Graphene-promoted 3,4,9,10-perylenetetracarboxylic acid nanocomposite as redox probe in label-free electrochemical aptasensor

Biosensors and Bioelectronics 30 (2011) 123–127

Contents lists available at SciVerse ScienceDirect

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

Graphene-promoted 3,4,9,10-perylenetetracarboxylic acid nanocomposite as redox probe in label-free electrochemical aptasensor Yali Yuan a , Xuxu Gou b , Ruo Yuan a,∗ , Yaqin Chai a , Ying Zhuo a , Xiaoya Ye c , Xianxue Gan a a

Key Laboratory on Luminescence and Real-Time Analysis, Ministry of Education, College of Chemistry and Chemical Engineering, Southwest University, Chongqing 400715, PR China Department of Medical Lab Science, Zhongshan School of Medicine, Sun Yat-Sen University, Guangzhou 510080, PR China c Southwest University, Chongqing 400715, PR China b

a r t i c l e

i n f o

Article history: Received 21 July 2011 Received in revised form 27 August 2011 Accepted 29 August 2011 Available online 6 September 2011 Keywords: Graphene 3,4,9,10-Perylenetetracarboxylic acid (PTCA) Redox-activity ␲–␲ stacking Electrochemical aptasensor

a b s t r a c t Graphene/3,4,9,10-perylenetetracarboxylic acid (GPD) with three-dimensional porous structure has been successfully synthesized and served as redox probe to construct ultrasensitive electrochemical aptasensor. The GPD nanocomposite shows promoted electrochemical redox-activity of 3,4,9,10perylenetetracarboxylic acid (PTCA) with an obvious well-defined cathodic peak from −0.7 to 0 V that never been seen from graphene or PTCA, which avoids miscellaneous redox peaks of PTCA in electrochemical characterization, offering a novel redox probe for electrochemical sensors with highly electrochemical active area and conductivity. To the best of our knowledge, this is the first study that utilizes PTCA self-derived redox-activity as redox probe in electrochemical sensors. Moreover, the interesting GPD possesses the advantages of membrane-forming property, providing a direct immobilization of redox probes on electrode surface. This simple process not only diminishes the conventional fussy immobilization of redox probes on the electrode surface, but also reduces the participation of the membrane materials that acted as a barrier of the electron propagation in redox probe immobilization. With thrombin as a model target, the redox probe-GPD based label-free electrochemical aptasensor shows a much higher sensitivity (a detection range from 0.001 nM to 40 nM with a detection limit of 200 fM) to that of analogous aptasensors produced from other redox probes. © 2011 Elsevier B.V. All rights reserved.

1. Introduction 3,4,9,10-Perylenetetracarboxylic acid (PTCA), an archetypal ␲stacking organic perylene dye with outstanding photo and chemical stability as well as desirable organic electronic and optical properties, has been widely recognized as one of the most promising and rapidly emerging research areas for advanced materials (Eremtchenko et al., 2003; Forrest, 1997; Jones et al., 2004; Gregg and Cormier, 2001; Li et al., 2004a). Nevertheless, to the best of our knowledge, few researches have been reported on the application of PTCA-based nanomaterials in electrochemical sensors, besides the exploration of PTCA self-derived redox-activity. In fact, the perylene dye indeed has redox-activity owing to the electronic property, however, there would be miscellaneous redox peaks in electrochemical characterization (e.g. cyclic voltammograms) (Asir et al., 2010), which largely limits its application of redoxactivity in electrochemical sensors. The conventional methods to fabricate PTCA based materials for electrochemical sensors are involved in the provision of PTCA just as semiconducting template, for instance, the porous organometallic materials synthesized for

∗ Corresponding author. Tel.: +86 23 68252277; fax: +86 23 68252277. E-mail address: [email protected] (R. Yuan). 0956-5663/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.bios.2011.08.041

immunosensor in our previous work (Zhuo et al., 2008). In this context, it is significant to improve the property of PTCA with efficient electrochemical redox-activity in the absence of other redox groups for electrochemical sensor. Graphene, an atomical layer of sp2 carbon atoms in a densely packed honeycomb two-dimensional lattice, has recently attracted great attention from both the experimental and theoretical scientific communities (Novoselov et al., 2004; Yang et al., 2010b; Tang et al., 2010). Due to its unique properties, such as excellent electronic transport properties, exceptional electrical conductivity, good fracture strength and high aspect ratio properties, graphene has been substantially applied in the areas of electrochemical sensors (Shan et al., 2009; Du et al., 2010a; Yang et al., 2010a; Dong et al., 2010; Zeng et al., 2010). In particular, owing to the fact that nanocomposites can exploit the novel properties of parent constituents, producing a desired material with improved performance, the fabrication of graphene-based nanocomposites is of great interest and importance. Zhang et al. (2010b) for example, developed biocompatible graphene-single stranded DNA nanocomposite and first used as an electrode material for the immobilization and biosensing of redox enzymes. Wang et al. (2010) displayed nitrogen-doped graphene with high electrocatalytic activity for reduction of hydrogen peroxide and fast direct electron transfer kinetics for glucose oxidase. Choi et al.

124

Y. Yuan et al. / Biosensors and Bioelectronics 30 (2011) 123–127

Fig. 1. A schematic diagram of the procedure used to prepare stable GPD nanocomposite.

(2010) reported graphene–nafion nanohybrid films with highperformance for electrochemical biosensing. Although remarkable progress has been made in the fabrication of graphene-based nanocomposites, for electrochemical sensor, utilizing the completely novel functional properties that were produced from the synergistic effect of graphene and organic material has not been reported. In the present study, we proposed a graphene promoted 3,4,9,10-perylenetetracarboxylic acid nanocomposite (GPD) with three-dimensional porous structure. The planar graphene and perylene dye face-to-face formed a larger delocalized area through ␲–␲ stacking, increased the resonance energy and delocalized the charge effectively (Lee et al., 1999). Significantly, the resulting nanocomposite showed enhanced electrochemical activity of PTCA with an obvious well-defined cathodic peak between −0.7 and 0 V that never been seen from the parent constituents (graphene or PTCA) in cyclic voltammetry, which provided an novel redox probe for electrochemical sensors with high electrochemical active area and conductivity. With thrombin aptamer and thrombin as model systems, a redox probe-label-free electrochemical aptasensor composed of GPD as platform and redox probe was constructed. The SEM, XPS, UV–vis, electrochemical characterization, reduction mechanism of the GPD and the performance of the resulted aptasensor are discussed as follows.

2. Experimental 2.1. Materials and reagents 3,4,9,10-Perylenetetracarboxylic acid (PTCA) was purchased from Lian Gang Dyestuff Chemical Industry Co. Ltd. (Liaoning, China). Graphene oxide was obtained from Nanjing xianfeng nano Co. (Nanjing, China). Thrombin, hemin bovine serum albumin (BSA), hemoglobin (Hb), hexanethiol (96%, HT), poly(ethylene imine) (PEI) and gold chloride (HAuCl4 ) were obtained from Sigma Chemical Co. (St. Louis, MO, USA). N-hydroxy succinimide (NHS) and N-(3dimethylaminopropyl)-N -ethylcarbodiimidehydrochloride (EDC) were acquired from Shanghai Medpep Co. (Shanghai, China). Trishydroxymethylaminomethane hydrochloride (tris) was purchased from Roche (Switzerland). Thrombin aptamer (TBA): 5 -(CH2)6 GGT TGG TGT GGT TGG-3 was purchased from TaKaRa (Dalian. China). All other chemicals were of reagent grade and used as received. Double distilled water was used throughout this study.

2.2. Apparatus The UV–vis absorption was recorded with an UV-3600 UV–vis–near infrared (NIR) spectrophotometer (Shimadzu, Japan). The scanning electron micrographs were taken with scanning electron microscope (SEM, S-4800, Hitachi). X-ray photoelectron spectroscopy (XPS) measurements were performed with a VG Scientific ESCALAB 250 spectrometer, using Al Ka X-ray (1486.6 eV) as the light source. All electrochemical measurements were performed on a CHI 660C electrochemical workstation (Shanghai Chenhua Instrument, China) with a conventional three electrode system composed of platinum wire as auxiliary, saturated calomel electrode as reference, and a modified gold electrode as working electrode. 2.3. Preparation of graphene/PTCA hybrid nanocomposite (GPD) The overall process involved in fabricating the GPD is shown schematically in Fig. 1. Briefly, reduced-graphene oxide sheets were firstly prepared according to literature with little modification (Cao et al., 2010). A stable dispersion of exfoliated GO sheets (1 mg/mL, 10 mL) was mixed with PEI (3%, 1 mL) and heated under reflux at 135 ◦ C for about 3 h. The PEI here served as reductant to reduce graphene oxide sheets and provided abundant amino for further association of PTCA. The final black product (PEI-rGO) was collected by centrifugation. Subsequently, EDC and NHS were used as coupling agents which catalyze the formation of amide bond between the carboxyl of PTCA and the amino of PEI-rGO. After centrifugation at 12,000 rpm for 15 min, the sediment of resulted GPD was resuspended in double distilled water and stored at 4 ◦ C for further use. 2.4. Fabrication of the electrochemical aptasensor The GPD here was served as a redox probe to construct an electrochemical aptasensor. Thrombin, a serine protease that has many effects in coagulation cascade (Rahman et al., 2009), was chosen as model analyte in this work. The stepwise self-assemble procedure of electrochemical aptasensor is shown in Fig. S1 (see Supplementary material). A gold electrode was carefully polished with 1.0 and 0.3 ␮m alumina powders separately, after rinsing with distilled water the electrode was dried in a nitrogen stream before use. Then, 3 ␮L as-prepared GPD was coated on the electrode surface and dried at room temperature. Following that, another electrodeposited nano-Au layer was employed for the immobilization of thrombin

Y. Yuan et al. / Biosensors and Bioelectronics 30 (2011) 123–127

125

Fig. 2. XPS (A), UV–vis spectra (B) and SEMs (C) of PEI-rGO (a), PTCA (b) and GPD (c).

aptamer. After the drop of 20 ␮L thrombin aptamer (2.5 ␮M) on electrode surface, the electrode was incubated for 16 h at moist environment. Finally, 20 ␮L of 1.0 mM hexanethiol was used for eliminating nonspecific binding effects and obtaining an optimal nucleic acid orientation.

well porous structure. After the formation of GPD nanomaterials, the porous structure and quadrateshaped molecular islands are reserved. However, the surface of quadrateshaped perylene nanomaterials became rough, which suggesting the PEI-rGO nanosheet matrix was highly coated on the PTCA surface (Fig. 2C-c).

3. Results and discussion

3.2. Electrochemical redox behavior of GPD

3.1. Characteristics of the GPD by XPS, UV–vis and SEM

In order to demonstrate that the interaction between PTCA and PEI-rGO would bring in a novel redox probe, PTCA, PEI-rGO and as-prepared GPD were immobilized on the gold electrode surfaces and subjected to cyclic voltammetry analysis (Fig. 3A). No obvious peaks are observed at PEI-rGO modified electrode (curve a). Due to the photochemical property of PTCA, there are three welldefined cathodic peaks [three one-electron (1e) transfers] (curve b). However, interestingly, for the GPD, the spacing among the three successive 1e cathodic peaks decreases until a single analogous 3e peak results (curve c) and the current is largely increased compared to equivalent PTCA. Three reasons may be attributed to the interesting phenomenon: firstly, PTCA polycrystalline films exhibit substantial anisotropies in its electronic transport properties. It transports holes perpendicular to graphene, while graphene transports electrons to PTCA (Ostrick et al., 1997). Secondly, the planar structure of PEI-rGO and PTCA could form a larger delocalized area face-to-face through ␲–␲ stacking, which increases the resonance energy and delocalizes the charge effectively. Lastly, ␲ stacking of black perylene dye polycrystalline shows a lager overlapped area, providing an avail for electron transport (Li et al., 2004b). The delocalized ␲ electrons then shuttle quickly and freely in the stacking system and finally bring in a remarkable synergistic action. Additionally, owing to the strong electron-accepting nature of the carboximide, the first 1e reduction potential of GPD (−0.537 V) is less negative than that of PTCA (−0.647 V), showing that the formation of carboximide makes the molecule much easier to be reduced. Additionally, owing to the strong electron-accepting

XPS analysis provides effective information on the chemical composition of the as-prepared GPD. As shown in Fig. 2A, the appearance of an N peak in the XPS survey spectrum for GPD was analogous to that of PEI-rGO, indicates the possible combination of PEI-rGO with PTCA. The surface nitrogen in PEI-rGO was estimated to be 13.06 at.%, whereas the percentage of nitrogen decreased to 7.32 at.% after treatment with PTCA, to some extent, suggesting some of the PTCA have been linked to the PEI-rGO. The UV–vis absorption spectra of PEI-rGO, PTCA and GPD were compared in order to further verify the interaction between PTCA and PEI-rGO. As can been seen from Fig. 2B, PEI-rGO shows an optical absorption spectrum of 262 nm, which is very similar to that of reported rGO (a) (Zhang et al., 2010a; Kong et al., 2009). In the meantime, the optical absorption spectrum of PTCA in double distilled water contains a characteristic maximum at 512 nm (attributed to the perylene core ␲–␲* transition) as well as a series of weak maxima at 228 and 254 nm (b). A new absorption band appear at 258 nm with the disappearance of weak maxima at 228, 254 and 262 nm after the formation of GPD, indicating that ground state interaction occurs successfully between PTCA and PEI-rGO (c). The morphology and microstructure of as-prepared nanomaterials were investigated by scanning electron microscopy (SEM) (Fig. 2C). Fig. 2C-a displays a nanosheet PEI-rGO structure with morphology similar to that of graphene sheet. In Fig. 2C-b, PTCA film grows in irregularly quadrateshaped molecular islands and exhibits

126

Y. Yuan et al. / Biosensors and Bioelectronics 30 (2011) 123–127

Fig. 3. (A) Cyclic voltammograms of (a) PEI-rGO, (b) PTCA and (c) GPD modified electrodes investigated in 0.10 M PBS (pH 7.4). And GPD modified electrode in PBS (pH 7.4) at various scan ranges: (B) negative potential was unaltered and (C) positive potential was unaltered. The scan rates were 50 mV/s.

Fig. 4. (A) Cyclic voltammograms in 0.10 M PBS (pH 7.4) using (a) bare gold electrode; (b) GPD film electrode; (c) the nano-Au electrodeposited GPD film electrode; (d) the electrode assembled with thrombin aptamer; (e) the modified electrode blocked with hexanethiol; (f) the as-prepared aptasensor incubated in 5 nM thrombin. The scan rates were 50 mV/s. (B) Cyclic voltammograms of proposed aptasensor in 0.10 M PBS (pH 7.4) at different scan rates. From above down: 10, 20, 40, 50, 80, 100, 120, 150 and 200 mV/s. (C) The relationships between the cathodic peak currents and the scan rate.

nature of the carboximide, the first 1e reduction potential of GPD (−0.537 V) is less negative than that of PTCA (−0.647 V), showing that the formation of carboximide makes the molecule much easier to be reduced. The proposed reduction mechanism for the GPD systems is shown in Fig. S2 (see Supplementary material). An electrochemical active compound can exhibit better redox property under suitable potential. Therefore, the GPD modified electrode was scanned under different potential area which clearly displayed in Fig. 3B and C. It is reasonable to assume that the redox reaction of GPD occurs in the potential area equivalent to or larger than −0.7 to 0 V due to the well-defined reversible peaks as indicated in possible resonance forms in Fig. 3B and C. 3.3. Characteristics of GPD-based electrochemical aptasensor With GPD as platform and redox probe, a redox probe-label-free electrochemical aptasensor was constructed. As shown in Fig. 4A, the cyclic voltammogram of bare gold electrode did not show any detectable signal in pH 7.4 PBS (curve a). After the immobilization of GPD, the electrode exhibited a stable and well-defined cathodic peak at −0.537 V (curve b), which correspond to the electrochemical behavior of GPD. In order to further immobilize thio-determined thrombin aptamer, a nano-Au layer was electrodeposited on the modified electrode, leading to an increase of response currents (curve c). It is reasonable that aptamers assembled on the electrode can block the electron transfer of the redox probe. Therefore, an obvious decrease of current was obtained after the self-assembly of thio-determined thrombin aptamer on modified electrode (curve d). Subsequently, HT was served as blocking reagent to possible remaining active sites on the electrode surface and a decreased peak current was acquired (curve e). Then the proposed electrochemical aptasensor was incubated in 5 nM thrombin for 40 min, the bulky thrombin molecules attached onto the electrode surface and blocked the electrode surface with a decreased current (curve f). The CVs of the GPD-based electrochemical aptasensor in PBS at different scan rates are used to investigate the redox behavior of GPD. The cathodic current increases with the increase of scan

rate and exhibits a peak-to-peak separation from 10 to 200 mV/s (Fig. 4B), implied the facile charge transfer kinetics in this range of scan rates. As expected for a diffusion-controlled redox reaction, the cathodic current exhibits a linear relationship with the square root of sweep rate (Fig. 4C). 3.4. Calibration curves of electrochemical aptasensor The as prepared electrochemical aptasensor was subjected to thrombin incubation with various concentrations. From the inset in Fig. 5, it can be seen that the cathodic peak decreased with the increase of thrombin concentration. The reason for that could ascribe to the bulky thrombin molecules which attached onto the electrode surface make a barrier for electrons and inhibit the electro-transfer significantly. Fig. 5 displays the corresponding calibration plots, the cathodic peak currents were proportional to thrombin concentration over the range from 0.001 to 40 nM. The linear equation was I = 31.73 lg Cthrombin − 304.2 with a correlation

Fig. 5. Calibration plots of the cathodic peak current response vs. concentration of thrombin. The inset shows the cyclic voltammograms of the cathodic peak with different concentrations (from the top down: 0, 0.001, 0.005, 0.01, 0.1, 0.5, 5, 20, and 40 nM).

Y. Yuan et al. / Biosensors and Bioelectronics 30 (2011) 123–127

coefficient of 0.991 and a detection limit of 200 fM (defined as 3, where  is the standard deviation of the blank). The value of detection limit was much lower than the detection limits obtained from analogous electrochemical thrombin detections with fussy layerby-layer technique where thionine (40 pM) (Yuan et al., 2010) and ferrocene (4 pM) (Du et al., 2010b) were served as redox probes. The GPD as redox probe in electrochemical sensors not only diminished the conventional fussy immobilization of redox probes on the electrode surface but also reduced the participation of the membrane materials that acted as a barrier of the electron propagation in redox probe immobilization, resulted a higher sensitivity of proposed aptasensors.

127

Acknowledgements This work was financially supported by the Postgraduate Science and Technology Innovation Program of Southwest China University (Grant No. KB2010006), the NNSF of China (21075100), State Key Laboratory of Electroanalytical Chemistry (SKLEAC2010009), Ministry of Education of China (Project 708073), Natural Science Foundation of Chongqing City (CSTC-2009BA1003, CSTC-2010BB4121), Specialized Research Fund for the Doctoral Program of Higher Education (20100182110015) and High Technology Project Foundation of Southwest University (XSGX02). Appendix A. Supplementary data

3.5. Reproducibility, stability and specificity of the GPD-based aptasensor Five freshly prepared aptasensors were used for the detection of 5 nM thrombin. All electrodes exhibited approximate electrochemical response and the relative standard deviation was 3.8%, implying acceptable reproducibility of the proposed aptasensor. The stability of the successive assays was investigated by 100 cycles CV measurements in PBS after being incubated in 5 nM thrombin and decreased only 3.79% of its initial current. When the aptasensor was stored at 4 ◦ C, it retained 91.2% of its previous response after a storage period of 15 days. The above experimental data effectively demonstrated that the GPD films could be directly assembled on bare gold electrode surface with satisfying stability. The specificity of the aptasensor was also examined after the incubation of thrombin and three interfering agents, such as BSA, hemoglobin (HB), and IgG. As a result, thrombin (5 nM) showed larger decreased cathodic peak while others had almost negligible electrochemical changes, suggesting that the aptasensor was specific to thrombin and possesses high selectivity. 4. Conclusion In summary, with ␲–␲ stacking and bonding action between graphene and PTCA, we successfully improved the property of PTCA with efficient electrochemical redox-activity in the absence of other redox groups. The obtained GPD nanocomplexes showed an obvious well-defined cathodic peak from −0.7 to 0 V that never been seen from graphene or PTCA, which offered a novel redox probe for electrochemical sensors with high electrochemical active area and conductivity. This work opens a novelty application of PTCA derived redox-activity as redox probe in electrochemical sensors. Additionally, the direct membrane-forming property of redox probes GPD on electrode surface not only diminished the conventional fussy immobilization of redox probes, but also reduced the participation of the membrane materials that acted as a barrier of the electron propagation in redox probe immobilization, which provided ideal platform and redox probe in electrochemical sensors.

Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.bios.2011.08.041. References Asir, S., Demir, A.S., Icil, H., 2010. Dyes Pigments 84, 1–13. Cao, L.Y., Liu, Y.L., Zhang, B.H., Lu, L.H., 2010. ACS Appl. Mater. Interfaces 2, 2339–2346. Choi, B.G., Park, H., Park, T.J., Yang, M.H., Kim, J.S., Jang, S.Y., Heo, N.S., Lee, S.Y., Kong, J., Hong, W.H., 2010. ACS Nano 4, 2910–2918. Dong, X.C., Shi, Y.M., Huang, W., Chen, P., Li, L.J., 2010. Adv. Mater. 22, 1649–1653. Du, D., Zou, Z.X., Shin, Y.S., Wang, J., Wu, H., Engelhard, H., Liu, J., Aksay, I.A., Lin, Y.H., 2010a. Anal. Chem. 82, 2989–2995. Du, Y., Chen, C.G., Li, B.L., Zhou, M., Wang, E.K., Dong, S.J., 2010b. Biosens. Bioelectron. 25, 1902–1907. Eremtchenko, M., Schaefer, J.A., Tautz, F.S., 2003. Nature 425, 602–605. Forrest, S.R., 1997. Chem. Rev. 97, 1793–1796. Gregg, B.A., Cormier, R.A., 2001. J. Am. Chem. Soc. 23, 7959. Jones, B.A., Ahrens, M.J., Yoon, M.H., Facchetti, A., Marks, T.J., Wasielewski, M.R., 2004. Angew. Chem. Int. Ed. 43, 6363–6366. Kong, B.S., Geng, J.X., Jung, H.T., 2009. Commun. Chem. 16, 2174–2176. Lee, S.K., Zu, Y.B., Herrmann, A., Geerts, Y., Mullen, K., Bard, A.J., 1999. J. Am. Chem. Soc. 121, 3513–3520. Li, X., Sinks, L.E., Rybtchinski, B., Wasielewski, M.R., 2004a. J. Am. Chem. Soc. 126, 10810–108011. Li, X.G., Huang, M.R., Li, S.X., 2004b. Acta Mater. 52, 5363–5374. 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. Ostrick, J.R., Dodabalapur, A., Torsi, L., Lovinger, A.J., Kwock, E.W., Miller, T.M., Galvin, M., Berggren, M., Katz, H.E., 1997. J. Appl. Phys. 81, 6804–6808. Rahman, M.A., Son, J.I., Won, M.S., Shim, Y.B., 2009. Anal. Chem. 81, 6604–6611. Shan, C.S., Yang, H.F., Han, D.X., Zhang, Q.X., Ivaska, A., Niu, L., 2009. Langmuir 25, 12030–12033. Tang, Z.H., Shen, S.L., Zhuang, J., Wang, X., 2010. Angew. Chem. Int. Ed. 49, 4603–4607. Wang, Y., Shao, Y.Y., Matson, D.W., Li, J.H., Lin, Y.H., 2010. ACS Nano 4, 1790–1798. Yang, M.H., Javadi, A., Li, H., Gong, S.Q., 2010a. Biosens. Bioelectron. 26, 560–565. Yang, S.B., Feng, X.L., Wang, L., Tang, K., Maier, J., Mullen, K., 2010b. Angew. Chem. Int. Ed. 49, 4795–4799. Yuan, Y.L., Yuan, R., Chai, Y.Q., Zhuo, Y., Liu, Z.Y., Mao, L., Guan, S., Qian, X.Q., 2010. Anal. Chim. Acta 668, 171–176. Zeng, Q., Cheng, J.S., Tang, L.H., Liu, X.F., Liu., Y.Z., Li, J.H., Jiang, J.H., 2010. Adv. Funct. Mater. 20, 3366–3372. Zhang, J.L., Yang, H.J., Shen, G.X., Cheng, P., Zhang, J.Y., Guo, S.W., 2010a. Chem. Commun. 46, 1112–1114. Zhang, Q., Qiao, Y., Hao, F., Zhang, L., Wu, S.Y., Li, Y., Li, J.H., Song, X.M., 2010b. Chem. Eur. J. 16, 8133–8139. Zhuo, Y., Yuan, P.X., Yuan, R., Chai, Y.Q., Hong, C.L., 2008. Biomaterials 29, 1501–1508.