Talanta 144 (2015) 1059–1064
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
Ultrasensitive detection of amifostine and alkaline phosphatase based on the growth of CdS quantum dots Weidan Na, Siyu Liu, Xiaotong Liu, Xingguang Su n Department of Analytical Chemistry, College of Chemistry, Jilin University, Changchun 130012, China
art ic l e i nf o
a b s t r a c t
Article history: Received 15 March 2015 Received in revised form 16 July 2015 Accepted 19 July 2015 Available online 20 July 2015
In this study, we reported a simple and sensitive fluorescence nanosensor for rapid detection of amifostine and alkaline phosphatase (ALP). The novel nanosensor was based on the fluorescence “turn onoff” of CdS quantum dots (QDs). Firstly, Cd2 þ cation could react with S2 anion to generate fluorescent CdS QDs in the presence of amifostine. The fluorescence (FL) intensity of amifostine-capped CdS QDs (Amifostine-CdS QDs) was increased with the increasing amounts of amifostine, and could be used for amifostine detection. However, amifostine could be converted to 2-(3-aminopropylamino) ethanethiol (WR1065) in the presence of ALP based on the dephosphorylation of ALP. Under the optimum conditions, the affinity of WR1065 to CdS QDs was weaker than that of amifostine. Therefore the new generation of WR1065-CdS QDs would reduce the FL intensity with the increase of ALP concentration, and the fluorescence of CdS QDs was turn off. The metabolic process of amifostine in the presence of alkaline phosphatase could be also studied via the change of FL intensity of CdS QDs. The present method was cost-effective, convenient, and does not require any complicated synthetic procedures. & 2015 Elsevier B.V. All rights reserved.
Keywords: CdS quantum dots Fluorescence detection Amifostine Alkaline phosphatase
1. Introduction Amifostine [S-2-(3-aminopropylamino)ethylphosphorothioic acid, WR2721] was developed by the U.S. Army Anti-Radiation Drug Development Program for its potential to protect soldiers against radioactive radiation in case of nuclear attack. Amifostine was used as a cytoprotective agent with cisplatin-based chemotherapy against ovarian cancer and radiotherapy for head and neck cancer [1,2]. Amifostine has been shown to specifically protect normal tissues from damage caused by radiation and chemotherapy and exhibits no protection function towards almost all tumor tissues [3]. It has been widely employed to improve the tolerance of patients to chemotherapy and radiotherapy to improve their life quality. Therefore, it is of high interest to develop a simple and sensitive approach for monitoring the levels of amifostine in body fluids. The alkaline phosphatases (ALP) is one of the most studied enzymes that are widely distributed in many tissues and fluids throughout the body. ALP is a phospho-monoesterase enzyme that can catalyze the hydrolysis of phosphate monoesters. ALP is used as a diagnostic indicator of variety diseases, because the changes of ALP levels in serum and other body fluids may reflect physiologic or pathologic changes [4,5]. Amifostine, as a prodrug, can be n
Corresponding author. E-mail address: [email protected] (X. Su).
http://dx.doi.org/10.1016/j.talanta.2015.07.057 0039-9140/& 2015 Elsevier B.V. All rights reserved.
converted to the active metabolite 2-(3-aminopropylamino)ethanethiol (WR-1065) by ALP. Within the cell, WR-1065 protects subcellular components like membranes from damage through free-radical scavenging, hydrogen donation, and DNA damage [6]. Therefore, it is very important to develop a simple and sensitive approach for simultaneous monitoring the levels of amifostine and its metabolic process. But, so far, there are only rare reports on this. Liquid chromatography is the most widely employed method for the simultaneous determination of amifostine and its metabolites in biological samples in combination with fluorimetric [7,8] and electrochemical detection [9]. These methods are time-consuming, laborious, and need the separation of be-surveyed substance. Fluorescent methods for amifostine assay will offer new alternatives since they are rapid, continuous, and in real time. In the past few years, fluorescence analysis for some small molecules has shown several unique advantages such as high sensitivity, high selectivity, and easy to operate. Liu et al. reported a novel strategy for Pb2 þ detection based on the DNA sequences PS2.M capped CdS QDs [10]. The one-pot synthesized aptamer-functionalized QDs have been developed as a new class of fluorophores that could be applied for the detection of various analytes based on different signal-transducing mechanisms. Recently, a series of fluorescence assays based on the formation of CdS QDs were reported. Saa's group introduced a new method to detect the enzymatic activities based on the enzymatic generation of CdS QDs [11,12]. Acetylcholinesterase could hydrolyze acetylthiocholine and yield the
1060
W. Na et al. / Talanta 144 (2015) 1059–1064
thiol-containing compound thiocholine which acted as a thiol to trigger the formation of CdS QDs. The increasing amounts of acetylcholinesterase resulted in the increase of fluorescence intensity [11]. In this paper, CdS QDs was generated in the presence of Cd2 þ and S2 with amifostine as stabilizers, and its fluorescence intensity was enhanced with the increasing of the amifostine concentration in a certain range. With the addition of ALP, amifostine was rapidly hydrolyzed to WR1065 and phosphoric acid via the dephosphorylation process, and the fluorescence of CdS QDs would be inhibited. Therefore, the assay system could be utilized to highly sensitive detection of amifostine and ALP. The assay system of modulating the growth of CdS quantum dots in situ could avoid the high background signal which was the drawbacks of pre-synthesized QDs, and could improved sensitivity, and reduced the costs of assays [13].
2. Experimental 2.1. Apparatus The fluorescence spectra were obtained by using a Shimadzu RF-5301 PC spectrofluorophotometer equipped with a xenon lamp using right-angle geometry. UV–vis absorption spectra were obtained by a Varian GBC Cintra 10e UV–vis spectrometer. In both experiments, a 1 cm path-length quartz cuvette was used. FT-IR spectra were recorded with a Bruker IFS66V FT-IR spectrometer equipped with a DGTS detector. 2.2. Reagents All reagents were of at least analytical grade. The water used in all experiments had a resistivity higher than 18 MΩ cm 1. Cadmium(II) chloride (CdCl2), sodium hydroxide (NaOH), Sodiumsulfide nonahydrate (Na2S 9H2O), trihydroxymethyl aminomethane (Tris) and hydrochloric acid were purchased from Shanghai Qingxi Technology Co., Ltd. ALP were purchased from Sigma-Aldrich Corporation. Amifostine was purchased from Changchun Dingguo biotechnology Co., Ltd. The 0.1 mol/L Tris–HCl buffered solution (pH 8.2) was used as the medium for detection process. The human serum was obtained as a gift from the university hospital. 2.3. Determination of amifostine based on the generation of CdS QDs Varying amounts of amifostine, 120 μL 50 mmol/L CdCl2 solution and 0.1 mol/L Tris–HCl buffer (pH 8.2, 150 μL) were added into 2 mL calibrated test tube, and shaken thoroughly for 10 min. After that, 120 μL 3.75 mmol/L Na2S solution was added into the test tube and diluted to the mark with deionized water followed by the thoroughly shaking and equilibrated for 15 min. The fluorescence spectra were recorded from 405 nm to 650 nm with the excitation wavelength of 340 nm. The slit widths of excitation and emission were both 10 nm. The FL intensity of the maximum emission peak was used for the quantitative analysis of amifostine. 2.4. Determination of ALP based on the generation of CdS QDs 250 μL 0.15 mmol/L amifostine, 120 μL 50 mmol/L CdCl2 solution, 0.1 mol/L Tris–HCl buffer (pH 8.2, 150 μL) and different concentrations of ALP were successively added into 2 mL calibrated test tube, and shaken thoroughly for 40 min. After that, 120 μL 3.75 mmol/L Na2S solution was added into the test tube and diluted to the mark with deionized water followed by the thoroughly shaking and equilibrated for 15 min. The fluorescence
spectra were recorded from 405 nm to 650 nm with the excitation wavelength of 340 nm. The slit widths of excitation and emission were both 10 nm. The FL intensity of the maximum emission peak was used for the quantitative analysis of ALP. 2.5. Human serum samples detection For serum samples detection, drug-free human blood samples were collected from healthy volunteer through venipuncture at the Hospital of Changchun China–Japan Union Hospital. Some necessary processes were conducted to remove large molecules and proteins to get the serum samples. The blood samples were segregated by adding acetonitrile (the volume of acetonitrile and blood was 1.5:1) and centrifuged at 10,000 rpm for 5 min after stored for 2 h at room temperature. The obtained supernatant was diluted by 50,000 times with deionized water before detection. 10 μL diluted human serum and 1 nmol/L Sodium orthovanadate (the inhibitor of ALP) were used for the detection of different concentrations of amifostine. 10, 15 and 20 μL diluted human serum were used for the detection of ALP under optimum condition.
3. Results and discussion 3.1. Spectra of the amifostine-CdS QDs The principle of amifostine and ALP assay was illustrated in Scheme 1. As shown in Scheme 1, the fluorescent CdS QDs could be generated in the presence of Cd2 þ and S2 with amifostine as stabilizers. The thiophosphate (PO3S3 ) group of amifostine has a powerful coordination ability and was employed to render the CdS QDs stable against aggregation, water-soluble and bio-compatible. The ALP could effectively quench the fluorescence of CdS QDs based on its hydrolysis ability for amifostine. Fig. 1 showed the UV–vis absorption (Dash line) and fluorescence emission spectra (Solid line) of the amifostine capped CdS QDs. As shown in Fig. 1, there was an increased absorption below 400 nm and a shoulder around 330 nm that was the result of 1Sh– 1Se excitonic transition characteristic of semiconductor nanoparticles [14]. The fluorescence emission spectra of the amifostine capped CdS QDs showed a fluorescent peak with maximum emission wavelength of 535 nm that was consistent with some previous reports [15]. The interaction between CdS QDs and amifostine molecules are mainly through interaction between surface Cd2 þ ions and the thiophosphate group of amifostine [16]. If there is no amifostine molecule as stabilizers in the assay system, Cd2 þ and S2 would generate some yellow precipitation, and it is hardly observed the fluorescence signal of CdS QDs. The FT-IR spectra of the CdS crystals with and without amifostine as stabilizers were compared to confirm the coordination of the amifostine on the surface of the CdS QDs. As shown in Fig. S1 curve b, the majority of amifostine functional groups could be clearly found through the stretching vibration of N–H(3190 cm 1) and the –NH2 feature (1630 cm 1), the asymmetric stretching vibrations of the PO2 (1290 cm 1), the out of phase symmetrical stretches (1040 cm 1), and the P–O stretches of the main chain (910 cm 1). And these characteristic peaks were not observed in the FT-IR spectra of uncapped CdS crystals (Fig. S1 curve a) which indicated the successful capping of amifostine on the surface of the CdS QDs [17,18]. 3.2. Optimization of CdS QDs for amifostine and ALP detection We firstly studied the effect of Na2S concentration on the fluorescence intensity of CdS QDs. As shown in Fig. 2(A), the
W. Na et al. / Talanta 144 (2015) 1059–1064
1061
Scheme 1. Schematic illustration for the detection of amifostine and ALP based on the formation of fluorescent CdS QDs.
Fig. 1. The UV–vis absorption (Dash line) and fluorescence emission spectra of amifostine-capped CdS QDs (Solid line) generated in the presence of CdCl2 (4 mmol/L), amifostine (25 μmol/L) and Na2S (0.3 mmol/L) in 10 mmol/L Tris–HCl buffer solution (pH 8.2).
fluorescence intensity of CdS QDs increased sharply when the concentration of Na2S ranged from 0.1 to 0.3 mmol/L, and then decreased obviously with increasing the concentration of Na2S from 0.3 to 0.5 mmol/L. The fluorescence intensity kept almost constant when the level Na2S was higher than 0.5 mmol/L. These results indicated 0.3 mmol/L Na2S was the optimum concentration for generation of CdS QDs. Fig. 2(B) showed the temporal evolution of the fluorescence intensity of Cd2 þ -amifostine solution after the addition of 0.3 mmol/L Na2S. It could be seen that the fluorescence intensity of the system gradually increased with the increase of incubation time, and reached equilibrium after 15 min due to the formation of amifostine-CdS QDs. Therefore, the reaction time of 15 min was used in the generation of amifostine-CdS QDs. Fig. 2(C) showed the fluorescence changes of CdS QDs-amifostine system upon the addition of ALP (nU/mL) at different incubation time. It could be seen that the fluorescence intensity gradually decreased with the increase of incubation time due to the dephosphorylation of the ALP, and reached equilibrium after 40 min. Therefore the incubation time of 40 min was used in the further experiments. At last, we systematically investigated the influence of pH value
1062
W. Na et al. / Talanta 144 (2015) 1059–1064
Fig. 2. (A) The effect of Na2S concentration (0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6 mmol/L) in the presence of 4 mmol/L Cd2 þ and 25 μmol/L amifostine on the fluorescence intensity of CdS QDs. (B) Fluorescence intensity of the assay system in the presence of 4 mmol/L Cd2 þ and 0.3 mmol/L S2 with 25 μmol/L amifostine in different incubation time. (C) The fluorescence of Cd2 þ –amifostine–S2 system in the presence of (a) 7.5, (b)15 and (c) 40 nU/mL ALP at different incubation time. (D) Fluorescence intensity of assay system in the presence of 4 mmol/L Cd2 þ , 0.3 mmol/L S2 and 25 μmol/L amifostine without ALP (curve a) and with 40 nU/mL ALP (curve b) in different pH environments (pH 6.6–9.0).
on the fluorescence intensity of amifostine-CdS QDs and the CdSamifostine-ALP system. From Fig. 2(D) curve a, it could be seen that the fluorescence intensity of amifostine-CdS QDs solution showed a downward tendency in the pH value range from 6.6 to 7.8 and an upward tendency in the pH value range from 7.8 to 9.0. For curve b, the fluorescence intensity of amifostine-CdS-ALP system had minor change from 6.6 to 7.8 and increased dramatically from 8.2 to 9.0, which is due to the binding ability of sulfhydryl group of WR 1065 had a surge from 8.2 to 9.0 and reached a peak at 9.0. At the point of pH 8.2, the fluorescence quenching effect of ALP reached the maximum value. So pH 8.2 was chosen as the optimum pH value. 3.3. Amifostine detection The influence of amifostine concentration on the fluorescence of CdS QDs was studied. As shown in Fig. 3, the fluorescence intensity (F) of assay system obviously increased with the increase of amifostine concentration. From the inset we could find there was a good linear relationship between the fluorescence intensity of the assay system and the amifostine concentration. The regression equation could be described as follows: F¼0.03673 þ0.05279 [amifostine], (μmol/L) for 0.5–10.0 μmol/ L amifostine and F¼0.3012 þ0.02793 [amifostine], (μmol/L) for 10–25 μmol/L amifostine The correlation coefficient were R2 ¼0.9970 and R2 ¼0.9995,
Fig. 3. The fluorescence spectra of assay system in the presence of 4 mmol/L Cd2 þ , different concentration of amifostine in the range from 0 to 25.0 μmol/L (0,0.5, 1.0, 2.5, 7.5, 10, 12.5, 15, 20, 25 μmol/L) and 0.3 mmol/L Na2S. Inset: Plot of fluorescence intensity (F) of assay system versus the concentration of amifostine. 10 mmol/L Tris–HCl buffer solution incubation for 15 min.
W. Na et al. / Talanta 144 (2015) 1059–1064
1063
3.5. Effect of biological molecules on the fluorescence of CdS QDs In this work, we investigated the effect of a series of biological molecules on the fluorescence of 4 mmol/L Cd2 þ –0.3 mmol/L S2 system. As shown in Fig. S2, after the addition of 50 μmol/L L-cystein hydrochloride, L-arginine, L-aspartic acid, threonine, glucine, lactamic acid, L-tyrosine, acetylcholine ascorbic acid, sodium pyruvate, L-DOPA, L-tryptophan, carnitine and lysine, the fluorescence intensity of CdS QDs were very weak, and only the addition of amifostine could induce the generation of strong fluorescent CdS QDs. These results indicated that the assay system is highly selective to amifostine. 3.6. Selectivity assay for ALP based on the amifostine capped CdS QDs
Fig. 4. Fluorescence quenching ratio F/F0 of Cd2 þ –25 μM amifostine with different concentration of ALP (0, 4.67, 9.34, 18.68, 23.35, 28.02, 37.36, 46.7 nU/mL) upon addition of 0.3 mmol/L Na2S. Inset: The linear plots of F/F0 versus the ALP concentration in the range of 0–46.7 nU/mL. Reaction condition: 10 mmol/L Tris–HCl buffer solution (pH 8.2) at 25 °C.
respectively. The limit of detection (LOD) for amifostine was 0.12 μmol/L, calculated following the 3s IUPAC criteria. The standard deviation for nine replicate measurements of 2 μmol/L amifostine is 3.9%. This is the first time to detect amifostine in combination with QDs. Compared with the previous reports about amifostine assay [7,19], our present method offered a comparable detection limit and dynamic range. 3.4. The fluorescence quenching of amifostine-CdS QDs by ALP It is well known that ALP is a hydrolase that could effectively remove the phosphate functional groups from amifostine molecules [20]. Due to the more powerful affinity of amifostine to CdS QDs than its metabolite WR-1065, the generation of fluorescent CdS QDs would be inhibited in the presence of ALP. Different concentrations of ALP were firstly added to the 4 mmol/L Cd2 þ – 25 μmol/L amifostine system before the addition of Na2S. It could be seen from Fig. 4 that the FL intensity of the mixed system decreased gradually with the increase of ALP concentration. As shown in Fig. 4 inset, there was a good linear relationship between the relative fluorescence quenching ratio F/F0 (F0 is the fluorescence intensity of Cd2 þ –25 μmol/L amifostine-S2 system, and F is the fluorescence intensity of Cd2 þ –25 μmol/L amifostineS2 after the addition of ALP) and the ALP concentrations was in the range from 0 to 46.7 nU/mL. The regression equation could be described as follow:
F /F0 = 0.9958 − 0.0144[ALP] (nU/mL) The corresponding regression coefficient is 0.999, and the detection limit for ALP was 2.0 nU/mL, calculated following the 3s IUPAC criteria. The standard deviation for nine replicate measurements of 10 nU/mL ALP was 0.63%. The conventional fluorimetric methodology which usually exploited some fluorescent orthophosphoric monoesters such as 8-quinolyl phosphate, 2-carboxy-1-naphthyl phosphate as substrate were incapable of achieving such a low detection limit [21,22]. Compared with the previous reported fluorescence detection methods for ALP [23– 25], our method obtained a similar or superior detection limit and wider dynamic range. This method could be utilized to build up a real-time analytical method to detect the enzyme activity.
Selectivity is a very important parameter to evaluate the performance of a new sensor, especially for ones with potential applications in biomedical samples, a highly selective response to the target over other potentially competing species is necessary. Therefore,we further evaluated the selectivity of our nanosensor with various coexistence substances added. As shown in Fig. S3, to investigate the influence of different enzymes on the fluorescence intensity of CdS QDs, a series of 250 nU/mL enzymes including horseradish peroxidase (HRP), trypsin (Try), glucose oxidase (Gox), urase (Ura), pepsase (Pep), lysozyme (Lys), lactate dehydrogenase (LDH), and 150 ng/mL albumin bovine (BSA), human serum albumin (HSA) were tested instead of 25 nU/mL ALP. As shown from Fig. S3, only ALP have a highly specific response to amifostine and the generation of fluorescent CdS QDs could be effectively inhibited. The results showed that this ALP detection system could provide well ability of resisting interference.
4. The detection of amifostine and ALP in human serum samples In order to demonstrate the practical application of the fluorescence nanosensor, we performed detection of amifostine and ALP in human serum samples under the optimal conditions. The results obtained by standard addition method were listed in Tables 1–2. From Tables 1–2, we can see that the average recoveries of amifostine and ALP in the real samples were in the range of 100.5–108.5% and 96.4–105.8% respectively, and the RSD was lower than 3.5%, indicating that the accuracy and precision of the proposed method were satisfactory. Moreover, it also held potential application in screening the ALP inhibitors.
5. Conclusion In summary, we developed a novel, simple and fast assay for amifostine based on the growth of CdS QDs. The amifostine concentration could be associated with the fluorescence intensity of the generated CdS QDs. Furthermore, upon the addition of ALP, the generation of fluorescent CdS QDs could be effectively inhibited due to the enzymatic dephosphorylation of ALP. Therefore the Table 1 Results of amifostine determination in adult human serum samples. Samples
Added (μmol/L)
Founded (μmol/L)
Recovery (%)
RSD (n¼ 3, %)
1 2 3
2.0 10.0 20.0
2.177 0.02 10.2 7 0.03 20.17 0.05
108.5 101.0 100.5
3.5 3.2 2.6
1064
W. Na et al. / Talanta 144 (2015) 1059–1064
References
Table 2 Results of ALP determination in adult human serum samples. Samples
Found (nU/mL)
Added (nU/mL)
Total found ALP Recovery (%) RSD (nU/mL) (n ¼3, %)
1 2 3
8.90 13.46 17.72
9.34 14.01 18.68
18.63 7 0.21 26.497 0.13 37.29 7 0.27
102.14 96.43 105.76
3.11 2.94 2.67
present method could also be applied to selectively to detect ALP and the obtained detection limit for ALP was as low as 2.0 nU/mL.
Acknowledgments This work was financially supported by the National Natural Science Foundation of China (Nos. 21075050 and 21275063) and the Science and Technology Development Project of Jilin Province, China (No. 20110334).
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.talanta.2015.07. 057.
[1] C.N. Andreassen, C. Gran, J.C. Lindegaard, Semin. Radiat. Oncol. 13 (2003) 62–72. [2] J.R. Kouvaris, V.E. Kouloulias, L.J. Vlahos, Oncologist 12 (2007) 738–747. [3] C. Uzal, G. Durmus-Altun, M. Caloglu, A. Ergulen, S. Altaner, N.O. Yigitbasi, Int. J. Radiat. Oncol. Biol. Phys. 60 (2004) 564–569. [4] J. Nicole, B.A.K. Fernandez, Vet. Clin. Pathol. 36 (2007) 223–233. [5] P.S. Henthorn, M. Raducha, K.N. Fedde, M.A. Lafferty, M.P. Whyte, Proc. Natl. Acad. Sci. USA 89 (1992) 9924–9928. [6] I. Majsterek, E. Gloc, J. Blasiak, R.J. Reiter, J. Pineal Res. 38 (2005) 254–263. [7] J. Chen, Z. Lu, T.S. Lawrence, D.E. Smith, J. Chromatogr. B – Anal. Technol. Biomed. Life Sci. 819 (2005) 161–167. [8] A.K. Souid, G.L. Newton, R.L. Dubowy, R.C. Fahey, M.L. Bernstein, Cancer Chemother. Pharmacol. 42 (1998) 400–406. [9] A.E. Korst, J.B. Vermorken, W.J. Van der Vijgh, J. Chromatogr. B Biomed. Sci. Appl. 693 (1997) 167–174. [10] S. Liu, W. Na, S. Pang, X. Su, Biosens. Bioelectron. 58 (2014) 17–21. [11] L. Saa, A. Virel, J. Sanchez-Lopez, V. Pavlov, Chemistry 16 (2010) 6187–6192. [12] G. Garai-Ibabe, L. Saa, V. Pavlov, Anal. Chem. 85 (2013) 5542–5546. [13] G. Garai-Ibabe, M. Möller, V. Pavlov, Anal. Chem. 84 (2012) 8033–8037. [14] H. Matsumoto, T. Sakata, H. Mori, H. Yoneyama, J. Phys. Chem. 100 (1996) 13781–13785. [15] L. Saa, J.M. Mato, V. Pavlov, Anal. Chem. 84 (2012) 8961–8965. [16] C. Zhang, J. Xu, S. Zhang, X. Ji, Z. He, Chemistry 18 (2012) 8296–8300. [17] H.B. Ren, X.P. Yan, Talanta 97 (2012) 16–22. [18] S.S. Liji Sobhana, M. Vimala Devi, T.P. Sastry, A. Mandal, J. Nanopart. Res. 13 (2011) 1747–1757. [19] T. Huang, N. Chen, L. Zhang, G. Chen, Chromatographia 76 (2013) 1739–1745. [20] R. Capizzi, Semin. Oncology 23 (1996) 2–17. [21] X. Zhu, Q. Liu, C. Jiang, Clin. Chim. Acta 377 (2007) 150–153. [22] X. Zhu, Q. Liu, C. Jiang, Anal. Chim. Acta 570 (2006) 29–33. [23] N. Malashikhina, G. Garai-Ibabe, V. Pavlov, Anal. Chem. 85 (2013) 6866–6870. [24] F. Zheng, S. Guo, F. Zeng, J. Li, S. Wu, Anal. Chem. 86 (2014) 9873–9879. [25] N. Wu, J. Lan, L. Yan, J. You, Chem. Commun. 50 (2014) 4438–4441.
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
Ultrasensitive detection of amifostine and alkaline phosphatase based on the growth of CdS quantum dots Weidan Na, Siyu Liu, Xiaotong Liu, Xingguang Su n Department of Analytical Chemistry, College of Chemistry, Jilin University, Changchun 130012, China
art ic l e i nf o
a b s t r a c t
Article history: Received 15 March 2015 Received in revised form 16 July 2015 Accepted 19 July 2015 Available online 20 July 2015
In this study, we reported a simple and sensitive fluorescence nanosensor for rapid detection of amifostine and alkaline phosphatase (ALP). The novel nanosensor was based on the fluorescence “turn onoff” of CdS quantum dots (QDs). Firstly, Cd2 þ cation could react with S2 anion to generate fluorescent CdS QDs in the presence of amifostine. The fluorescence (FL) intensity of amifostine-capped CdS QDs (Amifostine-CdS QDs) was increased with the increasing amounts of amifostine, and could be used for amifostine detection. However, amifostine could be converted to 2-(3-aminopropylamino) ethanethiol (WR1065) in the presence of ALP based on the dephosphorylation of ALP. Under the optimum conditions, the affinity of WR1065 to CdS QDs was weaker than that of amifostine. Therefore the new generation of WR1065-CdS QDs would reduce the FL intensity with the increase of ALP concentration, and the fluorescence of CdS QDs was turn off. The metabolic process of amifostine in the presence of alkaline phosphatase could be also studied via the change of FL intensity of CdS QDs. The present method was cost-effective, convenient, and does not require any complicated synthetic procedures. & 2015 Elsevier B.V. All rights reserved.
Keywords: CdS quantum dots Fluorescence detection Amifostine Alkaline phosphatase
1. Introduction Amifostine [S-2-(3-aminopropylamino)ethylphosphorothioic acid, WR2721] was developed by the U.S. Army Anti-Radiation Drug Development Program for its potential to protect soldiers against radioactive radiation in case of nuclear attack. Amifostine was used as a cytoprotective agent with cisplatin-based chemotherapy against ovarian cancer and radiotherapy for head and neck cancer [1,2]. Amifostine has been shown to specifically protect normal tissues from damage caused by radiation and chemotherapy and exhibits no protection function towards almost all tumor tissues [3]. It has been widely employed to improve the tolerance of patients to chemotherapy and radiotherapy to improve their life quality. Therefore, it is of high interest to develop a simple and sensitive approach for monitoring the levels of amifostine in body fluids. The alkaline phosphatases (ALP) is one of the most studied enzymes that are widely distributed in many tissues and fluids throughout the body. ALP is a phospho-monoesterase enzyme that can catalyze the hydrolysis of phosphate monoesters. ALP is used as a diagnostic indicator of variety diseases, because the changes of ALP levels in serum and other body fluids may reflect physiologic or pathologic changes [4,5]. Amifostine, as a prodrug, can be n
Corresponding author. E-mail address: [email protected] (X. Su).
http://dx.doi.org/10.1016/j.talanta.2015.07.057 0039-9140/& 2015 Elsevier B.V. All rights reserved.
converted to the active metabolite 2-(3-aminopropylamino)ethanethiol (WR-1065) by ALP. Within the cell, WR-1065 protects subcellular components like membranes from damage through free-radical scavenging, hydrogen donation, and DNA damage [6]. Therefore, it is very important to develop a simple and sensitive approach for simultaneous monitoring the levels of amifostine and its metabolic process. But, so far, there are only rare reports on this. Liquid chromatography is the most widely employed method for the simultaneous determination of amifostine and its metabolites in biological samples in combination with fluorimetric [7,8] and electrochemical detection [9]. These methods are time-consuming, laborious, and need the separation of be-surveyed substance. Fluorescent methods for amifostine assay will offer new alternatives since they are rapid, continuous, and in real time. In the past few years, fluorescence analysis for some small molecules has shown several unique advantages such as high sensitivity, high selectivity, and easy to operate. Liu et al. reported a novel strategy for Pb2 þ detection based on the DNA sequences PS2.M capped CdS QDs [10]. The one-pot synthesized aptamer-functionalized QDs have been developed as a new class of fluorophores that could be applied for the detection of various analytes based on different signal-transducing mechanisms. Recently, a series of fluorescence assays based on the formation of CdS QDs were reported. Saa's group introduced a new method to detect the enzymatic activities based on the enzymatic generation of CdS QDs [11,12]. Acetylcholinesterase could hydrolyze acetylthiocholine and yield the
1060
W. Na et al. / Talanta 144 (2015) 1059–1064
thiol-containing compound thiocholine which acted as a thiol to trigger the formation of CdS QDs. The increasing amounts of acetylcholinesterase resulted in the increase of fluorescence intensity [11]. In this paper, CdS QDs was generated in the presence of Cd2 þ and S2 with amifostine as stabilizers, and its fluorescence intensity was enhanced with the increasing of the amifostine concentration in a certain range. With the addition of ALP, amifostine was rapidly hydrolyzed to WR1065 and phosphoric acid via the dephosphorylation process, and the fluorescence of CdS QDs would be inhibited. Therefore, the assay system could be utilized to highly sensitive detection of amifostine and ALP. The assay system of modulating the growth of CdS quantum dots in situ could avoid the high background signal which was the drawbacks of pre-synthesized QDs, and could improved sensitivity, and reduced the costs of assays [13].
2. Experimental 2.1. Apparatus The fluorescence spectra were obtained by using a Shimadzu RF-5301 PC spectrofluorophotometer equipped with a xenon lamp using right-angle geometry. UV–vis absorption spectra were obtained by a Varian GBC Cintra 10e UV–vis spectrometer. In both experiments, a 1 cm path-length quartz cuvette was used. FT-IR spectra were recorded with a Bruker IFS66V FT-IR spectrometer equipped with a DGTS detector. 2.2. Reagents All reagents were of at least analytical grade. The water used in all experiments had a resistivity higher than 18 MΩ cm 1. Cadmium(II) chloride (CdCl2), sodium hydroxide (NaOH), Sodiumsulfide nonahydrate (Na2S 9H2O), trihydroxymethyl aminomethane (Tris) and hydrochloric acid were purchased from Shanghai Qingxi Technology Co., Ltd. ALP were purchased from Sigma-Aldrich Corporation. Amifostine was purchased from Changchun Dingguo biotechnology Co., Ltd. The 0.1 mol/L Tris–HCl buffered solution (pH 8.2) was used as the medium for detection process. The human serum was obtained as a gift from the university hospital. 2.3. Determination of amifostine based on the generation of CdS QDs Varying amounts of amifostine, 120 μL 50 mmol/L CdCl2 solution and 0.1 mol/L Tris–HCl buffer (pH 8.2, 150 μL) were added into 2 mL calibrated test tube, and shaken thoroughly for 10 min. After that, 120 μL 3.75 mmol/L Na2S solution was added into the test tube and diluted to the mark with deionized water followed by the thoroughly shaking and equilibrated for 15 min. The fluorescence spectra were recorded from 405 nm to 650 nm with the excitation wavelength of 340 nm. The slit widths of excitation and emission were both 10 nm. The FL intensity of the maximum emission peak was used for the quantitative analysis of amifostine. 2.4. Determination of ALP based on the generation of CdS QDs 250 μL 0.15 mmol/L amifostine, 120 μL 50 mmol/L CdCl2 solution, 0.1 mol/L Tris–HCl buffer (pH 8.2, 150 μL) and different concentrations of ALP were successively added into 2 mL calibrated test tube, and shaken thoroughly for 40 min. After that, 120 μL 3.75 mmol/L Na2S solution was added into the test tube and diluted to the mark with deionized water followed by the thoroughly shaking and equilibrated for 15 min. The fluorescence
spectra were recorded from 405 nm to 650 nm with the excitation wavelength of 340 nm. The slit widths of excitation and emission were both 10 nm. The FL intensity of the maximum emission peak was used for the quantitative analysis of ALP. 2.5. Human serum samples detection For serum samples detection, drug-free human blood samples were collected from healthy volunteer through venipuncture at the Hospital of Changchun China–Japan Union Hospital. Some necessary processes were conducted to remove large molecules and proteins to get the serum samples. The blood samples were segregated by adding acetonitrile (the volume of acetonitrile and blood was 1.5:1) and centrifuged at 10,000 rpm for 5 min after stored for 2 h at room temperature. The obtained supernatant was diluted by 50,000 times with deionized water before detection. 10 μL diluted human serum and 1 nmol/L Sodium orthovanadate (the inhibitor of ALP) were used for the detection of different concentrations of amifostine. 10, 15 and 20 μL diluted human serum were used for the detection of ALP under optimum condition.
3. Results and discussion 3.1. Spectra of the amifostine-CdS QDs The principle of amifostine and ALP assay was illustrated in Scheme 1. As shown in Scheme 1, the fluorescent CdS QDs could be generated in the presence of Cd2 þ and S2 with amifostine as stabilizers. The thiophosphate (PO3S3 ) group of amifostine has a powerful coordination ability and was employed to render the CdS QDs stable against aggregation, water-soluble and bio-compatible. The ALP could effectively quench the fluorescence of CdS QDs based on its hydrolysis ability for amifostine. Fig. 1 showed the UV–vis absorption (Dash line) and fluorescence emission spectra (Solid line) of the amifostine capped CdS QDs. As shown in Fig. 1, there was an increased absorption below 400 nm and a shoulder around 330 nm that was the result of 1Sh– 1Se excitonic transition characteristic of semiconductor nanoparticles [14]. The fluorescence emission spectra of the amifostine capped CdS QDs showed a fluorescent peak with maximum emission wavelength of 535 nm that was consistent with some previous reports [15]. The interaction between CdS QDs and amifostine molecules are mainly through interaction between surface Cd2 þ ions and the thiophosphate group of amifostine [16]. If there is no amifostine molecule as stabilizers in the assay system, Cd2 þ and S2 would generate some yellow precipitation, and it is hardly observed the fluorescence signal of CdS QDs. The FT-IR spectra of the CdS crystals with and without amifostine as stabilizers were compared to confirm the coordination of the amifostine on the surface of the CdS QDs. As shown in Fig. S1 curve b, the majority of amifostine functional groups could be clearly found through the stretching vibration of N–H(3190 cm 1) and the –NH2 feature (1630 cm 1), the asymmetric stretching vibrations of the PO2 (1290 cm 1), the out of phase symmetrical stretches (1040 cm 1), and the P–O stretches of the main chain (910 cm 1). And these characteristic peaks were not observed in the FT-IR spectra of uncapped CdS crystals (Fig. S1 curve a) which indicated the successful capping of amifostine on the surface of the CdS QDs [17,18]. 3.2. Optimization of CdS QDs for amifostine and ALP detection We firstly studied the effect of Na2S concentration on the fluorescence intensity of CdS QDs. As shown in Fig. 2(A), the
W. Na et al. / Talanta 144 (2015) 1059–1064
1061
Scheme 1. Schematic illustration for the detection of amifostine and ALP based on the formation of fluorescent CdS QDs.
Fig. 1. The UV–vis absorption (Dash line) and fluorescence emission spectra of amifostine-capped CdS QDs (Solid line) generated in the presence of CdCl2 (4 mmol/L), amifostine (25 μmol/L) and Na2S (0.3 mmol/L) in 10 mmol/L Tris–HCl buffer solution (pH 8.2).
fluorescence intensity of CdS QDs increased sharply when the concentration of Na2S ranged from 0.1 to 0.3 mmol/L, and then decreased obviously with increasing the concentration of Na2S from 0.3 to 0.5 mmol/L. The fluorescence intensity kept almost constant when the level Na2S was higher than 0.5 mmol/L. These results indicated 0.3 mmol/L Na2S was the optimum concentration for generation of CdS QDs. Fig. 2(B) showed the temporal evolution of the fluorescence intensity of Cd2 þ -amifostine solution after the addition of 0.3 mmol/L Na2S. It could be seen that the fluorescence intensity of the system gradually increased with the increase of incubation time, and reached equilibrium after 15 min due to the formation of amifostine-CdS QDs. Therefore, the reaction time of 15 min was used in the generation of amifostine-CdS QDs. Fig. 2(C) showed the fluorescence changes of CdS QDs-amifostine system upon the addition of ALP (nU/mL) at different incubation time. It could be seen that the fluorescence intensity gradually decreased with the increase of incubation time due to the dephosphorylation of the ALP, and reached equilibrium after 40 min. Therefore the incubation time of 40 min was used in the further experiments. At last, we systematically investigated the influence of pH value
1062
W. Na et al. / Talanta 144 (2015) 1059–1064
Fig. 2. (A) The effect of Na2S concentration (0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6 mmol/L) in the presence of 4 mmol/L Cd2 þ and 25 μmol/L amifostine on the fluorescence intensity of CdS QDs. (B) Fluorescence intensity of the assay system in the presence of 4 mmol/L Cd2 þ and 0.3 mmol/L S2 with 25 μmol/L amifostine in different incubation time. (C) The fluorescence of Cd2 þ –amifostine–S2 system in the presence of (a) 7.5, (b)15 and (c) 40 nU/mL ALP at different incubation time. (D) Fluorescence intensity of assay system in the presence of 4 mmol/L Cd2 þ , 0.3 mmol/L S2 and 25 μmol/L amifostine without ALP (curve a) and with 40 nU/mL ALP (curve b) in different pH environments (pH 6.6–9.0).
on the fluorescence intensity of amifostine-CdS QDs and the CdSamifostine-ALP system. From Fig. 2(D) curve a, it could be seen that the fluorescence intensity of amifostine-CdS QDs solution showed a downward tendency in the pH value range from 6.6 to 7.8 and an upward tendency in the pH value range from 7.8 to 9.0. For curve b, the fluorescence intensity of amifostine-CdS-ALP system had minor change from 6.6 to 7.8 and increased dramatically from 8.2 to 9.0, which is due to the binding ability of sulfhydryl group of WR 1065 had a surge from 8.2 to 9.0 and reached a peak at 9.0. At the point of pH 8.2, the fluorescence quenching effect of ALP reached the maximum value. So pH 8.2 was chosen as the optimum pH value. 3.3. Amifostine detection The influence of amifostine concentration on the fluorescence of CdS QDs was studied. As shown in Fig. 3, the fluorescence intensity (F) of assay system obviously increased with the increase of amifostine concentration. From the inset we could find there was a good linear relationship between the fluorescence intensity of the assay system and the amifostine concentration. The regression equation could be described as follows: F¼0.03673 þ0.05279 [amifostine], (μmol/L) for 0.5–10.0 μmol/ L amifostine and F¼0.3012 þ0.02793 [amifostine], (μmol/L) for 10–25 μmol/L amifostine The correlation coefficient were R2 ¼0.9970 and R2 ¼0.9995,
Fig. 3. The fluorescence spectra of assay system in the presence of 4 mmol/L Cd2 þ , different concentration of amifostine in the range from 0 to 25.0 μmol/L (0,0.5, 1.0, 2.5, 7.5, 10, 12.5, 15, 20, 25 μmol/L) and 0.3 mmol/L Na2S. Inset: Plot of fluorescence intensity (F) of assay system versus the concentration of amifostine. 10 mmol/L Tris–HCl buffer solution incubation for 15 min.
W. Na et al. / Talanta 144 (2015) 1059–1064
1063
3.5. Effect of biological molecules on the fluorescence of CdS QDs In this work, we investigated the effect of a series of biological molecules on the fluorescence of 4 mmol/L Cd2 þ –0.3 mmol/L S2 system. As shown in Fig. S2, after the addition of 50 μmol/L L-cystein hydrochloride, L-arginine, L-aspartic acid, threonine, glucine, lactamic acid, L-tyrosine, acetylcholine ascorbic acid, sodium pyruvate, L-DOPA, L-tryptophan, carnitine and lysine, the fluorescence intensity of CdS QDs were very weak, and only the addition of amifostine could induce the generation of strong fluorescent CdS QDs. These results indicated that the assay system is highly selective to amifostine. 3.6. Selectivity assay for ALP based on the amifostine capped CdS QDs
Fig. 4. Fluorescence quenching ratio F/F0 of Cd2 þ –25 μM amifostine with different concentration of ALP (0, 4.67, 9.34, 18.68, 23.35, 28.02, 37.36, 46.7 nU/mL) upon addition of 0.3 mmol/L Na2S. Inset: The linear plots of F/F0 versus the ALP concentration in the range of 0–46.7 nU/mL. Reaction condition: 10 mmol/L Tris–HCl buffer solution (pH 8.2) at 25 °C.
respectively. The limit of detection (LOD) for amifostine was 0.12 μmol/L, calculated following the 3s IUPAC criteria. The standard deviation for nine replicate measurements of 2 μmol/L amifostine is 3.9%. This is the first time to detect amifostine in combination with QDs. Compared with the previous reports about amifostine assay [7,19], our present method offered a comparable detection limit and dynamic range. 3.4. The fluorescence quenching of amifostine-CdS QDs by ALP It is well known that ALP is a hydrolase that could effectively remove the phosphate functional groups from amifostine molecules [20]. Due to the more powerful affinity of amifostine to CdS QDs than its metabolite WR-1065, the generation of fluorescent CdS QDs would be inhibited in the presence of ALP. Different concentrations of ALP were firstly added to the 4 mmol/L Cd2 þ – 25 μmol/L amifostine system before the addition of Na2S. It could be seen from Fig. 4 that the FL intensity of the mixed system decreased gradually with the increase of ALP concentration. As shown in Fig. 4 inset, there was a good linear relationship between the relative fluorescence quenching ratio F/F0 (F0 is the fluorescence intensity of Cd2 þ –25 μmol/L amifostine-S2 system, and F is the fluorescence intensity of Cd2 þ –25 μmol/L amifostineS2 after the addition of ALP) and the ALP concentrations was in the range from 0 to 46.7 nU/mL. The regression equation could be described as follow:
F /F0 = 0.9958 − 0.0144[ALP] (nU/mL) The corresponding regression coefficient is 0.999, and the detection limit for ALP was 2.0 nU/mL, calculated following the 3s IUPAC criteria. The standard deviation for nine replicate measurements of 10 nU/mL ALP was 0.63%. The conventional fluorimetric methodology which usually exploited some fluorescent orthophosphoric monoesters such as 8-quinolyl phosphate, 2-carboxy-1-naphthyl phosphate as substrate were incapable of achieving such a low detection limit [21,22]. Compared with the previous reported fluorescence detection methods for ALP [23– 25], our method obtained a similar or superior detection limit and wider dynamic range. This method could be utilized to build up a real-time analytical method to detect the enzyme activity.
Selectivity is a very important parameter to evaluate the performance of a new sensor, especially for ones with potential applications in biomedical samples, a highly selective response to the target over other potentially competing species is necessary. Therefore,we further evaluated the selectivity of our nanosensor with various coexistence substances added. As shown in Fig. S3, to investigate the influence of different enzymes on the fluorescence intensity of CdS QDs, a series of 250 nU/mL enzymes including horseradish peroxidase (HRP), trypsin (Try), glucose oxidase (Gox), urase (Ura), pepsase (Pep), lysozyme (Lys), lactate dehydrogenase (LDH), and 150 ng/mL albumin bovine (BSA), human serum albumin (HSA) were tested instead of 25 nU/mL ALP. As shown from Fig. S3, only ALP have a highly specific response to amifostine and the generation of fluorescent CdS QDs could be effectively inhibited. The results showed that this ALP detection system could provide well ability of resisting interference.
4. The detection of amifostine and ALP in human serum samples In order to demonstrate the practical application of the fluorescence nanosensor, we performed detection of amifostine and ALP in human serum samples under the optimal conditions. The results obtained by standard addition method were listed in Tables 1–2. From Tables 1–2, we can see that the average recoveries of amifostine and ALP in the real samples were in the range of 100.5–108.5% and 96.4–105.8% respectively, and the RSD was lower than 3.5%, indicating that the accuracy and precision of the proposed method were satisfactory. Moreover, it also held potential application in screening the ALP inhibitors.
5. Conclusion In summary, we developed a novel, simple and fast assay for amifostine based on the growth of CdS QDs. The amifostine concentration could be associated with the fluorescence intensity of the generated CdS QDs. Furthermore, upon the addition of ALP, the generation of fluorescent CdS QDs could be effectively inhibited due to the enzymatic dephosphorylation of ALP. Therefore the Table 1 Results of amifostine determination in adult human serum samples. Samples
Added (μmol/L)
Founded (μmol/L)
Recovery (%)
RSD (n¼ 3, %)
1 2 3
2.0 10.0 20.0
2.177 0.02 10.2 7 0.03 20.17 0.05
108.5 101.0 100.5
3.5 3.2 2.6
1064
W. Na et al. / Talanta 144 (2015) 1059–1064
References
Table 2 Results of ALP determination in adult human serum samples. Samples
Found (nU/mL)
Added (nU/mL)
Total found ALP Recovery (%) RSD (nU/mL) (n ¼3, %)
1 2 3
8.90 13.46 17.72
9.34 14.01 18.68
18.63 7 0.21 26.497 0.13 37.29 7 0.27
102.14 96.43 105.76
3.11 2.94 2.67
present method could also be applied to selectively to detect ALP and the obtained detection limit for ALP was as low as 2.0 nU/mL.
Acknowledgments This work was financially supported by the National Natural Science Foundation of China (Nos. 21075050 and 21275063) and the Science and Technology Development Project of Jilin Province, China (No. 20110334).
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.talanta.2015.07. 057.
[1] C.N. Andreassen, C. Gran, J.C. Lindegaard, Semin. Radiat. Oncol. 13 (2003) 62–72. [2] J.R. Kouvaris, V.E. Kouloulias, L.J. Vlahos, Oncologist 12 (2007) 738–747. [3] C. Uzal, G. Durmus-Altun, M. Caloglu, A. Ergulen, S. Altaner, N.O. Yigitbasi, Int. J. Radiat. Oncol. Biol. Phys. 60 (2004) 564–569. [4] J. Nicole, B.A.K. Fernandez, Vet. Clin. Pathol. 36 (2007) 223–233. [5] P.S. Henthorn, M. Raducha, K.N. Fedde, M.A. Lafferty, M.P. Whyte, Proc. Natl. Acad. Sci. USA 89 (1992) 9924–9928. [6] I. Majsterek, E. Gloc, J. Blasiak, R.J. Reiter, J. Pineal Res. 38 (2005) 254–263. [7] J. Chen, Z. Lu, T.S. Lawrence, D.E. Smith, J. Chromatogr. B – Anal. Technol. Biomed. Life Sci. 819 (2005) 161–167. [8] A.K. Souid, G.L. Newton, R.L. Dubowy, R.C. Fahey, M.L. Bernstein, Cancer Chemother. Pharmacol. 42 (1998) 400–406. [9] A.E. Korst, J.B. Vermorken, W.J. Van der Vijgh, J. Chromatogr. B Biomed. Sci. Appl. 693 (1997) 167–174. [10] S. Liu, W. Na, S. Pang, X. Su, Biosens. Bioelectron. 58 (2014) 17–21. [11] L. Saa, A. Virel, J. Sanchez-Lopez, V. Pavlov, Chemistry 16 (2010) 6187–6192. [12] G. Garai-Ibabe, L. Saa, V. Pavlov, Anal. Chem. 85 (2013) 5542–5546. [13] G. Garai-Ibabe, M. Möller, V. Pavlov, Anal. Chem. 84 (2012) 8033–8037. [14] H. Matsumoto, T. Sakata, H. Mori, H. Yoneyama, J. Phys. Chem. 100 (1996) 13781–13785. [15] L. Saa, J.M. Mato, V. Pavlov, Anal. Chem. 84 (2012) 8961–8965. [16] C. Zhang, J. Xu, S. Zhang, X. Ji, Z. He, Chemistry 18 (2012) 8296–8300. [17] H.B. Ren, X.P. Yan, Talanta 97 (2012) 16–22. [18] S.S. Liji Sobhana, M. Vimala Devi, T.P. Sastry, A. Mandal, J. Nanopart. Res. 13 (2011) 1747–1757. [19] T. Huang, N. Chen, L. Zhang, G. Chen, Chromatographia 76 (2013) 1739–1745. [20] R. Capizzi, Semin. Oncology 23 (1996) 2–17. [21] X. Zhu, Q. Liu, C. Jiang, Clin. Chim. Acta 377 (2007) 150–153. [22] X. Zhu, Q. Liu, C. Jiang, Anal. Chim. Acta 570 (2006) 29–33. [23] N. Malashikhina, G. Garai-Ibabe, V. Pavlov, Anal. Chem. 85 (2013) 6866–6870. [24] F. Zheng, S. Guo, F. Zeng, J. Li, S. Wu, Anal. Chem. 86 (2014) 9873–9879. [25] N. Wu, J. Lan, L. Yan, J. You, Chem. Commun. 50 (2014) 4438–4441.
