Biosensors and Bioelectronics 69 (2015) 316–320

Contents lists available at ScienceDirect

Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios

Highly sensitive and doubly orientated selective molecularly imprinted electrochemical sensor for Cu2 þ Jianping Li a,n, Lianming Zhang a,b, Ge Wei b, Yun Zhang a, Ying Zeng b a Guangxi Key Laboratory of Electrochemical and Magnetochemical Function Materials, College of Chemistry and Bioengineering, Guilin University of Technology, Guilin 541004, China b College of Materials and Chemistry & Chemical Engineering, Chengdu University of Tehcnology, Chengdu 610000, China

art ic l e i nf o

a b s t r a c t

Article history: Received 30 November 2014 Received in revised form 28 February 2015 Accepted 3 March 2015 Available online 5 March 2015

Studies on molecularly imprinted electrochemical sensors for metal ions determination have been widely reported. However, the sensitivity and selectivity of the sensors needs to be improved urgently. In the current work, a novel molecularly imprinted electrochemical sensor was originally developed for selective determination of ultratrace Cu2 þ by combining the metal–ligand chelate orientated recognition with enzyme amplification effect. The detection relied on a competition reaction between Cu2 þ -glycine (Cu-Gly) and horse radish peroxidase (HRP)-labeled Cu-Gly on the imprinted polymer membrane modified electrode. The sensitivity of this sensor was promoted by enzyme amplification. Selectivity was improved by the double-specificity derived from ligand-to-metal ion and metal–ligand chelate orientated recognition of 3D imprinted cavities. This technique was quantitatively sensitive to Cu2 þ concentrations ranging from 0.5 nmol/L to 30 nmol/L, with a detection limit of 42.4 pmol/L. which was lower than those in most of the reported methods. The allowable amounts of interference ions were higher when it compared to other common molecularly imprinted sensors. Moreover, the results of assaying several real samples have proven its feasibility for practical applications. & 2015 Elsevier B.V. All rights reserved.

Keywords: Molecular imprinted Sensor Cu2 þ Doubly orientated selectivity Enzyme amplification

1. Introduction Molecularly imprinted electrochemical sensors (MIECS) have received a great deal of attention in determination of trace organic compounds (Nopper et al., 2003; Zhang et al., 2006; Sergeyeva et al., 2010; Bui et al., 2010) and proteins (Whitcombe et al., 2011; Bossi et al., 2007). In the recent years, a series of MIECS have been successfully developed for determination of metal ions such as Cu2 þ (Li et al. 2011b), Pb2 þ (Alizadeh and Amjadi 2011a; Wang et al., 2012), Hg2 þ (Alizadeh et al., 2011c), Cd2 þ (Alizadeh et al., 2011b; Li et al., 2011), [UO2] 2 þ (Metilda et al., 2007), and Dy2 þ (Prasad et al., 2006). In these assays, metal ions directly enter the cavities and form covalent or noncovalent bonds with the active sites in the molecularly imprinted membrane to introduce changes in current or potential responses. The sensitivities of the MIECS are restricted because of the weak signal changes of the current or the trace amounts of metal ions rebinding to the cavities (Li et al., 2011a; Ganjali et al., 2011). Significant efforts have been dedicated to improve the sensitivities of MIECS for trace metal ion assay (Gurtova et al., 2013; Roy et al., 2013). A method based on enzyme n

Corresponding author. Fax: þ86 773 899 0404. E-mail address: [email protected] (J. Li).

http://dx.doi.org/10.1016/j.bios.2015.03.010 0956-5663/& 2015 Elsevier B.V. All rights reserved.

amplification effect has also been proposed in this study (Li et al., 2010, 2012). The selectivity of MIECS relies on the template size, cavity structure, and distribution of the binding sites in molecularly imprinted polymer (MIP) cavities (Metilda et al., 2007; Poma et al., 2013). Given that the binding sites in the 3D cavities can coordinate with various kinds of metal ions, MIP cavities are incompetent to recognize the target metal ions such as organic compounds that possess steric configuration. As a result, the selectivity of the MIPs sensor is still not significantly in favor of the metal ions. Moreover, the functional groups in the cavities that bond with the target metal ions by coordination bonds can bond with other kinds of metal ions with similar properties. Therefore, severe interferences are present for detection of metal ions by MIPs sensors (Liang et al., 2011a; Huan et al., 2004). Highly sensitive and selective determination of trace heavy metal ions is critical in many areas including food safety testing (Davis et al., 1987). In this work, Cu2 þ ion was selected as an example to be detected via this strategy. Some researchers have reported on trace metal determination yet (Ajtony et al., 2008; Şahin et al., 2010; Moreno et al., 2008; Lara et al., 2005); However, the sensitivities of the methods used still need to be imrpoved. For example, as one of the most sensitive methods, the detection limit of graphite furnace atomic adsorption spectrometry (GFAAS) can

J. Li et al. / Biosensors and Bioelectronics 69 (2015) 316–320

317

Scheme 1. Schematic of the molecular imprinting technique.

only reach to 2 nmol/L. In this study, a novel MIECS for trace metal determination has been developed for the first time. Cu2 þ has been chosen as a sample, while Cu2 þ -glycine (Cu-Gly) has been chosen as the template molecule. The selectivity has been greatly improved by the so-called “double-specificity” effect. In other words, glycine was added into the sample solution to capture Cu2 þ and form CuGly, which is referred to the first selective recognition of ligand-tometal ions. The imprinted polymer on the electrode selectively then recognizes the formed Cu-Gly as the second selective recognition of cavities-to-templates. The amplification effect is performed by employing the labeled HRP (Li and Jiang et al., 2010); thus, the sensitivity is also improved. The detection limit can be reduced by three order of magnitudes compared with common ionic imprinted polymer sensors. The strategy for the assay is shown in Scheme 1. Imprinting cavity is formed after the removal of the temple molecular of the Cu-Gly complex from the electro-polymerized MIP. A step named ‘isolation’ is then introduced to occupy the vacant binding cavities (Li et al., 2011a). After ‘isolation’, cavities located in both shallow and deep of MIP membrane are occupied. The MIP electrode is then incubated in an HRP–Cu-Gly solution to make HRP–Cu-Gly replace Cu-Gly and then occupy the cavities. When the MIP electrode is dipped into sample solutions, the analytes Cu-Gly molecules will replace HRP–Cu-Gly and the amount of HRP–Cu-Gly decreases. The current changes produced by the HRPH2O2-hydroquinone system are obtained, providing the basis of Cu2 þ assay.

2. Experimental 2.1. Preparation of MIP and Non-MIP (nMIP) Sensors The process of synthesis of Cu-Gly is shown in supplementary materials, Fig. S5. Prior to the preparation of the modified electrodes, the surface of the GC electrode was polished with successively finer-grade aqueous alumina slurries (1.0, 0.3, and 0.05 μm grain sizes) on a chamois leather, followed by alternately washing with ethanol-HNO3 (V:V¼ 1:1) and water. Deposition of the MIPMB film was performed by electrochemical polymerization method in a boric acid buffer solution (pH ¼9.18, 25 °C) containing 12 mmol/L MB and 3 mmol/L Cu-Gly (dissolved in boric acid buffer solution). CVs were then performed for 25 cycles in the potential

range of  0.4 V to þ1.4 V at a scanning rate of 50 mV/s (Supplementary material, Fig. S1). Afterward, the GC electrode was gently stirred in ethanol for 4 min to remove the template molecule and other physical adsorbates on the surface of the imprinted membrane. The fabrication procedure of the nMIP sensor was the same as that of the MIP sensor, but without the template molecular Cu-Gly. 2.2. Isolation, incubation, competition, and enzyme reaction After template elution, the MB-MIP sensor was placed in 10 mL of 20 mmol/L Cu-Gly solution for 15 min to block the cavities in the MIP membrane. The sensor was then incubated in 2 mL of 10 μg/mL HRP–Cu-Gly solution for 8 min for competitive adsorption of Cu-Gly by HRP–Cu-Gly. Cu2 þ was mixed with excess amount of glycine and TEA in the range of 0.5 nmol/L to 30 nmol/L. The sensor was subsequently immersed in the Cu-Gly complex solutions for 10 min to allow the competitive binding of the analyte complex to the specific sites that had been occupied by HRP– Cu-Gly (Fig. S4). The electrode was then immersed into the base solution (containing 1.5 mmol/L HQ and 3.5 mmol/L H2O2) for 2 min while stirring for enzyme reaction. Electrochemical measurements were then performed. Comparative experiments were conducted according to the same process using a bare GC electrode and an nMIP sensor instead of the MIP sensor. 2.3. Electrochemical Measurements CV measurement was performed for electrochemical characterization. The scanning potential range for CVs was set from  0.2 V to þ0.6 V with a scan rate of 50 mV/s in 0.01 mol/L K3 [Fe(CN)6] solution containing 0.5 mol/L KCl. DPV was performed in 10 mL of PBS (0.05 mol/L, pH 7.4) containing 1.5 mmol/L HQ and 3.5 mmol/L H2O2 in the potential range from  0.4 V to 0.3 V with a scan rate of 50 mV/s and an amplitude of 50 mV.

3. Results and discussion 3.1. Molecular recognition of Cu-Gly by the PMB-MIP film CV was employed to verify the binding ability of the MIP membrane to the Cu-Gly, using K3[Fe(CN)6] as probe. As shown in

J. Li et al. / Biosensors and Bioelectronics 69 (2015) 316–320

1000

90

a

d

30

b

0

c

-30 -60 0.6

0.4

0.2

0.0

i / μA

500

I / µA

60

-0.2

i / µA

E/V

0

d

b

-500

-1000 0.6

0.4

0.2

0.0

-0.2

E/V Fig. 1. CVs of the MIP films. The CV measurements were performed in 0.01 mol/L K3[Fe(CN)6] solution containing 0.5 mol/L KCl. Curves a–d correspond to bare electrode, MIP electrode, MIP electrode after template elution, MIP electrode after rebinding, respectively. Scanning potential:  0.2 V to þ0.6 V, scan rate: 50 mV/s.

Fig. 1, a poor conductive MIP is formed on the surface of electrode from curve “a” to “b”. The speed of electron transfer between the electrode surface and the measurement solution was influenced by the increases of film thickness, whereas the peak current of the probe decreased. After elution of the Cu-Gly, some cavities in MIP appeared and the peak current of the probe increased (curves “b” to “c”). When the MIP electrode continuously rebound with Cu-Gly, the cavities were blocked again and the peak current decreased (curve “c” to curve “d”). However, the current was higher than that of curve “b” because few empty cavities existed under the adsorption balance of the template molecules between the solution and cavities. The EIS and E  t results exhibited similar results as shown in the supplementary material (Fig. S2 and Fig. S3). 3.2. Effect of functional monomer Other common electroactive compounds, including o-phenylenediamine, hydroquinone, Prussian blue, and acridine orange, have been investigated comparatively as functional monomer to create the imprinted polymer membrane. The results showed that these compounds can chelate directly with Cu2 þ ions, leading to poor formation of imprinted polymer on the electrode surface. Thus, unsatisfactory results for Cu2 þ determination were obtained using the resultant imprinting polymer sensors. MB was used throughout the experiments to fabricate the molecularly imprinted electrochemical sensors. 3.3. Calibration curve The main experimental factors for Cu2 þ ion determination were optimized in detail, including the used amounts of Cu-Gly and HRP–Cu-Gly; effect of the buffer solution and pH value; amounts of Gly, TEA, HQ and H2O2; incubation time, and competition time (Figs. S6 and S7). After incubation in HRP–Cu-Gly solution, the electrode was immersed into the solutions with different concentrations of CuGly, and then transferred into 10 mL of PBS (0.05 mol/L, pH ¼7.4) containing 1.5 mmol/L HQ and 3.5 mmol/L H2O2 for 2 min with constant stirring. The DPV curves were recorded to evaluate the relationship between the Cu2 þ concentration and oxidation peak value. The relationship between the concentration of Cu2 þ (c) and

40 35 30 25 20 15 10 5 0

35 30 25

Δi / μA

318

20 15 10 5 0

0

5

10

15

20

25

30

c / nmol·L-1

a j

-0.4 -0.3 -0.2 -0.1

0.0

0.1

0.2

0.3

0.4

Fig. 2. DPV after incubation in different concentration of Cu2 þ and calibration curve (insert). a-j: 0, 0.5, 2, 5, 8, 10, 15, 20, 25 and 30  10  9 mol/L Cu2 þ , respectively. Scanning in 10 mL of PBS (0.05 mol/L, pH ¼ 7.4) containing 1.5 mmol/L HQ and 3.5 mmol/L H2O2. Potential range:  0.4 V to þ 0.3 V, scan rate: 50 mV/s, amplitude: 50 mV, RSD (n¼ 5) of calibration curve point a–j correspond to 0, 1.96%, 1.89%, 2.01%, 1.72%, 1.98%, 2.02%, 1.64%, 1.57%, 1.45%.

the change of oxidation peak value (ΔI) is shown in Fig.2. The concentrations of Cu-Gly-TEA complex increased with the Cu2 þ concentration. More HRP-labeled complexes in the imprinted cavities were replaced by the Cu-Gly complex in the solution. Hence, the DPV peak currents of the HRP-H2O2HQ system decreased. The calibration curve between ΔI and c can be described by the following equation: ΔI (μA) ¼1.0186c þ0.8165 (r ¼0.9986), from 0.5 nmol/L to 30 nmol/L. A detection limit of 42.4 pmol/L was calculated according to the following equation: DL¼3δb/K, where DL is the detection limit at the 95% confidence level, δb is standard deviation of the blank measurements (n¼ 12), and K is slope of the calibration curve. Compared with the Cu2 þ imprinted sensors previously reported (Yang and Zhang 2009; Huan et al., 2004; Liu et al., 2006), the proposed sensor showed a more sensitive capability in Cu2 þ detection, with detection limit lower by almost three order of magnitudes. 3.4. Selectivity of the MIP sensor To evaluate the selectivity of the MB-MIP sensor, some interference ions possibly coexisting with 1  10  9 mol/L Cu2 þ were examined separately. With an allowable deviation of 5%, 1000 times of the concentration of alkali metals and alkaline-earth metals ions, such as K þ , Na þ , Mg2 þ , Ca2 þ ; 800 times of Fe3 þ , Al3 þ , Cr3 þ , Mn2 þ , Hg2 þ , Cd 2 þ ; 650 times of Ni2 þ and Zn2 þ ; 500 times of Co2 þ and Pb2 þ ; and 750 times of Ag þ had little interference to the measurement of 1  10  9 mol/L Cu2 þ . Compared with other MIECS, the selectivity of the proposed sensor improved significantly. In addition, the selectivity of the MIECS also increased compared with traditional metal ion selective electrode (Isa et al., 2012; Kamel et al., 2010). Some organic species that might cause interference to the result were tested separately. For 1  10  9 mol/L Cu2 þ , 3  10  7 mol/L vitamin A, ascorbic acid, alanine, glycine, sucrose, glucose, starch, Lcysteines were suited separately. The results showed that they did not influence on the response of the DPV signal. The high sensitivity of the sensor contributes for the specific recognization of the 3D cavities to the chelates in the MIP. Pb2 þ , Zn2 þ , Co2 þ , Ni2 þ , and Cu2 þ are border-line acids that can form chelates with the same complexing agent. In the assay, the fittings of the chelate structure formed between the ions and glycine, with the mole ratio of 1:2, were carried out by ChemBio3D. The results

J. Li et al. / Biosensors and Bioelectronics 69 (2015) 316–320

319

Table 1 Sample analysis results. Samples

Found (MIECS, nmol/L, n¼ 5) RSD (%) ICP-MS value (nmol/L) Added (nmol/L) Total found (MIECS, nmol/L, n¼5) RSD (%) Recovery % (MIECS)

Running water 10.32 Citric fruits juice 18.85 Rain water 3.15 Beer A 11.73 Beer B 11.66 GBW10031a 2.43 GBW10043b 26.05 a b

1.6% 5.3% 4.3% 2.7% 3.1% 2.5% 2.9%

11.86 17.80 3.13 11.22 11.95 2.41 26.03

2.00 10.00 10.00 3.00 3.00 – –

12.08 27.54 13.33 15.55 14.38 – –

3.3% 4.1% 4.6% 3.9% 2.2% – –

98.1% 95.5% 101.4% 105.6% 98.1% – –

GBW10031 is a wine reference sample (i.e. Chinese National Standard Reference Materials), in which the content of Cu (II) is 0.159  10  6 g/g. GBW10043 is a rice reference sample (i.e. Chinese National Standard Reference Materials), in which the content of Cu (II) is 1.7  10  6 g/g.

showed that the bond length, bond angle, and configuration of the chelates vary (Supplementary material, Table S1 and Fig. S8). The chelate can enhance the recognizing ability because the glycine has different bonding angle and length when it chelates with different metal ions. Thus, the differences are enhanced by the chelation process with little influence, which resulted from other metal ions in the strategy. The chelate with the metal ions can then be recognized by the MIP as a whole. Moreover, the specific recognization of the MIP is towards the chelate instead of metal ions. Compared with the common ion imprinted polymers, in the assay, the metal ions are chelated first with the complexing agent. The complex can be better recognized much well by the molecular imprinted polymer compared with the metal ions alone. Thus, the selectivity and recognizing ability of the sensor are enhanced, and the sensitivity is improved using enzyme amplification effect. 3.5. Determination of Cu2 þ in real samples To further demonstrate the application potential of Cu2 þ , MIECS was applied to determine Cu2 þ in several kinds of samples including running water, citric fruit juice, rain water, beer, and standard food sample. The results were compared to the ICP-MS methods. The beer sample was heated at 80 °C to remove dissolved CO2. The grinded rice reference sample (5 mg) was digested using 5 mL of HNO3, and then dissolved by 100 mL H2O. Before the assay, 15 μL of 5.0 mmol/L glycine and 10 μL of TEA were added to 10 mL of sample solution and stirred for 2 min. The MIP electrode was immersed into the above solution for 10 min after incubation with HRP–Cu-Gly to allow the competitive binding. DPV test was then carried out in the base solution containing 1.5 mmol/L HQ and 3.5 mmol/L H2O2 after 2 min. Recovery tests were conducted using the standard addition method. The results shown in Table 1 indicate that MIECS can offer comparable or even better performance than the ICP-MS methods. Moreover, the recoveries obtained by the new technique range from 95.5% to 105.6%, with an RSD of less than 5%, suggesting good recovery and practicability. The formation of Cu2 þ -Gly complexes in the solution of trace level copper ions and large amount of ligand is necessary to achieve low detection limit of the MIP sensor. The stability constant of the complex was evaluated by the method reported (Fang, 1995). The secondary stability constant K was lg K ¼16.3, performed in the solution of 7.5  10-6 mol/L Gly. As a theoretical result, 6.67  10  14 mol/L Cu2 þ was obtained, which was much lower than 4.24  10  11 mol/L, and can form Cu-Gly chelate with glycine in the solutions containing 10 μL of TEA. Therefore, the sensor is capable of being used to determine trace copper specifically. 4. Conclusions A novel MIP sensor that combines the enzyme amplification effect with double-specificity effect was developed for detection of

ultra-trace Cu2 þ ions. The results demonstrated that the proposed sensor could provide a new strategy for highly sensitive and selective determination of ultra-trace metal ions.

Acknowledgments The authors gratefully acknowledge the financial support of the National Natural Science Foundation of China (No. 21375031 and No. 21165007).

Appendix A. Suplementary Information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.bios.2015.03.010.

References Ajtony, Z., Szoboszlai, N., Suskó, E.K., Mezei, P., György, K., Bencs, L., 2008. Talanta 76, 627–634. Alizadeh, T., Amjadi, S., 2011a. J. Hazard. Mater. 190, 451–459. Alizadeh, T., Ganjali, M.R., Nourozi, P., Zare, M., Hoseini, M., 2011b. J. Electroanal. Chem. 657, 98–106. Alizadeh, T., Ganjali, M.R., Zare, M., 2011c. Anal. Chim. Acta 689, 52–59. Bossi, A., Bonini, F., Turner, A.P.F., Piletsky, S.A., 2007. Biosens. Bioelectron. 22, 1131–1137. Bui, B.T.S., Merlier, F., Haupt, K., 2010. Anal. Chem. 82, 4420–4427. Davis, G.K., Mertz, W., Copper, 1987. In: Mertz, W. (Ed.), Trace Elements in Human and Animal Nutrition, 10th ed. Academic Press, Inc., San Diego, pp. 328–346. Fang, F., 1995. Talanta 42, 317–321. Ganjali, H., Ganjali, M.R., Alizadeh, T., Faridbod, F., Norouzi, P.J., 2011. Electrochem. Sci. 6, 6085–6093. Gurtova, O., Ye, L., Chmilenko, F., 2013. Anal. Bioanal. Chem. 405, 287–295. Huan, S.Y., Jiao, C.X., Shen, Q., Jiang, J.H., Zeng, G.M., Huang, G.H., Shen, G.L., Yu, R.Q., 2004. Electrochim. Acta 49, 4273–4280. Isa, I.M., Saidin, M.I., Ahmad, M., Mustaffa, S., Ghani, S.A., 2012. Int. J. Electrochem. Sci. 7, 9526–9536. Kamel, A.H., Mahmoud, W.H., Mostafa, M.S., 2010. Electroanalysis 22, 2453–2459. Lara, R., Cerutti, S., Salonia, J.A., Olsina, R.A., Martinez, L.D., 2005. Food Chem. Toxicol. 43, 293–297. Liang, R.N., Zhang, R.M., Song, W.J., Hu, X.F., Qin, W., 2011. Sens. Lett. 9, 557–562. Li, J.P., Jiang, F.Y., Wei, X.P., 2010. Anal. Chem. 82, 6074–6078. Li, J.P., Li, Y.P., Zhang, Y., Wei, G., 2012. Anal. Chem. 84, 1888–1893. Li, J.P., Jiang, F.Y., Li, Y.P., Chen., Z.Q., 2011a. Biosens. Bioelectron. 26, 2097–2101. Li, Z.C., Fan, H.T., Zhang, Y., Chen, M.X., Yu, Z.Y., Cao, X.Q., Sun, T., 2011b. Chem. Eng. J. 171, 703–710. Liu, Z.H., Huan, S.Y., Jiang, J.H., Shen, G.L., Yu, R.Q., 2006. Talanta 68, 1120–1125. Metilda, P., Prasad, K., Kala, P., Gladis, J.M., Rao, T.P., Naidu, G.R.K., 2007. Anal. Chim. Acta 582, 147–153. Moreno, I.M., González-Weller, D., Gutierrez, V., Marino, M., Cameán, A.M., González, A.G., Hardisson, A., 2008. Microchem. J. 88, 56–61. Nopper, D., Lammershop, O., Wulff, G., Gauglitz, G., 2003. Anal. Bioanal. Chem. 377, 608–613. Poma, A., Guerreiro, A., Whitcombe, M.J., Piletska, E.V., Turner, A.P.F., Piletsky, S.A., 2013. Adv. Funct. Mater., 1–7. Prasad, K., Kala, R., Rao, P.T., Naidu, G.R.K., 2006. Anal. Chim. Acta 566, 69–74. Roy, A.C., Nisha, V.S., Dhand, C., Ali, M.A., Malhotra, B.D., 2013. Anal. Chim. Acta 777, 63–71. Şahin, Ç.A., Tokgöz, İ., Bektaş, S., 2010. Hazard. Mater. 181, 359–365.

320

J. Li et al. / Biosensors and Bioelectronics 69 (2015) 316–320

Sergeyeva, T.A., Slinchenko, O.A., Gorbach, L.A., Matyushov, V.F., Brovko, O.O., Piletsky, S.A., Sergeeva, L.M., Elska, G.V., 2010. Anal. Chim. Acta 659, 274–279. Wang, Z.H., Qin, Y.X., Wang, C., Sun, L.J., Lu, X.L., Lu, X.Q., 2012. Appl. Surf. Sci. 258, 2017–2021. Whitcombe, M.J., Chianella, I., Larcombe, L., Piletsky, S.A., Noble, J., Porter, R.,

Horgan, A., 2011. Chem. Soc. Rev. 40, 1547–1571. Yang, Z.P., Zhang, C.J., 2009. Sens. Actuators B 142, 210–215. Zhang, H.Q., Ye, L., Mosbach, K., 2006. J. Mol. Recognit. 19, 248–259.