Talanta 132 (2015) 191–196
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
A sensitive and selective chemosensor for ascorbic acid based on a fluorescent nitroxide switch Tian Yang a, Baozhan Zheng a, Hengxing Liang c, Yuping Wan c, Juan Du a,n, Dan Xiao a,b,n a
College of Chemistry, Sichuan University, No. 29 Wangjiang Road, Chengdu, PR China College of Chemical Engineering, Sichuan University, No. 29 Wangjiang Road, Chengdu, PR China c Chengdu Product Quality Supervision and Inspection Institute, Chengdu 610041, PR China b
art ic l e i nf o
a b s t r a c t
Article history: Received 15 June 2014 Received in revised form 20 August 2014 Accepted 27 August 2014 Available online 6 September 2014
Ascorbic acid (AsA), also known as vitamin C, is a vital small-molecule antioxidant with multiple functions in vivo. It’s the major natural antioxidant found in plants and is also an essential component of human nutrition. AsA plays a key role in many diseases-related biological metabolism. Therefore, sensitive and selective detection of AsA is greatly important in pharmaceutical, clinical and food industry. Here a sensitive and selective sensor for ascorbic acid detection based on the recovered fluorescence of NAPS-NO (N-propyl-triethoxysilane-4-(4-ylamino-1-oxy-2,2,6,6-tetramethylpiperdine)naphthalimide) probe is described. The fluorescence of the naphthalimide moiety of NAPS-NO is inhibited by the nitroxide group, which is covalently linked to the fluorophore. Then, ascorbic acid reacts rapidly with the nitroxide moiety of NAPS-NO to form hydroxylamine, and the fluorescence properties of the naphthalimide moiety are recovered and the ESR signal decayed. Over a wide range from 80 nM to 50 μM, a good linear relationship between the fluorescence intensity and the concentration of ascorbic acid was found and the detection limit was estimated to be as low as 20 nM. To confirm the practical usefulness of the fluorophore–nitroxide probe, we demonstrated the use of NAPS-NO for the measurement of AsA in human blood serum and also successfully determined the concentration of AsA in HEK 293 cell lysate. Results from confocal laser scanning microscopy experiments demonstrated that this chemosensor is cell permeable and can be used as a fluorescent probe for monitoring ascorbic acid in living cells. & 2014 Published by Elsevier B.V.
Keywords: Chemosensor Nitroxide Fluorescence AsA
1. Introduction Ascorbic acid (AsA), which cannot be synthesized in the human body, is used in large scale as an antioxidant in food, animal feed, beverages, pharmaceutical formulations and cosmetic applications [1]. AsA is the major natural antioxidant found in plants and is also an essential component of human nutrition [2,3]. It plays a key role in biological metabolism and has been commonly used for the prevention and treatment of common cold, mental illness, infertility, cancer and AIDS, etc [4]. Thus, the development of a simple and rapid method for the determination of AsA with high selectivity and sensitivity is desirable for diagnostic and food safe applications [5]. A wide variety of analytical methods is available for the determination of ascorbic acid, such as titrimetric [6,7], spectrophotometric
n Corresponding authors at: College of Chemical Engineering, Sichuan University, No. 29 Wangjiang Road, Chengdu, PR China. Tel.: þ 86 28 85415029; fax: þ86 28 85416029. E-mail addresses: [email protected] (J. Du), [email protected] (D. Xiao).
http://dx.doi.org/10.1016/j.talanta.2014.08.066 0039-9140/& 2014 Published by Elsevier B.V.
[8,9] and chromatographic [10–12] methods. These methods rely on different phenomena and registration techniques suffering from several limitations. The spectrophotometric methods measure either the resultant color of products of certain reactions involving ascorbate [13] or absorbance of ascorbic acid itself [14]. However, the low selectivity of the reactions involved decreases the accuracy of the ascorbate analysis. Thereby, the reliability of these methods is limited by relatively low specificity. A number of specific assays for reductants including ascorbic acid, uric acid, tocopherol etc. have been developed based on high performance liquid chromatography (HPLC) [15,16]. These techniques require several complicated preliminary steps. In addition, HPLC separations would consume relatively long time, therefore, could not be considered as an express-method. Early methods utilizing nitroxide radicals for ascorbate quantifications were based on measurements of only ESR signal from nitroxide, therefore, suffered from general drawbacks of ESR spectroscopy. Although this technique permits investigations in high optical density substance, the ESR measurements are limited by high cost and complexity of equipments and difficulties in acquiring and processing experimental data. At present, electrochemical approaches become
192
T. Yang et al. / Talanta 132 (2015) 191–196
the major analytical methods for AsA determination because of their convenience and speediness. Nevertheless, the interference from other similar redox potential molecules can be serious, such as uric acid and dopamine [17]. What’s more, the detection limit is always higher than that of fluorescence methods. Particularly, attempts have been made to detect ascorbic acid using nanoparticles. Wang et al. [17] developed a simple fluorescence sensor for sensitive turn-off detection of ascorbic acid by using protein-modified Au nanoclusters in aqueous media. Zhang et al. [18] fabricated an electrochemical sensor for selective detection of ascorbic acid in the presence of dopamine and uric acid by modifying the glassy carbon electrode with carbon-supported NiCoO2(NiCoO2/C) nanoparticles. Wang et al. [19] prepared a palladium nanoparticle/graphene/chitosan/glassy carbon electrode which displayed excellent electrochemical catalytic activities towards ascorbic acid, dopamine and uric acid. But these methods have relatively high detection limit towards ascorbic acid. Fluorometric sensing is a preferable approach for detection of AsA due to its comparatively high selectivity, sensitivity, rapidity, operational simplicity, nondestructive methodology and direct visual perception [20]. In recent years, many fluorescent chemosensors were designed for ascorbic acid. Matsuoka et al. [21] synthesised several fluorescent nitroxide compounds based on naphthalene which were sensitive to ascorbic acid. They also demonstrated the use of NaphDiPy nitroxide for the measurement of ascorbic acid in the plasma of osteogenic disorder Shionogi rats to confirm the practical usefulness of the fluorophore–nitroxide probe. Zheng et al. [22] demonstrated a fluorescent carbon dot probe, its fluorescence was quenched by Cr(VI) through inner filter effect, but Cr(VI) can be reduced to low valent chromium species easily by ascorbic acid, and then resulting in the elimination of the IFE and recovery of CD fluorescence. Therefore the CD Cr(VI) mixture could behave as an off-on type fluorescent probe for ascorbic acid. Biqing et al. [23] developed a fluorescent turn-on detection for ascorbic acid based on hyperbranched conjugated polyelectrolyte. The fluorescence was superquenched by Cu2 þ ions, upon addition of ascorbic acid, the paramagnetic Cu2 þ ions can be transformed into diamagnetic Cu1 þ ions, which inhibit the quenching and thus lead to fluorescence enhancement of the hyperbranched conjugate polymer. These fluorescent probes all have sensitive response to ascorbic acid, but none of them were used in intracellular environment. Fluorophore-linked nitroxides represent examples of such fluorescent sensors. The standard redox potential of nitroxides is high enough (E1 ¼0.54 V vs SHE for piperidine) to oxidize ascorbic acid which have been the only significant reducing agent studied and used in earlier applications of nitroxide radicals to blood and food product [24]. The fluorescence is intramolecular quenched by the linked nitroxide through electron exchange interactions [25–27], but can be recovered when the nitroxide moiety reacts with certain types of radicals, such as methyl radicals and some reducing agents, such as ascorbic acid [28]. Because of this, fluorophore-linked nitroxides have been used for the determination of reducing agents as well as radicals [29–32]. Nitroxide can shuttle between the nitroxyl radical form, the reduced hydroxylamine, and the oxidized oxoammonium cation form with one- and two-electron transfer reactions [21,26]. When the hydroxyl radical or superoxide reacts with the nitroxyl radical, the oxoammonium cation is produced. Most nitroxides are reduced to the corresponding hydroxylamine by reacting with ascorbic acid [24,33–35]. At the same time, the quenching effect of nitroxides disappeared, thus creating a decay of the ESR signal and enhancement of the fluorescence. This indicates that fluorophore–nitroxide coupled systems can act as promising reagents for the rapid and convenient fluorometric detection of ascorbic acid. Because of the strong yellow–green fluorescence and good photostability of 1,8-naphthalimide derivatives, they have found wide application in spectral area [36–38]. Here, we describe a new
naphthalimide-linked nitroxide probe (NAPS-NO) with high sensitivity and selectivity for the detection of ascorbic acid. According to the previous reports [39,40], the introduction of 3-aminopropyltriethoxysilane can increase the hydrophilicity of the probe, thus resulting in favorable biocompatibility. At the same time, the silane structure can provide a suitable binding site for the further construction of inorganic–organic complex probes, which have the obvious advantage in the efficient separation and analysis of the targets. A modified previously reported procedure [41,42] was employed for the synthesis of NAPS-NO. It was synthesized from the reaction of 4-Br-1,8-naphthalimide and APTES, followed by reaction with 4amino-tempo (Scheme 1). The structure of NAPS-NO was characterized by ESR, ESI-MS, FT-IR, and fluorophotometry. The fluorescent intensity of NAPS could be inhibited by the linked nitroxide through electron exchange interactions. When AsA reacted with NAPS-NO, the fluorescent intensity could be recovered. The asprepared chemosensor exhibited a linear response to AsA concentrations ranging from 80 nM to 50 μM with a detection limit of about 20 nM. The concentration of AsA in human blood serum and HEK 293 cell lysate was determined respectively by the internal standard addition method. Moreover, results of confocal laser scanning microscopy experiments demonstrated that NAPS-NO could be used for monitoring ascorbic acid in living cells.
2. Experimental section 2.1. Apparatus ESR determination was performed with a Bruker A300 ESR Spectrometer. Fluorescence spectra were measured on a Hitachi F-4500 spectrophotometer equipped with a 1 cm quartz cell. Mass spectra were obtained with MAT-261 spectrometer. FT-IR was recorded on a Thermo Scientific Nicolet 6700 FT-IR spectrometer (Sugar Land, TX, USA). Confocal fluorescence imaging was taken using a Leica TCS SP5 laser scanning confocal microscope (excitation wavelengths 468 nm).
Scheme 1. Synthetic routes to NAPS and NAPS-NO.
T. Yang et al. / Talanta 132 (2015) 191–196
2.2. Reagents We purchased 4-Bromo-1,8-naphthalimide from Adamas Reagent Co., Ltd. 4-amino-1-oxy-2,2,6,6-tetramethylpiperidine was purchased from TCI. 3-aminopropyl-triethoxysilane was purchased from Alfa Aesar. The human serum samples were obtained from Wangjiang Hospital of Sichuan University. HEK 293 cell lysate (2 106 cells) was provided by the state key laboratory at Sichuan University. All solvents used for synthesis and measurements were redistilled before use. All the other chemicals were of analytical-reagent grade and used without further purification.
2.3. Cell culture and cell imaging Hela cells were provided by the state key laboratory at Sichuan University (Sichuan, China). Confocal fluorescence imaging was performed with a Leica TCS SP5 laser scanning confocal microscope (excitation wavelengths 468 nm). Before the experiments, the Hela cells were exposed to 40 mmol L 1 ascorbic acid for 30 min at room temperature to allow the AsA to permeate into the cells. The cells were then centrifuged to remove excess AsA, and the treated cells were then incubated with 40 mmol L 1 NAPSNO for another 30 min. Following incubation, the cells were imaged.
193
2.4.2. (N-Propyl-triethoxysilane-4-(4-ylamino-1-oxy-2,2,6,6tetramethylpiperdine)-naphthalimide) NAPS (0.48 g, 1 mmol) and 4-amino-1-oxy-2,2,6,6-tetramethylpiperidine (0.34 g, 2 mmol) was dissolved in 20 mL of 1-methyl-2pyrrolidone. The mixture was heated for 24 h at 100 1C and excess of 4-amino-2,2,6,6-tetramethylpiperidine and 1-methyl-2-pyrrolidone were recovered by vacuum distillation. The crude product was purified by column chromatography on silica gel using a mixture of ethyl acetate and petroleum ether (1:1) yielding 0.2 g (37%) of orange crystals. As shown in Fig. S3, the HPLC spectra indicated that no obvious impurities were observed. HRMS (ESI þ ) calcd for C30H44N3SiO6 [MþH] þ : 571.3078. Found: 571.3077. FTIR, cm 1: υ(N–H) 3440, υ(C¼O) 1700, υ(C–N) 1386, υ(SiOsym) 816, υ(SiOasym) 1119 (Fig. S4). 2.5. X-band ESR measurement To determine the reaction between N–O groups and ascorbic acid, NAPS-NO (10 μM) was mixed with ascorbic acid solutions (0–50 μM). X-band ESR spectra were recorded at room temperature using a Bruker A300 ESR spectrometer (Germany; microwave power 9 mW, amplitude of 100-kHz field modulation 0.6 mT, time constant 2.56 ms) and were analyzed using an ESR data analyzer.
3. Results and discussion 2.4. Synthesis of intermediates and probes (Scheme 1) 3.1. Chemical properties of spin-labeled fluorescence probe 2.4.1. Synthesis of NAPS (N-propyl-triethoxysilane-4-Brnaphthalimide) NAPS was synthesized according to Jozef et al. [39,40]. Under nitrogen, a solution of 4-Br-1,8-naphthalimide (0.6 g, 2 mmol) and 3-aminopropyl-triethoxysilane (APTES) (560 mL, 2.4 mmol) in EtOH (40 mL) was refluxed for 5 h, 4-Br-1,8-naphthalimide was almost no more left. In this type of reaction, EtOH was widely used as solvent [39,40]. The mixture was evaporated by rotary evaporation and the intermediate was directly investment in the next reaction. As shown in Fig. S1, 1H-NMR (400 MHz, CDCl3): δ¼0.72– 0.76 (m, 2H), 1.21 (t, J ¼7.0 Hz, 9H), 1.80–1.89 (m, 2H), 3.81 (q, J ¼7.0 Hz, 6H), 4.16 (t, J ¼7.6 Hz, 2H), 7.81–7.86 (m, 1H), 8.01– 8.04 (m, 1H), 8.39–8.41 (m, 1H), 8.54–8.57 (m, 1H), 8.63–8.65 (m, 1H) ppm. Synthesis of NAPS-NO
Reduction of the fluorophore–nitroxide probe by ascorbic acid has been monitored by two independent spectroscopic techniques: ESR and fluorescence. When NAPS-NO was mixed with ascorbic acid, its ESR signal decayed and the fluorescence intensity gradually recovered with different concentrations of ascorbic acid (Fig. 1a and b). As shown in the inset of Fig. 1b, the fluorescence enhanced a lot with the addition of ascorbic acid. This increase in fluorescence intensity clearly validated that NAPS-NO can serve as a highly sensitive sensor for AsA. As expected, the fluorescence intensity at approximately 540 nm increased with increasing AsA concentrations. Fig. 1b inset illustrated that this chemosensor exhibited a linear response toward negative logarithm of AsA concentration from 80 nM to 50 μM, the linear equation was y¼2915–381.4x and the linear relative coefficient was R2 ¼ 0.993. The detection limit of AsA was estimated to
Fig. 1. (a) ESR signal decay and (b) fluorescence spectra of NAPS-NO (10 μM) in aqueous solution (50 mM PBS buffer, pH 7.0) upon addition of various concentrations of AsA (0–50 μM). Inset: plot of fluorescence intensity of NAPS-NO versus the negative log concentration of AsA (λex ¼ 468 nm).
194
T. Yang et al. / Talanta 132 (2015) 191–196
20 nM, which is lower than the AsA concentration detected by previous researches [15,28] in physiological sample. We also compared our technique with other methods (Table S1), and the results showed that our chemosensor provided lower detection limit and wider line range for the detection of AsA. 3.2. Evaluation of the response to other reductants and oxidants in vitro NAPS-NO showed no fluorescence enhancement in response to glucose and BSA. In the case of other reductants, such as cysteine, dopamine etc., there were only slightly fluorescence enhancements could be detected (Fig. 2a). The oxoammonium cation is generated when a hydroxyl radical or superoxide reacts with the nitroxyl group at almost the diffusion-controlled rate [21]. Hence, we examined the reactivity of NAPS-NO towards ROS by fluorescence spectrometry (Fig. 2b). NAPS-NO did not show any obvious fluorescence enhancement in response to hydrogen peroxide, hydroxyl radical or high selectivity towards AsA over the other coexisting competitive species is a very important feature to evaluate the performance of NAPS-NO. Therefore, the competition experiments were also conducted. AsA was respectively added into the NAPS-NO solution containing different reductants or ROS. Relative fluorescence intensity was calculated by F/F0, where F0 and F were the fluorescence intensity of NAPS-NO containing AsA without and with the presence of reductants or ROS, respectively. As shown in Fig. 2c and d, glucose and BSA have almost no effect, whereas cysteine, dopamine, uric acid and GSH have relatively small effect on
the fluorescence intensity of NAPS-NO-AsA system. H2O2 and ClO play no role, whereas OH has a little impact on AsA detection. Fig. 2c and d showed that the extra addition of other reductants or ROS produced no obvious interference for AsA detection. These results indicated that NAPS-NO can act as a sensitive and selective AsAspecific fluorescent sensor, and it is expected to find potential applications in selectively and efficiently detecting AsA in biological to the previous reports [21,32], the fluorescence of fluorophore is quenched by the linked nitroxide through intramolecular electron exchange interactions, but can be recovered when the nitroxide moiety reacts with certain reactive oxygen species, such as hydroxyl radicals and some reducing agents, such as ascorbic acid. But they react in a different mechanism. Nitroxide can either accept one electron to be reduced to its diamagnetic counterpart hydroxylamine or donate one electron to be oxidized to its diamagnetic counterpart oxoammonium cation (Scheme S1) [43]. Based on the results of selectivity experiment, we proposed that the short life time of ROS species may result in its lower possibility of reaction with nitroxide than that of the reductants, which resulted in the selectivity of nitroxide towards reactive oxygen species. As for the selectivity towards other reductants, further research works are needed, and we are studying the reasonable mechanism constantly. In this paper, phosphate buffer solution (PBS) was used throughout because it has been proved to be a good buffer for application of chemosensor in organism. The influence of pH value on the fluorescence intensity of NAPS-NO within the pH range of 5.5–9.0 was evaluated. As shown in Fig. S5, the fluorescence intensity increased with increasing pH until it reached the maximum at
Fig. 2. Fluorescence responses of NAPS-NO (10 μM) to (a) various reductants (5 equiv.), AsA (5 equiv.) and (b) ROS (5 equiv.); fluorescence responses of NAPS-NO (10 μM) to AsA (5 equiv.) in the presence of (c) other reductants (5 equiv.) and (d) ROS (5 equiv.) (λex ¼ 468 nm). Reductants including cysteine, glucose, dopamine, uric acid, GSH, BSA. ROS including hydrogen peroxide, hydroxyl radical, hypochloride.
T. Yang et al. / Talanta 132 (2015) 191–196
pH 7.0 and thereafter decreased, thus we chose pH 7.0 as the optimal pH value to accord with that of neutral media and biological samples. 3.3. Determining the concentration of AsA in blood serum and recovery test With the purpose of illustrating the reliability and accuracy of the proposed method, the developed sensor was applied to determine AsA in blood serum samples. Herein, 5 μL of blood serum was added to 500 μL of 10 μM NAPS-NO in PBS (50 mM, pH 7.0). Different concentrations of AsA standards were then added to the sample and the mixtures were measured. The fluorescence intensity of NAPS-NO with blood serum was enhanced following additions of different amounts of AsA. After linearly fitting the Table 1 Recoveries of AsA from spiked blood serum samples. Sample
Level added (μM)
Level found (μM)
Recovery (%)
RSD (%)
1
5
2
6
3
8
4.95 4.84 4.96 6.14 6.17 6.23 7.86 7.89 7.81 9.09 9.05 9.04
98.9 96.8 99.2 102.3 102.9 103.9 98.3 98.6 97.6 101.0 100.5 100.4
1.1 3.2 0.8 2.3 2.9 3.9 1.7 1.4 2.4 1.0 0.5 0.4
4
9
195
fluorescence intensity with the concentration of added AsA (Fig. S6), we successfully determined the concentration of AsA in blood serum to be 25 μM, which matched with that of the previous report [44]. Recovery experiments were also carried out by adding known amounts of AsA to the human serum sample. As shown in Table 1, the recovery of AsA was between 96 and 104% with RSDo 5%, which showed that the sensor was suited to detect AsA in blood serum samples. 3.4. Determining the concentration of AsA in cell lysate To further show the potential utility of NAPS-NO in biological samples, the probe was studied to detect AsA in cell lysate. A volume of 1 mL of HEK 293 cell lysate was transferred to a centrifuge tube and the large biomolecules were separated from the cell lysate by centrifugation at 12,000 rpm for 10 min. Then, 2 μL of cell lysate supernatant was added to 500 μL of 10 μM NAPSNO in PBS (50 mM, pH 7.0). Different concentrations of AsA standards were added to the sample and the mixtures were measured. The fluorescence intensity of NAPS-NO with cell lysate was enhanced after additions of different amounts of AsA (Fig. S7). After linearly fitting, we successfully determined the concentration of AsA in cell lysate to be 1.51 μM, which accord with that of former report [45]. 3.5. Cell labeling and imaging We also proposed that NAPS-NO would permeate into cells because of its favorable amphiphilic properties. Laser scanning confocal microscopy was used to investigate this proposition.
Fig. 3. Confocal fluorescence and brightfield images of Hela cells. (a) Fluorescence image of Hela cells stained with 40 mmol L 1 AsA for 30 min at 37 1C. (b) Confocal image of Hela cells that were further incubated with 40 mmol L 1 NAPS-NO for another 30 min. (c) Bright field image of cells shown in panel b. (d) The overlay image of b and c. The excitation wavelength was 468 nm.
196
T. Yang et al. / Talanta 132 (2015) 191–196
thanks to Analytical & Testing Centre of Sichuan University for the MS measurements.
Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.talanta.2014.08.066. References
Fig. 4. Cellular toxicity of NAPS-NO determined by MTT assay. Effect of different concentration of NAPS-NO on viability of Hela cells.
Cultured Hela cells were incubated with 40 mmol L 1AsA for 30 min at 37 1C, and an insignificant amount of AsA fluorescence was detected in the cells’ interior (Fig. 3a). However, when the cells were supplemented with 40 mmol L 1 NAPS-NO for another 30 min, an enormous fluorescence increase from the intracellular area was observed (Fig. 3b). Bright field microscopic images (Fig. 3c) confirmed that the cells were viable throughout the imaging experiments. An overlay of fluorescence and bright field images revealed that the fluorescence signals were localized in the perinuclear region of the cytosol (Fig. 3d), indicating the subcellular distribution of AsA. These results demonstrated that NAPSNO was cell permeable and might be used for monitoring AsA within biological samples. For future biological applications, the inherent cytotoxicity of NAPS-NO was tested using Hela cells by MTT assay. As shown in Fig. 4, the viability of Hela cells retained more than 90% after treated with NAPS-NO at 20 mmol L 1 for 48 h, even at 40 mmol L 1 cell viability remained 89%. The results suggested low cytotoxicity and good biocompatibility of NAPS-NO. 4. Conclusion In conclusion, we proposed a new and simple fluorescent turn-on sensor to detect ascorbic acid in aqueous solution with very high sensitivity and wide detection range by using naphthalimide linked nitroxide as probe. The coordination of nitroxide with AsA was formed more easily, therefore, a recovered fluorescence intensity of naphthalimide moiety was observed. A good linear relationship was obtained from 80 nM to 50 μM for AsA and the detection limit was 20 nM. The chemosensor was successfully used to determine the concentration of AsA in human serum sample and HEK 293 cell lysate by using of standard addition methods. Furthermore, the excellent biological value of NAPS-NO was demonstrated by the fluorescence imaging with Hela cells. The developed method is expected to design new chemosensors based on organic fluorescent molecules and inorganic materials and applicate in clinical diagnostics and oxidative stress monitoring. Acknowledgments The work described in this paper was supported by the National Natural Science Foundation of China (21377089, 20927007 and 21177090). We would like to express our sincere
[1] S.A. Kumar, P.-H. Lo, S.-M. Chen, Biosens. Bioelectron. 24 (2008) 518–523. [2] Z. Chen, T.E. Young, J. Ling, S.-C. Chang, D.R. Gallie, Proc. Natl. Acad. Sci. U.S.A. 100 (2003) 3525–3530. [3] L. Zhang, Z.N. Wang, Y. Xia, G.Y. Kai, W.S. Chen, K.X. Tang, Crit. Rev. Biotechnol. 27 (2007) 173–182. [4] A. Sevanian, K.J.A. Davies, P. Hochstein, Am. J. Clin. Nutr. 54 (1991) S1129–S1134. [5] D. Wen, S. Guo, S. Dong, E. Wang, Biosens. Bioelectron. 26 (2010) 1056–1061. [6] D.E. Hughes, J. Pharm. Sci. 72 (1983) 126–129. [7] B. Vahid, J. Chem. Soc. Pak. 34 (2012) 1510–1512. [8] L. Kukoc-Modun, M. Biocic, N. Radic, Talanta 96 (2012) 174–179. [9] B. Shyla, G. Nagendrappa, Food Chem. 138 (2013) 2036–2342. [10] M.A. Kutnink, W.C. Hawkes, E.E. Schaus, S.T. Omaye, Anal. Biochem. 166 (1987) 424–430. [11] S.P. Sood, L.E. Sartori, D.P. Wittmer, W.G. Haney, Anal. Chem. 48 (1976) 796–798. [12] A.J. Speek, J. Schrijver, W.H. Schreurs, J. Chromatogr. 305 (1984) 53–60. [13] S. Moore, K.A. Calder, N.J. Miller, C.A. Rice-Evans, Free Radical Res. 21 (1994) 417–425. [14] G.R. Buettner, Free Radical Res. Commun. 10 (1990) 5–9. [15] R. Ferin, M.L. Pavao, J. Baptista, Clin. Biochem. 46 (2013) 665–669. [16] P.A. Motchnik, B. Frei, B.N. Ames, Methods Enzymol 234 (1994) 269–279. [17] X.X. Wang, P. Wu, X.D. Hou, Y. Lv, Analyst 138 (2013) 229–233. [18] X. Zhang, S. Yu, W.Y. He, H. Uyama, Q.J. Xie, L. Zhang, F.C. Yang, Biosens. Bioelectron. 55 (2014) 446–451. [19] X. Wang, M. Wu, W.R. Tang, Y. Zhu, L.W. Wang, Q.J. Wang, P.G. He, Y.Z. Fang, J. Electroanal. Chem. 695 (2013) 10–16. [20] D. Karak, A. Banerjee, A. Sahana, S. Guha, S. Lohar, S.S. Adhikari, D. Das, J. Hazard. Mater. 188 (2011) 274–280. [21] Y. Matsuoka, M. Yamato, T. Yamasaki, F. Mito, K.-i. Yamada, Free Radical Biol. Med 53 (2012) 2112–2118. [22] M. Zheng, Z. Xie, D. Qu, D. Li, P. Du, X. Jing, Z. Sun, ACS Appl. Mater. Interfaces 5 (2013) 13242–13247. [23] B. Biqing, T. Nanjiao, M. Mingfeng, Z. Lei, Y. Lihui, F. Quli, W. Lianhui, H. Wei, Soft Mater. 12 (2014) 73–78. [24] R.J. Mehlhorn, J. Biol. Chem. 266 (1991) 2724–2731. [25] J.P. Blinco, K.E. Fairfull-Smith, B.J. Morrow, S.E. Bottle, Aust. J. Chem. 64 (2011) 373–389. [26] S.A. Green, D.J. Simpson, G. Zhou, P.S. Ho, N.V. Blough, J. Am. Chem. Soc. 112 (1990) 7337–7346. [27] G.I. Likhtenshtein, Appl. Biochem. Biotechnol. 152 (2009) 135–155. [28] T. Maki, N. Soh, K. Nakano, T. Imato, Talanta 85 (2011) 1730–1733. [29] G.G. Borisenko, I. Martin, Q. Zhao, A.A. Amoscato, Y.Y. Tyurina, V.E. Kagan, J. Biochem. 279 (2004) 23453–23462. [30] J.L. Chen, S.J. Zhuo, Y.Q. Wu, F. Fang, L. Li, C.Q. Zhu, Spectrochim. Acta, Part A 63 (2006) 438–443. [31] M. Jia, Y. Tang, Y.F. Lam, S.A. Green, N.V. Blough, Anal. Chem. 81 (2009) 8033–8040. [32] E. Lozinsky, V.V. Martin, T.A. Berezina, A.I. Shames, A.L. Weis, G.I. Likhtenshtein, J. Biochem. Biophys. Methods 38 (1999) 29–42. [33] R.D. Kornberg, H.M. McConnell, Biochemistry 10 (1971) 1111–1120. [34] A. Sienkiewicz, A.M.D. Ferreira, B. Danner, C.P. Scholes, J. Magn. Reson. 136 (1999) 137–142. [35] B.S. Winkler, S.M. Orselli, T.S. Rex, Free Radical Biol. Med 17 (1994) 333–349. [36] G.R. Bardajee, A.Y. Li, J.C. Haley, M.A. Winnik, Dyes Pigm 79 (2008) 24–32. [37] V.B. Bojinov, I.P. Panova, Dyes Pigm 80 (2009) 61–66. [38] Y.G. Shi, J.H. Yao, Y.L. Duan, Q.L. Mi, J.H. Chen, Q.Q. Xu, Bioorg. Med. Chem. Lett. 23 (2013) 2538–2542. [39] T. Jesionowski, A. Krysztafkiewicz, Appl. Surf. Sci. 172 (2001) 18–32. [40] Q.T. Meng, X.L. Zhang, C. He, P. Zhou, W.P. Su, C.Y. Duan, Talanta 84 (2011) 53–59. [41] J. Kollar, P. Hrdlovic, S. Chmela, J. Photochem. Photobiol., A 195 (2008) 64–71. [42] J. Kollar, P. Hrdlovic, S. Chmela, J. Photochem. Photobiol., A 204 (2009) 191–199. [43] F. Lin, D. Pei, W. He, Z. Huang, Y. Huang, X. Guo, J. Mater. Chem. 22 (2012) 11801–11807. [44] K.R. Dhariwal, W.O. Hartzell, M. Levine, Am. J. Clin. Nutr. 54 (1991) 712–716. [45] A. Levin, C. DeSouza, C. Zaarour, W. Walsh, M.K. Chan, Z. Verjee, S. McIntyre, K. Adeli, Clin. Biochem. 43 (2010) 1411–1414.
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
A sensitive and selective chemosensor for ascorbic acid based on a fluorescent nitroxide switch Tian Yang a, Baozhan Zheng a, Hengxing Liang c, Yuping Wan c, Juan Du a,n, Dan Xiao a,b,n a
College of Chemistry, Sichuan University, No. 29 Wangjiang Road, Chengdu, PR China College of Chemical Engineering, Sichuan University, No. 29 Wangjiang Road, Chengdu, PR China c Chengdu Product Quality Supervision and Inspection Institute, Chengdu 610041, PR China b
art ic l e i nf o
a b s t r a c t
Article history: Received 15 June 2014 Received in revised form 20 August 2014 Accepted 27 August 2014 Available online 6 September 2014
Ascorbic acid (AsA), also known as vitamin C, is a vital small-molecule antioxidant with multiple functions in vivo. It’s the major natural antioxidant found in plants and is also an essential component of human nutrition. AsA plays a key role in many diseases-related biological metabolism. Therefore, sensitive and selective detection of AsA is greatly important in pharmaceutical, clinical and food industry. Here a sensitive and selective sensor for ascorbic acid detection based on the recovered fluorescence of NAPS-NO (N-propyl-triethoxysilane-4-(4-ylamino-1-oxy-2,2,6,6-tetramethylpiperdine)naphthalimide) probe is described. The fluorescence of the naphthalimide moiety of NAPS-NO is inhibited by the nitroxide group, which is covalently linked to the fluorophore. Then, ascorbic acid reacts rapidly with the nitroxide moiety of NAPS-NO to form hydroxylamine, and the fluorescence properties of the naphthalimide moiety are recovered and the ESR signal decayed. Over a wide range from 80 nM to 50 μM, a good linear relationship between the fluorescence intensity and the concentration of ascorbic acid was found and the detection limit was estimated to be as low as 20 nM. To confirm the practical usefulness of the fluorophore–nitroxide probe, we demonstrated the use of NAPS-NO for the measurement of AsA in human blood serum and also successfully determined the concentration of AsA in HEK 293 cell lysate. Results from confocal laser scanning microscopy experiments demonstrated that this chemosensor is cell permeable and can be used as a fluorescent probe for monitoring ascorbic acid in living cells. & 2014 Published by Elsevier B.V.
Keywords: Chemosensor Nitroxide Fluorescence AsA
1. Introduction Ascorbic acid (AsA), which cannot be synthesized in the human body, is used in large scale as an antioxidant in food, animal feed, beverages, pharmaceutical formulations and cosmetic applications [1]. AsA is the major natural antioxidant found in plants and is also an essential component of human nutrition [2,3]. It plays a key role in biological metabolism and has been commonly used for the prevention and treatment of common cold, mental illness, infertility, cancer and AIDS, etc [4]. Thus, the development of a simple and rapid method for the determination of AsA with high selectivity and sensitivity is desirable for diagnostic and food safe applications [5]. A wide variety of analytical methods is available for the determination of ascorbic acid, such as titrimetric [6,7], spectrophotometric
n Corresponding authors at: College of Chemical Engineering, Sichuan University, No. 29 Wangjiang Road, Chengdu, PR China. Tel.: þ 86 28 85415029; fax: þ86 28 85416029. E-mail addresses: [email protected] (J. Du), [email protected] (D. Xiao).
http://dx.doi.org/10.1016/j.talanta.2014.08.066 0039-9140/& 2014 Published by Elsevier B.V.
[8,9] and chromatographic [10–12] methods. These methods rely on different phenomena and registration techniques suffering from several limitations. The spectrophotometric methods measure either the resultant color of products of certain reactions involving ascorbate [13] or absorbance of ascorbic acid itself [14]. However, the low selectivity of the reactions involved decreases the accuracy of the ascorbate analysis. Thereby, the reliability of these methods is limited by relatively low specificity. A number of specific assays for reductants including ascorbic acid, uric acid, tocopherol etc. have been developed based on high performance liquid chromatography (HPLC) [15,16]. These techniques require several complicated preliminary steps. In addition, HPLC separations would consume relatively long time, therefore, could not be considered as an express-method. Early methods utilizing nitroxide radicals for ascorbate quantifications were based on measurements of only ESR signal from nitroxide, therefore, suffered from general drawbacks of ESR spectroscopy. Although this technique permits investigations in high optical density substance, the ESR measurements are limited by high cost and complexity of equipments and difficulties in acquiring and processing experimental data. At present, electrochemical approaches become
192
T. Yang et al. / Talanta 132 (2015) 191–196
the major analytical methods for AsA determination because of their convenience and speediness. Nevertheless, the interference from other similar redox potential molecules can be serious, such as uric acid and dopamine [17]. What’s more, the detection limit is always higher than that of fluorescence methods. Particularly, attempts have been made to detect ascorbic acid using nanoparticles. Wang et al. [17] developed a simple fluorescence sensor for sensitive turn-off detection of ascorbic acid by using protein-modified Au nanoclusters in aqueous media. Zhang et al. [18] fabricated an electrochemical sensor for selective detection of ascorbic acid in the presence of dopamine and uric acid by modifying the glassy carbon electrode with carbon-supported NiCoO2(NiCoO2/C) nanoparticles. Wang et al. [19] prepared a palladium nanoparticle/graphene/chitosan/glassy carbon electrode which displayed excellent electrochemical catalytic activities towards ascorbic acid, dopamine and uric acid. But these methods have relatively high detection limit towards ascorbic acid. Fluorometric sensing is a preferable approach for detection of AsA due to its comparatively high selectivity, sensitivity, rapidity, operational simplicity, nondestructive methodology and direct visual perception [20]. In recent years, many fluorescent chemosensors were designed for ascorbic acid. Matsuoka et al. [21] synthesised several fluorescent nitroxide compounds based on naphthalene which were sensitive to ascorbic acid. They also demonstrated the use of NaphDiPy nitroxide for the measurement of ascorbic acid in the plasma of osteogenic disorder Shionogi rats to confirm the practical usefulness of the fluorophore–nitroxide probe. Zheng et al. [22] demonstrated a fluorescent carbon dot probe, its fluorescence was quenched by Cr(VI) through inner filter effect, but Cr(VI) can be reduced to low valent chromium species easily by ascorbic acid, and then resulting in the elimination of the IFE and recovery of CD fluorescence. Therefore the CD Cr(VI) mixture could behave as an off-on type fluorescent probe for ascorbic acid. Biqing et al. [23] developed a fluorescent turn-on detection for ascorbic acid based on hyperbranched conjugated polyelectrolyte. The fluorescence was superquenched by Cu2 þ ions, upon addition of ascorbic acid, the paramagnetic Cu2 þ ions can be transformed into diamagnetic Cu1 þ ions, which inhibit the quenching and thus lead to fluorescence enhancement of the hyperbranched conjugate polymer. These fluorescent probes all have sensitive response to ascorbic acid, but none of them were used in intracellular environment. Fluorophore-linked nitroxides represent examples of such fluorescent sensors. The standard redox potential of nitroxides is high enough (E1 ¼0.54 V vs SHE for piperidine) to oxidize ascorbic acid which have been the only significant reducing agent studied and used in earlier applications of nitroxide radicals to blood and food product [24]. The fluorescence is intramolecular quenched by the linked nitroxide through electron exchange interactions [25–27], but can be recovered when the nitroxide moiety reacts with certain types of radicals, such as methyl radicals and some reducing agents, such as ascorbic acid [28]. Because of this, fluorophore-linked nitroxides have been used for the determination of reducing agents as well as radicals [29–32]. Nitroxide can shuttle between the nitroxyl radical form, the reduced hydroxylamine, and the oxidized oxoammonium cation form with one- and two-electron transfer reactions [21,26]. When the hydroxyl radical or superoxide reacts with the nitroxyl radical, the oxoammonium cation is produced. Most nitroxides are reduced to the corresponding hydroxylamine by reacting with ascorbic acid [24,33–35]. At the same time, the quenching effect of nitroxides disappeared, thus creating a decay of the ESR signal and enhancement of the fluorescence. This indicates that fluorophore–nitroxide coupled systems can act as promising reagents for the rapid and convenient fluorometric detection of ascorbic acid. Because of the strong yellow–green fluorescence and good photostability of 1,8-naphthalimide derivatives, they have found wide application in spectral area [36–38]. Here, we describe a new
naphthalimide-linked nitroxide probe (NAPS-NO) with high sensitivity and selectivity for the detection of ascorbic acid. According to the previous reports [39,40], the introduction of 3-aminopropyltriethoxysilane can increase the hydrophilicity of the probe, thus resulting in favorable biocompatibility. At the same time, the silane structure can provide a suitable binding site for the further construction of inorganic–organic complex probes, which have the obvious advantage in the efficient separation and analysis of the targets. A modified previously reported procedure [41,42] was employed for the synthesis of NAPS-NO. It was synthesized from the reaction of 4-Br-1,8-naphthalimide and APTES, followed by reaction with 4amino-tempo (Scheme 1). The structure of NAPS-NO was characterized by ESR, ESI-MS, FT-IR, and fluorophotometry. The fluorescent intensity of NAPS could be inhibited by the linked nitroxide through electron exchange interactions. When AsA reacted with NAPS-NO, the fluorescent intensity could be recovered. The asprepared chemosensor exhibited a linear response to AsA concentrations ranging from 80 nM to 50 μM with a detection limit of about 20 nM. The concentration of AsA in human blood serum and HEK 293 cell lysate was determined respectively by the internal standard addition method. Moreover, results of confocal laser scanning microscopy experiments demonstrated that NAPS-NO could be used for monitoring ascorbic acid in living cells.
2. Experimental section 2.1. Apparatus ESR determination was performed with a Bruker A300 ESR Spectrometer. Fluorescence spectra were measured on a Hitachi F-4500 spectrophotometer equipped with a 1 cm quartz cell. Mass spectra were obtained with MAT-261 spectrometer. FT-IR was recorded on a Thermo Scientific Nicolet 6700 FT-IR spectrometer (Sugar Land, TX, USA). Confocal fluorescence imaging was taken using a Leica TCS SP5 laser scanning confocal microscope (excitation wavelengths 468 nm).
Scheme 1. Synthetic routes to NAPS and NAPS-NO.
T. Yang et al. / Talanta 132 (2015) 191–196
2.2. Reagents We purchased 4-Bromo-1,8-naphthalimide from Adamas Reagent Co., Ltd. 4-amino-1-oxy-2,2,6,6-tetramethylpiperidine was purchased from TCI. 3-aminopropyl-triethoxysilane was purchased from Alfa Aesar. The human serum samples were obtained from Wangjiang Hospital of Sichuan University. HEK 293 cell lysate (2 106 cells) was provided by the state key laboratory at Sichuan University. All solvents used for synthesis and measurements were redistilled before use. All the other chemicals were of analytical-reagent grade and used without further purification.
2.3. Cell culture and cell imaging Hela cells were provided by the state key laboratory at Sichuan University (Sichuan, China). Confocal fluorescence imaging was performed with a Leica TCS SP5 laser scanning confocal microscope (excitation wavelengths 468 nm). Before the experiments, the Hela cells were exposed to 40 mmol L 1 ascorbic acid for 30 min at room temperature to allow the AsA to permeate into the cells. The cells were then centrifuged to remove excess AsA, and the treated cells were then incubated with 40 mmol L 1 NAPSNO for another 30 min. Following incubation, the cells were imaged.
193
2.4.2. (N-Propyl-triethoxysilane-4-(4-ylamino-1-oxy-2,2,6,6tetramethylpiperdine)-naphthalimide) NAPS (0.48 g, 1 mmol) and 4-amino-1-oxy-2,2,6,6-tetramethylpiperidine (0.34 g, 2 mmol) was dissolved in 20 mL of 1-methyl-2pyrrolidone. The mixture was heated for 24 h at 100 1C and excess of 4-amino-2,2,6,6-tetramethylpiperidine and 1-methyl-2-pyrrolidone were recovered by vacuum distillation. The crude product was purified by column chromatography on silica gel using a mixture of ethyl acetate and petroleum ether (1:1) yielding 0.2 g (37%) of orange crystals. As shown in Fig. S3, the HPLC spectra indicated that no obvious impurities were observed. HRMS (ESI þ ) calcd for C30H44N3SiO6 [MþH] þ : 571.3078. Found: 571.3077. FTIR, cm 1: υ(N–H) 3440, υ(C¼O) 1700, υ(C–N) 1386, υ(SiOsym) 816, υ(SiOasym) 1119 (Fig. S4). 2.5. X-band ESR measurement To determine the reaction between N–O groups and ascorbic acid, NAPS-NO (10 μM) was mixed with ascorbic acid solutions (0–50 μM). X-band ESR spectra were recorded at room temperature using a Bruker A300 ESR spectrometer (Germany; microwave power 9 mW, amplitude of 100-kHz field modulation 0.6 mT, time constant 2.56 ms) and were analyzed using an ESR data analyzer.
3. Results and discussion 2.4. Synthesis of intermediates and probes (Scheme 1) 3.1. Chemical properties of spin-labeled fluorescence probe 2.4.1. Synthesis of NAPS (N-propyl-triethoxysilane-4-Brnaphthalimide) NAPS was synthesized according to Jozef et al. [39,40]. Under nitrogen, a solution of 4-Br-1,8-naphthalimide (0.6 g, 2 mmol) and 3-aminopropyl-triethoxysilane (APTES) (560 mL, 2.4 mmol) in EtOH (40 mL) was refluxed for 5 h, 4-Br-1,8-naphthalimide was almost no more left. In this type of reaction, EtOH was widely used as solvent [39,40]. The mixture was evaporated by rotary evaporation and the intermediate was directly investment in the next reaction. As shown in Fig. S1, 1H-NMR (400 MHz, CDCl3): δ¼0.72– 0.76 (m, 2H), 1.21 (t, J ¼7.0 Hz, 9H), 1.80–1.89 (m, 2H), 3.81 (q, J ¼7.0 Hz, 6H), 4.16 (t, J ¼7.6 Hz, 2H), 7.81–7.86 (m, 1H), 8.01– 8.04 (m, 1H), 8.39–8.41 (m, 1H), 8.54–8.57 (m, 1H), 8.63–8.65 (m, 1H) ppm. Synthesis of NAPS-NO
Reduction of the fluorophore–nitroxide probe by ascorbic acid has been monitored by two independent spectroscopic techniques: ESR and fluorescence. When NAPS-NO was mixed with ascorbic acid, its ESR signal decayed and the fluorescence intensity gradually recovered with different concentrations of ascorbic acid (Fig. 1a and b). As shown in the inset of Fig. 1b, the fluorescence enhanced a lot with the addition of ascorbic acid. This increase in fluorescence intensity clearly validated that NAPS-NO can serve as a highly sensitive sensor for AsA. As expected, the fluorescence intensity at approximately 540 nm increased with increasing AsA concentrations. Fig. 1b inset illustrated that this chemosensor exhibited a linear response toward negative logarithm of AsA concentration from 80 nM to 50 μM, the linear equation was y¼2915–381.4x and the linear relative coefficient was R2 ¼ 0.993. The detection limit of AsA was estimated to
Fig. 1. (a) ESR signal decay and (b) fluorescence spectra of NAPS-NO (10 μM) in aqueous solution (50 mM PBS buffer, pH 7.0) upon addition of various concentrations of AsA (0–50 μM). Inset: plot of fluorescence intensity of NAPS-NO versus the negative log concentration of AsA (λex ¼ 468 nm).
194
T. Yang et al. / Talanta 132 (2015) 191–196
20 nM, which is lower than the AsA concentration detected by previous researches [15,28] in physiological sample. We also compared our technique with other methods (Table S1), and the results showed that our chemosensor provided lower detection limit and wider line range for the detection of AsA. 3.2. Evaluation of the response to other reductants and oxidants in vitro NAPS-NO showed no fluorescence enhancement in response to glucose and BSA. In the case of other reductants, such as cysteine, dopamine etc., there were only slightly fluorescence enhancements could be detected (Fig. 2a). The oxoammonium cation is generated when a hydroxyl radical or superoxide reacts with the nitroxyl group at almost the diffusion-controlled rate [21]. Hence, we examined the reactivity of NAPS-NO towards ROS by fluorescence spectrometry (Fig. 2b). NAPS-NO did not show any obvious fluorescence enhancement in response to hydrogen peroxide, hydroxyl radical or high selectivity towards AsA over the other coexisting competitive species is a very important feature to evaluate the performance of NAPS-NO. Therefore, the competition experiments were also conducted. AsA was respectively added into the NAPS-NO solution containing different reductants or ROS. Relative fluorescence intensity was calculated by F/F0, where F0 and F were the fluorescence intensity of NAPS-NO containing AsA without and with the presence of reductants or ROS, respectively. As shown in Fig. 2c and d, glucose and BSA have almost no effect, whereas cysteine, dopamine, uric acid and GSH have relatively small effect on
the fluorescence intensity of NAPS-NO-AsA system. H2O2 and ClO play no role, whereas OH has a little impact on AsA detection. Fig. 2c and d showed that the extra addition of other reductants or ROS produced no obvious interference for AsA detection. These results indicated that NAPS-NO can act as a sensitive and selective AsAspecific fluorescent sensor, and it is expected to find potential applications in selectively and efficiently detecting AsA in biological to the previous reports [21,32], the fluorescence of fluorophore is quenched by the linked nitroxide through intramolecular electron exchange interactions, but can be recovered when the nitroxide moiety reacts with certain reactive oxygen species, such as hydroxyl radicals and some reducing agents, such as ascorbic acid. But they react in a different mechanism. Nitroxide can either accept one electron to be reduced to its diamagnetic counterpart hydroxylamine or donate one electron to be oxidized to its diamagnetic counterpart oxoammonium cation (Scheme S1) [43]. Based on the results of selectivity experiment, we proposed that the short life time of ROS species may result in its lower possibility of reaction with nitroxide than that of the reductants, which resulted in the selectivity of nitroxide towards reactive oxygen species. As for the selectivity towards other reductants, further research works are needed, and we are studying the reasonable mechanism constantly. In this paper, phosphate buffer solution (PBS) was used throughout because it has been proved to be a good buffer for application of chemosensor in organism. The influence of pH value on the fluorescence intensity of NAPS-NO within the pH range of 5.5–9.0 was evaluated. As shown in Fig. S5, the fluorescence intensity increased with increasing pH until it reached the maximum at
Fig. 2. Fluorescence responses of NAPS-NO (10 μM) to (a) various reductants (5 equiv.), AsA (5 equiv.) and (b) ROS (5 equiv.); fluorescence responses of NAPS-NO (10 μM) to AsA (5 equiv.) in the presence of (c) other reductants (5 equiv.) and (d) ROS (5 equiv.) (λex ¼ 468 nm). Reductants including cysteine, glucose, dopamine, uric acid, GSH, BSA. ROS including hydrogen peroxide, hydroxyl radical, hypochloride.
T. Yang et al. / Talanta 132 (2015) 191–196
pH 7.0 and thereafter decreased, thus we chose pH 7.0 as the optimal pH value to accord with that of neutral media and biological samples. 3.3. Determining the concentration of AsA in blood serum and recovery test With the purpose of illustrating the reliability and accuracy of the proposed method, the developed sensor was applied to determine AsA in blood serum samples. Herein, 5 μL of blood serum was added to 500 μL of 10 μM NAPS-NO in PBS (50 mM, pH 7.0). Different concentrations of AsA standards were then added to the sample and the mixtures were measured. The fluorescence intensity of NAPS-NO with blood serum was enhanced following additions of different amounts of AsA. After linearly fitting the Table 1 Recoveries of AsA from spiked blood serum samples. Sample
Level added (μM)
Level found (μM)
Recovery (%)
RSD (%)
1
5
2
6
3
8
4.95 4.84 4.96 6.14 6.17 6.23 7.86 7.89 7.81 9.09 9.05 9.04
98.9 96.8 99.2 102.3 102.9 103.9 98.3 98.6 97.6 101.0 100.5 100.4
1.1 3.2 0.8 2.3 2.9 3.9 1.7 1.4 2.4 1.0 0.5 0.4
4
9
195
fluorescence intensity with the concentration of added AsA (Fig. S6), we successfully determined the concentration of AsA in blood serum to be 25 μM, which matched with that of the previous report [44]. Recovery experiments were also carried out by adding known amounts of AsA to the human serum sample. As shown in Table 1, the recovery of AsA was between 96 and 104% with RSDo 5%, which showed that the sensor was suited to detect AsA in blood serum samples. 3.4. Determining the concentration of AsA in cell lysate To further show the potential utility of NAPS-NO in biological samples, the probe was studied to detect AsA in cell lysate. A volume of 1 mL of HEK 293 cell lysate was transferred to a centrifuge tube and the large biomolecules were separated from the cell lysate by centrifugation at 12,000 rpm for 10 min. Then, 2 μL of cell lysate supernatant was added to 500 μL of 10 μM NAPSNO in PBS (50 mM, pH 7.0). Different concentrations of AsA standards were added to the sample and the mixtures were measured. The fluorescence intensity of NAPS-NO with cell lysate was enhanced after additions of different amounts of AsA (Fig. S7). After linearly fitting, we successfully determined the concentration of AsA in cell lysate to be 1.51 μM, which accord with that of former report [45]. 3.5. Cell labeling and imaging We also proposed that NAPS-NO would permeate into cells because of its favorable amphiphilic properties. Laser scanning confocal microscopy was used to investigate this proposition.
Fig. 3. Confocal fluorescence and brightfield images of Hela cells. (a) Fluorescence image of Hela cells stained with 40 mmol L 1 AsA for 30 min at 37 1C. (b) Confocal image of Hela cells that were further incubated with 40 mmol L 1 NAPS-NO for another 30 min. (c) Bright field image of cells shown in panel b. (d) The overlay image of b and c. The excitation wavelength was 468 nm.
196
T. Yang et al. / Talanta 132 (2015) 191–196
thanks to Analytical & Testing Centre of Sichuan University for the MS measurements.
Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.talanta.2014.08.066. References
Fig. 4. Cellular toxicity of NAPS-NO determined by MTT assay. Effect of different concentration of NAPS-NO on viability of Hela cells.
Cultured Hela cells were incubated with 40 mmol L 1AsA for 30 min at 37 1C, and an insignificant amount of AsA fluorescence was detected in the cells’ interior (Fig. 3a). However, when the cells were supplemented with 40 mmol L 1 NAPS-NO for another 30 min, an enormous fluorescence increase from the intracellular area was observed (Fig. 3b). Bright field microscopic images (Fig. 3c) confirmed that the cells were viable throughout the imaging experiments. An overlay of fluorescence and bright field images revealed that the fluorescence signals were localized in the perinuclear region of the cytosol (Fig. 3d), indicating the subcellular distribution of AsA. These results demonstrated that NAPSNO was cell permeable and might be used for monitoring AsA within biological samples. For future biological applications, the inherent cytotoxicity of NAPS-NO was tested using Hela cells by MTT assay. As shown in Fig. 4, the viability of Hela cells retained more than 90% after treated with NAPS-NO at 20 mmol L 1 for 48 h, even at 40 mmol L 1 cell viability remained 89%. The results suggested low cytotoxicity and good biocompatibility of NAPS-NO. 4. Conclusion In conclusion, we proposed a new and simple fluorescent turn-on sensor to detect ascorbic acid in aqueous solution with very high sensitivity and wide detection range by using naphthalimide linked nitroxide as probe. The coordination of nitroxide with AsA was formed more easily, therefore, a recovered fluorescence intensity of naphthalimide moiety was observed. A good linear relationship was obtained from 80 nM to 50 μM for AsA and the detection limit was 20 nM. The chemosensor was successfully used to determine the concentration of AsA in human serum sample and HEK 293 cell lysate by using of standard addition methods. Furthermore, the excellent biological value of NAPS-NO was demonstrated by the fluorescence imaging with Hela cells. The developed method is expected to design new chemosensors based on organic fluorescent molecules and inorganic materials and applicate in clinical diagnostics and oxidative stress monitoring. Acknowledgments The work described in this paper was supported by the National Natural Science Foundation of China (21377089, 20927007 and 21177090). We would like to express our sincere
[1] S.A. Kumar, P.-H. Lo, S.-M. Chen, Biosens. Bioelectron. 24 (2008) 518–523. [2] Z. Chen, T.E. Young, J. Ling, S.-C. Chang, D.R. Gallie, Proc. Natl. Acad. Sci. U.S.A. 100 (2003) 3525–3530. [3] L. Zhang, Z.N. Wang, Y. Xia, G.Y. Kai, W.S. Chen, K.X. Tang, Crit. Rev. Biotechnol. 27 (2007) 173–182. [4] A. Sevanian, K.J.A. Davies, P. Hochstein, Am. J. Clin. Nutr. 54 (1991) S1129–S1134. [5] D. Wen, S. Guo, S. Dong, E. Wang, Biosens. Bioelectron. 26 (2010) 1056–1061. [6] D.E. Hughes, J. Pharm. Sci. 72 (1983) 126–129. [7] B. Vahid, J. Chem. Soc. Pak. 34 (2012) 1510–1512. [8] L. Kukoc-Modun, M. Biocic, N. Radic, Talanta 96 (2012) 174–179. [9] B. Shyla, G. Nagendrappa, Food Chem. 138 (2013) 2036–2342. [10] M.A. Kutnink, W.C. Hawkes, E.E. Schaus, S.T. Omaye, Anal. Biochem. 166 (1987) 424–430. [11] S.P. Sood, L.E. Sartori, D.P. Wittmer, W.G. Haney, Anal. Chem. 48 (1976) 796–798. [12] A.J. Speek, J. Schrijver, W.H. Schreurs, J. Chromatogr. 305 (1984) 53–60. [13] S. Moore, K.A. Calder, N.J. Miller, C.A. Rice-Evans, Free Radical Res. 21 (1994) 417–425. [14] G.R. Buettner, Free Radical Res. Commun. 10 (1990) 5–9. [15] R. Ferin, M.L. Pavao, J. Baptista, Clin. Biochem. 46 (2013) 665–669. [16] P.A. Motchnik, B. Frei, B.N. Ames, Methods Enzymol 234 (1994) 269–279. [17] X.X. Wang, P. Wu, X.D. Hou, Y. Lv, Analyst 138 (2013) 229–233. [18] X. Zhang, S. Yu, W.Y. He, H. Uyama, Q.J. Xie, L. Zhang, F.C. Yang, Biosens. Bioelectron. 55 (2014) 446–451. [19] X. Wang, M. Wu, W.R. Tang, Y. Zhu, L.W. Wang, Q.J. Wang, P.G. He, Y.Z. Fang, J. Electroanal. Chem. 695 (2013) 10–16. [20] D. Karak, A. Banerjee, A. Sahana, S. Guha, S. Lohar, S.S. Adhikari, D. Das, J. Hazard. Mater. 188 (2011) 274–280. [21] Y. Matsuoka, M. Yamato, T. Yamasaki, F. Mito, K.-i. Yamada, Free Radical Biol. Med 53 (2012) 2112–2118. [22] M. Zheng, Z. Xie, D. Qu, D. Li, P. Du, X. Jing, Z. Sun, ACS Appl. Mater. Interfaces 5 (2013) 13242–13247. [23] B. Biqing, T. Nanjiao, M. Mingfeng, Z. Lei, Y. Lihui, F. Quli, W. Lianhui, H. Wei, Soft Mater. 12 (2014) 73–78. [24] R.J. Mehlhorn, J. Biol. Chem. 266 (1991) 2724–2731. [25] J.P. Blinco, K.E. Fairfull-Smith, B.J. Morrow, S.E. Bottle, Aust. J. Chem. 64 (2011) 373–389. [26] S.A. Green, D.J. Simpson, G. Zhou, P.S. Ho, N.V. Blough, J. Am. Chem. Soc. 112 (1990) 7337–7346. [27] G.I. Likhtenshtein, Appl. Biochem. Biotechnol. 152 (2009) 135–155. [28] T. Maki, N. Soh, K. Nakano, T. Imato, Talanta 85 (2011) 1730–1733. [29] G.G. Borisenko, I. Martin, Q. Zhao, A.A. Amoscato, Y.Y. Tyurina, V.E. Kagan, J. Biochem. 279 (2004) 23453–23462. [30] J.L. Chen, S.J. Zhuo, Y.Q. Wu, F. Fang, L. Li, C.Q. Zhu, Spectrochim. Acta, Part A 63 (2006) 438–443. [31] M. Jia, Y. Tang, Y.F. Lam, S.A. Green, N.V. Blough, Anal. Chem. 81 (2009) 8033–8040. [32] E. Lozinsky, V.V. Martin, T.A. Berezina, A.I. Shames, A.L. Weis, G.I. Likhtenshtein, J. Biochem. Biophys. Methods 38 (1999) 29–42. [33] R.D. Kornberg, H.M. McConnell, Biochemistry 10 (1971) 1111–1120. [34] A. Sienkiewicz, A.M.D. Ferreira, B. Danner, C.P. Scholes, J. Magn. Reson. 136 (1999) 137–142. [35] B.S. Winkler, S.M. Orselli, T.S. Rex, Free Radical Biol. Med 17 (1994) 333–349. [36] G.R. Bardajee, A.Y. Li, J.C. Haley, M.A. Winnik, Dyes Pigm 79 (2008) 24–32. [37] V.B. Bojinov, I.P. Panova, Dyes Pigm 80 (2009) 61–66. [38] Y.G. Shi, J.H. Yao, Y.L. Duan, Q.L. Mi, J.H. Chen, Q.Q. Xu, Bioorg. Med. Chem. Lett. 23 (2013) 2538–2542. [39] T. Jesionowski, A. Krysztafkiewicz, Appl. Surf. Sci. 172 (2001) 18–32. [40] Q.T. Meng, X.L. Zhang, C. He, P. Zhou, W.P. Su, C.Y. Duan, Talanta 84 (2011) 53–59. [41] J. Kollar, P. Hrdlovic, S. Chmela, J. Photochem. Photobiol., A 195 (2008) 64–71. [42] J. Kollar, P. Hrdlovic, S. Chmela, J. Photochem. Photobiol., A 204 (2009) 191–199. [43] F. Lin, D. Pei, W. He, Z. Huang, Y. Huang, X. Guo, J. Mater. Chem. 22 (2012) 11801–11807. [44] K.R. Dhariwal, W.O. Hartzell, M. Levine, Am. J. Clin. Nutr. 54 (1991) 712–716. [45] A. Levin, C. DeSouza, C. Zaarour, W. Walsh, M.K. Chan, Z. Verjee, S. McIntyre, K. Adeli, Clin. Biochem. 43 (2010) 1411–1414.
