Biosensors and Bioelectronics 63 (2015) 506–512
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
A fluorescent biosensor based on carbon dots-labeled oligodeoxyribonucleotide and graphene oxide for mercury (II) detection Xin Cui, Lei Zhu, Jing Wu, Yu Hou, Peiyao Wang, Zhenni Wang, Mei Yang n College of Chemistry and Chemical Engineering, Liaoning Normal University, Dalian 116029, Liaoning, China
art ic l e i nf o
a b s t r a c t
Article history: Received 29 April 2014 Received in revised form 27 July 2014 Accepted 30 July 2014 Available online 7 August 2014
As the newest two members of the carbon materials family, carbon dots (CDs) and graphene oxide (GO) possess many excellent optical properties resulting in a wide range of applications. In this work, we successfully synthesized CDs with a high-quantum-yield, and labeled them on oligodeoxyribonucleotide (ODN). The fluorescence of resultant CDs-labeled oligodeoxyribonucleotide (ODN–CDs) was quenched by GO via fluorescence resonance energy transfer. In the presence of Hg2 þ , the fluorescence was recovered by the release of ODN–CDs from GO due to the formation of T–Hg2 þ –T duplex. In the light of this theory, we designed a simple, highly sensitive and selective fluorometric Hg2 þ sensor based on CDs-labeled oligodeoxyribonucleotide and GO without complicated, costly and time-consuming operations. Under the optimal conditions, a linear relationship was obtained between relative fluorescence intensity and the concentration of Hg2 þ in the range of 5–200 nM (R2 ¼ 0.974). The present GO-based sensor system is highly selective toward Hg2 þ over a wide range of metal ions and has a detection limit of 2.6 nM. This method is reliable, and has been successfully applied for the detection of Hg2 þ in practical samples. & 2014 Elsevier B.V. All rights reserved.
Keywords: Carbon dots Graphene oxide Oligodeoxyribonucleotide Mercury detection
1. Introduction Mercury (II) ion (Hg2 þ ) is a highly toxic heavy metal ion in physiology, and causes harm to the human body mainly via the central nervous system, digestive system and internal organs (Pytharopoulou et al., 2013). Even in low concentration, it is a threat to the environment and human health because mercury is non-biodegradable and can enter food chain (Monikh et al., 2013; Zhang et al., 2013). The maximum concentration levels of mercury in cosmetics which regulated by the FDA and Health Canada are 1 mg/kg and 3 mg/kg, respectively (USFDA, 2014; Health Canada, 2014). FAO/WHO has launched the provisional tolerable weekly intake (PTWI) of 0.3 mg (including methylmercuryo0.2 mg) mercury in food. A number of analytical protocols for Hg2 þ were reported, such as Auger-electron spectroscopy (Lu et al., 2012), AAS/AES (Ghaedia et al., 2006; Frentiu et al., 2013), ICP-MS (Jia et al., 2011), etc. For routine analysis, however, the development of simple but sensitive and selective procedures is highly desired. Since the first research of oligodeoxyribonucleotide rich in thymine (T) for recognizing Hg2 þ via the specific formation of a Hg2 þ -mediated base pair (T–Hg2 þ –T) (Ono and Togashi, 2004), several methods for Hg2 þ detection based on this property been developed in the past years (Li et al., 2011; Wei et al., 2014; Zhang n
Corresponding author.
http://dx.doi.org/10.1016/j.bios.2014.07.085 0956-5663/& 2014 Elsevier B.V. All rights reserved.
et al., 2012). However, these methods suffer from several limitations such as the usage of highly costly fluorescent dye-labeled oligodeoxyribonucleotide, low photostability of the traditional fluorescent dyes (e.g. fluorescein isothiocyanate, FITC), poor resistance to digestion by endogenous nuclease (He et al., 2011), and so on. Nowadays some CdSe and CdTe inorganic quantum dots (QDs) can also be used as biolabeling reagents for the functionalization of aptamer oligonucleotide (Freeman et al., 2011; Zhang et al., 2011). Unfortunately, the application of QDs is limited since the heavy metals exhibit serious health and environmental concerns. Therefore, searching for benign materials which can overcome these drawbacks has become an urgent challenge. The newly emerging carbon dots (CDs), mainly including graphite nanoparticles less than 10 nm in size (Xu et al., 2004; Baker and Baker, 2010), may be considered as a favorable candidate. Compared to the conventional heavy-metal-based QDs and organic dyes, CDs possess a variety of advantages, such as superiority in ready preparation, convenience in surface functionalization, low cytoxicity, good biocompatibility and photostability (Dong et al., 2012; Song et al., 2012; Yu et al., 2013). More attentions have been paid to these nanomaterials, including their synthesis and applications. Recent studies based on CDs’ fluorescent properties indicated their potential applications as fluorescent labels in bioimaging and sensing (Zong et al., 2014).
X. Cui et al. / Biosensors and Bioelectronics 63 (2015) 506–512
Nevertheless, to our best knowledge, few researches about the oligonucleotides labeled with CDs have been reported for Hg2 þ detection. Graphene oxide (GO) has recently emerged as a fascinating material with one-atom thick, two dimensional graphitic carbon system (Novoselov et al., 2004; Geim and Novoselov, 2007) and various surface oxygen-containing groups including carboxyl, hydroxyl and epoxy groups (Liu et al., 2011). These properties make GO be attractive for biological applications (Li et al., 2014; Weaver et al., 2014). In addition, GO could adsorb dye-labeled oligonucleotides via hydrophobic and π–π stacking interactions (Wu et al., 2011), and simultaneously quench the fluorescence of dyes by fluorescence resonance energy transfer (FRET). Such characteristic have made GO an effective oligonucleotide-based fluorescent sensing platform material for Hg2 þ detection with low background. Herein, we report a fluorometric biosensor for Hg2 þ detection based on CDs-labeled oligonucleotide and GO. In the proposed system, CDs labeled T-rich 22-mer oligonucleotide acted as the energy donor and molecular recognition probe, and GO served as the FRET acceptor. In the absence of Hg2 þ ions, oligonucleotide would be adsorbed on GO surface and the fluorescence of CDs was quenched. While in the presence of Hg2 þ , thymine selectively bonded with Hg2 þ ions to form T–Hg2 þ –T complex and oligonucleotide self-hybridized into duplex, which can avoid the adsorption of oligonucleotide by GO and thus recover the fluorescence of CDs. This facile and effective method will show a potential application in the Hg2 þ monitoring of environment and food.
507
The quantum yield of CDs–COOH decreased greatly than CDs, which was attributed to the increase of electron-withdrawing carboxyl group. In the present work, the CDs-labeled ODN (ODN–CDs) conjugates were prepared by the well-established carbodiimide chemistry (Zhang et al., 2010; Zhao et al., 2013). Specifically, the as-prepared CDs–COOH were dissolved in a 10 mM PBS (pH ¼7.5) to obtain a solution containing 2.0 mg/mL CDs–COOH. Then 1 mL of 50 mM NHS and 1 mL of 500 mM EDC in 10 mM PBS were added to 1 mL of 2.0 mg/mL CDs–COOH solution, and the mixture was sonicated for 2 h. Afterwards, 0.074 mg (2 OD) of ODN was added, and further incubated for 24 h at 4 °C. The excess of CDs– COOH were removed in a dialysis bag (retained molecular weight: 7000 Da), until the dialysate had no UV–vis absorption values and fluorescence signals. The concentration of ODN–CDs was quantified according to the UV–vis absorption of the certain concentration of CDs–COOH at 335 nm. 2.3. Instrumentation and characterization Fourier transform infrared (FT-IR) spectra were obtained under a transmission mode with a Tensor 27 spectrometer (Bruker Optik GmbH, Ettlingen, Germany). UV–visible absorption spectra were recorded with a Shimadzu UV-240 spectrometer (Shimadzu Co., Kyoto, Japan). The fluorescence (FL) spectra were collected with a HITACHI F-7000 fluorescence spectrophotometer (Hitachi HighTech, Tokyo, Japan) under excitation of 347 nm. 2.4. Detection of Hg2 þ
2. Materials and methods 2.1. Chemicals and materials An oligonucleotide (ODN) with a sequence of 5′-NH2-(CH2)6 -TTCTTTCTTCGCGTTGTTTGTT-3′ (MW 6849.5) was purchased from Takara Biotechnology Co., Ltd. (Dalian, China). Sodium chloroacetate (ClCH2COONa), N-hydroxysuccinimide (NHS) and 1-ethyl-3(3-dimethylaminopropyl)-carbodiimide (EDC) were purchased from GL Biochem Ltd. (Shanghai, China). HNO3, NaOH, KH2PO4, Na2HPO4 12H2O, Hg(NO3)2 and other metal salts were purchased from Tianjin Kermel Chemical Reagent Co., Ltd. (Tianjin, China). Dialysis bags (MW 3500, MW 7000) were supplied by Union Carbide Co. (USA). Double distilled water was used for the preparation of all the solutions. All chemicals and solvents were obtained from the commercial sources and used without any further purification. GO used here was synthesized according to the well-established Hummers method (Hummers and Offeman, 1958). 2.2. Preparation of ODN–CDs conjugates CDs were prepared according to a reported method (Zhu et al., 2013). In summary, citric acid (3.0 g) and ethylenediamine (1875 mL) were dissolved in double distilled water (30 mL). Then the solution was transferred to a poly (tetrafluoroethylene) (Teflon)-lined autoclave (50 mL) and heated at 150 °C for 5 h. The product was subjected to dialysis against the double distilled water (retained molecular weight: 3500 Da) in order to obtain the CDs. Then, 50.0 mg of as-prepared CDs was dispersed in 50.0 mL of aqueous solution containing 2.5 g of NaOH and 2.5 g of ClCH2COONa, followed by bath sonication for 3 h (Zhang et al., 2010). After these treatments, the resulting CDs–COONa was neutralized with HCl and dialysed to obtain CDs–COOH. Finally, the CDs and CDs–COOH obtained possess high quantum yield of 75.0% and 7.19%, respectively (quinine sulfate as the standard).
The assay solution contained 4.0 μg/mL of ODN–CDs conjugates, 0.1 mg/mL GO, 100 mM Na þ and 50 mM Mg2 þ in 10 mM PBS (pH¼8.0). The signal output value was calculated according to the relative fluorescence intensity (F/F0 1), in which F0 and F represented the fluorescence intensities of the GO-based sensor system at 445 nm in the absence and presence of Hg2 þ , respectively. The optimization of experiment conditions was defined by relative fluorescence intensity at 445 nm after adding 100 nM Hg2 þ into the sensor system. For studying the linear relationship between the amount of Hg2 þ and F/F0 1, different concentrations of Hg2 þ were added to the assay solution, and the fluorescence spectra were recorded. 2.5. Samples preparation The present method was applied to detect Hg2 þ in citrus leaf samples which were obtained from Dalian Center for Disease Control and Prevention (Dalian, China) and treated by microwave digestion method and nitric acid leaching method, respectively. The specific operations were carried out as follows: 2.5.1. Microwave digestion method The citrus leaf samples were named 1 (0.496 g), 2 (0.499 g) and 3 (0.502 g). Then each sample was added to 10 mL of concentrated HNO3. After soaking overnight, all samples were processed by microwave irradiation at 110 °C for 8 min, 150 °C for 8 min and 180 °C for 40 min, respectively (a heating rate of 10 °C/min). The excess HNO3 was removed by volatilizing at 135 °C, until the residual liquid was around 1–2 mL. Finally, the samples were diluted to 25 mL with double distilled water. 2.5.2. Nitric acid leaching method The citrus leaf samples were named as J1 (0.836 g) and J2 (0.897 g), respectively. 5 mL concentrated HNO3 was added to J1 and J2, and heated in the boiling water bath for 2 h. After being
508
X. Cui et al. / Biosensors and Bioelectronics 63 (2015) 506–512
cooled to room temperature, the samples were added with 2 mL of 30% H2O2, followed by heating in the boiling water bath for 1 h. Finally, the samples were volumed to 25 mL with double distilled water. Before carrying out fluorescence experiments, the pH value of samples solution was neutralized by solid NaOH.
3. Results and discussion 3.1. Spectral characterization Fig. 1A shows the FT-IR spectrum of the CDs–COOH. The absorption bands at 3400 cm 1 and 1650 cm 1 were characteristic of O–H and CQO stretching vibration, respectively (Zhu et al., 2013). The others absorption bands were assigned as follows: 2919 cm 1 for C–H stretching vibration, 1388 cm 1 for O–H deformation peak, 1239 cm 1 for C–OH stretching peak, and 1135 cm 1 and 1092 cm 1 for double C–O strength stretching, respectively. The results demonstrated that CDs–COOH had excellent water solubility due to the oxygen-containing functional groups, which also enables the facile functionalization with biomolecules such as proteins and ODN. The absorptions of ODN–CDs conjugates were significantly different from those of CDs–COOH at 3100 cm 1, 1560 cm 1 and 1242 cm 1. These peaks were characteristic of –NH, –N–CQO and NHCQO stretching vibration, respectively, indicating the grafting of ODN on CDs–COOH. The UV–vis absorption spectra of the CDs–COOH and ODN–CDs are shown in Fig. 1B. In contrast to CDs–COOH, ODN–CDs showed a more intensive absorption in the range below 300 nm due to the bonding between ODN and CDs. The FT-IR and UV–vis spectrum characterization confirmed that ODN was conjugated to CDs– COOH. 3.2. Working principle of GO-based sensor The fluorescence spectra of ODN–CDs and GO-based sensor system in the presence and absence of Hg2 þ with excitation at 347 nm are shown in Fig. 2. Using quinine sulfate as a standard,
Fig. 2. The fluorescence spectra of ODN–CDs and GO-based sensor system in the presence or absence of Hg2 þ . Conditions: ODN–CDs, 4.0 μg/mL; GO, 0.1 mg/mL; Hg 2þ , 100 nM; pH, 8.0.
the fluorescence quantum yield of ODN–CDs was calculated to be 7.01%, which was similar to that of CDs–COOH. According to the results shown in Fig. 2, the principle of the GO-based sensor is depicted in Scheme 1. In the absence of Hg2 þ ions, ODN would be adsorbed on GO surface via hydrophobic and π–π stacking, and the fluorescence of CDs was quenched via FERT owing to the wide acceptance ability of GO. While in the presence of Hg2 þ , thymine selectively bonded with Hg2 þ ions to form T–Hg2 þ –T complex and ODN self-hybridized into duplex. This could avoid the adsorption of ODN by GO, and thus recovered the fluorescence of CDs. The value of F/F0 1, the concentrations of Hg2 þ will be determined.
Fig. 1. FT-IR spectra (A) and UV–vis absorption spectra (B) of CDs–COOH and ODN–CDs.
X. Cui et al. / Biosensors and Bioelectronics 63 (2015) 506–512
Scheme 1. Schematic illustration of the GO-based sensor system for Hg2 þ detection.
3.3. Optimization for Hg2 þ detection Before quantitative analysis of Hg2 þ using the proposed method, we optimized the detection conditions, including the amounts of GO and ODN–CDs used in the proposed sensor, ionic strength and pH value of the system. From Fig. 3A, it could be seen that the value of F/F0 1 of the GO-based sensor increased with the increasing concentration of
509
ODN–CDs up to 4.0 μg/mL. However, the background signal was increased as well with higher concentration of ODN–CDs, which would cause the decrease in the value of F/F0 1. Thus, we chose 4.0 μg/mL of ODN–CDs as the optimal concentration for the GObased sensor formation. The similar experiment was used to optimize the concentration of GO. As shown in Fig. 3B, 0.1 mg/mL GO displayed the highest F/F0 1 value, and was used during subsequent experiments. The carboxyl groups on the surface of GO were deprotonated at neutral pH. The polyanionic ODN should be repelled by the negatively charged GO due to electrostatic repulsion. On the other hand, ODN bases contain aromatic rings that can be bound to GO through hydrophobic interaction and π–π stacking (Wu et al., 2011). Therefore, the addition of salts can screen the electrostatic repulsion and bring ODN close to GO surface, leading to the enhanced signal-to-noise ratio of GO-based sensor. As shown in Fig. 3C, in the presence of 25 mM Na þ , the value of F/F0 1 was only 0.160. At the higher Na þ concentrations, the recovery efficiencies were progressively improved. The maximum F/F0 1 value was 0.242 with 100 mM Na þ . And then, we optimized the
Fig. 3. Effects of (A) ODN–CDs, (B) GO, (C) Na þ and Mg2 þ concentrations and (D) pH on the F/F0 1 value of sensor system. Conditions: Hg2 þ , 100 nM.
510
X. Cui et al. / Biosensors and Bioelectronics 63 (2015) 506–512
Fig. 4. (A) Fluorescence spectra of the GO-based sensor system containing various concentrations (0, 2, 5, 10, 20, 50, 100, and 200) of Hg2 þ ions; (B) plot of the F/F0 1 value as a function of the Hg2 þ concentration; (C) the linear region of sensing response; (D) selectivity of the sensing system of Hg2 þ to other competing metal ions, 100 nM Hg2 þ and 10 μM other metal ions. (Under the optimal conditions.)
concentration of divalent Mg2 þ ions. With concentrations of Mg2 þ higher than 50 mM, the values of F/F0 1 were over 0.351 (Fig. 3C). This is consistent with the concept that divalent metal ions are much more effective in screening charges and acting as a bridge to connect two negatively charged molecules in comparison to monovalent ones. Accordingly, 100 mM Na þ and 50 mM Mg2 þ were used in the further experiments. The effect of pH on the fluorescence intensity of GO-based sensor system was also studied. The results in Fig. 3D showed that there is slight variation of the value of F/F0 1 in the pH range of 5.5–9.0. Considering the best F/F0 1 value, we chose the PBS buffer with a pH value of 8.0 in the present work. 3.4. Performance evaluation Under the optimized conditions, the fluorescence spectrum of the different concentrations of Hg2 þ in the GO-based sensor system was studied (Fig. 4A). As shown in Fig. 4B, the value of
F/F0 1 increased gradually with the enhancement of Hg2 þ concentration, then declined obviously with the concentration of Hg2 þ higher than 200 nM. This phenomenon may be attributed to the fact that the interaction between Hg2 þ and the carboxylate or hydroxyl groups makes CDs close to each other, which accelerates the non-radiative recombination of the excitons through an effective electron transfer process, leading to a substantial decrease of the fluorescence of CDs (Guo et al., 2013; Zhou et al., 2012). Fig. 4C reveals a good linear correlation between F/F0 1 and the concentration of Hg2 þ in the range of 5–200 nM. The linear regression equation for Hg2 þ is F/F0 1 ¼0.0023CHg2 þ (nM)þ0.075 with the correlation coefficient (R2) of 0.974. The Hg2 þ can be detected as low as 2.6 nM (equal to 0.52 μg/kg, S/N ¼3), which is lower than the maximum allowable levels of Hg2 þ regulated by the USFDA (1 mg/kg) and Health Canada (3 mg/ kg). The relative standard deviation (RSD) was 3.2% for the determination of 50 nM Hg2 þ (n ¼ 11).
X. Cui et al. / Biosensors and Bioelectronics 63 (2015) 506–512
511
Table 1 Determination and recovery tests of Hg2 þ in practical samples by GO-based sensor system. Sample
1 2 3 J1 J2
Real contents (nM)
14.84 14.92 15.01 25.00 26.83
Measured (nM) AFS
The proposed method
15.81 15.00 15.43 26.20 27.51
13.58 13.87 14.30 25.28 26.46
3.5. Selectivity and interference For real world samples, the matrix compositions are usually quite complex, which might cause interferences for the determination of trace level Hg2 þ . Thus, potential interfering effects from some of metal ions frequently encountered in sample were investigated by using the present sensor system, and the results are shown in Fig. 4D. For the assay of 100 nM Hg2 þ , no obvious interfering effects were observed in the presence of 100-fold of Cd2 þ , Ca2 þ , Ag þ , Sn2 þ , Pb2 þ , Al3 þ , and 10-fold of Fe3 þ . However, equal amount of Fe2 þ caused interfering effects. In some particular cases, Fe2 þ can be oxidized to Fe3 þ by several oxidants, so the present method can still detect Hg2 þ in the presence of other possible interference ions after adding oxidant. In addition, the fluorescence intensities of the GO-based sensor system with tested interference ions alone were also explored (Fig. 4D). The F/F0 1 value with 100-fold of interference ions was much lower than that obtained with Hg2 þ . Obviously, the sensor based on ODN–CDs and GO possessed outstanding selectivity against other interference metal ions. 3.6. Determination of Hg2 þ in practical samples In order to evaluate the feasibility of the proposed method, the GO-based sensor system was applied to the determination of practical samples. The citrus leaf samples were certified reference materials for the chemical composition of biological samples containing (1507 20) 10 3 mg/kg Hg2 þ (GBW10020, GSB-11). Herein we treated the samples with microwave digestion and nitric acid leaching methods, which were named “1, 2, 3” and “J1, J2” respectively. The concentrations of Hg2 þ in these samples detected by the present method were compared with real contents and those measured by standard approach (Atomic Fluorescence Spectrometry, AFS). Thereafter, the samples were determined by the standard addition method and the analytical results are summarized in Table 1. Agreements were achieved between the results obtained by the AFS method and the present sensor. In addition, reasonable spiking recoveries in the range of 94.68– 109.8% were also attained employing the present procedure. The results demonstrated that the proposed sensor is suitable for Hg2 þ detection in practical samples.
4. Conclusions In this work, we combined the special optical properties of GO and CDs with selectivity of DNA ODN to develop a fluorometric biosensor for Hg2 þ detection. The fluorescence of CDs-labeled ODN was quenched by GO via FRET in the absence of Hg2 þ , and was further recovered by forming T–Hg2 þ –T duplex in the presence of Hg2 þ . We used the as-prepared platform to realize the sensitive and selective detection of Hg2 þ and further used it for
Added (nM)
Total found (nM)
Recovery (%)
RSD (n¼ 3, %)
50 50 50 50 50
60.92 66.98 67.15 80.19 81.09
94.7 106.2 105.7 109.8 109.3
2.64 2.06 1.12 2.07 2.01
real sample. This work expanded CDs applications in the fluorescent label technology, and predicted the potential abilities of both GO and CDs in the analytical methodology.
Acknowledgment This research was supported by the National Natural Science Foundation of China (NSFC, 51072074). Citrus leaf samples and the date of atomic fluorescence spectroscopy were obtained from Dalian Center for Disease Control and Prevention (Dalian, China).
References Baker, S.N., Baker, G.A., 2010. Angew. Chem. Int. Ed. 49, 6726–6744. Dong, Y.Q., Wang, R.X., Li, G.L., Chen, C.Q., Chi, Y.W., Chen, G., 2012. Anal. Chem. 84, 6220–6224. Freeman, R., Liu, X.Q., Willner, I., 2011. J. Am. Chem. Soc. 133, 11597–11604. Frentiu, T., Mihaltan, A.I., Senila, M., Darvasi, E., Ponta, M., Frentiu, M., Pintican, B.P., 2013. Microchem. J. 110, 545–552. Geim, A.K., Novoselov, K.S., 2007. Nat. Mater. 6, 183–191. Ghaedia, M., Fathi, M.R., Shokrollahi, A., Shajarat, F., 2006. Anal. Lett. 39, 1171–1185. Guo, Y.M., Wang, Z., Shao, H.W., Jiang, X.Y., 2013. Carbon 52, 583–589. Health Canada, Canada, Guidance on Heavy Metal Impurities in Cosmetics, 〈http:// www.hc-sc.gc.ca/cps-spc/pubs/indust/heavy_metals-metaux_lourds/indexeng.php#a33〉. (accessed: 22.4.2014). He, Y., Wang, Z.G., Tang, H.W., Pang, D.W., 2011. Biosens. Bioelectron. 29, 76–81. Hummers, W.S., Offeman, R.E., 1958. J. Am. Chem. Soc. 80, 1339. Jia, X.Y., Han, Y., Liu, X.L., Duan, T.C., Chen, H.T., 2011. Spectrochim. Acta Part B: At. Spectrosc. 66, 88–92. Li, H.L., Zhai, J.F., Tian, J.Q., Luo, Y.L., Sun, X.P., 2011. Biosens. Bioelectron. 26, 4656– 4660. Li, M., Liu, Q., Jia, Z.J., Xu, X.C., Cheng, Y., Xi, T.F., Wei, S.C., 2014. Carbon 67, 185–197. Liu, M.C., Chen, C.L., Hu, J., Wu, X.L., Wang, X., 2011. J. Phys. Chem. C 115, 25234– 25240. Lu, W.B., Qin, X.Y., Liu, S., Chang, G.H., Zhang, Y.W., Luo, Y.L., Asiri, A.M., Al-Youbi, A. O., Sun, X.P., 2012. Anal. Chem. 84, 5351–5357. Monikh, F.A., Karami, O., Hosseini, M., Karami, N., Bastami, A.A., Ghasemi, A.F., 2013. Ecotoxicol. Environ. Saf. 94, 112–115. Novoselov, K.S., Geim, A.K., Morozov, S.V., Jiang, D., Zhang, Y., Dubonos, S.V., Grigorieva, I.V., Firsov, A.A., 2004. Science 306, 666–669. Ono, A., Togashi, H., 2004. Angew. Chem. Int. Ed. 43, 4300–4302. Pytharopoulou, S., Kournoutou, G.G., Leotsinidis, M., Georgiou, C.D., Kalpaxis, D.L., 2013. Aquat. Toxicol. 134–135, 23–33. Song, Y.C., Shi, W., Chen, W., Li, X.H., Ma, H., 2012. J. Mater. Chem. 22, 12568–12573. USFDA, U.S., Cosmetics: ingredients prohibited & restricted by FDA regulations, 〈http://www.fda.gov/cosmetics/guidanceregulation/lawsregulations/ ucm127406.htm〉. (accessed: 22.4.2014). Weaver, C.L., LaRosa, J.M., Luo, X.L., Cui, X.T., 2014. ACS Nano 8, 1834–1843. Wei, Y., Li, B.M., Wang, X., Duan, Y.X., 2014. Analyst 139, 1618–1621. Wu, M., Kempaiah, R., Huang, P.J.J., Maheshwari, V., Liu, J., 2011. Langmuir 27, 2731– 2738. Xu, X.Y., Ray, R., Gu, Y.L., Ploehn, H.J., Gearheart, L., Raker, K., Scrivens, W.A., 2004. J. Am. Chem. Soc. 126, 12736–12737. Yu, C.M., Li, X.Z., Zeng, F., Zheng, F.Y., Wu, S.Z., 2013. Chem. Commun. 49, 403–405. Zhang, L.M., Xia, J.G., Zhao, Q.H., Liu, L.W., Zhang, Z., 2010. Small 6, 537–544. Zhang, X.R., Li, S.G., Jin, X., Li, X.M., 2011. Biosens. Bioelectron. 26, 3674–3678. Zhang, J.R., Huang, W.T., Xie, W.Y., Wen, T., Luo, H.Q., Li, N.B., 2012. Analyst 137, 3300–3305. Zhang, Q.H., Huang, L., Zhang, Y., Ke, C.H., Huang, H.Q., 2013. J. Proteomics 94, 37–53.
512
X. Cui et al. / Biosensors and Bioelectronics 63 (2015) 506–512
Zhao, H.M., Chang, Y.Y., Liu, M., Gao, S., Yu, H.T., Quan, X., 2013. Chem. Commun. 49, 234–236. Zhou, L., Lin, Y.H., Huang, Z.Z., Ren, J.S., Qu, X.G., 2012. Chem. Commun. 48, 1147– 1149.
Zhu, S.J., Meng, Q.N., Wang, L., Zhang, J.H., Song, Y.B., Jin, H., Zhang, K., Sun, H.C., Wang, H.Y., Yang, B., 2013. Angew. Chem. Int. Ed. 52, 3953–3957. Zong, J., Yang, X.L., Trinchi, A., Hardin, S., Cole, I., Zhu, Y.H., Li, C.Z., Muster, T., Wei, G., 2014. Biosens. Bioelectron. 51, 330–335.
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
A fluorescent biosensor based on carbon dots-labeled oligodeoxyribonucleotide and graphene oxide for mercury (II) detection Xin Cui, Lei Zhu, Jing Wu, Yu Hou, Peiyao Wang, Zhenni Wang, Mei Yang n College of Chemistry and Chemical Engineering, Liaoning Normal University, Dalian 116029, Liaoning, China
art ic l e i nf o
a b s t r a c t
Article history: Received 29 April 2014 Received in revised form 27 July 2014 Accepted 30 July 2014 Available online 7 August 2014
As the newest two members of the carbon materials family, carbon dots (CDs) and graphene oxide (GO) possess many excellent optical properties resulting in a wide range of applications. In this work, we successfully synthesized CDs with a high-quantum-yield, and labeled them on oligodeoxyribonucleotide (ODN). The fluorescence of resultant CDs-labeled oligodeoxyribonucleotide (ODN–CDs) was quenched by GO via fluorescence resonance energy transfer. In the presence of Hg2 þ , the fluorescence was recovered by the release of ODN–CDs from GO due to the formation of T–Hg2 þ –T duplex. In the light of this theory, we designed a simple, highly sensitive and selective fluorometric Hg2 þ sensor based on CDs-labeled oligodeoxyribonucleotide and GO without complicated, costly and time-consuming operations. Under the optimal conditions, a linear relationship was obtained between relative fluorescence intensity and the concentration of Hg2 þ in the range of 5–200 nM (R2 ¼ 0.974). The present GO-based sensor system is highly selective toward Hg2 þ over a wide range of metal ions and has a detection limit of 2.6 nM. This method is reliable, and has been successfully applied for the detection of Hg2 þ in practical samples. & 2014 Elsevier B.V. All rights reserved.
Keywords: Carbon dots Graphene oxide Oligodeoxyribonucleotide Mercury detection
1. Introduction Mercury (II) ion (Hg2 þ ) is a highly toxic heavy metal ion in physiology, and causes harm to the human body mainly via the central nervous system, digestive system and internal organs (Pytharopoulou et al., 2013). Even in low concentration, it is a threat to the environment and human health because mercury is non-biodegradable and can enter food chain (Monikh et al., 2013; Zhang et al., 2013). The maximum concentration levels of mercury in cosmetics which regulated by the FDA and Health Canada are 1 mg/kg and 3 mg/kg, respectively (USFDA, 2014; Health Canada, 2014). FAO/WHO has launched the provisional tolerable weekly intake (PTWI) of 0.3 mg (including methylmercuryo0.2 mg) mercury in food. A number of analytical protocols for Hg2 þ were reported, such as Auger-electron spectroscopy (Lu et al., 2012), AAS/AES (Ghaedia et al., 2006; Frentiu et al., 2013), ICP-MS (Jia et al., 2011), etc. For routine analysis, however, the development of simple but sensitive and selective procedures is highly desired. Since the first research of oligodeoxyribonucleotide rich in thymine (T) for recognizing Hg2 þ via the specific formation of a Hg2 þ -mediated base pair (T–Hg2 þ –T) (Ono and Togashi, 2004), several methods for Hg2 þ detection based on this property been developed in the past years (Li et al., 2011; Wei et al., 2014; Zhang n
Corresponding author.
http://dx.doi.org/10.1016/j.bios.2014.07.085 0956-5663/& 2014 Elsevier B.V. All rights reserved.
et al., 2012). However, these methods suffer from several limitations such as the usage of highly costly fluorescent dye-labeled oligodeoxyribonucleotide, low photostability of the traditional fluorescent dyes (e.g. fluorescein isothiocyanate, FITC), poor resistance to digestion by endogenous nuclease (He et al., 2011), and so on. Nowadays some CdSe and CdTe inorganic quantum dots (QDs) can also be used as biolabeling reagents for the functionalization of aptamer oligonucleotide (Freeman et al., 2011; Zhang et al., 2011). Unfortunately, the application of QDs is limited since the heavy metals exhibit serious health and environmental concerns. Therefore, searching for benign materials which can overcome these drawbacks has become an urgent challenge. The newly emerging carbon dots (CDs), mainly including graphite nanoparticles less than 10 nm in size (Xu et al., 2004; Baker and Baker, 2010), may be considered as a favorable candidate. Compared to the conventional heavy-metal-based QDs and organic dyes, CDs possess a variety of advantages, such as superiority in ready preparation, convenience in surface functionalization, low cytoxicity, good biocompatibility and photostability (Dong et al., 2012; Song et al., 2012; Yu et al., 2013). More attentions have been paid to these nanomaterials, including their synthesis and applications. Recent studies based on CDs’ fluorescent properties indicated their potential applications as fluorescent labels in bioimaging and sensing (Zong et al., 2014).
X. Cui et al. / Biosensors and Bioelectronics 63 (2015) 506–512
Nevertheless, to our best knowledge, few researches about the oligonucleotides labeled with CDs have been reported for Hg2 þ detection. Graphene oxide (GO) has recently emerged as a fascinating material with one-atom thick, two dimensional graphitic carbon system (Novoselov et al., 2004; Geim and Novoselov, 2007) and various surface oxygen-containing groups including carboxyl, hydroxyl and epoxy groups (Liu et al., 2011). These properties make GO be attractive for biological applications (Li et al., 2014; Weaver et al., 2014). In addition, GO could adsorb dye-labeled oligonucleotides via hydrophobic and π–π stacking interactions (Wu et al., 2011), and simultaneously quench the fluorescence of dyes by fluorescence resonance energy transfer (FRET). Such characteristic have made GO an effective oligonucleotide-based fluorescent sensing platform material for Hg2 þ detection with low background. Herein, we report a fluorometric biosensor for Hg2 þ detection based on CDs-labeled oligonucleotide and GO. In the proposed system, CDs labeled T-rich 22-mer oligonucleotide acted as the energy donor and molecular recognition probe, and GO served as the FRET acceptor. In the absence of Hg2 þ ions, oligonucleotide would be adsorbed on GO surface and the fluorescence of CDs was quenched. While in the presence of Hg2 þ , thymine selectively bonded with Hg2 þ ions to form T–Hg2 þ –T complex and oligonucleotide self-hybridized into duplex, which can avoid the adsorption of oligonucleotide by GO and thus recover the fluorescence of CDs. This facile and effective method will show a potential application in the Hg2 þ monitoring of environment and food.
507
The quantum yield of CDs–COOH decreased greatly than CDs, which was attributed to the increase of electron-withdrawing carboxyl group. In the present work, the CDs-labeled ODN (ODN–CDs) conjugates were prepared by the well-established carbodiimide chemistry (Zhang et al., 2010; Zhao et al., 2013). Specifically, the as-prepared CDs–COOH were dissolved in a 10 mM PBS (pH ¼7.5) to obtain a solution containing 2.0 mg/mL CDs–COOH. Then 1 mL of 50 mM NHS and 1 mL of 500 mM EDC in 10 mM PBS were added to 1 mL of 2.0 mg/mL CDs–COOH solution, and the mixture was sonicated for 2 h. Afterwards, 0.074 mg (2 OD) of ODN was added, and further incubated for 24 h at 4 °C. The excess of CDs– COOH were removed in a dialysis bag (retained molecular weight: 7000 Da), until the dialysate had no UV–vis absorption values and fluorescence signals. The concentration of ODN–CDs was quantified according to the UV–vis absorption of the certain concentration of CDs–COOH at 335 nm. 2.3. Instrumentation and characterization Fourier transform infrared (FT-IR) spectra were obtained under a transmission mode with a Tensor 27 spectrometer (Bruker Optik GmbH, Ettlingen, Germany). UV–visible absorption spectra were recorded with a Shimadzu UV-240 spectrometer (Shimadzu Co., Kyoto, Japan). The fluorescence (FL) spectra were collected with a HITACHI F-7000 fluorescence spectrophotometer (Hitachi HighTech, Tokyo, Japan) under excitation of 347 nm. 2.4. Detection of Hg2 þ
2. Materials and methods 2.1. Chemicals and materials An oligonucleotide (ODN) with a sequence of 5′-NH2-(CH2)6 -TTCTTTCTTCGCGTTGTTTGTT-3′ (MW 6849.5) was purchased from Takara Biotechnology Co., Ltd. (Dalian, China). Sodium chloroacetate (ClCH2COONa), N-hydroxysuccinimide (NHS) and 1-ethyl-3(3-dimethylaminopropyl)-carbodiimide (EDC) were purchased from GL Biochem Ltd. (Shanghai, China). HNO3, NaOH, KH2PO4, Na2HPO4 12H2O, Hg(NO3)2 and other metal salts were purchased from Tianjin Kermel Chemical Reagent Co., Ltd. (Tianjin, China). Dialysis bags (MW 3500, MW 7000) were supplied by Union Carbide Co. (USA). Double distilled water was used for the preparation of all the solutions. All chemicals and solvents were obtained from the commercial sources and used without any further purification. GO used here was synthesized according to the well-established Hummers method (Hummers and Offeman, 1958). 2.2. Preparation of ODN–CDs conjugates CDs were prepared according to a reported method (Zhu et al., 2013). In summary, citric acid (3.0 g) and ethylenediamine (1875 mL) were dissolved in double distilled water (30 mL). Then the solution was transferred to a poly (tetrafluoroethylene) (Teflon)-lined autoclave (50 mL) and heated at 150 °C for 5 h. The product was subjected to dialysis against the double distilled water (retained molecular weight: 3500 Da) in order to obtain the CDs. Then, 50.0 mg of as-prepared CDs was dispersed in 50.0 mL of aqueous solution containing 2.5 g of NaOH and 2.5 g of ClCH2COONa, followed by bath sonication for 3 h (Zhang et al., 2010). After these treatments, the resulting CDs–COONa was neutralized with HCl and dialysed to obtain CDs–COOH. Finally, the CDs and CDs–COOH obtained possess high quantum yield of 75.0% and 7.19%, respectively (quinine sulfate as the standard).
The assay solution contained 4.0 μg/mL of ODN–CDs conjugates, 0.1 mg/mL GO, 100 mM Na þ and 50 mM Mg2 þ in 10 mM PBS (pH¼8.0). The signal output value was calculated according to the relative fluorescence intensity (F/F0 1), in which F0 and F represented the fluorescence intensities of the GO-based sensor system at 445 nm in the absence and presence of Hg2 þ , respectively. The optimization of experiment conditions was defined by relative fluorescence intensity at 445 nm after adding 100 nM Hg2 þ into the sensor system. For studying the linear relationship between the amount of Hg2 þ and F/F0 1, different concentrations of Hg2 þ were added to the assay solution, and the fluorescence spectra were recorded. 2.5. Samples preparation The present method was applied to detect Hg2 þ in citrus leaf samples which were obtained from Dalian Center for Disease Control and Prevention (Dalian, China) and treated by microwave digestion method and nitric acid leaching method, respectively. The specific operations were carried out as follows: 2.5.1. Microwave digestion method The citrus leaf samples were named 1 (0.496 g), 2 (0.499 g) and 3 (0.502 g). Then each sample was added to 10 mL of concentrated HNO3. After soaking overnight, all samples were processed by microwave irradiation at 110 °C for 8 min, 150 °C for 8 min and 180 °C for 40 min, respectively (a heating rate of 10 °C/min). The excess HNO3 was removed by volatilizing at 135 °C, until the residual liquid was around 1–2 mL. Finally, the samples were diluted to 25 mL with double distilled water. 2.5.2. Nitric acid leaching method The citrus leaf samples were named as J1 (0.836 g) and J2 (0.897 g), respectively. 5 mL concentrated HNO3 was added to J1 and J2, and heated in the boiling water bath for 2 h. After being
508
X. Cui et al. / Biosensors and Bioelectronics 63 (2015) 506–512
cooled to room temperature, the samples were added with 2 mL of 30% H2O2, followed by heating in the boiling water bath for 1 h. Finally, the samples were volumed to 25 mL with double distilled water. Before carrying out fluorescence experiments, the pH value of samples solution was neutralized by solid NaOH.
3. Results and discussion 3.1. Spectral characterization Fig. 1A shows the FT-IR spectrum of the CDs–COOH. The absorption bands at 3400 cm 1 and 1650 cm 1 were characteristic of O–H and CQO stretching vibration, respectively (Zhu et al., 2013). The others absorption bands were assigned as follows: 2919 cm 1 for C–H stretching vibration, 1388 cm 1 for O–H deformation peak, 1239 cm 1 for C–OH stretching peak, and 1135 cm 1 and 1092 cm 1 for double C–O strength stretching, respectively. The results demonstrated that CDs–COOH had excellent water solubility due to the oxygen-containing functional groups, which also enables the facile functionalization with biomolecules such as proteins and ODN. The absorptions of ODN–CDs conjugates were significantly different from those of CDs–COOH at 3100 cm 1, 1560 cm 1 and 1242 cm 1. These peaks were characteristic of –NH, –N–CQO and NHCQO stretching vibration, respectively, indicating the grafting of ODN on CDs–COOH. The UV–vis absorption spectra of the CDs–COOH and ODN–CDs are shown in Fig. 1B. In contrast to CDs–COOH, ODN–CDs showed a more intensive absorption in the range below 300 nm due to the bonding between ODN and CDs. The FT-IR and UV–vis spectrum characterization confirmed that ODN was conjugated to CDs– COOH. 3.2. Working principle of GO-based sensor The fluorescence spectra of ODN–CDs and GO-based sensor system in the presence and absence of Hg2 þ with excitation at 347 nm are shown in Fig. 2. Using quinine sulfate as a standard,
Fig. 2. The fluorescence spectra of ODN–CDs and GO-based sensor system in the presence or absence of Hg2 þ . Conditions: ODN–CDs, 4.0 μg/mL; GO, 0.1 mg/mL; Hg 2þ , 100 nM; pH, 8.0.
the fluorescence quantum yield of ODN–CDs was calculated to be 7.01%, which was similar to that of CDs–COOH. According to the results shown in Fig. 2, the principle of the GO-based sensor is depicted in Scheme 1. In the absence of Hg2 þ ions, ODN would be adsorbed on GO surface via hydrophobic and π–π stacking, and the fluorescence of CDs was quenched via FERT owing to the wide acceptance ability of GO. While in the presence of Hg2 þ , thymine selectively bonded with Hg2 þ ions to form T–Hg2 þ –T complex and ODN self-hybridized into duplex. This could avoid the adsorption of ODN by GO, and thus recovered the fluorescence of CDs. The value of F/F0 1, the concentrations of Hg2 þ will be determined.
Fig. 1. FT-IR spectra (A) and UV–vis absorption spectra (B) of CDs–COOH and ODN–CDs.
X. Cui et al. / Biosensors and Bioelectronics 63 (2015) 506–512
Scheme 1. Schematic illustration of the GO-based sensor system for Hg2 þ detection.
3.3. Optimization for Hg2 þ detection Before quantitative analysis of Hg2 þ using the proposed method, we optimized the detection conditions, including the amounts of GO and ODN–CDs used in the proposed sensor, ionic strength and pH value of the system. From Fig. 3A, it could be seen that the value of F/F0 1 of the GO-based sensor increased with the increasing concentration of
509
ODN–CDs up to 4.0 μg/mL. However, the background signal was increased as well with higher concentration of ODN–CDs, which would cause the decrease in the value of F/F0 1. Thus, we chose 4.0 μg/mL of ODN–CDs as the optimal concentration for the GObased sensor formation. The similar experiment was used to optimize the concentration of GO. As shown in Fig. 3B, 0.1 mg/mL GO displayed the highest F/F0 1 value, and was used during subsequent experiments. The carboxyl groups on the surface of GO were deprotonated at neutral pH. The polyanionic ODN should be repelled by the negatively charged GO due to electrostatic repulsion. On the other hand, ODN bases contain aromatic rings that can be bound to GO through hydrophobic interaction and π–π stacking (Wu et al., 2011). Therefore, the addition of salts can screen the electrostatic repulsion and bring ODN close to GO surface, leading to the enhanced signal-to-noise ratio of GO-based sensor. As shown in Fig. 3C, in the presence of 25 mM Na þ , the value of F/F0 1 was only 0.160. At the higher Na þ concentrations, the recovery efficiencies were progressively improved. The maximum F/F0 1 value was 0.242 with 100 mM Na þ . And then, we optimized the
Fig. 3. Effects of (A) ODN–CDs, (B) GO, (C) Na þ and Mg2 þ concentrations and (D) pH on the F/F0 1 value of sensor system. Conditions: Hg2 þ , 100 nM.
510
X. Cui et al. / Biosensors and Bioelectronics 63 (2015) 506–512
Fig. 4. (A) Fluorescence spectra of the GO-based sensor system containing various concentrations (0, 2, 5, 10, 20, 50, 100, and 200) of Hg2 þ ions; (B) plot of the F/F0 1 value as a function of the Hg2 þ concentration; (C) the linear region of sensing response; (D) selectivity of the sensing system of Hg2 þ to other competing metal ions, 100 nM Hg2 þ and 10 μM other metal ions. (Under the optimal conditions.)
concentration of divalent Mg2 þ ions. With concentrations of Mg2 þ higher than 50 mM, the values of F/F0 1 were over 0.351 (Fig. 3C). This is consistent with the concept that divalent metal ions are much more effective in screening charges and acting as a bridge to connect two negatively charged molecules in comparison to monovalent ones. Accordingly, 100 mM Na þ and 50 mM Mg2 þ were used in the further experiments. The effect of pH on the fluorescence intensity of GO-based sensor system was also studied. The results in Fig. 3D showed that there is slight variation of the value of F/F0 1 in the pH range of 5.5–9.0. Considering the best F/F0 1 value, we chose the PBS buffer with a pH value of 8.0 in the present work. 3.4. Performance evaluation Under the optimized conditions, the fluorescence spectrum of the different concentrations of Hg2 þ in the GO-based sensor system was studied (Fig. 4A). As shown in Fig. 4B, the value of
F/F0 1 increased gradually with the enhancement of Hg2 þ concentration, then declined obviously with the concentration of Hg2 þ higher than 200 nM. This phenomenon may be attributed to the fact that the interaction between Hg2 þ and the carboxylate or hydroxyl groups makes CDs close to each other, which accelerates the non-radiative recombination of the excitons through an effective electron transfer process, leading to a substantial decrease of the fluorescence of CDs (Guo et al., 2013; Zhou et al., 2012). Fig. 4C reveals a good linear correlation between F/F0 1 and the concentration of Hg2 þ in the range of 5–200 nM. The linear regression equation for Hg2 þ is F/F0 1 ¼0.0023CHg2 þ (nM)þ0.075 with the correlation coefficient (R2) of 0.974. The Hg2 þ can be detected as low as 2.6 nM (equal to 0.52 μg/kg, S/N ¼3), which is lower than the maximum allowable levels of Hg2 þ regulated by the USFDA (1 mg/kg) and Health Canada (3 mg/ kg). The relative standard deviation (RSD) was 3.2% for the determination of 50 nM Hg2 þ (n ¼ 11).
X. Cui et al. / Biosensors and Bioelectronics 63 (2015) 506–512
511
Table 1 Determination and recovery tests of Hg2 þ in practical samples by GO-based sensor system. Sample
1 2 3 J1 J2
Real contents (nM)
14.84 14.92 15.01 25.00 26.83
Measured (nM) AFS
The proposed method
15.81 15.00 15.43 26.20 27.51
13.58 13.87 14.30 25.28 26.46
3.5. Selectivity and interference For real world samples, the matrix compositions are usually quite complex, which might cause interferences for the determination of trace level Hg2 þ . Thus, potential interfering effects from some of metal ions frequently encountered in sample were investigated by using the present sensor system, and the results are shown in Fig. 4D. For the assay of 100 nM Hg2 þ , no obvious interfering effects were observed in the presence of 100-fold of Cd2 þ , Ca2 þ , Ag þ , Sn2 þ , Pb2 þ , Al3 þ , and 10-fold of Fe3 þ . However, equal amount of Fe2 þ caused interfering effects. In some particular cases, Fe2 þ can be oxidized to Fe3 þ by several oxidants, so the present method can still detect Hg2 þ in the presence of other possible interference ions after adding oxidant. In addition, the fluorescence intensities of the GO-based sensor system with tested interference ions alone were also explored (Fig. 4D). The F/F0 1 value with 100-fold of interference ions was much lower than that obtained with Hg2 þ . Obviously, the sensor based on ODN–CDs and GO possessed outstanding selectivity against other interference metal ions. 3.6. Determination of Hg2 þ in practical samples In order to evaluate the feasibility of the proposed method, the GO-based sensor system was applied to the determination of practical samples. The citrus leaf samples were certified reference materials for the chemical composition of biological samples containing (1507 20) 10 3 mg/kg Hg2 þ (GBW10020, GSB-11). Herein we treated the samples with microwave digestion and nitric acid leaching methods, which were named “1, 2, 3” and “J1, J2” respectively. The concentrations of Hg2 þ in these samples detected by the present method were compared with real contents and those measured by standard approach (Atomic Fluorescence Spectrometry, AFS). Thereafter, the samples were determined by the standard addition method and the analytical results are summarized in Table 1. Agreements were achieved between the results obtained by the AFS method and the present sensor. In addition, reasonable spiking recoveries in the range of 94.68– 109.8% were also attained employing the present procedure. The results demonstrated that the proposed sensor is suitable for Hg2 þ detection in practical samples.
4. Conclusions In this work, we combined the special optical properties of GO and CDs with selectivity of DNA ODN to develop a fluorometric biosensor for Hg2 þ detection. The fluorescence of CDs-labeled ODN was quenched by GO via FRET in the absence of Hg2 þ , and was further recovered by forming T–Hg2 þ –T duplex in the presence of Hg2 þ . We used the as-prepared platform to realize the sensitive and selective detection of Hg2 þ and further used it for
Added (nM)
Total found (nM)
Recovery (%)
RSD (n¼ 3, %)
50 50 50 50 50
60.92 66.98 67.15 80.19 81.09
94.7 106.2 105.7 109.8 109.3
2.64 2.06 1.12 2.07 2.01
real sample. This work expanded CDs applications in the fluorescent label technology, and predicted the potential abilities of both GO and CDs in the analytical methodology.
Acknowledgment This research was supported by the National Natural Science Foundation of China (NSFC, 51072074). Citrus leaf samples and the date of atomic fluorescence spectroscopy were obtained from Dalian Center for Disease Control and Prevention (Dalian, China).
References Baker, S.N., Baker, G.A., 2010. Angew. Chem. Int. Ed. 49, 6726–6744. Dong, Y.Q., Wang, R.X., Li, G.L., Chen, C.Q., Chi, Y.W., Chen, G., 2012. Anal. Chem. 84, 6220–6224. Freeman, R., Liu, X.Q., Willner, I., 2011. J. Am. Chem. Soc. 133, 11597–11604. Frentiu, T., Mihaltan, A.I., Senila, M., Darvasi, E., Ponta, M., Frentiu, M., Pintican, B.P., 2013. Microchem. J. 110, 545–552. Geim, A.K., Novoselov, K.S., 2007. Nat. Mater. 6, 183–191. Ghaedia, M., Fathi, M.R., Shokrollahi, A., Shajarat, F., 2006. Anal. Lett. 39, 1171–1185. Guo, Y.M., Wang, Z., Shao, H.W., Jiang, X.Y., 2013. Carbon 52, 583–589. Health Canada, Canada, Guidance on Heavy Metal Impurities in Cosmetics, 〈http:// www.hc-sc.gc.ca/cps-spc/pubs/indust/heavy_metals-metaux_lourds/indexeng.php#a33〉. (accessed: 22.4.2014). He, Y., Wang, Z.G., Tang, H.W., Pang, D.W., 2011. Biosens. Bioelectron. 29, 76–81. Hummers, W.S., Offeman, R.E., 1958. J. Am. Chem. Soc. 80, 1339. Jia, X.Y., Han, Y., Liu, X.L., Duan, T.C., Chen, H.T., 2011. Spectrochim. Acta Part B: At. Spectrosc. 66, 88–92. Li, H.L., Zhai, J.F., Tian, J.Q., Luo, Y.L., Sun, X.P., 2011. Biosens. Bioelectron. 26, 4656– 4660. Li, M., Liu, Q., Jia, Z.J., Xu, X.C., Cheng, Y., Xi, T.F., Wei, S.C., 2014. Carbon 67, 185–197. Liu, M.C., Chen, C.L., Hu, J., Wu, X.L., Wang, X., 2011. J. Phys. Chem. C 115, 25234– 25240. Lu, W.B., Qin, X.Y., Liu, S., Chang, G.H., Zhang, Y.W., Luo, Y.L., Asiri, A.M., Al-Youbi, A. O., Sun, X.P., 2012. Anal. Chem. 84, 5351–5357. Monikh, F.A., Karami, O., Hosseini, M., Karami, N., Bastami, A.A., Ghasemi, A.F., 2013. Ecotoxicol. Environ. Saf. 94, 112–115. Novoselov, K.S., Geim, A.K., Morozov, S.V., Jiang, D., Zhang, Y., Dubonos, S.V., Grigorieva, I.V., Firsov, A.A., 2004. Science 306, 666–669. Ono, A., Togashi, H., 2004. Angew. Chem. Int. Ed. 43, 4300–4302. Pytharopoulou, S., Kournoutou, G.G., Leotsinidis, M., Georgiou, C.D., Kalpaxis, D.L., 2013. Aquat. Toxicol. 134–135, 23–33. Song, Y.C., Shi, W., Chen, W., Li, X.H., Ma, H., 2012. J. Mater. Chem. 22, 12568–12573. USFDA, U.S., Cosmetics: ingredients prohibited & restricted by FDA regulations, 〈http://www.fda.gov/cosmetics/guidanceregulation/lawsregulations/ ucm127406.htm〉. (accessed: 22.4.2014). Weaver, C.L., LaRosa, J.M., Luo, X.L., Cui, X.T., 2014. ACS Nano 8, 1834–1843. Wei, Y., Li, B.M., Wang, X., Duan, Y.X., 2014. Analyst 139, 1618–1621. Wu, M., Kempaiah, R., Huang, P.J.J., Maheshwari, V., Liu, J., 2011. Langmuir 27, 2731– 2738. Xu, X.Y., Ray, R., Gu, Y.L., Ploehn, H.J., Gearheart, L., Raker, K., Scrivens, W.A., 2004. J. Am. Chem. Soc. 126, 12736–12737. Yu, C.M., Li, X.Z., Zeng, F., Zheng, F.Y., Wu, S.Z., 2013. Chem. Commun. 49, 403–405. Zhang, L.M., Xia, J.G., Zhao, Q.H., Liu, L.W., Zhang, Z., 2010. Small 6, 537–544. Zhang, X.R., Li, S.G., Jin, X., Li, X.M., 2011. Biosens. Bioelectron. 26, 3674–3678. Zhang, J.R., Huang, W.T., Xie, W.Y., Wen, T., Luo, H.Q., Li, N.B., 2012. Analyst 137, 3300–3305. Zhang, Q.H., Huang, L., Zhang, Y., Ke, C.H., Huang, H.Q., 2013. J. Proteomics 94, 37–53.
512
X. Cui et al. / Biosensors and Bioelectronics 63 (2015) 506–512
Zhao, H.M., Chang, Y.Y., Liu, M., Gao, S., Yu, H.T., Quan, X., 2013. Chem. Commun. 49, 234–236. Zhou, L., Lin, Y.H., Huang, Z.Z., Ren, J.S., Qu, X.G., 2012. Chem. Commun. 48, 1147– 1149.
Zhu, S.J., Meng, Q.N., Wang, L., Zhang, J.H., Song, Y.B., Jin, H., Zhang, K., Sun, H.C., Wang, H.Y., Yang, B., 2013. Angew. Chem. Int. Ed. 52, 3953–3957. Zong, J., Yang, X.L., Trinchi, A., Hardin, S., Cole, I., Zhu, Y.H., Li, C.Z., Muster, T., Wei, G., 2014. Biosens. Bioelectron. 51, 330–335.
