, C., Sinskey, A.J., Lodish, H.F., 1992. Science 257, 1496–1502. Jia, M.-Y., Niu, L.-Y., Zhang, Y., Yang, Q.-A., Tung, C.-H., Guan, Y.-F., Feng, L., 2015. ACS Appl. Mater. Interfaces 7, 5907–5914. Jung, H.S., Han, J.H., Pradhan, T., Kim, S., Lee, S.W., Sessler, J.L., Kim, T.W., Kang, C., Kim, J.S., 2012a. Biomaterials 33, 945–953. Jung, H.S., Pradhan, T., Han, J.H., Heo, K.J., Lee, J.H., Kang, C., Kim, J.S., 2012b. Biomaterials 33, 8495–8502. Klockow, J.L., Glass, T.E., 2013. Org. Lett. 15, 235–237. Lee, M.H., Han, J.H., Kwon, P.-S., Bhuniya, S., Kim, J.Y., Sessler, J.L., Kang, C., Kim, J.S., 2012. J. Am. Chem. Soc. 134, 1316–1322. Li, H., Jin, L.-Y., Kan, Y., Yin, B., 2014. Sens. Actuators B: Chem. 196, 546–554. Lim, S.-Y., Hong, K.-H., Kim, D.I., Kwon, H., Kim, H.-J., 2014. J. Am. Chem. Soc. 136, 7018–7025. Liu, J., Sun, Y.-Q., Zhang, H., Huo, Y., Shi, Y., Guo, W., 2014. Chem. Sci. 5, 3183–3188. Liu, Y., Lv, X., Liu, J., Yuan-Qiang Sun, Y.-Q., Guo., W., 2015. Chem. Eur. J. 21, 4747–4754.
281
Ma, L., Qian, J., Tian, H., Lan, M., Zhang, W., 2012. Analyst 137, 5046–5050. Meister, A., 1988. J. Biol. Chem. 263, 17205–17208. Miller, E.W., Bian, S.X., Chang, C.J., 2007. J. Am. Chem. Soc. 129, 3458–3459. Niu, L.-Y., Guan, Y.-S., Chen, Y.-Z., Wu, L.-Z., Tung, C.-H., Yang, Q.-Z., 2012. J. Am. Chem. Soc. 134, 18928–18931. Niu, L.-Y., Guan, Y.-S., Chen, Y.-Z., Wu, L.-Z., Tung, C.-H., Yang, Q.-Z., 2013. Chem. Commun. 49, 1294–1296. Pires, M.M., Chmielewski, J., 2008. Org. Lett. 10, 837–840. Rusin, O., Luce, N.N., Agbaria, R.A., Escobedo, J.O., jiang, S., Warner, I.M., Dawan, F.B., Lian, K., Strongin, R.M., 2004. J. Am. Chem. Soc. 126, 438–439. Secor, K.E., Glass, T.E., 2004. Org. Lett. 6, 3727–3730. Secor, K.E., Plante, J., Avetta, C., Glass, T., 2005. J. Mater. Chem. 15, 4073–4077. Shao, J., Sun, H., Guo, H., Ji, S., Zhao, J., Wu, W., Yuan, X., Zhang, C., James, T.D., 2012. Chem. Sci. 3, 1049–1061. Shao, N., Jin, J., Wang, H., Zheng, J., Yang, R., Chan, W., Abliz, Z., 2010. J. Am. Chem. Soc. 132, 725–736. Sun, Y., Liu, Y., Chen, M., Guo, W., 2009. Talanta 80, 996–1000. Tang, B., Xing, Y., Li, P., Zhang, N., Yu, F., Yang, G., 2007. J. Am. Chem. Soc. 129, 11666–11667. Townsend, D.M., Tew, K.D., Tapiero, H., 2003. Biomed. Pharmacother. 57, 145–155. Wang, F., Li Zhou, L., Zhao, C., Wang, R., Fei, Q., Luo, S., Guo, Z., Tian, H., Zhu, W.-H., 2015. Chem. Sci. 6, 2584–2589. Xu, Y., Li, B., Han, P., Sun, S., Pang, Y., 2013. Analyst 138, 1004–1007. Yang, X.-F., Huang, Q., Zhong, Y., Li, Z., Li, H., Lowry, M., Escobedoc, J.O., Strongin, R. M., 2014. Chem. Sci. 5, 2177–2183. Yang, X., Guo, Y., Strongin, R.M., 2011. Angew. Chem. Int. Ed. 50, 10690–10693. Yang, X., Guo, Y., Strongin, R.M., 2012a. Org. Biomol. Chem. 10, 2739–2741. Yang, Z., Zhao, N., Sun, Y., Miao, F., Liu, Y., Liu, X., Zhang, Y., Ai, W., Song, G., Shen, X., Yu, X., Sun, J., Wong, W.-Y., 2012b. Chem. Commun. 48, 3442–3444. Yin, J., Kwon, Y., Kim, D., Lee, D., Kim, G., Hu, Y., Ryu, J.-H., Yoon, J., 2014. J. Am. Chem. Soc. 136, 5351–5358. Yuan, L., Lin, W., Yang, Y., 2011. Chem. Commun. 47, 6275–6277. Zhang, J., Shibata, A., Ito, M., Shuto, S., Ito, Y., Mannervik, B., Abe, H., Morgenstern, R., 2011. J. Am. Chem. Soc. 133, 14109–14119. Zhang, M., Yu, M., Li, F., Zhu, M., Li, M., Gao, Y., Li, L., Liu, Z., Zhang, J., Zhang, D., Yi, T., Huang, C., 2007. J. Am. Chem. Soc. 129, 10322–10323. Zhang, M., Wu, Y., Zhang, S., Zhu, H., Wu, Q., Jiao, L., Hao, E., 2012. Chem. Commun. 48, 8925–8927. Zhou, X., Jin, X., Sun, G., Li, D., Wu, X., 2012. Chem. Commun. 48, 8793–8795.
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
A colorimetric and ratiometric fluorescent probe for selective detection and cellular imaging of glutathione Chang Xu a, Hongda Li b, Bingzhu Yin a,n a Key Laboratory of Natural Resources of Changbai Mountain & Functional Molecules of the Ministry of Education, Department of Chemistry, Yanbian University, Yanji 133002, PR China b Department of Forensic Chemistry, National Police University of China, Shenyang 110854, PR China
art ic l e i nf o
a b s t r a c t
Article history: Received 12 March 2015 Received in revised form 10 May 2015 Accepted 11 May 2015 Available online 13 May 2015
A new colorimetric and ratiometric fluorescent probe 1 based on a chlorinated coumarinyl aldehyde was developed for selective detection and cellular imaging of glutathione (GSH) over cysteine (Cys) and homocysteine (Hcy). Probe 1 exhibits a dramatic colorimetric and ratiometric fluorescence responses toward biothiols Cys, Hcy and GSH with high selectivity over other amino acids. Cys (or Hcy) induces a tandem SNAr-rearrangement reaction to form the corresponding amino-coumarins (2a or 2b), which result in about 75 nm and 35 nm blue-shifts in absorption and emission, respectively. By comparison, the thio-coumarin (3′) derived from the SNAr reaction with GSH, which does not occur rearrangement because of steric hindrance, undergoes an intramolecular aldimine condensation lead to a cyclic iminium cation (3) with 47 nm and 39 nm red-shifts in absorption and emission, respectively. The significantly difference of photophysical properties enable excellent selectivity towards GSH over Cys and Hcy. Further application to cellular imaging indicates that the probe has appreciable cell permeability and is highly responsive to the changes of GSH level. As a result, it is applicable to monitor GSH level in living cells. & 2015 Elsevier B.V. All rights reserved.
Keywords: Ratiometric fluorescent probe Glutathione detection Cysteine Homocysteine chlorinated coumarinyl aldehyde Cell imaging
1. Introduction Glutathione (GSH; γ-glutamylcysteinylglycine) is ubiquitous in mammalian and many prokaryotic cells and is the most abundant intracellular thiol (1–10 mM) (Meister, 1988). It is an essential endogenous antioxidant that plays a central role in cellular and defenses against toxins and free radicals (Miller et al., 2007). Aberrant levels of GSH have been associated with a number of diseases, including cancer, AIDS, Alzheimer's, cardiovascular disease and others (Townsend et al., 2003). Therefore, the accurate concentration measurement of GSH in physiological media has been considered as an essential factor in these diseases and their therapy because of its biological and clinical significance. Because of this, the considerable contemporary effort devoted to the development of an efficient method for the detection and quantification of GSH in living systems (Chen et al., 2010). Among the various analytical methods that are available, molecular imaging based on fluorescent probes is considered to be the most sensitive approach owing to its sensitivity and simplicity. Most of the existing probes utilize the strong nucleophilicity of the thiol group, operating by Michael addition (Guy et al., 2007), cleavage of n
Corresponding author. Fax: þ 86 433 2732456. E-mail address: [email protected] (B. Yin).
http://dx.doi.org/10.1016/j.bios.2015.05.030 0956-5663/& 2015 Elsevier B.V. All rights reserved.
disulfide (Pires and Chmielewski, 2008; Lee et al., 2012) and sulfonamide (Zhang et al., 2011; Shao et al., 2012), etc (Tang et al., 2007; Zhang et al., 2007). However, although these probes can highly selectively distinguish these biothiols from other amino acids, most of them cannot distinguish Cys/Hcy/GSH from each other due to their similar structure and reactivity. Up to now, the discrimination between them has been a focal point and also a tough challenge for researchers, albeit some advances have been obtained. By means of the cyclization of Cys/Hcy with aldehydes or acrylates, selective detection of Cys/Hcy over GSH was firstly achieved by the Strongin group (Rusin et al., 2004; Yang et al., 2011). Since then, as the extended version of the two strategies, some more specific probes for Cys or Hcy were also developed (Chen. et al. 2007; Li et al., 2014; Yuan et al., 2011; Yang et al., 2012b; Guo et al., 2012a; Yang et al., 2012a; Guo et al., 2012c). Recently, discrimination of Cys from Hcy/GSH was also achieved by taking advantage of either the Cys-induced substitution-rearrangement cascade reaction (Ma et al., 2012; Niu et al., 2013) or Michael addition combined with steric and electrostatic interactions (Jung et al., 2012a; Jung et al., 2012b; Zhou et al., 2012; Zhang et al., 2012). However, probes capable of discrimination of GSH from Cys/Hcy are still rare. As far as we know, only several strategies have been reported, including supramolecular interaction between a bis-spiropyran receptor and GSH (Shao et al., 2010), a micelle-catalyzed large ring formation of GSH with acrylate (Guo
276
C. Xu et al. / Biosensors and Bioelectronics 72 (2015) 275–281
standard procedures. Fluorescence spectra were carried out on a Shimadzu RF-5301PC fluorescence spectrophotometer. UV/vis spectra were recorded with a Shimadzu UV-2550 spectrophotometer. NMR spectra were recorded on a Bruker AV-300 spectrometer (300 MHz for 1H and 75 MHz for 13C), and chemical shifts were referenced relative to tetramethylsilane. Mass data were obtained by a Shimadzu AXIMA-CFR™ plus mass spectrometry using a 1, 8, 9-anthracenetriol (DITH) matrix. PBS buffer (10 mM, pH 7.4, 10% DMSO) prepared with deionized water and spectroscopic grade of DMSO were used as the solvent for all spectroscopic experiments. Stock solution of the probe (1 10 3 M) was prepared in DMSO, which was diluted to 1 10 5 M with the PBS buffer. Stock solution of the various amino acids were prepared with the PBS buffer for future use. For all fluorescence measurements, the excitations were at 476 nm for GSH and 385 nm for Cys and Hcy, and the excitation and emission slit widths were 3 and 5 nm, respectively.
et al., 2012b), a specific reaction of GSH with o-phthalaldehyde (Xu et al., 2013), and those GSH-induced tandem reactions, such as SNAr-rearrangement (Niu et al., 2012; Liu et al., 2014; Jia et al., 2015), SNAr-rearrangement-cyclization (Liu et al., 2014; Liu et al., 2015), and nucleophilic substitution (Lim et al., 2014; Yin et al., 2014) as well as a native chemical ligation (NCL) reaction (Yang et al., 2014). Very recently, a novel strategy for selectively sensing GSH by a dual-response mechanism was reported (Wang et al., 2015). This strategy integrates two independent reaction sites with a disulfide linker and a thioether function into a fluorescent BODIPY-based probe to guarantee the synergetic dual-response in an elegant fashion to address the discrimination of GSH. Even so, there still are some limitations when considering the practical applications in biological systems, such as use of organic solvent or surfactant, long response time and require high equivalents of GSH to reach a maximal fluorescent signal, relatively poor selectivity, undesired spectra overlap, and so on. Furthermore, almost all of them are based on fluorescence measurement at a single wavelength, which means that GSH detection only depends on the changes of emission intensity, and could be significantly influenced by the excitation power and the detector sensitivity. Moreover, fluorescent intensity changes are not easy to observe by the naked eye directly (Hu et al., 2010). By contrast, ratiometric fluorescent probes allow the measurement of emission intensities at two wavelengths, which should provide a built-in correction for environmental effects and the fluorescence color change, which can be measured directly with a colorimeter or even distinguished easily by eye (Fu et al., 2011). Therefore, it is of high interest to develop new ratiometric fluorescent probes for GSH, in particular, with rapid response and high sensitivity. With these considerations in mind, we developed a new colorimetric and ratiometric fluorescent probe 1 for discrimination of GSH over Cys and Hcy based on a chlorinated coumaryl aldehyde (Scheme 1). As we expected, probe 1 displayed a high selectivity for detection of GSH over Cys and Hcy based on a novel GSH-induced SNAr-aldimine condensation strategy and that it can be used to monitor GSH in cells.
HeLa cells were obtained from American Type Culture collection and grown in Dulbecco's modification of Eagle's medium Dulbecco (Free DMEM/high: with 4500 mg/L Glucose, 4.0 mM LGlutamine, and 110 mg/L Sodium Pyruvate). The cells were incubated in a 5% CO2 humidified incubator at 37 °C and typically passaged with sub-cultivation ratio of 1:4 for two days. The HeLa cells were seeded in 6-well culture plate overnight. Stock solutions of probe (3 mM) and NEM (0.15 M) in DMSO, GSH, Cys (0.15 M) in PBS buffer (10 mM, pH 7.4) were prepared at the same day of experiment. Then, the cells were treated without or with GSH (0.1, 0.2, 0.5, 1.0, and 2.0 mM) or Cys (0.5 mM) for 30 min and NEM (0.1 mM) for 20 min at 37 °C, respectively, in culture media. After washing with phosphate buffered saline to remove the remaining GSH or Cys, the cells were further incubated with 10 mM of probe in the culture media for 30 min at 37 °C. The cells were imaged using an inverted microscope (OLYMPUS IX73) and Leica SP5II confocal microscope.
2. Experimental
3. Results and discussion
2.1. Materials and apparatus
3.1. Design concept of the probe
Commercially available compounds were used without further purification. The reaction solvents were dried according to
The design rationale is depicted in Scheme 1, and illustrated as follows. Chlorine lies in 4-position of probe 1, where is doubly
2.2. Methods for cell culture and fluorescent imaging
Scheme 1. Proposed reaction mechanisms of probe 1 with Cys, Hcy and GSH.
C. Xu et al. / Biosensors and Bioelectronics 72 (2015) 275–281
277
activated and has been proved to be subject to the thiol-halogen SNAr reaction (Liu et al., 2014). Thus, it was expected that the chlorine atom of 1 could initially be replaced by thiol group of Cys (or Hcy) to produce corresponding thio-coumarins 2a′ (or 2b′), and the subsequent intramolecular rearrangement would lead to amino-coumarins 2 (Scheme 1). We speculated that the thiol group of 2 is hard to further attack the adjacent carbonyl group due to the unstable macrocyclic hemi-thioacetal product and the HSAB effect. In the case of GSH, the initial thiol-halogen SNAr reaction would similarly lead to thio-coumarin 3′ (Scheme 1), which does not occur rearrangement due to the bulkiness of its tripeptide and the unstable 10-membered macrocyclic transition state. 3′ would undergo a intramolecular aldimine condensation leading to a cyclic iminium cation 3 that enhances the ICT and modulates both the absorbance and fluorescence. Given the distinct photophysical properties of the amino-coumarin and the cyclic iminium cation, a ratiometric colorimetric and fluorescent dual response to GSH with high selectivity is expected using this probe. 3.2. Colorimetric and ratiometric response towards GSH, Cys and Hcy To realize the above design concept, we synthesized probe 1 according to the previously reported method by formylation of 7-diethylamino-4-hydroxycoumarin under Vilsmeier reaction condition (Supporting information). With probe 1 in hand, we explored the water solubility initially. Fortunately, probe 1 was observed to display satisfactory water solubility. However, the sensing processes of the probe toward thiols were more time consuming when the water content was more than 90%. Thus, we selected a pH 7.4 PBS buffer solution containing 10% (v/v) DMSO for further experiments. Next, we examined the reactivity of 1 towards Cys, Hcy and GSH through UV–vis spectra. As shown in Fig. 1a, the UV–vis spectrum of free 1 shows a main absorption at 457 nm (Klockow et al., 2013). Upon addition of 10 equiv. of Cys, the initial absorption peak decreased, along with the emergence of a blue-shifted peak at 380 nm. According to the aforementioned speculation and the well-established tandem SNAr-rearrangement reaction mechanism (Niu et al., 2012), the absorption at 380 nm should be assigned to the amino-coumarin (2a), which was further supported by both the data of MALDI-TOF mass experiment (Fig. S1) and control compound A and B (Scheme S1). The absorption maximums of A and B were found at around 376 (Fig. S2) and 377 nm, respectively, which very close to that of 2a. Interestingly, in the titration process of Cys we have observed the corresponding absorption band of the intermediate (2a′). As seen in Fig. 1b, when increasing amount of Cys was added to the solution of probe 1, the maximum absorption peak centered at 457 nm decreased and red-shifted gradually to 475 nm and disappeared finally, which should be assigned to the thio-coumarin (2a′). The assign could be supported by control compound C (Absmax: 473 nm, Fig. S3) and D (Absmax: 468 nm) (Scheme S1). A similar observation was made for Hcy under same experimental condition. By contrast, addition of GSH to the solution of 1 led to the decrease of the initial absorption band at 457 nm followed by the emergence of two red-shifted peaks at 490 and 517 nm (Fig. 1a). This represents an unprecedented red-shift of 60 nm compared to the 35 nm shift that is typical for the coumarin aldehyde class of sensors (Feuster et al., 2003; Secor and Glass, 2004; Secor et al., 2005). Note that the absorptions at 490 and 517 nm was initially assigned to thio-coumarin 3′. However, the assign did not match the absorption spectra of the control compound C and D, which is much shorter than that of our reaction product. Fortunately, this unprecedented red-shift makes us to mind the hydrogen-bonding iminium ion 3, which would enhance the ICT and modulate both
Fig. 1. (a) Absorbance changes of probe 1 (10 μM) 30 min after addition of 5 equiv. of various biothiols (inset: color changes of the solution of 1 in the presence of 5 equiv. of different biothiols and (b) upon addition of increasing concentrations of Cys (0–2 equiv.).
the absorbance and fluorescence. In fact, the two new absorption peaks are very similar to those of the previously reported the hydrogen-bonding iminium ions E and F (Absmax: 525 nm) (Scheme S1), and can thus be assigned to 3. Importantly, the formation of the iminium ion 3 was evidenced by mass spectrum analysis of the product generated from the incubation of 1 with GSH in CH3CN–H2O. Two prominent peaks at m/z 533.90 and 489.9 corresponding to [3 þ H] þ (calcd. 532.16 for C24H28N4O8S) and [3 – CO2 þ H] þ (calcd. 488.2 for C23H28N4O6S) are clearly observed in the MALDI-TOF mass spectrum (Fig. S4), which provides strong evidence for aforementioned assign. Notably, the reaction also likely led to a spot of the by-product amino-coumarin as a result of the subsequent SNAr reaction between 3′ and excess GSH, which could be indicated by the newly emerged absorption peak around 379 nm in the titration process (Fig. S5). In a word, these results are in good accordance with our proposed reaction mechanism. Encouraged by the above results, subsequently, we examined the emission behavior of 1 upon addition of Cys, Hcy and GSH by use of two different excitations at 385 nm and 476 nm, respectively. Firstly, we selected the excitation wavelength at 385 nm to probe Cys. Upon 385 nm excitation, probe 1 is almost non-emissive. Addition of 10 equiv. of Cys to a solution of 1 elicited a big fluorescence enhancement (18-fold) at 477 nm due to the
278
C. Xu et al. / Biosensors and Bioelectronics 72 (2015) 275–281
Fig. 2. Fluorescence response of probe 1 (10 μM) to 5 equiv. of various biothiols with λex ¼476 nm (a) and λex ¼385 nm (b). Inset: the fluorescence photographs of 1 in the presence of GSH, Cys and Hcy exposed to a UV lamp at 365 nm.
production of amino-coumarin 2a (Fig. 2b). Despondently, under this excitation wavelength, addition of 10 equiv. of GSH similarly emerges a minor emission around 477 nm, which might be attributed to the amino-coumarin derivative. This phenomenon is in accordance with the results of UV spectral titration experiment. Next, we selected the excitation wavelength at 476 nm to probe GSH, which near the isosbestic point (Fig. 1a). As shown in Fig. 2a, probe 1 emits a green fluorescence with an emission maximum at 507 nm. Upon addition of 10 equiv. of GSH, the initial emission at 507 nm disappeared along with the emergence of a red-shifted emission at 546 nm, indicating a clear ratiometric fluorescence response. As mentioned before, this should be attributed to the production of the iminium ion 3. In contrast, addition of Cys/Hcy caused hardly any significant fluorescence changes in probe 1, indicating the strong ability of probe 1 to discriminate GSH from Cys/Hcy at the excitation wavelength at 476 nm, and thus it can serve as a selective ratiometric fluorescence probe for GSH in living cells. In addition, GSH can be easily distinguished from Cys/Hcy by the naked eye when illuminated by a hand-held UV lamp at 365 nm. Introduction of GSH to the solution of 1 results in a color change from yellow to red–orange and a fluorescence change from green to orange. By contrast, Cys (or Hcy) induces a color change from yellow to colorless and a fluorescence change from green to blue. These can clearly be perceived by naked eyes (Fig. 1a, inset and Fig. 2a, inset). Therefore, the colorimetric probe 1 can serve as
Fig. 3. (a) Fluorescence changes of probe 1 (10 μM) on the incremental addition of GSH, and (b) Job's plot between 1 and GSH based on data of absorption spectra. [1] þ [GSH] ¼ 40 μM. Inset: ratio of the fluorescence intensities at 546 and 507 nm (I546/I507) as a function of the GSH concentration.
a micromolar level “naked-eye” indicator for GSH. Subsequently, we performed the spectroscopic titration experiments under the same conditions. Upon treatment with the increasing concentrations of GSH, the emissions at 507 nm decreased gradually along with the simultaneous emergence and increase of the red-shifted emission at 546 nm, indicating a clear ratiometric fluorescence response as seen in Fig. 3a. Similarly, this should be attributed to the production of the iminium 3. Notably, the maximal spectral signals in both absorption and emission spectra were achieved in the presence of only 1 equiv. of GSH (Figs. S5 and S6). Moreover, the ratio of emission intensities (I546/I507) varies from 0.29 to 2.09 with a 7.2-fold ratiometric enhancement (Fig. 3a, inset). Time-dependent ratiometric changes in the fluorescence spectra showed that 10 equiv. of GSH is sufficient to complete the sensing reaction in about 20 min under the experimental condition (Fig. S7), and the observed rate constant (Kobs) (Dale et al., 2006) for sensing reaction was determined to be 0.1447 0.004 min 1 under pseudo-first-order kinetic conditions (Fig. S8). In addition, a Job's plot exhibits a maximum at 0.5 molecular fraction, which indicates that the reaction between 1 and GSH has a 1:1 stoichiometry (Fig. 3b). The result is in accordance with the data of mass experiment. More importantly, the fluorescence intensity ratio (I546/I507) was linearly proportional to the
C. Xu et al. / Biosensors and Bioelectronics 72 (2015) 275–281
279
virtually no influence on the fluorescence detection of GSH except for Cys and Hcy. Cys or Hcy might form stable amino-substituted covalent complexes with probe 1 through a sequential thiol substitution and rearrangement reaction as observed in Yang's BODIPY-based probe (Niu et al., 2012). These inspiring results suggest that probe 1 can be established as promising prototypes of ratiometric fluorescence GSH probe in actual living cells. 3.4. The effect of pH The fluorescent response of probe 1 toward GSH in PBS buffer (10 mM, 10%, DMSO) at different pH conditions were further investigated to apply probe 1 in complicated systems, such as environmental or biological systems. The results indicate that the response of probe 1 towards GSH was pH dependent. The fluorescence spectra of probe 1 had a relatively weak emission band at 546 nm and remained unaffected between pH levels 2 and 10, as shown in Fig. S10. Upon addition of GSH, the maximal fluorescence signals were observed in the pH range of 6–9 without the interference by protons, and the stable fluorescence of the probe at a pH level of approximately 7.4 was favorable for the sensing assays of GSH in environmental and biological samples. 3.5. Ratiometric cell imaging
Fig. 4. (a) Fluorescence response (λex ¼476 nm) of probe 1 (10 μM) to 5 equiv. of various amino acids and (b) changes in fluorescence intensity of 1 (10 μM) toward GSH (2 equiv.) in the presence of the competing analytes.
GSH concentration in the 0–10 μM range (R2 ¼0.998), indicating the suitability of 1 for quantitative detection of GSH. The detection limit (Sun et al., 2009 and 1.2. detection limit in Supporting information) is determined to be 0.38 μM (Fig. 3a, inset), which is significantly below the physiological levels of GSH in live cells (1– 10 mM) (Hwang et al., 1992). This result is indeed important in view of the very few fluorescent probes that could selectively and quantificationally detect GSH over Cys/Hcy in the literature so far. 3.3. Selective response of probe 1 to GSH To evaluate the specific nature of probe 1 for GSH, the spectral changes in probe 1 toward other amino acids and the biologically related analytes were explored. Compared with GSH, no obvious changes in absorption and emission spectra were observed upon addition of other amino acids and the related analytes, indicating probe 1 is highly selective to GSH over the other biologically relevant analytes tested (Figs. 4a and S9). On the other hand, whether the probe 1 could still retain the selective response to GSH in the presence of the competing analytes is very important for a fluorescent probe. Thus, competition experiments of probe 1 were conducted. As shown in Fig. 4b, probe 1 was treated with GSH in the presence of the competing analytes (10 equiv). All of the competing analytes tested have
For biological applications, the cytotoxicity in HeLa cells was firstly determined by the MTT assay with different concentrations of the probe (Supporting information). Fig. S11 shows the cell viability after incubation with probe 1 at the concentrations of 0 μM, 1 μM, 2.5 μM, 5 μM and 10 μM for 24 h, respectively. HeLa cell viabilities of more than 95% were observed during the tested time, indicative of low cytotoxicity of probe 1. Finally, we studied the capacity of probe 1 for ratiometric imaging of GSH in living cells. When HeLa cells were incubated with 1 (10 μM) for 20 min and then washed with PBS buffer, a bright fluorescence in the green channel, and weak fluorescence in both red channel and blue channel were observed simultaneously, indicating that the probe is cell-permeable (Fig. 5a–c) and is responsive to intracellular Cys and GSH. However, when the cells were pretreated with 0.1 mM N-ethylmaleimide (NEM), which is a trapping reagent for thiol species, and then treated with probe 1 (10 μM), the fluorescence in the red and blue channels are both decreased slightly while the fluorescence in green channel did not caused any detectable fluorescence suppression (Fig. 5d–f), suggesting that ratiometric imaging of 1 is responsive to changing intracellular GSH and Cys levels in the blue and red channels, respectively. Furthermore, when HeLa cells were pre-treated with 0.5 mM GSH, a significant increase in red emission and a minor decrease in green emission were observed (Fig. 5g and h). When higher concentrations of GSH (1.0 mM) were employed to image the HeLa cells, an obvious increase in red emission was found as seen in Fig. 5i, which implies that 1 is highly sensitive to concentration levels of GSH even in living cells. Indeed, concentrationdependent ratiometric intensity changes in confocal fluorescent images was linearly proportional to the GSH concentration (Figs. S12 and S13), indicating the suitability of the probe for quantitative detection of GSH in biological samples. To explore the specific nature of probe 1 toward GSH in living cells, HeLa cells were pretreated with Cys (0.5 mM), and then treated with probe 1 (10 μM). As expected, a marked enhancement in blue emission and a minor reduction in green emission were observed, respectively, compared with that of HeLa cells incubated with probe 1 only (Fig. 5j and k). More meaningfully, the fluorescence in red channel decreased slightly rather than increased (Fig. 5l), conforming that probe 1 is specific to GSH detection over Cys/Hcy in actual living cells. These results showed that the probe 1 not only
280
C. Xu et al. / Biosensors and Bioelectronics 72 (2015) 275–281
Fig. 5. Fluorescence images of living HeLa cells. (a–c) 1 (10 μM) for 30 min; (d–f) NEM (0.1 mM) for 20 min, and then 1 (10 μM) for 30 min; (g–i) GSH (0.5 mM and 1.0 mM, respectively) for 30 min, and then 1 (10 μM) for 30 min; (j–l) Cys (0.5 mM) for 30 min, and then 1 (10 μM) for 30 min. (a, d, g and j: green channel; b, e, h, i and l: red channel; c, f and k: blue channel).
discriminates GSH from Cys and Hys but also displays high sensitivity towards concentration levels of GSH, and thus, 1 can be employed to monitor GSH level in living cells.
4. Conclusions In summary, a structurally simple colorimetric and ratiometric fluorescent probe 1 was developed for selective detection and cellular imaging of GSH. The probe exhibits unique selectivity and sensitivity for ratiometric detection of GSH over Cys and Hcy in buffer environments. The probe operates by undergoing a displacement of chloride with thiolate. The discrimination of GSH from Cys/Hcy is attributed to subsequent aldimine condensation of the carbonyl of the probe with the amino group of GSH to form a
macrocyclic iminium cation, which exhibits dramatically different photophysical properties compared with the amino-coumarin derivative produced by the reaction with Cys (Hcy). In addition, the probe is cell-membrane-permeable and has low cytotoxicity. Thus, the probe is suitable for selective ratiometric fluorescent imaging of GSH in living cells. We hope that this new strategy could inspire the exploration of new systems with improved sensitivity and greater ratiometric response for probing biothiols function in biological systems.
Acknowledgments We acknowledge the National Natural Science Foundation of China (No. 21262039).
C. Xu et al. / Biosensors and Bioelectronics 72 (2015) 275–281
Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.bios.2015.05.030.
References Chen., H., Zhao, Q., Wu, Y., Li, F., Yang, H., Yi, T., Huang, C., 2007. Inorg. Chem. 46, 11075–11081. Chen, X., Zhou, Y., Peng, X., Yoon, J., 2010. Chem. Soc. Rev. 39, 2120–2135. Dale, T.J., Rebek Jr., J., 2006. J. Am. Chem. Soc. 128, 4500–4501. Feuster, E.K., Glass, T.E., 2003. J. Am. Chem. Soc. 125, 16174–16175. Fu, L., Jiang, F.-L., Fortin, D., Harvey, P.D., Liu, Y., 2011. Chem. Commun. 47, 5503–5505. Guo, Y., Wang, H., Sun, Y., Qu, B., 2012a. Chem. Commun. 48, 3221–3223. Guo, Y., Yang, X., Hakuna, L., Barve, A., Escobedo, J.O., Lowry, M., Strongin, R.M., 2012b. Sensors 12, 5940–5950. Guo, Z., Nam, S.W., Park, S., Yoon, J., 2012c. Chem. Sci. 3, 2760–2765. Guy, J., Caron, K., Dufresne, S., Michnick, S.W., Skene, W.G., Keillor, J.W., 2007. J. Am. Chem. Soc. 129, 11969–11977. Hu, R., Feng, J., Hu, D., Wang, S., Li, S., Li, Y., Yang, G., 2010. Angew. Chem. Int. Ed. 49, 4915–4918. Hwang, C., Sinskey, A.J., Lodish, H.F., 1992. Science 257, 1496–1502. Jia, M.-Y., Niu, L.-Y., Zhang, Y., Yang, Q.-A., Tung, C.-H., Guan, Y.-F., Feng, L., 2015. ACS Appl. Mater. Interfaces 7, 5907–5914. Jung, H.S., Han, J.H., Pradhan, T., Kim, S., Lee, S.W., Sessler, J.L., Kim, T.W., Kang, C., Kim, J.S., 2012a. Biomaterials 33, 945–953. Jung, H.S., Pradhan, T., Han, J.H., Heo, K.J., Lee, J.H., Kang, C., Kim, J.S., 2012b. Biomaterials 33, 8495–8502. Klockow, J.L., Glass, T.E., 2013. Org. Lett. 15, 235–237. Lee, M.H., Han, J.H., Kwon, P.-S., Bhuniya, S., Kim, J.Y., Sessler, J.L., Kang, C., Kim, J.S., 2012. J. Am. Chem. Soc. 134, 1316–1322. Li, H., Jin, L.-Y., Kan, Y., Yin, B., 2014. Sens. Actuators B: Chem. 196, 546–554. Lim, S.-Y., Hong, K.-H., Kim, D.I., Kwon, H., Kim, H.-J., 2014. J. Am. Chem. Soc. 136, 7018–7025. Liu, J., Sun, Y.-Q., Zhang, H., Huo, Y., Shi, Y., Guo, W., 2014. Chem. Sci. 5, 3183–3188. Liu, Y., Lv, X., Liu, J., Yuan-Qiang Sun, Y.-Q., Guo., W., 2015. Chem. Eur. J. 21, 4747–4754.
281
Ma, L., Qian, J., Tian, H., Lan, M., Zhang, W., 2012. Analyst 137, 5046–5050. Meister, A., 1988. J. Biol. Chem. 263, 17205–17208. Miller, E.W., Bian, S.X., Chang, C.J., 2007. J. Am. Chem. Soc. 129, 3458–3459. Niu, L.-Y., Guan, Y.-S., Chen, Y.-Z., Wu, L.-Z., Tung, C.-H., Yang, Q.-Z., 2012. J. Am. Chem. Soc. 134, 18928–18931. Niu, L.-Y., Guan, Y.-S., Chen, Y.-Z., Wu, L.-Z., Tung, C.-H., Yang, Q.-Z., 2013. Chem. Commun. 49, 1294–1296. Pires, M.M., Chmielewski, J., 2008. Org. Lett. 10, 837–840. Rusin, O., Luce, N.N., Agbaria, R.A., Escobedo, J.O., jiang, S., Warner, I.M., Dawan, F.B., Lian, K., Strongin, R.M., 2004. J. Am. Chem. Soc. 126, 438–439. Secor, K.E., Glass, T.E., 2004. Org. Lett. 6, 3727–3730. Secor, K.E., Plante, J., Avetta, C., Glass, T., 2005. J. Mater. Chem. 15, 4073–4077. Shao, J., Sun, H., Guo, H., Ji, S., Zhao, J., Wu, W., Yuan, X., Zhang, C., James, T.D., 2012. Chem. Sci. 3, 1049–1061. Shao, N., Jin, J., Wang, H., Zheng, J., Yang, R., Chan, W., Abliz, Z., 2010. J. Am. Chem. Soc. 132, 725–736. Sun, Y., Liu, Y., Chen, M., Guo, W., 2009. Talanta 80, 996–1000. Tang, B., Xing, Y., Li, P., Zhang, N., Yu, F., Yang, G., 2007. J. Am. Chem. Soc. 129, 11666–11667. Townsend, D.M., Tew, K.D., Tapiero, H., 2003. Biomed. Pharmacother. 57, 145–155. Wang, F., Li Zhou, L., Zhao, C., Wang, R., Fei, Q., Luo, S., Guo, Z., Tian, H., Zhu, W.-H., 2015. Chem. Sci. 6, 2584–2589. Xu, Y., Li, B., Han, P., Sun, S., Pang, Y., 2013. Analyst 138, 1004–1007. Yang, X.-F., Huang, Q., Zhong, Y., Li, Z., Li, H., Lowry, M., Escobedoc, J.O., Strongin, R. M., 2014. Chem. Sci. 5, 2177–2183. Yang, X., Guo, Y., Strongin, R.M., 2011. Angew. Chem. Int. Ed. 50, 10690–10693. Yang, X., Guo, Y., Strongin, R.M., 2012a. Org. Biomol. Chem. 10, 2739–2741. Yang, Z., Zhao, N., Sun, Y., Miao, F., Liu, Y., Liu, X., Zhang, Y., Ai, W., Song, G., Shen, X., Yu, X., Sun, J., Wong, W.-Y., 2012b. Chem. Commun. 48, 3442–3444. Yin, J., Kwon, Y., Kim, D., Lee, D., Kim, G., Hu, Y., Ryu, J.-H., Yoon, J., 2014. J. Am. Chem. Soc. 136, 5351–5358. Yuan, L., Lin, W., Yang, Y., 2011. Chem. Commun. 47, 6275–6277. Zhang, J., Shibata, A., Ito, M., Shuto, S., Ito, Y., Mannervik, B., Abe, H., Morgenstern, R., 2011. J. Am. Chem. Soc. 133, 14109–14119. Zhang, M., Yu, M., Li, F., Zhu, M., Li, M., Gao, Y., Li, L., Liu, Z., Zhang, J., Zhang, D., Yi, T., Huang, C., 2007. J. Am. Chem. Soc. 129, 10322–10323. Zhang, M., Wu, Y., Zhang, S., Zhu, H., Wu, Q., Jiao, L., Hao, E., 2012. Chem. Commun. 48, 8925–8927. Zhou, X., Jin, X., Sun, G., Li, D., Wu, X., 2012. Chem. Commun. 48, 8793–8795.
