Biosensors and Bioelectronics 55 (2014) 350–354
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
Electrogenerated chemiluminescence of magnesium chlorophyllin a aqueous solution and its sensitive response to the carcinogen aflatoxin B1 Xiuting Li a, Hongyan Yuan b,n, Lan Li a, Dan Xiao a,b,nn a b
College of Chemistry, Sichuan University, Chengdu 610064, PR China College of Chemical Engineering, Sichuan University, Chengdu 610065, PR China
art ic l e i nf o
a b s t r a c t
Article history: Received 29 September 2013 Received in revised form 2 December 2013 Accepted 9 December 2013 Available online 25 December 2013
The chlorophylls, crucial participants of photosynthesis, are large conjugated molecules with special electron donor–acceptor properties. Based on this, many investigations on the electrogenerated chemiluminescence (ECL) of chlorophyll a (Chl a) were performed in organic solvents, but no efficient signals were detected. Herein, ECL research of magnesium chlorophyllins a (Chlorins a) from simple saponification of the natural Chl a was carried out, and highly efficient and stable ECL signal was obtained for the first time. The mechanism study indicated that the ECL resulted from radical ion annihilation. Under the optimal conditions, the effect of the key gas O2 on the ECL of Chlorins a aqueous solution was investigated, and the recoverable inhibition of O2 was obviously observed. What is more, owing to the strong non-covalent interaction between Chlorins a and the carcinogen aflatoxin B1 (AFB1), ECL intensity of Chlorins a aqueous solution exhibited fast, sensitive and selective response to AFB1 with a low detection limit of 0.027 ppb at the signal-to-noise ratio of 3. The costless and environmentally friendly ECL method opens a new potential way of the rapid detection of AFB1 in practical application. & 2013 Elsevier B.V. All rights reserved.
Keywords: Electrogenerated chemiluminescence Magnesium chlorophyllins a Annihilation mechanism Sensitive response Aflatoxin B1
1. Introduction Chlorophyll has attracted significant attention as an important topic for a long time because of its prominent role in photosynthesis. Much information on chlorophyll is being accumulated to give a complete and clear elucidation of the photosynthesis mechanism (Nagata et al., 2012; Springer et al., 2012; Bhuyan et al., 2011; Saito et al., 2011; Wang et al., 2004). In order to better understand the electron transfer processes of chlorophyll a (Chl a) in the initial oxidation–reduction reaction following light-harvesting, many researchers studied the electrochemical behavior of Chl a (Allakhverdiev et al., 2010; Kato et al., 2009; Kuroiwa et al., 2009; Cotton and Van Duyne, 1979). Furthermore, as a large conjugated molecule with special electron donor–acceptor properties, the excited state of Chl a might be produced from the annihilation reaction between its corresponding cationic radical and anionic radical. Based on this, extensive electrogenerated chemiluminescence (ECL) studies of Chl a in organic solvents were conducted in the 1970 s and 1980 s (Krasnovskii and Litvin, 1972; Saji and Bard, 1977; Wasielewski et al., 1980). However, efficient ECL from radical ion annihilation of Chl a was not observed in n
Corresponding author. Tel.: þ 86-28-85415029; fax: þ86-28-85416029 Corresponding author at: College of Chemistry, Sichuan University, Chengdu 610064, PR China. Tel.: þ86 28 85415029; fax: þ86 28 85416029. E-mail address: [email protected] (D. Xiao). nn
0956-5663/$ - see front matter & 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.bios.2013.12.026
these investigations. In recent years, the electroluminescence of Chl a was studied and some corresponding light-emitting diodes were fabricated (Matsuda et al., 2008; Tajima et al., 2006), while few efforts have been spent on studying the ECL of chlorophylls and its derivatives, and the charge conversion and energy transduction of them in aqueous solution have not been investigated with ECL technology. Our attempts to study the optical–electric properties of chlorophylls and its derivatives lead to the discovery of the excellent ECL performance of the natural chlorophylls derivatives in aqueous solution. Since the photosynthesis occurs in water environment, the fat-soluble Chl a was transformed to the water-soluble chlorophyllin a with a simple saponification reaction. Being a diester, chlorophyll a can be saponified by alkali treatment, which will result in de-esterification of the methyl and phytyl esters. Moreover, chlorophyll a is a ß-ketoeater and susceptible to be attacked by hydroxide ions, leading to destruction of the isocyclic ring. Thus, the saponification product chlorophyllins a is a water-soluble mixture of numbers of compounds, and most of them comprise the porphyrin ring combined with active carboxylic anions (Mínguez-Mosquera and Gandul-Rojas, 1995; Mortensen and Geppel, 2007), as shown in Fig. S1. Analogically, the ECL behavior of hemin with a similar structure has been reported (Nan Chen et al., 2000). In the present work, the ECL behavior of magnesium chlorophyllins a (Chlorins a) aqueous solution was investigated, and the ECL mechanism was studied and established.
X. Li et al. / Biosensors and Bioelectronics 55 (2014) 350–354
For decades of years, chlorophyllins have been used for medicinal purposes as accelerant of wound healing and for internal deodorization (Kephart, 1955; Young and Beregi, 1980). And as constituents of the human diet, chlorophyll and chlorophyllins have been shown to exert antimutagenic and anticarcinogenic behavior against a wide range of dietary and environmental agents, including aflatoxin B1 (AFB1), heterocyclic amines, and polycyclic aromatic hydrocarbons. Antimutagenic/anticarcinogenic mechanistic studies show that chlorophyllin can form tight molecular complexes with the carcinogens (Breinholt et al., 1995; Dashwood et al., 1996; Reddy et al., 1999). The formation of complex impedes carcinogen absorption and reduces bioavailability to the target tissue, thereby leading to less DNA adduction and lower tumor incidence (Dashwood et al., 1998). Based on the formation of the Chlorins a–AFB1 complex, the ECL response of Chlorins a aqueous solution for the carcinogen AFB1 was studied. There are a lot of chemical and biochemical methods for determination of AFB1. Official methods include high performance liquid chromatography (HPLC) coupled with fluorescence detectors (Nguyen et al., 2007) or mass spectrometry (Soleimany et al., 2012). These methods require expensive instruments and are time-consuming. Recently, a variety of simpler and faster analytical procedures for AFB1 analyses have emerged, such as immune-chromatography (Li et al., 2013) and electrochemical biosensors (Vidal et al., 2013). ECL as a fast and sensitive electrochemical technique has been widely used in analytical field. In this paper, an ECL method with Chlorins a as probe was developed to detect AFB1 for the first time, and the proposed ECL method was employed to detect AFB1 in real sample.
351
2.3. Saponification of chlorophyll a According to the previous report (Khachik et al., 1986), the saponification of high-purity Chl a was carried out under a nitrogen atmosphere for 3 h at room temperature using methanolic potassium hydroxide (30%). The required amount of potassium hydroxide was estimated and an excess was added to ensure complete conversion of chlorophyll a to chlorophyllin a, which was indicated by the complete solubility of the blue-green product in water and the absence of any blue-green ether-soluble residue. After saponification the methanol was distilled out under reduced pressure and the resultant solution was partitioned into water and petroleum ether. The organic layer was removed and the aqueous layer was washed repeatedly with petroleum ether to remove the phytol released from the saponification reaction. The visible absorption spectra were collected to characterize Chlorins a. Fig. S2 illustrates the visible light absorption spectra of Chl a (curve A) and Chlorins a (curve B) in methanol. Clearly, the principal absorption bands of Chl a at 430 and 662 nm are shifted to 417 and 641 nm for Chlorins a, which is well consistent with the previous report (Oster et al., 1964). According to the report, the shoulder on the blue band of Chl a is absent in the Chlorins a spectra, presumably due to loss of the cyclopentanone ring since it is also absent in the allomerized chlorophyll, where the ring has been transformed (Holt and Jacobs, 1954).
3. Results and discussion 3.1. Possible ECL mechanism of Chlorins a aqueous solution
2. Experimental section 2.1. Materials The high-purity chlorophyll a (Chl a) and aflatoxins (aflatoxin B1 (AFB1), aflatoxin B2 (AFB2), aflatoxin M1 (AFM1), aflatoxin M2 (AFM2), aflatoxin G1 (AFG1) and aflatoxin G2 (AFG2)) were supplied by Sigma-Aldrich (St. Louis, MO, USA) and Supleco (Bellefonte, PA, USA) respectively. KOH and all organic solvents, including acetone, and methyl alcohol, were purchased from Chengdu Chemicals (Sichuan, China) and used as received without further purification. The corn sample solution with 82.6 ng/mL of AFB1 (determined by official HPLC method after immunoaffinity columns clear-up (Romer Labs, Inc.)) was provided by Sichuan Provincial Agricultural Department (Chengdu, China). All aqueous solutions were prepared with doubly distilled water, which was used throughout the whole experiment.
The ECL of Chlorins a aqueous solution purged with a constant flow (20 mL/min) of N2 was investigated between 0.50 V and þ1.50 V, displaying a peak at around 1.45 V in forward scan (the solid curve in Fig. 1). The ECL intensity–time curves in Fig. 1 (inset) demonstrated that the ECL of Chlorins a exhibited high intensity and stability. In order to study the ECL mechanism of Chlorins a aqueous solution, a double potential pulse sweep was performed. As shown in Fig. 2, ECL signal was not detected at the first positive (or negative) potential step but evident ECL signal was found at the subsequent potential steps. Obviously, both Chlorins a þ and Chlorins a are requisites for the ECL of Chlorins a, implying that the ECL occurs upon radical ion annihilation, which is in agreement with the ECL mechanism of Chl a in organic solvents reported in the previous studies (Saji and Bard, 1977; Wasielewski et al., 1980). Thus, it can be
2.2. Apparatus ECL measurements were carried out with a Remax MPI-E ECL instrument (Xi0 an Remax Electronic Science Tech. Co. Ltd., Xi0 an, China) using a conventional three-electrode system. A glass carbon electrode 3 mm in diameter was employed as the working electrode, with an Ag/AgCl electrode (saturated KCl) and a platinum wire as the reference and counter-electrodes respectively. The ECL emission spectrum as well as photoluminescence (PL) spectrum were measured on a Hitachi F-4500 fluorescence spectrophotometer (Tokyo, Japan). The visible absorption spectra were acquired on a UV-1100 spectrophotometer (Shanghai, China) and recorded from 400 to 700 nm with a bandwidth of 4.0 nm and a scan rate of 800 nm min 1. The pH of PBS was measured with a pH meter (Thermo, Orion 920A). The rate of gases was controlled by mass flow controllers (Brooks Instrument, 0154).
Fig. 1. ECL (the solid curve), CV (the dash curve) and ECL intensity–time curves (inset) of Chlorins a aqueous solution (5.0 10 5 M; pH 13) at a scan rate of 0.25 V/s.
352
X. Li et al. / Biosensors and Bioelectronics 55 (2014) 350–354
Fig. 2. The double potential pulse sweep curves of Chlorins a aqueous solution (5.0 10 5 M) in ECL study. (a) The first potential is positive; and (b) the first potential is negative. The positive and negative potentials were set at þ1.5 V, 0.5 V respectively and employed by turns with a length of time 1 s.
concluded that in the ECL reaction Chlorins a were reduced and oxidized to Chlorins a and Chlorins a þ respectively (Eqs. (1) and (2)), which subsequently interacted and produced excited-state 1 Chlorins an (Eq. (3)) or 3Chlorins an (Eq. (4)). If 1Chlorins an was produced, it directly decayed to ground state with the simultaneous release of luminescence (Eq. (6)). If 3Chlorins an was generated, the triplet–triplet annihilation proceeded to produce 1Chlorins an (Eq. (5)) before it decayed to ground state and radiated light (Eq. (6)). Chlorins a þe -Chlorins a U
ð1Þ
Chlorins a e -Chlorins a U þ
ð2Þ
Chlorins a U þ Chlorins a U þ -1 Chlorins an þ Chlorins a
ð3Þ
or Chlorins a U þ Chlorins a U þ -3 Chlorins an þ Chlorins a
ð4Þ
3
Chlorins an þ 3 Chlorins an -1 Chlorins an þ Chlorins a
ð5Þ
1
Chlorins an -Chlorins a þ hv
ð6Þ
Chlorins an þ 3 O2 -Chlorins a þ 1 O2
ð7Þ
The ECL emission spectrum of Chlorins a aqueous solution was collected by directly placing the cell for ECL in front of the window in the spectrofluorometer. As shown in Fig. 3, an emission peak of ca. 650 nm was obtained, which is quite similar with the photoluminescence of Chlorins a aqueous solution, suggesting that the same excited state Chlorins an was generated in the electrochemical reactions and photo-excitation process. In previous reports, the authors analyzed in detail the reason why no ECL of Chl a was observed in organic solvents. It can be concluded that the dominant inefficient step was most likely the formation of excited state in the radical ion annihilation step (Saji and Bard, 1977; Wasielewski et al., 1980). In the present work, high and steady ECL signal in aqueous solution was observed, demonstrating that the efficiency of the radical ion annihilation was probably highly enhanced after saponification of Chl a. This may be ascribed to the facilitation for the electron transfer in ECL reaction of the porphyrin ring directly combined with active carboxylic anions.
Fig. 3. Photoluminescence (a) and ECL emission spectrum (b) of Chlorins a aqueous solution (5.0 10 5 M). The ECL spectrum was obtained under CV scan between 0.5 V and 1.5 V by putting the cell for ECL in front of the window in the spectrofluorometer.
3.2. Optimization of the ECL experimental conditions In order to obtain the optimal ECL performance, the effects of pH and scan rate on the ECL of Chlorins a aqueous solution were investigated. The strong dependence of ECL signal of Chlorins a aqueous solution on the pH of the solution is shown in Fig. 4(a). The suitable pH value of the solution to the ECL analysis of Chlorins a is up to 13. This is likely because much more active carboxylic anions directly linking with the porphyrin ring were formed in strong alkaline solution, leading to more efficient ECL. The effect of the scan rates on the ECL intensity of Chlorins a was also investigated. Fig. 4(b) shows the ECL peak intensity at different scan rates in the range from 0.05 to 0.50 V/s. It can be observed that the ECL intensity increased sharply along with the increase of scan rate over the range of 0.05–0.30 V/s and then remained almost constant from 0.30 to 0.50 V/s. Fig. S3 shows the effects of scan rate on currents at 1.45 V where ECL peaks. It can be observed that the currents are proportional to the square root of the scan rate in the range of 0.05–0.50 V/s, indicating a diffusion controlled process. Owing to the fast electrochemical oxidation reduction reactions, differences in the ECL intensity in response to diverse scan rates mainly depended on the diffusion rate of the luminophor Chlorins a to the electrode surface. In the range of 0.05–0.30 V/s, the reaction between Chlorins a was easier and easier with the increase of scan rate, leading to the sharply increased ECL signals. However, the diffusion rate of Chlorins a remained almost constant at the very high scan rates and consequently no ECL increase was observed over 0.30 V/s. On the other hand, the ECL intensity exhibited slight fluctuation when the scan rate was over 0.25 V/s. Therefore, the optimum scan rate of 0.25 V/ s is employed to the ECL experiment. Thus, the solution of pH 13 and the scan rate 0.25 V/s were the optimal conditions and resultantly employed for the ECL investigation in the following experiments. 3.3. Effect of O2 on ECL of Chlorins a aqueous solution The solution was bubbled successively by N2 and O2 with a constant flow (20 mL/min) to study the effect of O2 on ECL of Chlorins a since it plays important role in photosynthesis. It was observed that the ECL intensity was increased evidently when N2
X. Li et al. / Biosensors and Bioelectronics 55 (2014) 350–354
353
Fig. 4. (a) Effect of pH value of the solution on the ECL intensity of Chlorins a aqueous solution (5.0 10 5 M; Scan rate: 0.25 V/s); (b) Effect of scan rate on the ECL intensity of Chlorins a aqueous solution (5.0 10 5 M; pH 13).
Fig. 5. Effect of O2 on the ECL of Chlorins a aqueous solution (5.0 10 5 M; pH 13) at a scan rate of 0.25 V/s.
was injected into the air-saturated solution. This is due to the removal of O2, which has been proved to inhibit the ECL signal of Chlorins a (Fig. 5). The inhibition may result from the quenching of molecular oxygen for Chlorins a excited state and probably be facilitated by the electrostatic bonding between oxygen atom and the partly electropositive magnesium atom. To be specific, Chlorins an can activate the triplet (ground) state oxygen molecule (3O2) dissolved in aqueous solution to become singlet (excited) state oxygen molecule (1O2), decaying to its ground state simultaneously (Eq. (7)) (Wang et al., 2011). It also can be seen that the ECL intensity recovered after a sharp decrease when O2 was replaced by N2, indicating that the combination of Chlorins a and O2 was reversible.
3.4. ECL response of Chlorins a aqueous solution to AFB1 As illustrated in Fig. 6, the ECL intensity decreased gradually with the increasing of AFB1 concentration and a low concentration (0.24 ppb (ng/mL)) of AFB1 was detected. In a previous work, it has been proved that the 8,9-double bond of AFB1 oriented 1801 with respect to the carboxyl groups of Cu Chlorins a and a strong noncovalent complex was formed consequently (Dashwood et al., 1996). Moreover, the 4-coordinate central metal in the chlorophyllin molecule was proved having no effect on complex formation (Arimoto et al., 1993; Shelnutt, 1984). Accordingly, it can be inferred that the sensitive inhibition of AFB1 on ECL of Chlorins a resulted from its combination with Chlorins a, and probably the
Fig. 6. ECL intensity–time curves of Chlorins a aqueous solution (5.0 10 5 M) in different concentrations (ppb (ng/mL)) of AFB1: 0, 0.24, 0.48, 0.60, 0.72, 0.84, 0.96, 1.08, 1.32, 1.56, 1.80 at a scan rate of 0.25 V/s. Inset: the correlativity of ECL intensity and the concentrations of AFB1.
strong non-covalent interaction caused the change of the electrochemical property of the porphyrin ring in Chlorins a. Fig. 6 inset showed that the ECL intensity exhibited non-linear response toward the concentration of AFB1 from 0.24 ppb to 1.8 ppb, the non-linear fitting equation was y¼1.208 e 0.756x– 0.183 and the relative coefficient was R2 ¼0.9914. The relative standard deviation was 2.0% for three repeated measurements of 0.84 ng/mL AFB1. Compared to other methods, a lower detection limit of 0.027 ppb was obtained with a signal-to-noise ratio of 3, as shown in Table S1. In addition, the ECL response of Chlorins a to other common aflatoxins (AFB2, AFM1, AFM2, AFG1, and AFG2) were also investigated. The same concentration AFB2, AFM1, AFM2, AFG1 and AFG2 were added into the Chlorins a aqueous solution with 0.84 ppb AFB1. As shown in Fig. S4, no obvious change of ECL intensity was observed when AFB2, AFM2 and AFG2 were included, while AFM1 and AFG1 exhibited slight inhibition to ECL intensity of the system, which might be ascribed to the existence of the 8,9 double bond. Overall, presumably structural differences resulted in their weaker interaction with Chlorins a than AFB1, leading to the good specificity of ECL response of Chlorins a to AFB1. It means that a fast, sensitive and selective ECL method was developed here for the detection of the carcinogen AFB1.
354
X. Li et al. / Biosensors and Bioelectronics 55 (2014) 350–354
In order to evaluate the analytical reliability and application potential of the proposed ECL method in practical analysis, the concentration of AFB1 in corn sample solution was determined with Chlorins a as ECL probe. The determination results (83.9 ng/mL, 82.0 ng/mL, 84.3 ng/mL) are well consistent with the value of 82.6 ng/mL from the official HPLC method (Romer Labs, Inc.) conducted by Sichuan Provincial Agricultural Department (Chengdu, China). In addition, acceptable recoveries (96.7–105.3%) were obtained, implying that the analytical performance of the proposed ECL method is satisfactory.
4. Conclusions In conclusion, the Chlorins a aqueous solution formed by saponification of Chl a exhibits highly efficient and stable ECL signal, which most probably resulted from effective radical ion annihilation reaction of Chlorins a with active carboxylic anions. The effect of the key gas O2 on the ECL of Chlorins a aqueous solution was studied and recoverable inhibition of O2 was obviously observed here. What is more, fast, sensitive and selective ECL response of Chlorins a aqueous solution to AFB1 was presented here for the first time, and a low detection limit of 0.027 ppb was obtained. The costless and environmentally friendly ECL method opens a new potential way of the rapid detection of AFB1 in practical application.
Acknowledgment This work is financially supported by the National Natural Science Foundation of China (21075083, 21275104 and 21177090).
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.2013.12.026.
References Allakhverdiev, S.I., Tomo, T., Shimada, Y., Kindo, H., Nagao, R., Klimov, V.V., Mimuro, M., 2010. Proc. Natl. Acad. Sci. 107, 3924–3929. Arimoto, S., Fukuoka, S., Itome, C., Nakano, H., Rai, H., Hayatsu, H., 1993. Mutat. Res./ Fundam. Mol. Mech. Mutagenesis 287, 293–305. Bhuyan, J., Sarkar, R., Sarkar, S., 2011. Angew. Chem. Int. Ed. 50, 10603–10607. Breinholt, V., Schimerlik, M., Dashwood, R., Bailey, G., 1995. Chem. Res. Toxicol. 8, 506–514. Cotton, T.M., Van Duyne, R.P., 1979. J. Am. Chem. Soc. 101, 7605–7612. Dashwood, R., Negishi, T., Hayatsu, H., Breinholt, V., Hendricks, J., Bailey, G., 1998. Mutat. Res./Fundam. Mol. Mech. Mutagenesis 399, 245–253. Dashwood, R., Yamane, S., Larsen, R., 1996. Environ. Mol. Mutagenesis 27, 211–218. Holt, A.S., Jacobs, E.E., 1954. Am. J. Bot. 41, 710–717. Kato, Y., Sugiura, M., Oda, A., Watanabe, T., 2009. Proc. Natl. Acad. Sci. 106, 17365–17370. Kephart, J., 1955. Econ. Bot. 9, 3–38. Khachik, F., Beecher, G.R., Whittaker, N.F., 1986. J. Agric. Food Chem. 34, 603–616. Krasnovskii, A.A., Litvin, F.F., 1972. Biofizika 17, 764–768. Kuroiwa, Y., Kato, Y., Watanabe, T., 2009. J. Photochem. Photobiol. A: Chem. 202, 191–195. Li, X., Li, P., Zhang, Q., Li, R., Zhang, W., Zhang, Z., Ding, X., Tang, X., 2013. Biosens. Bioelectron. 49, 426–432. Mínguez-Mosquera, M.I., Gandul-Rojas, B., 1995. J. Chromatogr. A 690, 161–176. Matsuda, M., Isozaki, H., Tajima, H., 2008. Thin Solid Films 517, 1465–1467. Mortensen, A., Geppel, A., 2007. Innov. Food Sci. Emerg. Technol. 8, 419–425. Nagata, M., Amano, M., Joke, T., Fujii, K., Okuda, A., Kondo, M., Ishigure, S., Dewa, T., Iida, K., Secundo, F., Amao, Y., Hashimoto, H., Nango, M., 2012. ACS Macro Lett. 1, 296–299. Nan Chen, G., Zhang, L., Er Lin, R., Cong Yang, Z., Ping Duan, J., Qing Chen, H., Brynn Hibbert, D., 2000. Talanta 50, 1275–1281. Nguyen, M.T., Tozlovanu, M., Tran, T.L., Pfohl-Leszkowicz, A., 2007. Food Chem. 105, 42–47. Reddy, A.P., Harttig, U., Barth, M.C., Baird, W.M., Schimerlik, M., Hendricks, J.D., Bailey, G.S., 1999. Carcinogenesis 20, 1919–1926. Saito, K., Ishida, T., Sugiura, M., Kawakami, K., Umena, Y., Kamiya, N., Shen, J.-R., Ishikita, H., 2011. J. Am. Chem. Soc. 133, 14379–14388. Saji, T., Bard, A.J., 1977. J. Am. Chem. Soc. 99, 2235–2240. Shelnutt, J.A., 1984. J. Phys. Chem. 88, 6121–6127. Soleimany, F., Jinap, S., Faridah, A., Khatib, A., 2012. Food Control 25, 647–653. Springer, J.W., Parkes-Loach, P.S., Reddy, K.R., Krayer, M., Jiao, J., Lee, G.M., Niedzwiedzki, D.M., Harris, M.A., Kirmaier, C., Bocian, D.F., Lindsey, J.S., Holten, D., Loach, P.A., 2012. J. Am. Chem. Soc. 134, 4589–4599. Tajima, H., Shimatani, K., Komino, T., Ikeda, S., Matsuda, M., Ando, Y., Akiyama, H., 2006. Coll. Surf. A: Physicochem. Eng. Asp. 284–285, 61–65. Vidal, J.C., Bonel, L., Ezquerra, A., Hernández, S., Bertolín, J.R., Cubel, C., Castillo, J.R., 2013. Biosens. Bioelectron. 49, 146–158. Wang, C., Xing, D., Chen, Q., 2004. Biosens.Bioelectron. 20, 454–459. Wang, J., Guo, Y., Gao, J., Jin, X., Wang, Z., Wang, B., Li, K., Li, Y., 2011. Ultrason. Sonochem. 18, 1028–1034. Wasielewski, M.R., Smith, R.L., Kostka, A.G., 1980. J. Am. Chem. Soc. 102, 6923–6928. Young, R.W., Beregi Jr., J.S., 1980. J. Am. Geriatr. Soc. 28, 46–47.
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
Electrogenerated chemiluminescence of magnesium chlorophyllin a aqueous solution and its sensitive response to the carcinogen aflatoxin B1 Xiuting Li a, Hongyan Yuan b,n, Lan Li a, Dan Xiao a,b,nn a b
College of Chemistry, Sichuan University, Chengdu 610064, PR China College of Chemical Engineering, Sichuan University, Chengdu 610065, PR China
art ic l e i nf o
a b s t r a c t
Article history: Received 29 September 2013 Received in revised form 2 December 2013 Accepted 9 December 2013 Available online 25 December 2013
The chlorophylls, crucial participants of photosynthesis, are large conjugated molecules with special electron donor–acceptor properties. Based on this, many investigations on the electrogenerated chemiluminescence (ECL) of chlorophyll a (Chl a) were performed in organic solvents, but no efficient signals were detected. Herein, ECL research of magnesium chlorophyllins a (Chlorins a) from simple saponification of the natural Chl a was carried out, and highly efficient and stable ECL signal was obtained for the first time. The mechanism study indicated that the ECL resulted from radical ion annihilation. Under the optimal conditions, the effect of the key gas O2 on the ECL of Chlorins a aqueous solution was investigated, and the recoverable inhibition of O2 was obviously observed. What is more, owing to the strong non-covalent interaction between Chlorins a and the carcinogen aflatoxin B1 (AFB1), ECL intensity of Chlorins a aqueous solution exhibited fast, sensitive and selective response to AFB1 with a low detection limit of 0.027 ppb at the signal-to-noise ratio of 3. The costless and environmentally friendly ECL method opens a new potential way of the rapid detection of AFB1 in practical application. & 2013 Elsevier B.V. All rights reserved.
Keywords: Electrogenerated chemiluminescence Magnesium chlorophyllins a Annihilation mechanism Sensitive response Aflatoxin B1
1. Introduction Chlorophyll has attracted significant attention as an important topic for a long time because of its prominent role in photosynthesis. Much information on chlorophyll is being accumulated to give a complete and clear elucidation of the photosynthesis mechanism (Nagata et al., 2012; Springer et al., 2012; Bhuyan et al., 2011; Saito et al., 2011; Wang et al., 2004). In order to better understand the electron transfer processes of chlorophyll a (Chl a) in the initial oxidation–reduction reaction following light-harvesting, many researchers studied the electrochemical behavior of Chl a (Allakhverdiev et al., 2010; Kato et al., 2009; Kuroiwa et al., 2009; Cotton and Van Duyne, 1979). Furthermore, as a large conjugated molecule with special electron donor–acceptor properties, the excited state of Chl a might be produced from the annihilation reaction between its corresponding cationic radical and anionic radical. Based on this, extensive electrogenerated chemiluminescence (ECL) studies of Chl a in organic solvents were conducted in the 1970 s and 1980 s (Krasnovskii and Litvin, 1972; Saji and Bard, 1977; Wasielewski et al., 1980). However, efficient ECL from radical ion annihilation of Chl a was not observed in n
Corresponding author. Tel.: þ 86-28-85415029; fax: þ86-28-85416029 Corresponding author at: College of Chemistry, Sichuan University, Chengdu 610064, PR China. Tel.: þ86 28 85415029; fax: þ86 28 85416029. E-mail address: [email protected] (D. Xiao). nn
0956-5663/$ - see front matter & 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.bios.2013.12.026
these investigations. In recent years, the electroluminescence of Chl a was studied and some corresponding light-emitting diodes were fabricated (Matsuda et al., 2008; Tajima et al., 2006), while few efforts have been spent on studying the ECL of chlorophylls and its derivatives, and the charge conversion and energy transduction of them in aqueous solution have not been investigated with ECL technology. Our attempts to study the optical–electric properties of chlorophylls and its derivatives lead to the discovery of the excellent ECL performance of the natural chlorophylls derivatives in aqueous solution. Since the photosynthesis occurs in water environment, the fat-soluble Chl a was transformed to the water-soluble chlorophyllin a with a simple saponification reaction. Being a diester, chlorophyll a can be saponified by alkali treatment, which will result in de-esterification of the methyl and phytyl esters. Moreover, chlorophyll a is a ß-ketoeater and susceptible to be attacked by hydroxide ions, leading to destruction of the isocyclic ring. Thus, the saponification product chlorophyllins a is a water-soluble mixture of numbers of compounds, and most of them comprise the porphyrin ring combined with active carboxylic anions (Mínguez-Mosquera and Gandul-Rojas, 1995; Mortensen and Geppel, 2007), as shown in Fig. S1. Analogically, the ECL behavior of hemin with a similar structure has been reported (Nan Chen et al., 2000). In the present work, the ECL behavior of magnesium chlorophyllins a (Chlorins a) aqueous solution was investigated, and the ECL mechanism was studied and established.
X. Li et al. / Biosensors and Bioelectronics 55 (2014) 350–354
For decades of years, chlorophyllins have been used for medicinal purposes as accelerant of wound healing and for internal deodorization (Kephart, 1955; Young and Beregi, 1980). And as constituents of the human diet, chlorophyll and chlorophyllins have been shown to exert antimutagenic and anticarcinogenic behavior against a wide range of dietary and environmental agents, including aflatoxin B1 (AFB1), heterocyclic amines, and polycyclic aromatic hydrocarbons. Antimutagenic/anticarcinogenic mechanistic studies show that chlorophyllin can form tight molecular complexes with the carcinogens (Breinholt et al., 1995; Dashwood et al., 1996; Reddy et al., 1999). The formation of complex impedes carcinogen absorption and reduces bioavailability to the target tissue, thereby leading to less DNA adduction and lower tumor incidence (Dashwood et al., 1998). Based on the formation of the Chlorins a–AFB1 complex, the ECL response of Chlorins a aqueous solution for the carcinogen AFB1 was studied. There are a lot of chemical and biochemical methods for determination of AFB1. Official methods include high performance liquid chromatography (HPLC) coupled with fluorescence detectors (Nguyen et al., 2007) or mass spectrometry (Soleimany et al., 2012). These methods require expensive instruments and are time-consuming. Recently, a variety of simpler and faster analytical procedures for AFB1 analyses have emerged, such as immune-chromatography (Li et al., 2013) and electrochemical biosensors (Vidal et al., 2013). ECL as a fast and sensitive electrochemical technique has been widely used in analytical field. In this paper, an ECL method with Chlorins a as probe was developed to detect AFB1 for the first time, and the proposed ECL method was employed to detect AFB1 in real sample.
351
2.3. Saponification of chlorophyll a According to the previous report (Khachik et al., 1986), the saponification of high-purity Chl a was carried out under a nitrogen atmosphere for 3 h at room temperature using methanolic potassium hydroxide (30%). The required amount of potassium hydroxide was estimated and an excess was added to ensure complete conversion of chlorophyll a to chlorophyllin a, which was indicated by the complete solubility of the blue-green product in water and the absence of any blue-green ether-soluble residue. After saponification the methanol was distilled out under reduced pressure and the resultant solution was partitioned into water and petroleum ether. The organic layer was removed and the aqueous layer was washed repeatedly with petroleum ether to remove the phytol released from the saponification reaction. The visible absorption spectra were collected to characterize Chlorins a. Fig. S2 illustrates the visible light absorption spectra of Chl a (curve A) and Chlorins a (curve B) in methanol. Clearly, the principal absorption bands of Chl a at 430 and 662 nm are shifted to 417 and 641 nm for Chlorins a, which is well consistent with the previous report (Oster et al., 1964). According to the report, the shoulder on the blue band of Chl a is absent in the Chlorins a spectra, presumably due to loss of the cyclopentanone ring since it is also absent in the allomerized chlorophyll, where the ring has been transformed (Holt and Jacobs, 1954).
3. Results and discussion 3.1. Possible ECL mechanism of Chlorins a aqueous solution
2. Experimental section 2.1. Materials The high-purity chlorophyll a (Chl a) and aflatoxins (aflatoxin B1 (AFB1), aflatoxin B2 (AFB2), aflatoxin M1 (AFM1), aflatoxin M2 (AFM2), aflatoxin G1 (AFG1) and aflatoxin G2 (AFG2)) were supplied by Sigma-Aldrich (St. Louis, MO, USA) and Supleco (Bellefonte, PA, USA) respectively. KOH and all organic solvents, including acetone, and methyl alcohol, were purchased from Chengdu Chemicals (Sichuan, China) and used as received without further purification. The corn sample solution with 82.6 ng/mL of AFB1 (determined by official HPLC method after immunoaffinity columns clear-up (Romer Labs, Inc.)) was provided by Sichuan Provincial Agricultural Department (Chengdu, China). All aqueous solutions were prepared with doubly distilled water, which was used throughout the whole experiment.
The ECL of Chlorins a aqueous solution purged with a constant flow (20 mL/min) of N2 was investigated between 0.50 V and þ1.50 V, displaying a peak at around 1.45 V in forward scan (the solid curve in Fig. 1). The ECL intensity–time curves in Fig. 1 (inset) demonstrated that the ECL of Chlorins a exhibited high intensity and stability. In order to study the ECL mechanism of Chlorins a aqueous solution, a double potential pulse sweep was performed. As shown in Fig. 2, ECL signal was not detected at the first positive (or negative) potential step but evident ECL signal was found at the subsequent potential steps. Obviously, both Chlorins a þ and Chlorins a are requisites for the ECL of Chlorins a, implying that the ECL occurs upon radical ion annihilation, which is in agreement with the ECL mechanism of Chl a in organic solvents reported in the previous studies (Saji and Bard, 1977; Wasielewski et al., 1980). Thus, it can be
2.2. Apparatus ECL measurements were carried out with a Remax MPI-E ECL instrument (Xi0 an Remax Electronic Science Tech. Co. Ltd., Xi0 an, China) using a conventional three-electrode system. A glass carbon electrode 3 mm in diameter was employed as the working electrode, with an Ag/AgCl electrode (saturated KCl) and a platinum wire as the reference and counter-electrodes respectively. The ECL emission spectrum as well as photoluminescence (PL) spectrum were measured on a Hitachi F-4500 fluorescence spectrophotometer (Tokyo, Japan). The visible absorption spectra were acquired on a UV-1100 spectrophotometer (Shanghai, China) and recorded from 400 to 700 nm with a bandwidth of 4.0 nm and a scan rate of 800 nm min 1. The pH of PBS was measured with a pH meter (Thermo, Orion 920A). The rate of gases was controlled by mass flow controllers (Brooks Instrument, 0154).
Fig. 1. ECL (the solid curve), CV (the dash curve) and ECL intensity–time curves (inset) of Chlorins a aqueous solution (5.0 10 5 M; pH 13) at a scan rate of 0.25 V/s.
352
X. Li et al. / Biosensors and Bioelectronics 55 (2014) 350–354
Fig. 2. The double potential pulse sweep curves of Chlorins a aqueous solution (5.0 10 5 M) in ECL study. (a) The first potential is positive; and (b) the first potential is negative. The positive and negative potentials were set at þ1.5 V, 0.5 V respectively and employed by turns with a length of time 1 s.
concluded that in the ECL reaction Chlorins a were reduced and oxidized to Chlorins a and Chlorins a þ respectively (Eqs. (1) and (2)), which subsequently interacted and produced excited-state 1 Chlorins an (Eq. (3)) or 3Chlorins an (Eq. (4)). If 1Chlorins an was produced, it directly decayed to ground state with the simultaneous release of luminescence (Eq. (6)). If 3Chlorins an was generated, the triplet–triplet annihilation proceeded to produce 1Chlorins an (Eq. (5)) before it decayed to ground state and radiated light (Eq. (6)). Chlorins a þe -Chlorins a U
ð1Þ
Chlorins a e -Chlorins a U þ
ð2Þ
Chlorins a U þ Chlorins a U þ -1 Chlorins an þ Chlorins a
ð3Þ
or Chlorins a U þ Chlorins a U þ -3 Chlorins an þ Chlorins a
ð4Þ
3
Chlorins an þ 3 Chlorins an -1 Chlorins an þ Chlorins a
ð5Þ
1
Chlorins an -Chlorins a þ hv
ð6Þ
Chlorins an þ 3 O2 -Chlorins a þ 1 O2
ð7Þ
The ECL emission spectrum of Chlorins a aqueous solution was collected by directly placing the cell for ECL in front of the window in the spectrofluorometer. As shown in Fig. 3, an emission peak of ca. 650 nm was obtained, which is quite similar with the photoluminescence of Chlorins a aqueous solution, suggesting that the same excited state Chlorins an was generated in the electrochemical reactions and photo-excitation process. In previous reports, the authors analyzed in detail the reason why no ECL of Chl a was observed in organic solvents. It can be concluded that the dominant inefficient step was most likely the formation of excited state in the radical ion annihilation step (Saji and Bard, 1977; Wasielewski et al., 1980). In the present work, high and steady ECL signal in aqueous solution was observed, demonstrating that the efficiency of the radical ion annihilation was probably highly enhanced after saponification of Chl a. This may be ascribed to the facilitation for the electron transfer in ECL reaction of the porphyrin ring directly combined with active carboxylic anions.
Fig. 3. Photoluminescence (a) and ECL emission spectrum (b) of Chlorins a aqueous solution (5.0 10 5 M). The ECL spectrum was obtained under CV scan between 0.5 V and 1.5 V by putting the cell for ECL in front of the window in the spectrofluorometer.
3.2. Optimization of the ECL experimental conditions In order to obtain the optimal ECL performance, the effects of pH and scan rate on the ECL of Chlorins a aqueous solution were investigated. The strong dependence of ECL signal of Chlorins a aqueous solution on the pH of the solution is shown in Fig. 4(a). The suitable pH value of the solution to the ECL analysis of Chlorins a is up to 13. This is likely because much more active carboxylic anions directly linking with the porphyrin ring were formed in strong alkaline solution, leading to more efficient ECL. The effect of the scan rates on the ECL intensity of Chlorins a was also investigated. Fig. 4(b) shows the ECL peak intensity at different scan rates in the range from 0.05 to 0.50 V/s. It can be observed that the ECL intensity increased sharply along with the increase of scan rate over the range of 0.05–0.30 V/s and then remained almost constant from 0.30 to 0.50 V/s. Fig. S3 shows the effects of scan rate on currents at 1.45 V where ECL peaks. It can be observed that the currents are proportional to the square root of the scan rate in the range of 0.05–0.50 V/s, indicating a diffusion controlled process. Owing to the fast electrochemical oxidation reduction reactions, differences in the ECL intensity in response to diverse scan rates mainly depended on the diffusion rate of the luminophor Chlorins a to the electrode surface. In the range of 0.05–0.30 V/s, the reaction between Chlorins a was easier and easier with the increase of scan rate, leading to the sharply increased ECL signals. However, the diffusion rate of Chlorins a remained almost constant at the very high scan rates and consequently no ECL increase was observed over 0.30 V/s. On the other hand, the ECL intensity exhibited slight fluctuation when the scan rate was over 0.25 V/s. Therefore, the optimum scan rate of 0.25 V/ s is employed to the ECL experiment. Thus, the solution of pH 13 and the scan rate 0.25 V/s were the optimal conditions and resultantly employed for the ECL investigation in the following experiments. 3.3. Effect of O2 on ECL of Chlorins a aqueous solution The solution was bubbled successively by N2 and O2 with a constant flow (20 mL/min) to study the effect of O2 on ECL of Chlorins a since it plays important role in photosynthesis. It was observed that the ECL intensity was increased evidently when N2
X. Li et al. / Biosensors and Bioelectronics 55 (2014) 350–354
353
Fig. 4. (a) Effect of pH value of the solution on the ECL intensity of Chlorins a aqueous solution (5.0 10 5 M; Scan rate: 0.25 V/s); (b) Effect of scan rate on the ECL intensity of Chlorins a aqueous solution (5.0 10 5 M; pH 13).
Fig. 5. Effect of O2 on the ECL of Chlorins a aqueous solution (5.0 10 5 M; pH 13) at a scan rate of 0.25 V/s.
was injected into the air-saturated solution. This is due to the removal of O2, which has been proved to inhibit the ECL signal of Chlorins a (Fig. 5). The inhibition may result from the quenching of molecular oxygen for Chlorins a excited state and probably be facilitated by the electrostatic bonding between oxygen atom and the partly electropositive magnesium atom. To be specific, Chlorins an can activate the triplet (ground) state oxygen molecule (3O2) dissolved in aqueous solution to become singlet (excited) state oxygen molecule (1O2), decaying to its ground state simultaneously (Eq. (7)) (Wang et al., 2011). It also can be seen that the ECL intensity recovered after a sharp decrease when O2 was replaced by N2, indicating that the combination of Chlorins a and O2 was reversible.
3.4. ECL response of Chlorins a aqueous solution to AFB1 As illustrated in Fig. 6, the ECL intensity decreased gradually with the increasing of AFB1 concentration and a low concentration (0.24 ppb (ng/mL)) of AFB1 was detected. In a previous work, it has been proved that the 8,9-double bond of AFB1 oriented 1801 with respect to the carboxyl groups of Cu Chlorins a and a strong noncovalent complex was formed consequently (Dashwood et al., 1996). Moreover, the 4-coordinate central metal in the chlorophyllin molecule was proved having no effect on complex formation (Arimoto et al., 1993; Shelnutt, 1984). Accordingly, it can be inferred that the sensitive inhibition of AFB1 on ECL of Chlorins a resulted from its combination with Chlorins a, and probably the
Fig. 6. ECL intensity–time curves of Chlorins a aqueous solution (5.0 10 5 M) in different concentrations (ppb (ng/mL)) of AFB1: 0, 0.24, 0.48, 0.60, 0.72, 0.84, 0.96, 1.08, 1.32, 1.56, 1.80 at a scan rate of 0.25 V/s. Inset: the correlativity of ECL intensity and the concentrations of AFB1.
strong non-covalent interaction caused the change of the electrochemical property of the porphyrin ring in Chlorins a. Fig. 6 inset showed that the ECL intensity exhibited non-linear response toward the concentration of AFB1 from 0.24 ppb to 1.8 ppb, the non-linear fitting equation was y¼1.208 e 0.756x– 0.183 and the relative coefficient was R2 ¼0.9914. The relative standard deviation was 2.0% for three repeated measurements of 0.84 ng/mL AFB1. Compared to other methods, a lower detection limit of 0.027 ppb was obtained with a signal-to-noise ratio of 3, as shown in Table S1. In addition, the ECL response of Chlorins a to other common aflatoxins (AFB2, AFM1, AFM2, AFG1, and AFG2) were also investigated. The same concentration AFB2, AFM1, AFM2, AFG1 and AFG2 were added into the Chlorins a aqueous solution with 0.84 ppb AFB1. As shown in Fig. S4, no obvious change of ECL intensity was observed when AFB2, AFM2 and AFG2 were included, while AFM1 and AFG1 exhibited slight inhibition to ECL intensity of the system, which might be ascribed to the existence of the 8,9 double bond. Overall, presumably structural differences resulted in their weaker interaction with Chlorins a than AFB1, leading to the good specificity of ECL response of Chlorins a to AFB1. It means that a fast, sensitive and selective ECL method was developed here for the detection of the carcinogen AFB1.
354
X. Li et al. / Biosensors and Bioelectronics 55 (2014) 350–354
In order to evaluate the analytical reliability and application potential of the proposed ECL method in practical analysis, the concentration of AFB1 in corn sample solution was determined with Chlorins a as ECL probe. The determination results (83.9 ng/mL, 82.0 ng/mL, 84.3 ng/mL) are well consistent with the value of 82.6 ng/mL from the official HPLC method (Romer Labs, Inc.) conducted by Sichuan Provincial Agricultural Department (Chengdu, China). In addition, acceptable recoveries (96.7–105.3%) were obtained, implying that the analytical performance of the proposed ECL method is satisfactory.
4. Conclusions In conclusion, the Chlorins a aqueous solution formed by saponification of Chl a exhibits highly efficient and stable ECL signal, which most probably resulted from effective radical ion annihilation reaction of Chlorins a with active carboxylic anions. The effect of the key gas O2 on the ECL of Chlorins a aqueous solution was studied and recoverable inhibition of O2 was obviously observed here. What is more, fast, sensitive and selective ECL response of Chlorins a aqueous solution to AFB1 was presented here for the first time, and a low detection limit of 0.027 ppb was obtained. The costless and environmentally friendly ECL method opens a new potential way of the rapid detection of AFB1 in practical application.
Acknowledgment This work is financially supported by the National Natural Science Foundation of China (21075083, 21275104 and 21177090).
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.2013.12.026.
References Allakhverdiev, S.I., Tomo, T., Shimada, Y., Kindo, H., Nagao, R., Klimov, V.V., Mimuro, M., 2010. Proc. Natl. Acad. Sci. 107, 3924–3929. Arimoto, S., Fukuoka, S., Itome, C., Nakano, H., Rai, H., Hayatsu, H., 1993. Mutat. Res./ Fundam. Mol. Mech. Mutagenesis 287, 293–305. Bhuyan, J., Sarkar, R., Sarkar, S., 2011. Angew. Chem. Int. Ed. 50, 10603–10607. Breinholt, V., Schimerlik, M., Dashwood, R., Bailey, G., 1995. Chem. Res. Toxicol. 8, 506–514. Cotton, T.M., Van Duyne, R.P., 1979. J. Am. Chem. Soc. 101, 7605–7612. Dashwood, R., Negishi, T., Hayatsu, H., Breinholt, V., Hendricks, J., Bailey, G., 1998. Mutat. Res./Fundam. Mol. Mech. Mutagenesis 399, 245–253. Dashwood, R., Yamane, S., Larsen, R., 1996. Environ. Mol. Mutagenesis 27, 211–218. Holt, A.S., Jacobs, E.E., 1954. Am. J. Bot. 41, 710–717. Kato, Y., Sugiura, M., Oda, A., Watanabe, T., 2009. Proc. Natl. Acad. Sci. 106, 17365–17370. Kephart, J., 1955. Econ. Bot. 9, 3–38. Khachik, F., Beecher, G.R., Whittaker, N.F., 1986. J. Agric. Food Chem. 34, 603–616. Krasnovskii, A.A., Litvin, F.F., 1972. Biofizika 17, 764–768. Kuroiwa, Y., Kato, Y., Watanabe, T., 2009. J. Photochem. Photobiol. A: Chem. 202, 191–195. Li, X., Li, P., Zhang, Q., Li, R., Zhang, W., Zhang, Z., Ding, X., Tang, X., 2013. Biosens. Bioelectron. 49, 426–432. Mínguez-Mosquera, M.I., Gandul-Rojas, B., 1995. J. Chromatogr. A 690, 161–176. Matsuda, M., Isozaki, H., Tajima, H., 2008. Thin Solid Films 517, 1465–1467. Mortensen, A., Geppel, A., 2007. Innov. Food Sci. Emerg. Technol. 8, 419–425. Nagata, M., Amano, M., Joke, T., Fujii, K., Okuda, A., Kondo, M., Ishigure, S., Dewa, T., Iida, K., Secundo, F., Amao, Y., Hashimoto, H., Nango, M., 2012. ACS Macro Lett. 1, 296–299. Nan Chen, G., Zhang, L., Er Lin, R., Cong Yang, Z., Ping Duan, J., Qing Chen, H., Brynn Hibbert, D., 2000. Talanta 50, 1275–1281. Nguyen, M.T., Tozlovanu, M., Tran, T.L., Pfohl-Leszkowicz, A., 2007. Food Chem. 105, 42–47. Reddy, A.P., Harttig, U., Barth, M.C., Baird, W.M., Schimerlik, M., Hendricks, J.D., Bailey, G.S., 1999. Carcinogenesis 20, 1919–1926. Saito, K., Ishida, T., Sugiura, M., Kawakami, K., Umena, Y., Kamiya, N., Shen, J.-R., Ishikita, H., 2011. J. Am. Chem. Soc. 133, 14379–14388. Saji, T., Bard, A.J., 1977. J. Am. Chem. Soc. 99, 2235–2240. Shelnutt, J.A., 1984. J. Phys. Chem. 88, 6121–6127. Soleimany, F., Jinap, S., Faridah, A., Khatib, A., 2012. Food Control 25, 647–653. Springer, J.W., Parkes-Loach, P.S., Reddy, K.R., Krayer, M., Jiao, J., Lee, G.M., Niedzwiedzki, D.M., Harris, M.A., Kirmaier, C., Bocian, D.F., Lindsey, J.S., Holten, D., Loach, P.A., 2012. J. Am. Chem. Soc. 134, 4589–4599. Tajima, H., Shimatani, K., Komino, T., Ikeda, S., Matsuda, M., Ando, Y., Akiyama, H., 2006. Coll. Surf. A: Physicochem. Eng. Asp. 284–285, 61–65. Vidal, J.C., Bonel, L., Ezquerra, A., Hernández, S., Bertolín, J.R., Cubel, C., Castillo, J.R., 2013. Biosens. Bioelectron. 49, 146–158. Wang, C., Xing, D., Chen, Q., 2004. Biosens.Bioelectron. 20, 454–459. Wang, J., Guo, Y., Gao, J., Jin, X., Wang, Z., Wang, B., Li, K., Li, Y., 2011. Ultrason. Sonochem. 18, 1028–1034. Wasielewski, M.R., Smith, R.L., Kostka, A.G., 1980. J. Am. Chem. Soc. 102, 6923–6928. Young, R.W., Beregi Jr., J.S., 1980. J. Am. Geriatr. Soc. 28, 46–47.
