Biosensors and Bioelectronics 61 (2014) 9–13
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
A novel ultrasensitive carboxymethyl chitosan-quantum dot-based fluorescence “turn on–off” nanosensor for lysozyme detection Yu Song a, Yang Li a, Ziping Liu a, Linlin Liu a, Xinyan Wang b, Xingguang Su a, Qiang Ma a,n a b
Department of Analytical Chemistry, College of Chemistry, Jilin University, Changchun 130012, China Changchun Institute of Applied Chemistry Chinese Academy of Sciences, Changchun 230022, China
art ic l e i nf o
a b s t r a c t
Article history: Received 16 January 2014 Received in revised form 18 April 2014 Accepted 21 April 2014 Available online 30 April 2014
In this work, we developed an ultrasensitive “turn on–off” fluorescence nanosensor for lysozyme (Lyz) detection. The novel nanosensor was constructed with the carboxymethyl chitosan modified CdTe quantum dots (CMCS-QDs). Firstly, the CMCS-QDs were fabricated via the electrostatic interaction between amino groups in CMCS polymeric chains and carboxyl groups on the surface of QDs. In the fluorescence “turn-on” step, the strong binding ability between Zn2 þ and CMCS on the surface of QDs can enhance the photoluminescence intensity (PL) of QDs. In the following fluorescence “turn-off” step, the N-acetyl-glucosamine (NAG) section along the CMCS chains was hydrolyzed by Lyz. As a result, Zn2 þ was released from the surface of QDs, and the Lyz–QDs complexes were formed to quench the QDs PL. Under the optimal conditions, there was a good linear relationship between the PL of QDs and the Lyz concentration (0.1–1.2 ng/mL) with the detection limit of 0.031 ng/mL. The developed method was ultrasensitive, highly selective and fast. It has been successfully employed in the detection of Lyz in the serum with satisfactory results. & 2014 Elsevier B.V. All rights reserved.
Keywords: Nanosensor Quantum dots Lysozyme Fluorescence turn on–off
1. Introduction As a kind of ideal nanomaterial, luminescent quantum dots (QDs) have attracted great attention in the past decades due to their unique properties (Hanif et al., 2002; Pradhan et al., 2005; Yang et al., 2006), including narrow, symmetrical and size-tunable emission spectrum, broad excitation spectrum and so on. QDs have been employed widely in biosensing and biolabeling applications (Wu et al., 2008) to detect multifarious substances (e.g. small molecules and proteins) (Wang et al., 2013). QD-based nanosensors are becoming the most important nanomaterials for biological and medical applications (Vannoy et al., 2010). However, there are still some concerned problems with how to prepare QD-based sensing systems for high sensitive and selective analysis. Therefore, various techniques have been employed in the assembled strategies of QDs-based sensors. These assembled strategies not only effected the efficiency and stability of sensors, but also determined the cost and practicability of application. Currently, more and more surface modified reagents have been employed to render diverse affinity and specificity of QDs towards different targets. As an ideal functional material, the natural polysaccharide—carboxymethyl
n
Corresponding author. Tel.: þ 86 431 85168352. E-mail address: [email protected] (Q. Ma).
http://dx.doi.org/10.1016/j.bios.2014.04.036 0956-5663/& 2014 Elsevier B.V. All rights reserved.
chitosan (CMCS) which composed of N-acetyl-glucosamine (NAG) have many attractive physical and biological properties such as hydrophilicity, good biocompatibility, biodegradability, low toxicity, and remarkable affinity for metal ions (Gaberc and Menart , 2001; Lee et al., 2009; Zhou et al., 2006). These properties made CMCS become a promising modification material (Chen et al., 2002; Zhu and Fang, 2005; Chenn et al., 2006). The amino (– NH2), carboxyl (–COOH) and hydroxyl (–OH) groups on CMCS chains can serve as electrostatic interaction and coordination sites (Zeng and Ruckenstein 1999). So, CMCS had strong binding ability towards metal ions, especially with Zn2 þ (Shen et al., 2012; Upadhyaya et al., 2013). Furthermore, we found a strong photoluminescence-activation effect after Zn2 þ binding with CMCS-QDs (Ma and Lin et al., 2014). Lyz (Mw: 14.3 kDa) is a relatively small single chain protein with only 129 amino acids. Lyz can only recognize the NAG section and hydrolyze the similar glycan backbone, such as chitin, partially deacetylated chitosan (Kurita et al., 2000, Amano and Ito, 1978). As a tool enzyme, Lyz has been used in gene and cell engineering, which exists in body tissues and secretions (e.g. serum, urine and tears). Because the increased Lyz concentration in urine and serum are potential indicators for leukemia and meningitis (Klockars et al., 1978), the reliable and sensitive methods for the analysis of Lyz are required. To date, a variety of strategies for the Lyz detection have been reported, such as surface plasmon resonance (SPR)
10
Y. Song et al. / Biosensors and Bioelectronics 61 (2014) 9–13
the excessive CMCS. The collected CMCS-QDs were washed with ethanol/water and stored with PBS (pH 7.4) in dark. The final concentration was 3.56 nM QDs/0.05 g CMCS. 2.3. Preparation of Zn2 þ –CMCS-QDs Scheme 1. The schematic illustration of the novel Zn2 þ –CMCS-QDs nanosensor for Lyz detection.
(Vasilescu et al., 2013), Resonance Rayleigh-scattering (Cai et al., 2011), quartz crystal microbalance (QCM) (Senera et al., 2010), fluorescence anisotropic sensing (Zou et al., 2012), colorimetric & fluorometric dual-readout sensor (Zheng et al., 2013) and so on. However, many methods suffered from some drawbacks including interfering substances, unsatisfactory detection limit and time consuming. In present study, we prepared CMCS-QDs to establish a fluorescence “turn on–off” sensing system, as shown in Scheme 1. Firstly, CMCS-QDs can absorb Zn2 þ to enhance the PL of QDs in the fluorescence “turn on” step. In the next fluorescence “turn off” step, Lyz was added into Zn2 þ –CMCS-QDs system. CMCS was degraded quickly to small molecule by hydrolyzing of Lyz, which can release Zn2 þ from the QDs surface and form Lyz–QDs complexes. As a result, the QDs PL was quenched. The established fluorescence “turn on–off” sensing system has high selectivity and ultrasensitivity for Lyz detection, and this sensing system has been successfully applied to detect Lyz in the human serum samples with good precision and accuracy.
In the fluorescence “turn on” step, 1 mL CMCS-QDs (3.56 nM QDs/0.05 g CMCS) were mixed with a series of different concentrations of Zn2 þ solution. In this process, all measurements were performed under the same condition. The slit widths of the excitation and emission were 5 nm and 10 nm. The excitation wavelength was set at 380 nm. And the fluorescence spectra were recorded in the 500–700 nm emission wavelength range. 2.4. Lyz detection In the fluorescence “turn off” step, a series of different concentrations of Lyz and the Zn2 þ –CMCS-QDs complexes in phosphate-buffered saline solutions (PBS, 2 mM, pH 7.4) were mixed in 2.0 mL calibrated test tubes. Subsequently, the mixtures were gently shaken for 30 min at room temperature. The human blood samples were segregated by centrifugation at 10,000 rpm for 10 min after adding CH3COOH in samples (CH3COOH:serum 1.5:1). Finally, all supernatant serum samples were subjected to a 1000-fold dilution with PBS before analysis, and the different concentrations of Lyz were added to prepare the spiked samples.
3. Results and discussion 2. Experiment section
3.1. Spectra characteristics of CMCS-QDs.
2.1. Materials and apparatus
The fluorescence spectra of QDs and CMCS-QDs were shown in Fig. 1A. After coated with CMCS, the slight blue shifts of the QDs PL peaks (from 579 nm to 570 nm) can be observed (Fig. 1A b and d). It was due to the photo-induced surface reconstruction of QDs surface atoms (Ma et al., 2014). Then, the enhancement effect of Zn2 þ on the PL of QDs and CMCS-QDs was studied. It can be seen that the PL of CMCS-QDs was significantly enhanced to about 130% with the increase of Zn2 þ concentration (Fig. 1A d and e). The fluorescence quantum yield of CMCS-QDs and Zn2 þ –CMCS-QDs were 26.31% and 35.02%, respectively. It also proved that Zn2 þ had the ability to enhance the PL of CMCS-QDs. By comparison, the PL of QDs without CMCS did not change (Fig. 1A a and b). It indicated that CMCS played a significant role in enhancing the PL of QDs. CMCS was bound on the surface of QDs via the electrostatic interaction between amino groups in its polymeric chains and carboxyl groups on the QDs surface. The Zn2 þ ions can be captured by CMCS-QDs due to the high affinity of CMCS to Zn2 þ ions. Zn2 þ can prevent the surface nonradiative relaxation of QDs. As a result, the fluorescence signal of CMCS-QDs was “turned on”. With the addition of Lyz, the PL of Zn2 þ –QDs (c) and Zn2 þ –CMCSQDs (f) were “turned off”. The PL quenching effect of Lyz on Zn2 þ – CMCS-QDs (from 130% to 90%) was stronger than that of Zn2 þ – QDs (from 101% to 92%), and there was no significant shift of QDs PL peak after the addition of Lyz. It indicated that Lyz can quench the PL of both Zn2 þ –QDs and Zn2 þ –CMCS-QDs systems. In the Zn2 þ –QDs system, Lyz can bind with QDs to form Lyz–QDs complexes (Wu et al., 2008). The complexes can quench the PL of Zn2 þ –QDs system. In the Zn2 þ –CMCS-QDs system, Lyz can recognize the NAG section along the CMCS chains which act as a hydrolyzing site. So, CMCS was hydrolyzed to small molecules by Lyz, and the captured Zn2 þ was released from the surface of CMCS-QDs firstly. After CMCS was hydrolyzed, the formation of Lyz–QDs complexes can further quench the PL of Zn2 þ –CMCSQDs. From Fig. S1a, the size of QDs (3–5 nm) were well-distributed
3-Mercaptopropyl acid (MPA) (99%) was purchased from J&K Chemical Co.. And tellurium powder ( 200 mesh, 99.8%), CdCl2 (99%) and NaBH4 (99%) were purchased from Aldrich Chemical Co. Carboxymethyl chitosan, NaH2PO4, Na2HPO4, NaCl and ZnCl2 were purchased from Beijing Dingguo Chemicals Co. (Beijing, China). All chemicals used were of analytical grade. The water used in the study had a resistivity higher than 18 MΩ cm. Lyz were purchased from Genview. The human serum was obtained as a gift from the university hospital. The fluorescence measurements were performed on a Shimadzu RF-5301 PC spectrofluorophotometer (Shimadzu Co., Kyoto, Japan) equipped with a xenon lamp using right-angle geometry and a 1 cm path-length quartz cell. All pH measurements were taken with a PHS-3C pH meter (Tuopu Co., Hangzhou, China). 2.2. Synthesis of CMCS-QDs QDs used in our work were synthesized by refluxing routes as the method described in our previous papers (Ma et al., 2011). In brief, CdCl2 solution (1.25 mM) with MPA as stabilizer was added into a 250 mL three-necked flask under N2 atmosphere. NaHTe was produced in an aqueous solution by the reaction of tellurium powder with NaBH4 at a molar ratio of 1:2. Later, the fresh NaHTe was added into the CdCl2 solution. The molar ratio of Cd2 þ /MPA/NaHTe was at 1:1.5:0.2. The solution was subjected to a reflux at 100 1C under open-air conditions to obtain watercompatible MPA-capped QDs. The fluorescence emission wavelength of the QDs used in present experiments was 579 nm. Then, 3.56 nM QDs was added into 10 mL CMCS stock solution (5 g/L). The reaction mixtures were sonicated for 5 min, stirried and vibrated overnight at room temperature in dark. The resultant mixture was transferred into ethanol and centrifuged to remove
Y. Song et al. / Biosensors and Bioelectronics 61 (2014) 9–13
11
Fig. 1. A: The fluorescence spectrum of (a) QDs, (b)Zn2 þ –QDs, (c) Lyz–Zn2 þ -QDs, (d) CMCS-QDs, (e) Zn2 þ –CMCS-QDs, and (f) Lyz–Zn2 þ –CMCS-QDs. (0.1 mM Zn2 þ and 1 mg/L Lyz). B: The photographs of (a) CMCS-QDs (b) Zn2 þ –CMCS-QDs and (c) Lyz–Zn2 þ –CMCS-QDs under UV light.
in solution. After Zn2 þ added (Fig. S1b), the size of Zn2 þ –CMCSQDs was about 10 nm. With the addition of Lyz, there was an obvious aggregation of QD complexes (Fig. S1c). This conclusion was in accord with the former documents reported (Wu et al., 2008; Vannoy et al., 2010). Fig. 1B showed the photographs of CMCS-QDs, Zn2 þ –CMCS-QDs and Lyz–Zn2 þ –CMCS-QDs under UV light, respectively. It can be seen that the luminescence signal of Zn2 þ –CMCS-QDs (b) was brighter than that of CMCS-QDs (a) in the fluorescence “turn on” step. After the addition of Lyz in the system (c), the luminescence signal became weaker than that of CMCS-QDs (a). In order to investigate the selectivity of this novel sensing system, we added Ca2 þ , Mg2 þ or Na þ in the CMCS-QDs and CMCS-QDs–Zn2 þ system. These ions also can chelate with CMCS. From Fig. S2, it can be seen that the PL of CMCS-QDs was easily influenced by Ca2 þ , Mg2 þ and Na þ without Zn2 þ (Fig. S2c and d). But, the PL of CMCS-QDs–Zn2 þ system can keep stable with those ions (Fig. S2a and b). Because Zn2 þ had strong binding ability with CMCS, almost all carboxyl (–COOH) groups on CMCS chains that served as electrostatic interaction and coordination sites can be bound. So other ions can hardly bind with CMCS to influence the PL of QDs. It was obvious that the selectivity and stability of the CMCS-QDs–Zn2 þ sensing system were improved largely.
Fig. 2. The influence of different Zn2 þ concentrations on the PL intensity of CMCSQDs (a–h) 0, 0.03, 0.05, 0.06, 0.08, 0.15, 0.2, set: a plot of normalized PL intensity against the concentration of Zn2 þ .
3.2. Optimization for Lyz detection Firstly, we studied the effect of Zn2 þ concentration on the PL of CMCS-QDs. As shown in Fig. 2, the PL of CMCS-QDs increased with the increasing Zn2 þ concentration. In this fluorescence “turn on” step, the surface nonradiative relaxation centers of CMCS-QDs were passivated efficiently via Zn2 þ binding at a certain concentration. The nonradiative recombination of “filled” surface trap sites was effectively eliminated. When Zn2 þ concentration increased to 0.2 mM, the QDs PL reached maximum and kept stable. So, 0.2 mM Zn2 þ was used in the further experiments. The effect of pH on the PL of Zn2 þ –CMCS-QDs system with Lyz (1 ng/mL) was also studied. It can be seen (Fig. 3) that the system PL was decreased to about 62% at pH 6. The “turn-off” efficiency at pH 6 was more obvious than at other pH circumstances. As a kind of amphoteric proteins, the type (positive or negative) and the number of charges on the Lyz surface can vary with the pH of the medium. The pI of Lyz is 11.2 as reported (Meltem et al., 2005). So, the Lyz molecules can be cationic when pH o11.2 and anionic at pH 411.2. At pH ¼6, the carboxyl groups on the QDs were
Fig. 3. The influence of pH on the PL intensity of Lyz–Zn2 þ –CMCS-QDs (a) pH9.5 (b) pH8 (c) pH4 (d) pH7 (e) pH5 and (f) pH6. Inset: a plot of I/I0 against the pH. I and I0 were the fluorescence intensity of Zn2 þ –CMCS-QDs in the presence and absence of Lyz, respectively.
12
Y. Song et al. / Biosensors and Bioelectronics 61 (2014) 9–13
1.0
0.8
I/I0
k
I/I0
concentration in range of 0.1–1.2 ng/mL. The linear regression equation was I/I0 ¼1.01 0.42 [Lyz] (ng/mL). The corresponding regression coefficient was 0.998. The detection limit (LOD) was 0.031 ng/mL (LOD¼3s/s). s was the slope of the calibration curve. s was the standard deviation of the corrected blank PL signals of the Zn2 þ –CMCS-QDs. Some other reported methods for the determination of Lyz were compared with the proposed method. The comparison was summed up in Table S1. It clearly showed that the analytical performance of this Zn2 þ –CMCS-QDs based nanosensor for Lyz detection was better than most of the reported methods. The novel nanosensor for Lyz detection based on Zn2 þ – CMCS-QDs was ultrasensitive and high selective.
1.0
a
0.6
0.5 0.0 0.2 0.4 0.6 0.8 1.0 1.2
Lysozyme (ng/mL)
0.0
550
600
650
700
Wavelength/nm 2þ
Fig. 4. The PL intensity of Zn –CMCS-QDs with different Lyz concentrations from 0 to 1.2 ng/mL. (a) 0 (b) 0.1 (c) 0.2 (d) 0.3 (e) 0.4 (f) 0.5 (g) 0.6 (h) 0.8 (i) 0.9 (j) 1 and (k) 1.2. The inset showed the relationship between I/I0. I and I0 were the fluorescence intensity of Zn2 þ –CMCS-QDs in the presence and absence of Lyz, respectively.
Table 1 Determination of Lyz in serum samples (n¼ 3). Samples
Found (ng/mL)
Added (ng/mL)
Total found (ng/mL)
Recovery (%)
RSD (%, n ¼3)
1 2 3
0.319 0.453 0.338
0.4 0.5 0.4
0.720 0.943 0.736
100.18 97.97 99.48
0.97 0.92 1.24
negatively charged. So, the weak acid environment promoted that the PL of Zn2 þ –CMCS-QDs system further decreasing. The PL quenching degree was different along with temperature changing. The different reaction temperature in the range of 4–60 1C was studied as shown in Fig. S3. The PL of QDs gradually decreased with the increasing temperature and reached the lowest at 40 1C. When the temperature increased to 60 1C, the PL of QDs increased again. The results indicated that the temperature had a close relationship with Lyz activity. If the temperature was too high or too low, Lyz can have conformational changes and be inactivation during the hydrolysis process. As former article reported, Lyz showed a rapid loss of the tertiary structure after 55 1C. It can form amorphous aggregation when temperature reached to 65 1C (Yang and Dunstan, 2013). This kind of conformational changes can influence the action between Lyz and CMCSQDs. Regarding above results, 40 1C was chosen for the PL quenching experimenst as the optimal temperature for the Zn2 þ –CMCS-QDs system. Fig. S4 showed the effect of reaction time on the PL of QDs. The PL of QDs decreased with time increasing until 30 min. It indicated that Lyz hydrolyzing process and the formation of Lyz–QDs complexes were finished within 30 min. As shown in Fig. S5, the salt concentration in a certain range had no significant influence on the PL of Lyz–Zn2 þ -CMCSQDs system. Under the above optimum conditions, the PL of Zn2 þ – CMCS-QDs system with Lyz can be quenched to about 60%. 3.3. Detection of Lyz Under the optimal conditions, the PL of Zn2 þ –CMCS-QDs system at 572 nm decreased gradually with the increasing concentration of Lyz (Fig. 4). Furthermore, the inset of Fig. 4 showed that there was a good linearity between I/I0 (I and I0 were the PL of Zn2 þ –CMCS-QDs system in the presence and absence of Lyz, respectively) and the Lyz
3.4. Interference study To illustrate the validity and test the selectivity of the nanosensor for Lyz, a systematic study of various common mental ions, amino acids and proteins were added into the system when the determination of Lyz (1 mg/L) was carried out under the optimum conditions. Table S2 showed the interference effect. A relative error of 75.0% was considered to be tolerable. As shown in Table S2, the tolerable concentration ratios of coexisting substances was over 1000 fold for Na þ , K þ , Ca2 þ , Mg2 þ , 50 fold for Fe2 þ , Fe3 þ , Ba2 þ , 1200 fold for Thr, 1500 fold for Lys, 1600 fold for Phe, 500 fold for Cys, Gly, Arg, Tyr, human serum albumin, transferring, immunoglobulin and 100 fold for BSA. With these high concentrations of commonly existing substances, there was little interference on the PL of Zn2 þ –CMCS-QDs. It was obvious that the novel nanosensor based on Zn2 þ –CMCS-QDs was ultrasensitive and high selective for Lyz detection. In order to assess the feasibility of the proposed fluorescence nanosensor, the amount of Lyz in human serum was detected. The results obtained by standard addition method were shown in Table 1. The RSD was lower than 1.5% and the recoveries of Lyz in the real samples were between 97.97% and 100.18%.
4. Conclusion In summary, we have developed a novel ultrasensitive “turn on–off”fluorescence nanosensor for Lyz detection. The fluorescence signals of CMCS-QDs were enhanced by Zn2 þ and quenched by Lyz successively. The reaction conditions including the concentrations of Zn2 þ , pH, temperature, reaction time and concentrations of salt have been discussed. A good linearity for Lyz detection in the range of 0.1–1.2 ng/mL with the detection limit of 0.031 ng/mL was found. Moreover, the new analytical method has been employed to detect the concentration of Lyz in human serum samples successfully. The “turn on–off” fluorescence nanosensor for Lyz detection was ultrasensitive, high selective and convenient in the real analysis application.
Acknowledgments This work was financially supported by the National Natural Science Foundation of China, People's Republic of China (nos. 21075050 and 21005029).
Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.bios.2014.04.036.
Y. Song et al. / Biosensors and Bioelectronics 61 (2014) 9–13
References Amano, K., Ito, E., 1978. Eur. J. Biochem. 85, 97–104. Cai, Z.X., Chen, G.X., Huang, X., Ma, M.H., 2011. Sens. Actuators B 157, 368–373. Chenn, R., Wang, G., Chen, C., Ho, H., Sheu, M., 2006. Biomacromolecules 7, 1058–1064. Chen, X., Wang, Z., Liu, W., Park, H., 2002. Biomaterials 23, 4609–4614. Gaberc, P.V., Menart, V., 2001. J. Biochem. Biophys. Methods 49, 335–360. Hanif, K.M., Meulenberg, R.W., Strouse, G.F., 2002. J. Am. Chem. Soc. 124, 11495–11502. Kurita, K., Kaji, Y., Mori, T., Nishiyama, Y., 2000. Carbohydr. Polym. 42, 19–21. Klockars, M., Reitamo, S., Weber, T., Kerttula, Y., 1978. Acta Med. Scand. 203, 71–74. Lee, C.M., Jong, H.J., Kim, S.L., Kim, E.M., Kim, D.W., Lim, S.T., Jang, K.Y., Jeon, Y.Y., Nah, J.W., Sohn, M.H., 2009. Int. J. Pharm. 371, 163–169. Ma, Q., Ha, E., Yang, F.P., Su, X.G., 2011. Anal. Chim. Acta 701, 60–65. Ma, Q., Lin, Z.H., Yang, N., Li, Y., Su, X.G., 2014. Acta Biomater. 10, 868–874. Meltem, Y., Gülay, B., Arıca, M.Y., 2005. Food Chem. 89, 11–18. Pradhan, N., Goorskey, D., Thessing, J., Peng, X.G., 2005. J. Am. Chem. Soc. 127, 17586–17587. Senera, G., Ozgur, E., Yılmaz, E., Uzun, L., Say, R., Denizli, A., 2010. Biosens. Bioelectron. 26, 815–821.
13
Shen, J.M., Tang, W.J., Zhang, X.L., Chen, T., Zhang, H.X., 2012. Carbohydr. Polym. 88, 239–249. Upadhyaya, L., Singh, J., Agarwal, V., Tewaria, R.P., 2013. Carbohydr. Polym. 91, 452–466. Vannoy, C.H., Xu, J.N., Leblanc, R.M., 2010. J. Phys. Chem. C 114, 766–773. Vasilescu, A., Gaspar, S., Mihai, I., Tache, A., Litescu, S.C., 2013. Analyst 138, 3530–3537. Wang, J.L., Chen, G.H., Jiang, H., Li, Z.Y., Wang, X.M., 2013. Analyst 138, 4427–4435. Wu, Y.L., He, F., He, X.W., Li, W.Y., Zhang, Y.K., 2008. Spectrochim. Acta Part A 71, 1199–1203. Yang, S.O., Dunstan, D.E., 2013. Soft Matter 9, 9692–9701. Yang, Y., Chen, O., Angerhofer, A., Cao, Y.C., 2006. J. Am. Chem. Soc. 128, 12428–12429. Zeng, X., Ruckenstein, E., 1999. Biotechnol. Prog. 15, 1003–1019. Zheng, H.Y., Qiu, S.Y., Xu, K.F., Luo, L.G., Song, Y.B., Lin, Z.Y., Guo, L.H., Qiu, B., Chen, G.N., 2013. Analyst 138, 6517–6522. Zhou, L.M., Wang, Y.P., Liu, Z.R., Huang, Q.W., 2006. Acta Phys. Chim. Sin. 22, 1342–1346. Zhu, A., Fang, N., 2005. Biomacromolecules 6, 2607–2614. Zou, M.J., Chen, Y., Xu, X., Huang, H.D., Liu, F., Li, N., 2012. Biosens. Bioelectron. 32, 148–154.
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
A novel ultrasensitive carboxymethyl chitosan-quantum dot-based fluorescence “turn on–off” nanosensor for lysozyme detection Yu Song a, Yang Li a, Ziping Liu a, Linlin Liu a, Xinyan Wang b, Xingguang Su a, Qiang Ma a,n a b
Department of Analytical Chemistry, College of Chemistry, Jilin University, Changchun 130012, China Changchun Institute of Applied Chemistry Chinese Academy of Sciences, Changchun 230022, China
art ic l e i nf o
a b s t r a c t
Article history: Received 16 January 2014 Received in revised form 18 April 2014 Accepted 21 April 2014 Available online 30 April 2014
In this work, we developed an ultrasensitive “turn on–off” fluorescence nanosensor for lysozyme (Lyz) detection. The novel nanosensor was constructed with the carboxymethyl chitosan modified CdTe quantum dots (CMCS-QDs). Firstly, the CMCS-QDs were fabricated via the electrostatic interaction between amino groups in CMCS polymeric chains and carboxyl groups on the surface of QDs. In the fluorescence “turn-on” step, the strong binding ability between Zn2 þ and CMCS on the surface of QDs can enhance the photoluminescence intensity (PL) of QDs. In the following fluorescence “turn-off” step, the N-acetyl-glucosamine (NAG) section along the CMCS chains was hydrolyzed by Lyz. As a result, Zn2 þ was released from the surface of QDs, and the Lyz–QDs complexes were formed to quench the QDs PL. Under the optimal conditions, there was a good linear relationship between the PL of QDs and the Lyz concentration (0.1–1.2 ng/mL) with the detection limit of 0.031 ng/mL. The developed method was ultrasensitive, highly selective and fast. It has been successfully employed in the detection of Lyz in the serum with satisfactory results. & 2014 Elsevier B.V. All rights reserved.
Keywords: Nanosensor Quantum dots Lysozyme Fluorescence turn on–off
1. Introduction As a kind of ideal nanomaterial, luminescent quantum dots (QDs) have attracted great attention in the past decades due to their unique properties (Hanif et al., 2002; Pradhan et al., 2005; Yang et al., 2006), including narrow, symmetrical and size-tunable emission spectrum, broad excitation spectrum and so on. QDs have been employed widely in biosensing and biolabeling applications (Wu et al., 2008) to detect multifarious substances (e.g. small molecules and proteins) (Wang et al., 2013). QD-based nanosensors are becoming the most important nanomaterials for biological and medical applications (Vannoy et al., 2010). However, there are still some concerned problems with how to prepare QD-based sensing systems for high sensitive and selective analysis. Therefore, various techniques have been employed in the assembled strategies of QDs-based sensors. These assembled strategies not only effected the efficiency and stability of sensors, but also determined the cost and practicability of application. Currently, more and more surface modified reagents have been employed to render diverse affinity and specificity of QDs towards different targets. As an ideal functional material, the natural polysaccharide—carboxymethyl
n
Corresponding author. Tel.: þ 86 431 85168352. E-mail address: [email protected] (Q. Ma).
http://dx.doi.org/10.1016/j.bios.2014.04.036 0956-5663/& 2014 Elsevier B.V. All rights reserved.
chitosan (CMCS) which composed of N-acetyl-glucosamine (NAG) have many attractive physical and biological properties such as hydrophilicity, good biocompatibility, biodegradability, low toxicity, and remarkable affinity for metal ions (Gaberc and Menart , 2001; Lee et al., 2009; Zhou et al., 2006). These properties made CMCS become a promising modification material (Chen et al., 2002; Zhu and Fang, 2005; Chenn et al., 2006). The amino (– NH2), carboxyl (–COOH) and hydroxyl (–OH) groups on CMCS chains can serve as electrostatic interaction and coordination sites (Zeng and Ruckenstein 1999). So, CMCS had strong binding ability towards metal ions, especially with Zn2 þ (Shen et al., 2012; Upadhyaya et al., 2013). Furthermore, we found a strong photoluminescence-activation effect after Zn2 þ binding with CMCS-QDs (Ma and Lin et al., 2014). Lyz (Mw: 14.3 kDa) is a relatively small single chain protein with only 129 amino acids. Lyz can only recognize the NAG section and hydrolyze the similar glycan backbone, such as chitin, partially deacetylated chitosan (Kurita et al., 2000, Amano and Ito, 1978). As a tool enzyme, Lyz has been used in gene and cell engineering, which exists in body tissues and secretions (e.g. serum, urine and tears). Because the increased Lyz concentration in urine and serum are potential indicators for leukemia and meningitis (Klockars et al., 1978), the reliable and sensitive methods for the analysis of Lyz are required. To date, a variety of strategies for the Lyz detection have been reported, such as surface plasmon resonance (SPR)
10
Y. Song et al. / Biosensors and Bioelectronics 61 (2014) 9–13
the excessive CMCS. The collected CMCS-QDs were washed with ethanol/water and stored with PBS (pH 7.4) in dark. The final concentration was 3.56 nM QDs/0.05 g CMCS. 2.3. Preparation of Zn2 þ –CMCS-QDs Scheme 1. The schematic illustration of the novel Zn2 þ –CMCS-QDs nanosensor for Lyz detection.
(Vasilescu et al., 2013), Resonance Rayleigh-scattering (Cai et al., 2011), quartz crystal microbalance (QCM) (Senera et al., 2010), fluorescence anisotropic sensing (Zou et al., 2012), colorimetric & fluorometric dual-readout sensor (Zheng et al., 2013) and so on. However, many methods suffered from some drawbacks including interfering substances, unsatisfactory detection limit and time consuming. In present study, we prepared CMCS-QDs to establish a fluorescence “turn on–off” sensing system, as shown in Scheme 1. Firstly, CMCS-QDs can absorb Zn2 þ to enhance the PL of QDs in the fluorescence “turn on” step. In the next fluorescence “turn off” step, Lyz was added into Zn2 þ –CMCS-QDs system. CMCS was degraded quickly to small molecule by hydrolyzing of Lyz, which can release Zn2 þ from the QDs surface and form Lyz–QDs complexes. As a result, the QDs PL was quenched. The established fluorescence “turn on–off” sensing system has high selectivity and ultrasensitivity for Lyz detection, and this sensing system has been successfully applied to detect Lyz in the human serum samples with good precision and accuracy.
In the fluorescence “turn on” step, 1 mL CMCS-QDs (3.56 nM QDs/0.05 g CMCS) were mixed with a series of different concentrations of Zn2 þ solution. In this process, all measurements were performed under the same condition. The slit widths of the excitation and emission were 5 nm and 10 nm. The excitation wavelength was set at 380 nm. And the fluorescence spectra were recorded in the 500–700 nm emission wavelength range. 2.4. Lyz detection In the fluorescence “turn off” step, a series of different concentrations of Lyz and the Zn2 þ –CMCS-QDs complexes in phosphate-buffered saline solutions (PBS, 2 mM, pH 7.4) were mixed in 2.0 mL calibrated test tubes. Subsequently, the mixtures were gently shaken for 30 min at room temperature. The human blood samples were segregated by centrifugation at 10,000 rpm for 10 min after adding CH3COOH in samples (CH3COOH:serum 1.5:1). Finally, all supernatant serum samples were subjected to a 1000-fold dilution with PBS before analysis, and the different concentrations of Lyz were added to prepare the spiked samples.
3. Results and discussion 2. Experiment section
3.1. Spectra characteristics of CMCS-QDs.
2.1. Materials and apparatus
The fluorescence spectra of QDs and CMCS-QDs were shown in Fig. 1A. After coated with CMCS, the slight blue shifts of the QDs PL peaks (from 579 nm to 570 nm) can be observed (Fig. 1A b and d). It was due to the photo-induced surface reconstruction of QDs surface atoms (Ma et al., 2014). Then, the enhancement effect of Zn2 þ on the PL of QDs and CMCS-QDs was studied. It can be seen that the PL of CMCS-QDs was significantly enhanced to about 130% with the increase of Zn2 þ concentration (Fig. 1A d and e). The fluorescence quantum yield of CMCS-QDs and Zn2 þ –CMCS-QDs were 26.31% and 35.02%, respectively. It also proved that Zn2 þ had the ability to enhance the PL of CMCS-QDs. By comparison, the PL of QDs without CMCS did not change (Fig. 1A a and b). It indicated that CMCS played a significant role in enhancing the PL of QDs. CMCS was bound on the surface of QDs via the electrostatic interaction between amino groups in its polymeric chains and carboxyl groups on the QDs surface. The Zn2 þ ions can be captured by CMCS-QDs due to the high affinity of CMCS to Zn2 þ ions. Zn2 þ can prevent the surface nonradiative relaxation of QDs. As a result, the fluorescence signal of CMCS-QDs was “turned on”. With the addition of Lyz, the PL of Zn2 þ –QDs (c) and Zn2 þ –CMCSQDs (f) were “turned off”. The PL quenching effect of Lyz on Zn2 þ – CMCS-QDs (from 130% to 90%) was stronger than that of Zn2 þ – QDs (from 101% to 92%), and there was no significant shift of QDs PL peak after the addition of Lyz. It indicated that Lyz can quench the PL of both Zn2 þ –QDs and Zn2 þ –CMCS-QDs systems. In the Zn2 þ –QDs system, Lyz can bind with QDs to form Lyz–QDs complexes (Wu et al., 2008). The complexes can quench the PL of Zn2 þ –QDs system. In the Zn2 þ –CMCS-QDs system, Lyz can recognize the NAG section along the CMCS chains which act as a hydrolyzing site. So, CMCS was hydrolyzed to small molecules by Lyz, and the captured Zn2 þ was released from the surface of CMCS-QDs firstly. After CMCS was hydrolyzed, the formation of Lyz–QDs complexes can further quench the PL of Zn2 þ –CMCSQDs. From Fig. S1a, the size of QDs (3–5 nm) were well-distributed
3-Mercaptopropyl acid (MPA) (99%) was purchased from J&K Chemical Co.. And tellurium powder ( 200 mesh, 99.8%), CdCl2 (99%) and NaBH4 (99%) were purchased from Aldrich Chemical Co. Carboxymethyl chitosan, NaH2PO4, Na2HPO4, NaCl and ZnCl2 were purchased from Beijing Dingguo Chemicals Co. (Beijing, China). All chemicals used were of analytical grade. The water used in the study had a resistivity higher than 18 MΩ cm. Lyz were purchased from Genview. The human serum was obtained as a gift from the university hospital. The fluorescence measurements were performed on a Shimadzu RF-5301 PC spectrofluorophotometer (Shimadzu Co., Kyoto, Japan) equipped with a xenon lamp using right-angle geometry and a 1 cm path-length quartz cell. All pH measurements were taken with a PHS-3C pH meter (Tuopu Co., Hangzhou, China). 2.2. Synthesis of CMCS-QDs QDs used in our work were synthesized by refluxing routes as the method described in our previous papers (Ma et al., 2011). In brief, CdCl2 solution (1.25 mM) with MPA as stabilizer was added into a 250 mL three-necked flask under N2 atmosphere. NaHTe was produced in an aqueous solution by the reaction of tellurium powder with NaBH4 at a molar ratio of 1:2. Later, the fresh NaHTe was added into the CdCl2 solution. The molar ratio of Cd2 þ /MPA/NaHTe was at 1:1.5:0.2. The solution was subjected to a reflux at 100 1C under open-air conditions to obtain watercompatible MPA-capped QDs. The fluorescence emission wavelength of the QDs used in present experiments was 579 nm. Then, 3.56 nM QDs was added into 10 mL CMCS stock solution (5 g/L). The reaction mixtures were sonicated for 5 min, stirried and vibrated overnight at room temperature in dark. The resultant mixture was transferred into ethanol and centrifuged to remove
Y. Song et al. / Biosensors and Bioelectronics 61 (2014) 9–13
11
Fig. 1. A: The fluorescence spectrum of (a) QDs, (b)Zn2 þ –QDs, (c) Lyz–Zn2 þ -QDs, (d) CMCS-QDs, (e) Zn2 þ –CMCS-QDs, and (f) Lyz–Zn2 þ –CMCS-QDs. (0.1 mM Zn2 þ and 1 mg/L Lyz). B: The photographs of (a) CMCS-QDs (b) Zn2 þ –CMCS-QDs and (c) Lyz–Zn2 þ –CMCS-QDs under UV light.
in solution. After Zn2 þ added (Fig. S1b), the size of Zn2 þ –CMCSQDs was about 10 nm. With the addition of Lyz, there was an obvious aggregation of QD complexes (Fig. S1c). This conclusion was in accord with the former documents reported (Wu et al., 2008; Vannoy et al., 2010). Fig. 1B showed the photographs of CMCS-QDs, Zn2 þ –CMCS-QDs and Lyz–Zn2 þ –CMCS-QDs under UV light, respectively. It can be seen that the luminescence signal of Zn2 þ –CMCS-QDs (b) was brighter than that of CMCS-QDs (a) in the fluorescence “turn on” step. After the addition of Lyz in the system (c), the luminescence signal became weaker than that of CMCS-QDs (a). In order to investigate the selectivity of this novel sensing system, we added Ca2 þ , Mg2 þ or Na þ in the CMCS-QDs and CMCS-QDs–Zn2 þ system. These ions also can chelate with CMCS. From Fig. S2, it can be seen that the PL of CMCS-QDs was easily influenced by Ca2 þ , Mg2 þ and Na þ without Zn2 þ (Fig. S2c and d). But, the PL of CMCS-QDs–Zn2 þ system can keep stable with those ions (Fig. S2a and b). Because Zn2 þ had strong binding ability with CMCS, almost all carboxyl (–COOH) groups on CMCS chains that served as electrostatic interaction and coordination sites can be bound. So other ions can hardly bind with CMCS to influence the PL of QDs. It was obvious that the selectivity and stability of the CMCS-QDs–Zn2 þ sensing system were improved largely.
Fig. 2. The influence of different Zn2 þ concentrations on the PL intensity of CMCSQDs (a–h) 0, 0.03, 0.05, 0.06, 0.08, 0.15, 0.2, set: a plot of normalized PL intensity against the concentration of Zn2 þ .
3.2. Optimization for Lyz detection Firstly, we studied the effect of Zn2 þ concentration on the PL of CMCS-QDs. As shown in Fig. 2, the PL of CMCS-QDs increased with the increasing Zn2 þ concentration. In this fluorescence “turn on” step, the surface nonradiative relaxation centers of CMCS-QDs were passivated efficiently via Zn2 þ binding at a certain concentration. The nonradiative recombination of “filled” surface trap sites was effectively eliminated. When Zn2 þ concentration increased to 0.2 mM, the QDs PL reached maximum and kept stable. So, 0.2 mM Zn2 þ was used in the further experiments. The effect of pH on the PL of Zn2 þ –CMCS-QDs system with Lyz (1 ng/mL) was also studied. It can be seen (Fig. 3) that the system PL was decreased to about 62% at pH 6. The “turn-off” efficiency at pH 6 was more obvious than at other pH circumstances. As a kind of amphoteric proteins, the type (positive or negative) and the number of charges on the Lyz surface can vary with the pH of the medium. The pI of Lyz is 11.2 as reported (Meltem et al., 2005). So, the Lyz molecules can be cationic when pH o11.2 and anionic at pH 411.2. At pH ¼6, the carboxyl groups on the QDs were
Fig. 3. The influence of pH on the PL intensity of Lyz–Zn2 þ –CMCS-QDs (a) pH9.5 (b) pH8 (c) pH4 (d) pH7 (e) pH5 and (f) pH6. Inset: a plot of I/I0 against the pH. I and I0 were the fluorescence intensity of Zn2 þ –CMCS-QDs in the presence and absence of Lyz, respectively.
12
Y. Song et al. / Biosensors and Bioelectronics 61 (2014) 9–13
1.0
0.8
I/I0
k
I/I0
concentration in range of 0.1–1.2 ng/mL. The linear regression equation was I/I0 ¼1.01 0.42 [Lyz] (ng/mL). The corresponding regression coefficient was 0.998. The detection limit (LOD) was 0.031 ng/mL (LOD¼3s/s). s was the slope of the calibration curve. s was the standard deviation of the corrected blank PL signals of the Zn2 þ –CMCS-QDs. Some other reported methods for the determination of Lyz were compared with the proposed method. The comparison was summed up in Table S1. It clearly showed that the analytical performance of this Zn2 þ –CMCS-QDs based nanosensor for Lyz detection was better than most of the reported methods. The novel nanosensor for Lyz detection based on Zn2 þ – CMCS-QDs was ultrasensitive and high selective.
1.0
a
0.6
0.5 0.0 0.2 0.4 0.6 0.8 1.0 1.2
Lysozyme (ng/mL)
0.0
550
600
650
700
Wavelength/nm 2þ
Fig. 4. The PL intensity of Zn –CMCS-QDs with different Lyz concentrations from 0 to 1.2 ng/mL. (a) 0 (b) 0.1 (c) 0.2 (d) 0.3 (e) 0.4 (f) 0.5 (g) 0.6 (h) 0.8 (i) 0.9 (j) 1 and (k) 1.2. The inset showed the relationship between I/I0. I and I0 were the fluorescence intensity of Zn2 þ –CMCS-QDs in the presence and absence of Lyz, respectively.
Table 1 Determination of Lyz in serum samples (n¼ 3). Samples
Found (ng/mL)
Added (ng/mL)
Total found (ng/mL)
Recovery (%)
RSD (%, n ¼3)
1 2 3
0.319 0.453 0.338
0.4 0.5 0.4
0.720 0.943 0.736
100.18 97.97 99.48
0.97 0.92 1.24
negatively charged. So, the weak acid environment promoted that the PL of Zn2 þ –CMCS-QDs system further decreasing. The PL quenching degree was different along with temperature changing. The different reaction temperature in the range of 4–60 1C was studied as shown in Fig. S3. The PL of QDs gradually decreased with the increasing temperature and reached the lowest at 40 1C. When the temperature increased to 60 1C, the PL of QDs increased again. The results indicated that the temperature had a close relationship with Lyz activity. If the temperature was too high or too low, Lyz can have conformational changes and be inactivation during the hydrolysis process. As former article reported, Lyz showed a rapid loss of the tertiary structure after 55 1C. It can form amorphous aggregation when temperature reached to 65 1C (Yang and Dunstan, 2013). This kind of conformational changes can influence the action between Lyz and CMCSQDs. Regarding above results, 40 1C was chosen for the PL quenching experimenst as the optimal temperature for the Zn2 þ –CMCS-QDs system. Fig. S4 showed the effect of reaction time on the PL of QDs. The PL of QDs decreased with time increasing until 30 min. It indicated that Lyz hydrolyzing process and the formation of Lyz–QDs complexes were finished within 30 min. As shown in Fig. S5, the salt concentration in a certain range had no significant influence on the PL of Lyz–Zn2 þ -CMCSQDs system. Under the above optimum conditions, the PL of Zn2 þ – CMCS-QDs system with Lyz can be quenched to about 60%. 3.3. Detection of Lyz Under the optimal conditions, the PL of Zn2 þ –CMCS-QDs system at 572 nm decreased gradually with the increasing concentration of Lyz (Fig. 4). Furthermore, the inset of Fig. 4 showed that there was a good linearity between I/I0 (I and I0 were the PL of Zn2 þ –CMCS-QDs system in the presence and absence of Lyz, respectively) and the Lyz
3.4. Interference study To illustrate the validity and test the selectivity of the nanosensor for Lyz, a systematic study of various common mental ions, amino acids and proteins were added into the system when the determination of Lyz (1 mg/L) was carried out under the optimum conditions. Table S2 showed the interference effect. A relative error of 75.0% was considered to be tolerable. As shown in Table S2, the tolerable concentration ratios of coexisting substances was over 1000 fold for Na þ , K þ , Ca2 þ , Mg2 þ , 50 fold for Fe2 þ , Fe3 þ , Ba2 þ , 1200 fold for Thr, 1500 fold for Lys, 1600 fold for Phe, 500 fold for Cys, Gly, Arg, Tyr, human serum albumin, transferring, immunoglobulin and 100 fold for BSA. With these high concentrations of commonly existing substances, there was little interference on the PL of Zn2 þ –CMCS-QDs. It was obvious that the novel nanosensor based on Zn2 þ –CMCS-QDs was ultrasensitive and high selective for Lyz detection. In order to assess the feasibility of the proposed fluorescence nanosensor, the amount of Lyz in human serum was detected. The results obtained by standard addition method were shown in Table 1. The RSD was lower than 1.5% and the recoveries of Lyz in the real samples were between 97.97% and 100.18%.
4. Conclusion In summary, we have developed a novel ultrasensitive “turn on–off”fluorescence nanosensor for Lyz detection. The fluorescence signals of CMCS-QDs were enhanced by Zn2 þ and quenched by Lyz successively. The reaction conditions including the concentrations of Zn2 þ , pH, temperature, reaction time and concentrations of salt have been discussed. A good linearity for Lyz detection in the range of 0.1–1.2 ng/mL with the detection limit of 0.031 ng/mL was found. Moreover, the new analytical method has been employed to detect the concentration of Lyz in human serum samples successfully. The “turn on–off” fluorescence nanosensor for Lyz detection was ultrasensitive, high selective and convenient in the real analysis application.
Acknowledgments This work was financially supported by the National Natural Science Foundation of China, People's Republic of China (nos. 21075050 and 21005029).
Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.bios.2014.04.036.
Y. Song et al. / Biosensors and Bioelectronics 61 (2014) 9–13
References Amano, K., Ito, E., 1978. Eur. J. Biochem. 85, 97–104. Cai, Z.X., Chen, G.X., Huang, X., Ma, M.H., 2011. Sens. Actuators B 157, 368–373. Chenn, R., Wang, G., Chen, C., Ho, H., Sheu, M., 2006. Biomacromolecules 7, 1058–1064. Chen, X., Wang, Z., Liu, W., Park, H., 2002. Biomaterials 23, 4609–4614. Gaberc, P.V., Menart, V., 2001. J. Biochem. Biophys. Methods 49, 335–360. Hanif, K.M., Meulenberg, R.W., Strouse, G.F., 2002. J. Am. Chem. Soc. 124, 11495–11502. Kurita, K., Kaji, Y., Mori, T., Nishiyama, Y., 2000. Carbohydr. Polym. 42, 19–21. Klockars, M., Reitamo, S., Weber, T., Kerttula, Y., 1978. Acta Med. Scand. 203, 71–74. Lee, C.M., Jong, H.J., Kim, S.L., Kim, E.M., Kim, D.W., Lim, S.T., Jang, K.Y., Jeon, Y.Y., Nah, J.W., Sohn, M.H., 2009. Int. J. Pharm. 371, 163–169. Ma, Q., Ha, E., Yang, F.P., Su, X.G., 2011. Anal. Chim. Acta 701, 60–65. Ma, Q., Lin, Z.H., Yang, N., Li, Y., Su, X.G., 2014. Acta Biomater. 10, 868–874. Meltem, Y., Gülay, B., Arıca, M.Y., 2005. Food Chem. 89, 11–18. Pradhan, N., Goorskey, D., Thessing, J., Peng, X.G., 2005. J. Am. Chem. Soc. 127, 17586–17587. Senera, G., Ozgur, E., Yılmaz, E., Uzun, L., Say, R., Denizli, A., 2010. Biosens. Bioelectron. 26, 815–821.
13
Shen, J.M., Tang, W.J., Zhang, X.L., Chen, T., Zhang, H.X., 2012. Carbohydr. Polym. 88, 239–249. Upadhyaya, L., Singh, J., Agarwal, V., Tewaria, R.P., 2013. Carbohydr. Polym. 91, 452–466. Vannoy, C.H., Xu, J.N., Leblanc, R.M., 2010. J. Phys. Chem. C 114, 766–773. Vasilescu, A., Gaspar, S., Mihai, I., Tache, A., Litescu, S.C., 2013. Analyst 138, 3530–3537. Wang, J.L., Chen, G.H., Jiang, H., Li, Z.Y., Wang, X.M., 2013. Analyst 138, 4427–4435. Wu, Y.L., He, F., He, X.W., Li, W.Y., Zhang, Y.K., 2008. Spectrochim. Acta Part A 71, 1199–1203. Yang, S.O., Dunstan, D.E., 2013. Soft Matter 9, 9692–9701. Yang, Y., Chen, O., Angerhofer, A., Cao, Y.C., 2006. J. Am. Chem. Soc. 128, 12428–12429. Zeng, X., Ruckenstein, E., 1999. Biotechnol. Prog. 15, 1003–1019. Zheng, H.Y., Qiu, S.Y., Xu, K.F., Luo, L.G., Song, Y.B., Lin, Z.Y., Guo, L.H., Qiu, B., Chen, G.N., 2013. Analyst 138, 6517–6522. Zhou, L.M., Wang, Y.P., Liu, Z.R., Huang, Q.W., 2006. Acta Phys. Chim. Sin. 22, 1342–1346. Zhu, A., Fang, N., 2005. Biomacromolecules 6, 2607–2614. Zou, M.J., Chen, Y., Xu, X., Huang, H.D., Liu, F., Li, N., 2012. Biosens. Bioelectron. 32, 148–154.
