Glucose oxidase-functionalized fluorescent gold nanoclusters as probes for glucose

Analytica Chimica Acta 772 (2013) 81–86

Contents lists available at SciVerse ScienceDirect

Analytica Chimica Acta journal homepage: www.elsevier.com/locate/aca

Glucose oxidase-functionalized fluorescent gold nanoclusters as probes for glucose Xiaodong Xia a,b , Yunfei Long b,∗ , Jianxiu Wang a,∗∗ a b

College of Chemistry and Chemical Engineering, Central South University, Changsha 410083, PR China School of Chemistry and Chemical Engineering, Hunan University of Science and Technology, Xiangtan 411201, China

h i g h l i g h t s

g r a p h i c a l

a b s t r a c t

 A glucose oxidase/gold nanocluster conjugates formed by etching chemistry.  Integration of the bioactivities and fluorescence properties within a single unit.  These conjugates serve as novel fluorescent probe for glucose.

a r t i c l e

i n f o

Article history: Received 3 January 2013 Received in revised form 8 February 2013 Accepted 13 February 2013 Available online 24 February 2013 Keywords: Gold nanocluster Glucose oxidase Etching chemistry Glucose Fluorescence

a b s t r a c t Creation and application of noble metal nanoclusters have received continuous attention. By integrating enzyme activity and fluorescence for potential applications, enzyme-capped metal clusters are more desirable. This work demonstrated a glucose oxidase (an enzyme for glucose)-functionalized gold cluster as probe for glucose. Under physiological conditions, such bioconjugate was successfully prepared by an etching reaction, where tetrakis (hydroxylmethyl) phosphonium-protected gold nanoparticle and thioctic acid-modified glucose oxidase were used as precursor and etchant, respectively. These bioconjugates showed unique fluorescence spectra (em max = 650 nm, ex max = 507 nm) with an acceptable quantum yield (ca. 7%). Moreover, the conjugated glucose oxidase remained active and catalyzed reaction of glucose and dissolved O2 to produce H2 O2 , which quenched quantitatively the fluorescence of gold clusters and laid a foundation of glucose detection. A linear range of 2.0 × 10−6 –140 × 10−6 M and a detection limit of 0.7 × 10−6 M (S/N = 3) were obtained. Also, another horseradish peroxidase/gold cluster bioconjugate was produced by such general synthesis method. Such enzyme/metal cluster bioconjugates represented a promising class of biosensors for biologically important targets in organelles or cells. Crown Copyright © 2013 Published by Elsevier B.V. All rights reserved.

1. Introduction Fluorescence behaviors of noble metal nanoclusters (NCs) have focused recently due to its good photostability and high brightness [1–11]. Moreover, their fluorescent properties are highly dependent on the size of the metal nanocluster and the ligands or scaffolds

∗ Corresponding author. Tel.: +86 7328372324; fax: +86 7328372324. ∗∗ Corresponding author. E-mail addresses: l [email protected], l [email protected] (Y. Long), [email protected] (J. Wang).

on cluster surface [5,12]. Such features allow metal clusters to be widely applied in analysis field for specific target readouts by various mechanisms, such as metal core leaching [13], cluster aggregations [8,14], ligand displacements [15,16], conversion of scaffold conformation [17], and adjustment of microenvironments around NCs [18–20]. Although various targets have been detected by these strategies, no enzyme bioactivity has been focused in these systems. Therefore, to create active enzyme-functionalized metal nanoclusters with direct application is more desirable. With appropriate scaffolds, various metal clusters have been created [5,11,21,22]. However, how to prepare enzyme-capped metal nanoclusters with intact enzyme activity for biosensing

0003-2670/$ – see front matter. Crown Copyright © 2013 Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.aca.2013.02.025

82

X. Xia et al. / Analytica Chimica Acta 772 (2013) 81–86

seemed difficult. For instance, the essential harsh conditions (e.g., pH ≥ 12) or very strong reducers (e.g., BH4 − ) used in the existing synthesis protocols would retard biomolecular activities in some wise [11,23,24]. Recently, etching chemistry has been widely used for creation of few-atom metal clusters by special etchants such as thiol compounds and polymers [8,25–28]. Motivated by these findings, we expected that thiol compound-modified enzyme would serve as efficient etchant to facilitate preparation of enzymefunctionalized metal clusters under mild etching condition. More importantly, the mild etching environment along with no strong reducer use favored keeping enzyme activity intact. Level of body fluid glucose is used for diagnosis of diabetes or hypoglycemia. Thus, accurate detection of glucose is very important. Recently, various nanomaterials have been used for detection of glucose with high sensitivity [29–36]. For instance, a glucose oxidase/gold nanoparticle bioconjugate has been developed as a colorimetric probe for glucose [37]. Although this assay is simple, the improvement of sensitivity was required. Mixing metal nanoclusters with glucose oxidase, facile glucose probes have been developed [36,38,39]. However, these mixtures disfavored applications in organelles or cells [32]. Therefore, glucose oxidase (GOD) was chosen as model enzyme for preparation of enzyme-functionalized fluorescent gold nanocluster (GOD-AuNC) for glucose detection. Thioctic acid-functionalized glucose oxidase (TA-GOD) was prepared by using carbodiimide chemistry and tetrakis (hydroxylmethyl) phosphonium-capped gold nanoparticle (THP-AuNP) was synthesized according previous report [40]. TA-GOD etched THP-AuNPs into fluorescent AuNCs. Such GOD-AuNC bioconjugate integrated catalysis function of enzyme and fluorescence of metal cluster in a single unit. Moreover, the conjugated GOD enzyme catalyzed reaction of glucose and dissolved O2 to produce H2 O2 , which quenched quantitatively the fluorescence of AuNCs. Therefore, such bioconjugates served as probe for direct detection of glucose. Also, a horseradish peroxidase/gold cluster bioconjugate (HRP-AuNC) was prepared by this synthesis strategy. These bioconjugates had potentials to detect targets within living cells [10,24,41]. 2. Experimental 2.1. Reagents Glucose oxidase, horseradish peroxidase (HRP), tetrakis (hydroxylmethyl) phosphonium (THP), 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC) and N-hydroxysuccinimide (NHS) were purchased from Sigma–Aldrich. Thioctic acid and other chemicals were obtained from Sinopharm Chemical Regaent Co. Ltd. Glucose and other carbohydrates were analytical grade. Serum and urine samples were obtained from Xiangtan Central Hospital (China). Informed consent was signed for each patient participating in this study and reviewed by the Joint Ethics Committee of the Xiangtan Health Authority and performed in accordance with national guidelines. 2.2. Instruments Hydrodynamic diameters of free GOD enzyme and GODAuNCs were analyzed by dynamic light scattering (DLS) (zetasizer, Malvern Instruments Ltd). Separation of free GOD and GOD-AuNCs by size exclusion chromatography (SEC) was performed on a highperformance liquid chromatography system (Shimadzu). HRTEM analyses of size and size distributions of AuNPs and AuNCs were carried out on Tecnai G2 20 AEM transmission electron microscopy. Binding energies of 4f electrons of Au on the surface of AuNCs were obtained from electron spectrometer (Thermo Fisher Scientific).

Absorption spectra of AuNPs and AuNCs were recorded by UVvis spectrophotometer (Shimadzu, UV-2450) and fluorescence spectra of AuNCs were measured by fluorescence spectrophotometer (Hitachi, F4600). Fluorescence lifetimes of AuNCs were measured with Fluorescence lifetime spectroscopy (FluoTime 100, PicoQuant). 2.3. Preparation of GOD-AuNCs AuNPs were prepared by reduction of HAuCl4 with THP [40]. Typically, 12 ␮L THP was injected to NaOH (45 mL, 10 mM) solution. Under vigorous shaking, HAuCl4 solution (1.0 wt%, 1.50 mL) was rapidly added to form AuNPs. The excess reactants and byproducts were removed by dialysis. TA-GOD was prepared by succinimide coupling. Typically, thioctic acid (1.00 mL, 50 mM) solution was added into a mixture of EDC and NHS (5.00 mL, both 10 mM, buffered by phosphate, pH 7.4). After 30 min incubation, 2.00 mL GOD solution (1.0 mg mL−1 ) was added. After another 2 h incubation, the mixture was kept overnight at 4 ◦ C and then was dialyzed by phosphate buffer. The as-prepared TA-GOD solution (8.00 mL) was mixed with the above AuNP solution (0.10 mL). This mixture was incubated under vigorous shaking for 2 d at room temperature and then GOD-AuNCs were prepared. After dialysis and further isolation of the GODAuNCs over free GOD by SEC, the as-prepared GOD-AuNCs were stored at 4 ◦ C for further use. 2.4. Glucose detection 0.10 mL purified GOD-AuNC solution was mixed with glucose solution of known concentration and diluted to 1.00 mL with phosphate buffer (7.4). After 15 min incubation, the mixture was subjected to fluorescence measurements. The fluorescence intensity was used to fit the calibration curve and evaluate the performance of the proposed assay. To test the practicability, glucose levels in urine samples (spiked with glucose) and serum samples were detected by the proposed method. 3. Results and discussion 3.1. Synthesis and characterization of GOD-AuNCs Currently, approaches of conversion larger metal nanoparticles to few-atom clusters have been focused [8,26,42–44]. Particularly, various thiol compounds are employed as etchants for creation of AuNCs [8,26,42]. For example, Au25 has been created by using glutathione as etchant [45]. Because of significant fragmentation energies, once the capping thiol agent is absorbed onto the particle surface, AuNP tends to dissociate into smaller Au and form Au–thiolate cluster [9]. Here, a facile, mild etching avenue to yield GOD-AuNCs was described (Scheme 1A). First, TA-GOD Enzyme was synthesized by using carbodiimide chemistry and THP-AuNPs were obtained by reducing HAuCl4 with THP that also severed as a stabilizer for AuNPs [40]. THP-AuNPs were employed as precursors mainly because organic phosphorous ligands are prone to be substituted by other stronger functional groups [46]. Second, GOD-AuNCs were synthesized through etching THP-AuNPs by TA-GOD. The development of the etching reaction was monitored by fluorescence assessment (Supplemental data Fig. S1). Third, purified GOD-AuNCs were achieved by SEC separation based on different hydrodynamic diameters of GOD-AuNC (ca. 35 nm) and free TA-GOD (ca. 6 nm) measured by DLS. The separation results (Supplemental Data Fig. S2) exhibited a high-performance etching process. The purified GOD-AuNC fraction was employed to perform further experiments.

X. Xia et al. / Analytica Chimica Acta 772 (2013) 81–86

83

Scheme 1. (A) THP-AuNPs were transformed into fluorescent GOD-AuNCs by an etching reaction. (B) Fluorescence quenching of GOD-AuNCs in presence of glucose.

No phosphor was found by element analysis (XPS), which confirmed that initial phosphine ligands were replaced by TA-GOD enzyme. A control experiment showed that thiol-modification of enzyme was essential to prepare enzyme-functionalized Au clusters and no fluorescent species formed without modification. The reproducibility of the synthesis method was validated by three separate creations of GOD-AuNCs with similar optical properties. Such etching strategy would be a general method for preparation other enzyme-functionalized metal clusters, as evident by creation of HRP-AuNCs. The size and size distribution of as-obtained THP-AuNPs and GOD-AuNCs were measured by HRTEM. THP-AuNPs exhibited size distribution range of 2.0–9.5 nm (Fig. 1A and D). GOD-Au NC exhibited size distribution of 0.50–2.5 nm (mainly 1.0 nm, Fig. 1B and E).

were important to biological applications. Fluorescence quantum yield and lifetime were measured. Biexponential decay kinetics ( 1 = 1.8 ns;  2 = 0.2 ns) and a decent quantum yield of 7.0% (relative to Rhodamine B in ethanol) were observed. Another control experimental showed that thioctic acid also directly etched the above AuNPs to AuNCs with weak fluorescence intensity, as evident by the results in Supplemental Data Fig. S3. According to previous report, ligands with electron-rich atoms (O, S, N, etc) or other functional groups would effectively enhance fluorescence of GOD-AuNCs [12]. Also, the GOD layer would protect the nanocluster from quenchers present in solution. Thus, the fluorescence of GOD-AuNCs was stronger than that of thioctic acid-capped AuNCs. Further, HRP-AuNCs were prepared under the identical conditions, while its emission spectrum was different from that of GOD-AuNCs (Supplemental Data Fig. S3), again supporting ligand-dependent optical properties of AuNCs.

3.2. Absorption and fluorescence spectra 3.3. Fluorescence quenching in presence of glucose The absorption spectra of THP-AuNPs and GOD-AuNCs were shown in Fig. 2A. THP-AuNPs obviously exhibited a characteristic SPR peak at ca. 520 nm for its relative larger size. For GOD-AuNCs, no characteristic peak was observed, as similar to other proteindirected AuNCs [11,38]. We noted that the shoulders at about 280 nm for both GOD-AuNCs in the absence and presence glucose were assigned to the characteristic absorption of the aromatic side groups in GOD units. Fluorescence spectra of GOD-AuNCs were indicated in Fig. 2B. The large Stokes-shift (ca. 140 nm) and approaching near-IR emission of these AuNCs enabled enhancing sensitivity and

Interestingly, the emission intensity of GOD-AuNCs was effectively quenched when glucose (substrate) presented (Scheme 1B and Fig. 3). The possible mechanism beyond such quenching was further investigated. The unchanged fluorescence lifetimes of GOD-AuNCs (both of  1 / 2 = 1.8/0.2 ns in the absence and presence of glucose) indicated that the fluorescence quench obeyed a static quenching mechanism. Further investigation showed that the emission of HRP-AuNCs was not quenched by glucose while both emissions of HRP-AuNCs and GOD-AuNCs were quenched by H2 O2 . Another finding was that the fluorescence of GOD-AuNCs

84

X. Xia et al. / Analytica Chimica Acta 772 (2013) 81–86

Fig. 2. (A) UV–vis absorption spectra of GOD-AuNCs in the absence (dotted) and presence (long dash) of glucose, THP-AuNPs (solid) and GOD enzyme (dash-dot). (B) Emission (excited at 507 nm) and excitation (emission at 650 nm) spectra of GOD-AuNCs.

of absorption region between 470 and 700 nm when glucose presented in GOD-Au NC system (Fig. 2A). 3.4. Detection of glucose Fig. 1. HRTEM images and size distributions analyses of THP-Au NPS (A and D), and GOD-Au NCs in the absence (B and E) and presence (C and F) of glucose, respectively; 240 particles are measured to get the size distribution.

was not quenched by glucose when the dissolved O2 in solutions was excluded by nitrogen gas bubbling. Thus, fluorescence of GODAuNCs was quenched mainly by enzymatically produced H2 O2 GOD

(glucose + O2 −→gluconic acid + H2 O2 ) when glucose presented, which was consistent with previous reports [38,39]. As well known, optical properties of Au clusters dramatically depend on the cluster size and hence aggregation of clusters always quenches its fluorescence [39,47–49]. Here, a significant fluorescence quenching occurred accompanying with the evolution of small clusters to larger ones in the presence of glucose (compared Fig. 1B, E and C, F). Two factors were possibly responsible for this cluster aggregation and fluorescence quenching. First, the oxidizedstate Au(I) on the cluster surface would be reduced into Au(0) in the presence of enzymatic product H2 O2 because H2 O2 can reduce oxidized-state gold to atom [50,51]. The XPS measurements (Fig. 4) testified the above reaction. It indicated that the binding energy of the Au on the surface of AuNCs had an Au 4f7/2 peak at 84.8 eV in the absence of glucose while that was shifted to 83.9 eV in the presence of glucose, which was assigned to Au(I) and Au(0), respectively [52,53]. Another possibility was the coordination structure of thiol group toward AuNC would suffer breakage and hence NC aggregation occurred when enzymatic product H2 O2 present, as evidenced by previous RS–Au bonds are oxidized into RS–SR and even RSO3 H by H2 O2 [38,39,54]. Also, the above oxidate-state change and nanocluster structure degradation were validated by the disappearance

With active enzyme and fluorescence in a single unit, recognition and quantification of glucose were accomplished by GOD-AuNC bioconjugate (Scheme 1B). Fig. 5A indicates that the fluorescence intensity decreased with the glucose concentration increase. The effects of temperature (25–60 ◦ C) and buffer solution pH (5.0–9.0) on the fluorescence intensity of GOD-AuNCs were slight (data not shown). For convenience, detection of glucose was performed under room temperature and pH 7.4. The fluorescence quenching reached completion within 15 min. This probe exhibited a linear range of 2.0 × 10−6 –140 × 10−6 M with a detection limit of 0.7 × 10−6 M (S/N = 3) and a correlation efficiency of 0.998 (Fig. 5B).

Fig. 3. Glucose quenched fluorescence of GOD-AuNCs. Glucose concentration: 0 (solid), 1.0 × 10−4 (dotted) and 1.0 × 10−3 M (long dash).

X. Xia et al. / Analytica Chimica Acta 772 (2013) 81–86

85

Table 1 Glucose content detection in serum samples (diluted by 50-folds, n = 3). Sample no.

Detection by instrument (glucose, ␮M)

Detection by our method (glucose, ␮M)

RSD (%)

1 2 3

99.0 45.0 120

96.5 42.8 113.2

−5.3 −4.6 −3.2

assay. For example, glucose contents of three serum samples were detected by the proposed method. The results (Table 1) of glucose detection obtained by our method are close to that by instrument (Hitachi 7080, Japan, glucose oxidase method). Here, we demonstrated a fully new method for creation of enzyme-functionalized metal clusters for targeted applications. In contrast to mixing metal clusters and GOD enzyme for glucose detection, this bioconjugate was expected to favor detecting targets in living cells. Fig. 4. XPS spectra of Au on the surface of GOD-AuNCs in the absence (dotted) and presence (solid) of glucose.

With high substrate specificity of GOD enzyme, the probe exhibited a good selectivity to glucose detection, as supported by the neglectable changes of the fluorescence responses when interferences (maltose, lactose, fructose, or mannose) present. To test the applicability of the proposed assay, this probe was employed to detect glucose levels in urine samples spiked with glucose. The presence of glucose in urine is a very dangerous condition, as it is an indication of worsening of diabetes. The results were collected in Supplemental Data Table S1. The acceptable recoveries indicated that a good practicability was obtained. The large Stokesshift and approaching near-infrared emission of AuNCs favored a satisfied sensitivity. Therefore, it was believed that glucose levels in other human fluids such as sera could be assessed by this proposed

4. Conclusions In this work, we presented a straightforward etching avenue to produce enzyme–AuNC bioconjugates. Integrating bioactivity and fluorescence, these bioconjugates enabled real time response to enzymatic reaction. As a proof-of-concept experiment, GOD enzyme–AuNCs were prepared and used as optical probe for substrates (i.e., glucose). Also, HRP-AuNCs were prepared. Other enzyme-functionalized metal clusters were expected to be created by this general synthesis protocol for potential applications. Acknowledgements Partial support of this work by the Program for New Century Excellent Talents in University (NCET-10-0796 to JW), Specialized Research Fund for the Doctoral Program of Higher Education (No. 20100162110018 to JW), National Natural Science Foundation of China (No. 21275047 to YL); Research Foundation of Department of Education of Hunan Province (11A033 to YL). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.aca.2013.02.025. References

Fig. 5. (A) Fluorescence responses of GOD-AuNCs to glucose (0–1.0 × 10−3 M). (B) Fluorescence ratio (F0 /F) versus the concentration of glucose. Inset: plot of the fluorescence ratio versus the glucose concentration over a range of 0–140 × 10−6 M. F0 and F denoted fluorescence intensity in the absence and presence of glucose, respectively. Error bars were derived from triplicate separate measurements.

[1] T. Vosch, Y. Antoku, J.C. Hsiang, C.I. Richards, J.I. Gonzalez, R.M. Dickson, Proc. Natl. Acad. Sci. U.S.A. 104 (2007) 12621. [2] L.A. Peyser, A.E. Vinson, A.P. Bartko, R.M. Dickson, Science 291 (2001) 106. [3] W. Wei, Y. Lu, W. Chen, S. Chen, J. Am. Chem. Soc. 133 (2011) 2063. [4] W. Luo, S.J. Pennycook, S.T. Pantelides, Nano Lett. 7 (2007) 3137. [5] Y. Negishi, Y. Takasugi, S. Sato, H. Yao, K. Kimura, T. Tsukuda, J. Am. Chem. Soc. 126 (2004) 6519. [6] K. Saha, S.S. Agasti, C. Kim, X. Li, V.M. Rotello, Chem. Rev. 112 (2012) 2779. [7] C.M. Aikens, J. Phys. Chem. Lett. 2 (2010) 104. [8] C.-C. Huang, Z. Yang, K.-H. Lee, H.-T. Chang, Angew. Chem. Int. Ed. 46 (2007) 6828. ˇ D. Marx, J. Am. Chem. Soc. 126 (2004) 12111. [9] M. Konôpka, R. Rousseau, I. Stich, [10] H.-C. Yeh, J. Sharma, J.J. Han, J.S. Martinez, J.H. Werner, Nano Lett. 10 (2010) 3110. [11] J. Xie, Y. Zheng, J.Y. Ying, J. Am. Chem. Soc. 131 (2009) 889. [12] Z. WuR. Jin, Nano Lett. 10 (2010) 2573. [13] Y.L. Liu, K.L. Ai, X.L. Cheng, L.H. Huo, L.H. Lu, Adv. Funct. Mater. 20 (2010) 956. [14] N. Goswami, A. Giri, M.S. Bootharaju, P.L. Xavier, T. Pradeep, S.K. Pal, Anal. Chem. 83 (2011) 9680. [15] Y.-T. Su, G.-Y. Lan, W.-Y. Chen, H.-T. Chang, Anal. Chem. 82 (2010) 8572. [16] C.-Y. Yang, C.-L. Li, C.-J. Lu, Anal. Chim. Acta 565 (2006) 26. [17] J.T. Petty, S.P. Story, S. Juarez, S.S. Votto, A.G. Herbst, N.N. Degtyareva, B. Sengupta, Anal. Chem. 84 (2011) 364. [18] H.-C. Yeh, J. Sharma, I.-M. Shih, D.M. Vu, J.S. Martinez, J.H. Werner, J. Am. Chem. Soc. 134 (2012) 11558. [19] Q. Cui, K. Ma, Y. Shao, S. Xu, F. Wu, G. Liu, N. Teramae, H. Bao, Anal. Chim. Acta 724 (2012) 91.

86

X. Xia et al. / Analytica Chimica Acta 772 (2013) 81–86

[20] J. Yuan, W. Guo, E. Wang, Anal. Chim. Acta 706 (2011) 342. [21] J. Zheng, J.T. Petty, R.M. Dickson, J. Am. Chem. Soc. 125 (2003) 7781. [22] Z. Tang, B. Xu, B. Wu, M.W. Germann, G. Wang, J. Am. Chem. Soc. 132 (2010) 3374. [23] C. GuoJ. Irudayaraj, Anal. Chem. 83 (2011) 2889. [24] F. Wen, Y. Dong, L. Feng, S. Wang, S. Zhang, X. Zhang, Anal. Chem. 83 (2011) 1196. [25] M.A. Habeeb Muhammed, P.K. Verma, S.K. Pal, A. Retnakumari, M. Koyakutty, S. Nair, T. Pradeep, Chem. Eur. J. 16 (2010) 10112. [26] C.-A.J. Lin, T.-Y. Yang, C.-H. Lee, S.H. Huang, R.A. Sperling, M. Zanella, J.K. Li, J.-L. Shen, H.-H. Wang, H.-I. Yeh, W.J. Parak, W.H. Chang, ACS Nano 3 (2009) 401. [27] C.-C. Huang, C.-T. Chen, Y.-C. Shiang, Z.-H. Lin, H.-T. Chang, Anal. Chem. 81 (2009) 882. [28] H. Duan, S. Nie, J. Am. Chem. Soc. 129 (2007) 2413. [29] P. Wu, Y. He, H.-F. Wang, X.-P. Yan, Anal. Chem. 82 (2010) 1433. [30] X. Li, Y. Zhou, Z. Zheng, X. Yue, Z. Dai, S. Liu, Z. Tang, Langmuir 25 (2009) 6586. [31] W. Luo, C. Zhu, S. Su, D. Li, Y. He, Q. Huang, C. Fan, ACS Nano 4 (2010) 7458. [32] J.T. Holland, C. Lau, S. Brozik, P. Atanassov, S. Banta, J. Am. Chem. Soc. 133 (2011) 19265. [33] L. Hu, Y. Yuan, L. Zhang, J. Zhao, S. Majeed, G. Xu, Anal. Chim. Acta 762 (2013) 86. [34] J. Luo, S. Jiang, H. Zhang, J. Jiang, X. Liu, Anal. Chim. Acta 709 (2012) 53. [35] W. Wang, L. Bao, J. Lei, W. Tu, H. Ju, Anal. Chim. Acta 744 (2012) 38. [36] T. Wen, F. Qu, N.B. Li, H.Q. Luo, Anal. Chim. Acta 749 (2012) 62.

[37] C. Radhakumary, K. Sreenivasan, Anal. Chem. 83 (2011) 2833. [38] L. Jin, L. Shang, S. Guo, Y. Fang, D. Wen, L. Wang, J. Yin, S. Dong, Biosens. Bioelectron. 26 (2011) 1969. [39] Y.-C. Shiang, C.-C. Huang, H.-T. Chang, Chem. Commun. (2009) 3439. [40] D.G. Duff, A. Baiker, P.P. Edwards, Langmuir 9 (1993) 2309. [41] C.-L. Liu, H.-T. Wu, Y.-H. Hsiao, C.-W. Lai, C.-W. Shih, Y.-K. Peng, K.-C. Tang, H.W. Chang, Y.-C. Chien, J.-K. Hsiao, J.-T. Cheng, P.-T. Chou, Angew. Chem. Int. Ed. 50 (2011) 7060. [42] X. Yuan, Z. Luo, Q. Zhang, X. Zhang, Y. Zheng, J.Y. Lee, J. Xie, ACS Nano 5 (2011) 8808. [43] T. Udaya Bhaskara Rao, T. Pradeep, Angew. Chem. Int. Ed. 49 (2010) 3929. [44] R. Zhou, M. Shi, X. Chen, M. Wang, H. Chen, Chem. Eur. J. 15 (2009) 4951. [45] E.S. ShibuT. Pradeep, Chem. Mater. 23 (2011) 999. [46] J.M. Pettibone, J.W. Hudgens, ACS Nano 5 (2011) 3002. [47] I. Diez, R.H.A. Ras, Nanoscale 3 (2011) 1970. [48] J.-a. Annie Ho, H.-C. Chang, W.-T. Su, Anal. Chem. 84 (2012) 3253. [49] P.-H. Li, J.-Y. Lin, C.-T. Chen, W.-R. Ciou, P.-H. Chan, L. Luo, H.-Y. Hsu, E.W.-G. Diau, Y.-C. Chen, Anal. Chem. 84 (2012) 5488. [50] M. Zayats, R. Baron, I. Popov, I. Willner, Nano Lett. 5 (2004) 25. [51] H. Pan, R. Cui, J.-J. Zhu, J. Phys. Chem. B 112 (2008) 16901. [52] Y. Negishi, K. Nobusada, T. Tsukuda, J. Am. Chem. Soc. 127 (2005) 5270. [53] W.-Y. Chen, G.-Y. Lan, H.-T. Chang, Anal. Chem. 83 (2011) 9455. [54] C. Vericat, M.E. Vela, G. Benitez, P. Carro, R.C. Salvarezza, Chem. Soc. Rev. 39 (2010) 1834.