- Email: [email protected]
Silver nanoparticles-enhanced time-resolved fluorescence sensor for VEGF165 based on Mn-doped ZnS quantum dots
Biosensors and Bioelectronics 74 (2015) 1053–1060
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
Silver nanoparticles-enhanced time-resolved fluorescence sensor for VEGF165 based on Mn-doped ZnS quantum dots Dong Zhu n, Wei Li n, Hong-Mei Wen, Sheng Yu, Zhao-Yi Miao, An Kang, Aihua Zhang School of Pharmacy, Nanjing University of Chinese Medicine, Nanjing 210023, PR China
art ic l e i nf o
a b s t r a c t
Article history: Received 28 May 2015 Received in revised form 4 August 2015 Accepted 5 August 2015 Available online 6 August 2015
A silver nanoparticles (AgNPs)-enhanced time-resolved fluorescence (TR-FL) sensor based on long-lived fluorescent Mn-doped ZnS quantum dots (QDs) is developed for the sensitive detection of vascular endothelial growth factor-165 (VEGF165), a predominant cancer biomarker in cancer angiogenesis. The aptamers bond with the Mn-doped ZnS QDs and the BHQ-2 quencher-labelling strands hybridized in duplex are coupled with streptavidin (SA)-functionalized AgNPs to form the AgNPs-enhanced TR-FL sensor, showing lower fluorescence intensity in the duplex state due to the fluorescence resonance energy transfer (FRET) between the Mn-doped ZnS QDs and quenchers. Upon the addition of VEGF165, the BHQ-2 quencher-labelling strands of the duplex are displaced, leading to the disruption of the FRET. As a result, the fluorescence of the Mn-doped QDs within the proximity of the AgNPs is recovered. The FL signal can be measured free of the interference of short-lived background by setting appropriate delay time and gate time, which offers a signal with high signal-to-noise ratio in photoluminescent biodetection. Compared with the bare TR-FL sensor, the AgNPs-based TR-FL sensor showed a huge improvement in fluorescence based on metal-enhanced fluorescence (MEF) effect, and the sensitivity increased 11-fold with the detection limit of 0.08 nM. In addition, the sensor provided a wide range of linear detection from 0.1 nM to 16 nM. & 2015 Elsevier B.V. All rights reserved.
Keywords: Metal-enhanced fluorescence Time-resolved Fluorescence resonance energy transfer Mn-doped ZnS quantum dots
1. Introduction The exploration of time-resolved fluorescence (TR-FL) biosensing systems is a topic of considerable interest (Eis and Millar, 1993; Sandin et al., 2008; Kishore et al., 2013; Huang et al., 2009; Ozers et al., 2007;) since TR-FL assay brings the advantage of near zero background signal from the time-resolved (TR) technique, provides an excellent solution to eliminate the interference of short-lived autofluorescence from cells, tissues, and assay multiwell plate and thus offers remarkably high sensitivity as compared with the conventional fluorometry (Liu et al., 2012). The probe is key in this technique, and the ideal probe should have long fluorescence life time, large Stoke's shift and narrow emission peaks. Currently, the TR-FL bio-probes are mainly lanthanide chelates and their corresponding nanoparticles (Liu et al., 2012; Laitala and Hemmila1, 2005; Hirata et al., 2004) owing to their long fluorescence lifetime (as long as 20–1000 μs). However, the development of lanthanide probe is limited by their shortcomings such as weak luminescence and photobleaching (Seveus et al., n
Corresponding authors. Fax: þ 86 2585811839. E-mail addresses: [email protected] (D. Zhu), [email protected] (W. Li). http://dx.doi.org/10.1016/j.bios.2015.08.005 0956-5663/& 2015 Elsevier B.V. All rights reserved.
1992), and thus, development of a fluorescent probe with stronger luminescence and better photostability is critical for the further application of TR-FL in biological studies. Quantum dots (QDs) have emerged as the desirable candidates for applications in biomolecular sensing (Zhang et al., 2005; Ho et al., 2005; Zhelev et al., 2006) and cellular imaging (Michalet et al., 2005; Hiroshi et al., 2012), mainly due to their unique luminescent quality such as strong photoluminescence and good photostability. Cadmium chalcogenide QDs are widely explored. However, their fluorescence lifetime is similar to that of the background autofluorescence (Michalet et al., 2005), which obstructs the application of these traditional QDs in the time-resolved fluorometry. Recently, Mn-doped ZnS QDs have attracted increasing attention (Wu and Yan, 2013; He et al., 2008; Zhu et al., 2014; Wu et al., 2010; Wang et al., 2010) because of their charming characteristics such as strong photoluminescence, high resistance to photobleaching and especially long fluorescence life time (as long as 1–2 ms), from the sharp 4T1(4G)-6A1(6S) emission of the Mn2 þ (Norris et al., 2008; Michalet et al., 2005; Xie et al., 2007; Xie and Peng , 2008). Some researchers have successfully employed the Mn-doped ZnS QDs for TR fluorescence detection (He et al., 2008; Zhu et al., 2014; Wu et al., 2010; Wang et al., 2010). However, the fluorescence lifetime of ms scale about Mn2 þ emission is too long for the only a few nanoseconds lifetime from
1054
D. Zhu et al. / Biosensors and Bioelectronics 74 (2015) 1053–1060
background autofluorescence (Michalet et al., 2005) and such a long decay time results in a low photon count rate, which may lead to a relatively low quantum yield and limit the improvement of detection sensitivity (Zhang et al., 2009). Therefore, development of a strategy that can increase the emission intensity and simultaneously shorten the fluorescence lifetime is critical for the further application of TR-FL detection in biological studies. Lately, the metal-enhanced fluorescence (MEF) effect, based on the surface plasmon resonance (SPR) of metallic nanostructures, seems to be very promising for solving the problem (Lakowicz, 2001). The SPR of metallic nanostructures, especially the Ag nanoparticles (NPs), can enhance the local electromagnetic field surrounding them and finally lead to the photoluminescence increase and lifetime decrease of nearby fluorophores (Geddes and Lakowicz, 2002; Sokolov et al., 1998). Most MEF materials developed thus far have been targeted towards enhancing fluorescence from organic fluorophores (Medintz et al., 2005; Mackowski et al., 2008; Muskens et al., 2007; Zhang et al., 2007), while few has been developed for inorganic fluorescent species, such as QDs (Akimov et al., 2007; Fu et al., 2009), and more importantly, there is very limited progress in the development of biomolecular sensing by this MEF approach. In this paper, based on the MEF principle, an AgNPs-enhanced TR-FL biosensor is developed using the Mn-doped ZnS QDs bond with aptamers that are hybridized with quencher BHQ-2 labelling strands for the detection of vascular endothelial growth factor-165 (VEGF165), a predominant cancer biomarker in cancer angiogenesis (Zhu et al., 2011). Aptamers are short, single-stranded oligonucleotides (RNA or DNA o 100 nt) or peptides (o 100-mer) generated from an in vitro method known as SELEX (systematic evolution of ligands by exponential enrichment) (Fang and Tan, 2010), which have high affinity, specificity and stability, therefore, they are regarded as alternative reagents to antibodies (Cho et al., 2009). Furthermore, aptamers change structures from an elongated metastable structure to a uniquely folded stable structure upon interaction with targets (Bock et al., 1992). As illustrated in Scheme 1, the aptamers modified with biotin on the 3′ end and the Mn-doped ZnS QDs on the 5′ end and BHQ-2 quencher-labelling strands which are hybridized in duplex are bond with streptavidin (SA)-functionalized AgNPs to form an AgNPs-enhanced TR-FL sensor. The TR-FL sensor shows lower fluorescence
in the duplex sate because the quencher BHQ-2 would quench the fluorescence of the Mn-doped ZnS QDs. However, the addition of VEGF165 results in the displacement of the quencher-labelling strands (BHQ-2), thereby disrupting the FRET, and the appearance of AgNPs enhanced fluorescence on the Mn-doped ZnS QDs. What's more, the FL signal can be measured free of the interference from short-lived background by setting appropriate delay time and gate time.
2. Experimental section 2.1. Chemicals 3-Mercaptopropionic acid (MPA) was from Fluka. AgNO3, Zn (NO3)2, Na2S, MnCl2, ascorbic acid, trisodium citrate and sodium palmitate were obtained from Shanghai Reagent Company. Platelet derived growth factor (PDGF)-BB, VEGF165 and VEGF121 were purchased from Shanghai Sangon Biotechnology Co. All the synthetic oligonucleotides used in this study were purchased from Shanghai Sangon Biotechnology Co. The sequences of oligonucleotides are as follows: Apt-1(31 bp): 5′-SH-(CH2)6-TTG TCC CGT CTT CCA GAC AAG AGT GCA GGG A Biotin-3′ Apt-1(42 bp):5′-SH-(CH2)6-TTG TCC CGT CTT CCA GAC AAG AGT GCA GGG ATG ACA AAA AAA-Biotin-3′ Apt-1(51 bp): 5′-SH-(CH2)6-TTG TCC CGT CTT CCA GAC AAG AGT GCA GGG ATG ACA AAA AAA AAA AAA AAA-Biotin-3′ BHQ2-strands (12 bp): 5′-AAG ACG GGA CAA-BHQ2-3′. 2.2. Apparatus and characterization TEM images were recorded on a Shimadzu JEM-2010 CX with an accelerating voltage of 100 kV. UV–vis absorption spectra were obtained by using an UV-3600 spectrophotometer (Shimadzu). Fluorescence measurements were performed by using a Shimadzu RF-5301 PC fluorescence spectrometer. The time-resolved fluorescence spectrum was performed on an LS-55 fluorometer (Perkin-Elmer). The fluorescence lifetime of the Mn-doped ZnS QDs was measured with a FLS 920 time-resolved spectroscope (Edinberge). The fluorescence decays were analyzed in terms of the multi-exponential model as the sum of individual single exponential decays: n
I=
∑ αiexp( − i=1
t ) τi
τi are the decay times and αi are the amplitudes. The fractional contribution of each component to the steady-state intensity is described by: fi =
αiτi ∑j αjτj
and the amplitude-weighted lifetime is given by:
τ=
∑ αiτi i
2.3. Preparation of AgNPs
Scheme 1. Schematic illustration of the preparation of the silver nanoparticleenhanced TR-FL sensor based on the Mn-doped ZnS QDs and the determination of vascular endothelial growth factor 165.
AgNPs were prepared according to the literature (Qin et al., 2010). Briefly, a 40 mL aqueous solution containing ascorbic acid (6.0 10 4 M) and trisodium citrate (3.0 10 3 M) was adjusted to pH 9.0 by addition of 0.2 mol/L citrate acid or 0.1 mol/L NaOH solution. The 0.4 mL of 0.1 M aqueous solution of AgNO3 was added under gentle stirring in a 30 °C water bath. The reaction
D. Zhu et al. / Biosensors and Bioelectronics 74 (2015) 1053–1060
solutions were observed to change from being colorless to dark yellow. After 15 min, no further change in color took place, indicating the reactions were complete. The AgNPs suspension was washed three times with a water-and-ethanol mixture (5:4) and centrifuged. And then the prepared AgNPs were dispersed to Tris– HCl buffer (pH 7.4, 50 mM). 2.4. Synthesis of the MPA-capped Mn-doped ZnS QDs The Mn-doped ZnS QDs were prepared according to the literature (Zhu et al. 2014). Briefly, 2.0 g sodium palmitate was added to the mixture of 15 mL water and 5 mL ethanol, and the pellucid solution was obtained. Then 10 mL of aqueous solution containing 0.25 g of zinc nitrate and 0.02 g of manganese chloride, and 5 mL fresh Na2S solution was added in turn to the sodium palmitate solution. The typical molar ratio of Zn2 þ :S2 was 2:1 in our experiments. A microwave synthesis system (CEM Discover) was used for the preparation. The mixture was transferred to an 80 mL cylindrical digestion vessel under agitation. The reaction was maintained at 170 °C and 180 psi for 35 min under microwave irradiation (260 W). The Mn-doped ZnS QDs collected at the bottom of the container were dispersed in 50 mL of chloroform. After centrifugation at 6000 rpm, the transparent upper Mn-doped ZnS QDs solution was collected. The solution was dried in vacuum, and the Mn-doped ZnS QDs powders of about 100 mg were obtained. The Mn-doped ZnS QDs coated with the original alkyl ligands were dissolved in 20 mL of chloroform and treated with 200 μL of MPA. The mixture was shaken for 60 min under sonication. The MPAcoated Mn-doped ZnS QDs precipitate was isolated by centrifugation and decantation. Excessive MPA was further removed by washing the precipitate with chloroform for three times. The final precipitate was dried in vacuum, and then the Mn-doped ZnS QDs powders were obtained. The powders can be dispersed to Tris–HCl buffer (pH 7.4, 50 mM). 2.5. SA-functionalized AgNPs AgNPs were functionalized with SA according to the reported methods with minor modifications (Chen et al., 2010). Briefly, one milliliter of AgNPs was mixed with 200 μL of 0.5 mg/mL SA for 3 h with gentle shaking at 37 °C. The mixture was centrifuged at 15,000 rpm for 15 min, washed, and then re-dispersed and stored in Tris buffer solution (pH 7.4, 50 mM) containing 0.1% BSA. 2.6. Bioconjugation of the Mn-doped ZnS QDs with Apt-1 Immobilization of Apt-1 onto the Mn-doped ZnS QDs was performed via ligand exchange according to the literature (Hu et al., 2010) with slight modification. Briefly, 1.2 nmol Mn-doped ZnS QDs was mixed with 12 nmol Apt-1. An appropriately calculated amount of 10 Tris buffer (0.1 M TrisHCl, 1 M NaCl, pH 7.6) was then added to obtain a final buffer concentration of 1 Tris buffer (10 mM TrisHCl, 0.1 M NaCl, pH 7.6). The solution was mixed and stored at 25 °C overnight in order to allow the complete exchange of the thiol group MPA associated with at the surface of the Mn-doped ZnS QDs with the thiolated oligonucleotides. To reduce nonspecific interaction, the rest active sites were blocked with 1% BSA solution. After reaction overnight, the free nonconjugated Apt-1 were then removed by ultrafiltration. The above mixture was subjected to ultrafiltration using a 50,000 MW filter; after the lower phase was removed, the upper phase containing Mn-doped ZnS QDs bond with Apt-1 conjugations (Mn-QDs/Apt1) was decanted and diluted with Tris buffer (pH 7.4, 50 mM), and the solution was stored at 4 °C.
1055
2.7. Fabrication of TR-FL biosensor One milliliter of SA-functionalized AgNPs was mixed with 200 μL of 1 μM Mn-QDs/Apt-1 with gentle shaking at 37 °C for 1 h. Then 200 μL of 10 μM BHQ-2 strands was added to the above solution, and the mixture was shaken gently for 1 h at 37 °C. Subsequently, the above mixture was subjected to ultrafiltration using a 50,000 MW filter; after the lower phase was removed, the upper phase containing TR-FL biosensor was decanted and diluted with PBS buffer (pH 7.4, 50 mM), and the solution was stored at 4 °C in the dark until use. 2.8. Strategy for detection of vascular endothelial growth factor-165 (VEGF165) Various concentrations of VEGF165 in the range of 0.1–50 nM were added to the TR-FL sensor (50 μL) for 55 min at 37 °C in a micro quartz cuvette. Then, the TR-FL spectra were recorded, and the fluorescence intensities at 585 nm were determined. The timeresolved fluorescence measurements were carried out with delay time of 0.2 ms, gate time of 0.4 ms and cycle time of 20 ms. The experiments for the selectivity test were performed with the VEGF165 analogs PDGF-BB, and VEGF121 under identical conditions.
3. Results and discussion 3.1. Characterization of the SA-functionalized AgNPs (SA-AgNPs) The AgNPs enhanced fluorescence effect reflects the scattering contribution, with relative proportions depending on the size, shape, and composition of the NPs (Lakowicz, 2005). Lakowicz et al. (2008) proved that the diameter of AgNPs is optimal from 50 to 70 nm to provide the strength distribution of the enhanced electromagnetic field of localized surface plasmon resonance and the depression of competitive quenching. The TEM image of Fig. 1 (A) shows generally spherical and proportioned AgNPs with a diameter at 5073 nm, a size that has been shown to be suitable for MEF and the radiating plasmon model. In Fig. 1(B), the AgNPs show an absorption peak at 445 nm in the plasmon resonance band. After SA was attached to the AgNPs, an obvious absorption peak occurs at 280 nm, which is a characteristic of the SA, indicating the successful binding between SA and the AgNPs. The zeta potential of the AgNPs decreased from 23.2 to 24.9 mV (n ¼3) when binding with SA occurred. The decrease of zeta potential is consistent with the binding of oligonucleotides since oligonucleotides are negatively charged. 3.2. Characterization of the Mn-doped ZnS QDs/aptamers (Apt-1). The Mn-doped ZnS QDs shows good monodispersity with particle sizes ranging from 4.5 to 5.5 nm as shown in Fig S1 (see the Supplementary information). In Fig. 2(A), the Mn-doped ZnS QDs show an absorption peak at 315 nm, and the two photoluminescence (PL) peaks, a strong peak around 585 nm from an internal electronic transition of the Mn (4T1–6A1), and a weak blue peak around 420 nm from the defect related emission (Suyver et al., 2001). After Apt-1 was bond onto the Mn-doped ZnS QDs, an obvious absorption peak occurs at 260 nm as shown in Fig. 2(C), which is a characteristic of the DNA strand, indicating the successful binding between Apt-1 and the Mn-doped ZnS QDs. In Fig. 2(B), the shape and intensity of the emission peak around 585 nm is almost the same before and after the modification of the QDs with Apt-1, suggesting good stability of covalent conjugation between the Mn-doped ZnS QDs and Apt-1, while the defect related emission around 420 nm is greatly suppressed. In Fig. 3(D),
1056
D. Zhu et al. / Biosensors and Bioelectronics 74 (2015) 1053–1060
Fig.1. (A) TEM image of the prepared AgNPs with an average diameter of 50 nm. (B) UV absorption spectra of the AgNPs and the AgNPs immobilized with SA.
the peaks at 1553 and 3428 cm 1 correspond to the stretching vibrations of C ¼ O and O–H, respectively, which demonstrates the carboxylic group capped on the Mn-doped ZnS QDs. The characteristic absorption peaks of Apt-1 at 1642, 1081, and 991 cm 1 are preserved after the formation of the Mn-doped ZnS QDs/Apt-1 conjugates, which further proves the successful connection. The
amount of conjugated Apt-1 is also estimated to be 8 Apt-1 per the Mn-doped ZnS QD (see the Supplementary information, Section S1).
Fig. 2. (A) Fluorescence and UV–vis absorption spectrum of the Mn-doped ZnS QDs. (B) Normalized fluorescence spectra of the Mn-doped ZnS QDs and the Mn-doped ZnS QDs/Apt-1. (C) UV–vis absorption spectra of the Mn-doped ZnS QDs and the Mn-doped ZnS QDs/Apt-1. (D) FT-IR spectra of the Mn-doped ZnS QDs and the Mn-doped ZnS QDs/Apt-1.
D. Zhu et al. / Biosensors and Bioelectronics 74 (2015) 1053–1060
1057
Fig. 3. (A) Emission spectrum (a) of donor (the Mn-doped ZnS QDs), the absorption spectrum (b) of acceptor (BHQ-2). (B) Loss of the Mn-doped ZnS fluorescence with increasing number of BHQ-2 quench oligonucleotides. (C) ΔF/F0 at different lengths of Apt-1; The concentration of VEGF165 was 6 nM. (ΔF/F0 ¼fluorescence intensity change at a specified concentration of VEGF165/fluorescence intensity with absence of VEGF165). (D) ΔF/F0 at different dilution ratio of SA-functionalized AgNPs for 55 min at 37 °C. (E) Fluorescence changes of the Mn-doped ZnS QDs at λ¼ 585 nm upon different incubation time in the treatment of the TR-FL sensor with VEGF165, 6 nM. Spectra were recorded at time intervals of 5 min. (F) Gel-shift analysis of the TR-FL probe–VEGF165 complex, as visualized with ethidium bromide. Lane 1: the Apt-1 bond with the Mndoped ZnS QDs; Lane 2: TR-FL sensor based on AgNPs–SA–Apt-1/BHQ-Mn-QDs complex; Lane 3: the TR-FL sensor/VEGF165 complex; Lane 4: DNA marker; Lane 5: BHQ-2 quencher-labelling strands.
3.3. Optimization of the Ag-enhanced TR-FL sensor The emission spectrum of the donor (the Mn-doped Zns QDs) and the absorption spectrum of the acceptor (BHQ-2) are shown in Fig. 3(A). BHQ-2 has a broad absorption band around 580 nm that matches well with the Mn (4T1-6A1) emission centered at 585 nm. FRET response relies on not only the overlap between the donor emission and acceptor absorption spectrum, but also the
donor–acceptor distances (R) and the number of quencher molecules interacting with donor. Fig. 3(B) shows a plot of the relative fluorescence at 585 nm emission versus increasing amounts of added BHQ-2 Apt (reported as BHQ-2 Apt to QD ratio, n) through hybridization with Apt-1/Mn-doped ZnS QDs, which indicates that saturation occurs at BHQ-2 to QD ratio of 10 with an ultimate quenching efficiency of 90%. The hybridization efficiency of BHQ-2 Apt with Apt-1 QDs can be calculated to 89% of the
1058
D. Zhu et al. / Biosensors and Bioelectronics 74 (2015) 1053–1060
Fig. 4. (A) Steady-state fluorescence spectra of BSA (a), VEGF165 (b), Apt-1 (c) a mixture of BSA, the Mn-doped ZnS QDs, Apt-1 and VEGF165 (d), and time-resolved fluorescence spectra of the mixture (e) in PBS solution (pH ¼7.4). (B) Comparison of ΔF/F0 of AgNPs-based TR-FL sensor and bare TR-FL sensor at different concentrations of VEGF165. (C) Fluorescence decay of AgNPs-based TR-FL sensor and bare TR-FL sensor at 8 nM VEGF165.
immobilized BHQ-2-labelling strands by means of measuring the absorbance of the BHQ-2-labelling strands solution before hybridization and that of the collected ultrafiltration solution after hybridization. So, Apt-1-binding sites on the quantum-dot surface occurs at 8 BHQ-2 Apt per dot with an ultimate quenching efficiency of 90%. From the basis of the equations (Clapp et al., 2004.; Goldman et al., 2005)
E=
nR 06 nR 06 + R6
R 0 = 0.211[k 2nd−4 YJ ]1/6 where E is the energy transfer efficiency, n is the number of quencher molecules, R0 is the critical transfer distance, R is actual distance, k is the orientation factor (k2 ¼2/3 for random collisions), nd is the index of refraction (n ¼1.33 in an aqueous medium), Y is the quantum yield of the Mn-doped Zns QDs (Y¼0.086) and the overlap integral J is obtained by a program that overlays the normalized emission of the donor and absorption spectra of the acceptor, J ¼2.58 1015 cm 1 M 1 nm4, the critical transfer distance R0 was calculated as 3.91 nm, and the separation distance (r) between the QD center and the acceptor was estimated to 3.83 nm.
Using the estimate for the donor–acceptor separation distance and the QD finite radii extracted from size measurements (5 nm) using transmission electron microscopy (TEM) and scattering techniques, we infer a distance between the QD surface and the BHQ-2 of 12–14 Å. In such close distance, the Mn-doped ZnS QDs could contact with BHQ-2 that contributes more of quenching mechanism than FRET because electron withdrawing groups and large conjugated groups of BHQ-2 could quench fluorescence of QDs, and such quenching by other similar strong electron transfer agents has been reported (Nieto et al., 2004; Tu. et al., 2008; Liu et al., 2010). The MEF is highly distance-dependent, and the length of the Apt-1 bond with the Mn-doped ZnS QDs, a single stranded DNA, can affect MEF effect. When the BHQ-2-labelling strands are dissociated from the Apt-1/Mn-doped ZnS QDs by VEGF165, the distance between the Mn-doped ZnS QDs and AgNPs is also changed, and this can be controlled by the length of the Apt-1. The Apt-1 of different lengths(31 bp, 42 bp and 51 bp) are evaluated as shown in Fig. 3(C), and AgNPs show better fluorescence enhancement effect on the Mn-doped ZnS QDs with 42 bp Apt-1. The concentration of SA-functionalized AgNPs was also optimized as shown in Fig. 3(D), a significant decrease of the ΔF/F0 ratio is observed with the increase in dilution ratio between 2 and
D. Zhu et al. / Biosensors and Bioelectronics 74 (2015) 1053–1060
8. When the SA-functionalized AgNPs was diluted with 1 mg/mL BSA by 2-fold (the volume ratio of BSA to SA-functionalized was 1:1), the ΔF/F0 ratio was almost equal to that of undiluted SAfunctionalized AgNPs. Therefore, SA-functionalized AgNPs with 2-fold dilution ratio was selected for the next experiments. In addition, the incubation time between TR-FL sensor and VEGF165 was estimated. Fig. 3(E) shows the fluorescence changes upon different incubation time in the treatment of the TR-FL sensor with VEGF165 at a concentration of 6 nM. A time-dependent increase in the fluorescence is observed, consistent with the dissociation of the BHQ-2 quencher-labelling strands from the Mndoped ZnS QDs, due to the formation of the aptamer–VEGF complex. The increases of fluorescence reached a saturation value after ca. 55 min and represent the kinetics of formation of the equilibrium state of the sensor–VEGF165 complex. Therefore, 55 min incubation time was chosen for the next experiments. Gel-shift confirmed the formation of the TR-FL sensor and the sensor–VEGF165 complex (Fig. 3(F)). Compared with the Apt-1 bond with the Mn-doped ZnS QDs (Lane 1) and BHQ-2 quencherlabelling strands (Lane 5), the TR-FL sensor complex band presented a much slower migration rate (Lane 3). After incubation with VEGF165, the sensor–VEGF165 complex also showed slower migration than the sensor (Lane 3). 3.4. Superiority of AgNPs-enhanced TR-FL sensor Fig. 4(A) compares the fluorescence spectra of a mixture of the Mn-doped ZnS QDs, BSA, DNA and VEGF165 in the steady-state PL mode and time-resolved PL technique. In the steady-state fluorescence spectra, the emissions of both BSA and VEGF165 from 400 to 620 nm interfered with the emission peaks at 420 and 585 nm of the Mn-doped ZnS QDs (curves a and b in Fig. 4(A)). The steadystate fluorescence interferences would inevitably lead to a serious limitation on analytical accuracy and testing sensitivity. In sharp contrast, the short-lived fluorescence of both BSA and VEGF165 from 400 to 620 nm disappeared, and only intense emission originating from the Mn-doped ZnS QDs was detected in the TR-FL spectra by setting up appropriate delay time and gate time. The MEF ability of TR-FL sensor was estimated by comparing the fluorescence intensity change (ΔF/F0) of the AgNPs-based TRFL sensor with bare sensor as shown in Fig. 4(B). The ΔF/F0 ratio of the AgNPs-based TR-FL sensor is much higher in the corresponding concentration range of VEGF165, which can be attributed to the fluorescence enhancement effect of the Mn-doped ZnS QDs based
1059
on AgNPs. The fluorescence intensity increases by 3.5 times when the concentration of VEGF165 is 16 nM. The MEF process is further confirmed by fluorescence decay measurements as shown in Fig. 4(C). The intensity decay of the bare TR-FL sensor was fitted with a multiple exponential with three lifetime components of 125 μs (11%), 450 μs (17% ) and 1330 μs (72%) with the total amplitude-weighted lifetime being 1.28 ms. The intensity decay of AgNPs-based TR-FL sensor was also fitted with a multiple exponential with three lifetimes of 67 μs (9%), 319 μs (12%) and 1130 μs (79%). Hence, the amplitudeweighted fluorescence lifetime of the AgNPs-based TR-FL sensor was calculated to be 1.03 ms. The reduction in the fluorescence lifetime of the Mn-doped ZnS QDs despite an increase in the fluorescence intensity is a characteristic of the fluorophore plasmon coupling effect (Zhang et al., 2009; Lakowicz 2001). 3.5. Assay performance on VEGF165 Fig. 5(A) depicts the resulting calibration curve corresponding to the fluorescence change of the Mn-doped ZnS QDs in the presence of different concentrations of VEGF165. The ΔF/F0 ratio increased with the VEGF165 concentration, and a linear relationship (r ¼0.995) between ΔF/F0 and VEGF165 concentration in the range of 0.1–16 nM was observed. Based on the averaged values of experiments, the concentration of VEGF165 was estimated by using the following equation:
ΔF /F0 = 0.1946 + 0.4167CVEGF165 The precision for seven repeated measurements of 4 nM VEGF165 was 3.3% (RSD), and the detection limit (3δ) was 0.08 nM. The specificity of the TR-FL sensor toward VEGF165 was also investigated in the presence of other cancer-related proteins such as platelet-derived growth factor (PDGF)-BB and VEGF121. As shown in Fig. 5(B), the TR-FL sensor produced a significant signal change with VEGF165 at 3 nM, while signal changes from 3 nM PDGF-BB and 3 nM VEGF121 were not significant. The presence of 10-fold human serum (10% serum), 50% serum and 100% serum did not show any interference. Then, VEGF165 in 100% human serum was also tested to examine whether the complex matrix affected the AgNPs-enhanced TR-FL sensor. The fluctuation of 1.8% was weak compared with VEGF165 as shown in Fig. 5(B), and the recovery of 92–106% of spiked VEGF165 in 100% human serum were obtained as shown in Table S1 in Supplementary information.
Fig. 5. (A) Response curve of ΔF/F0 to different VEGF165 concentrations, where ΔF/F0 represents the fluorescence intensity change of the modified Mn-doped ZnS QDs at λ ¼585 nm, in the absence and presence of VEGF165, respectively. (Error bars were evaluated from N ¼3 experiments.). (B) Specificity of the TR-FL for VEGF165. The concentrations of molecules are physiologically relevant values: 3 nM (VEGF165, PDGF-BB and VEGF121), 10% serum, 50% serum and 100% serum (error bars were evaluated from N ¼3 experiments.)
1060
D. Zhu et al. / Biosensors and Bioelectronics 74 (2015) 1053–1060
4. Conclusions A silver nanoparticles-enhanced TR-FL sensor based on the Mndoped ZnS QDs has been successfully developed for the detection of VEGF165. The TR-FL sensor shows good selectivity and sensitivity toward the target species in the presence of other proteins or in the complex matrix. Compared with the bare FL sensor, the AgNPs-enhanced TR-FL sensor clearly shows remarkable advantages in terms of sensitivity. What is more, the TR-FL sensor also offers a signal with high signal-to-noise ratio in luminescent biodetection as compared with conventional FL. The sensitivity was increases 11-fold compared with bare FL sensor with the detection limit of 0.08 nM. In addition, the sensor provided a wide range of linear detection from 0.1 nM to 16 nM. Furthermore, the principle of this TR-FL sensor can be extended to other proteins or new biomarkers for disease diagnosis and fundamental research.
Acknowledgments We greatly appreciate the National Natural Science Foundation of China for the financial support (21205064). This work has also been supported by Fund of State Key Laboratory of Analytical Chemistry for Life Science (SKLACLS1208). We also greatly appreciate “A Project Funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions” (PAPD).
Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.bios.2015.08.005.
References Akimov, A.V., Mukherjee, A., Yu, C.L., Chang, D.E., Zibrov, A.S., Hemmer, P.R., Park, H., Lukin, M.D., 2007. Nature 450, 402–406. Bock, L.C., Griffin, L.C., Latham, J.A., Vermaas, E.H., Toole, J.J., 1992. Nature 355, 564–566. Chen, L.Q., Xiao, S.J., Peng, W.T., Ling, J., Li, Y.F., Huang, C.Z., 2010. J. Phys. Chem. B 114, 3655–3659. Cho, E.J., Lee, J.-W., Ellington, A.D., 2009. Annu. Rev. Anal. Chem. 2, 241–264. Clapp, A.R., Medintz, I.L., Mauro, J.M., Fisher, B., Bawendi, M.G., Mattoussi, H., 2004. J. Am. Chem. Soc. 126, 301–310. Eis, P.S., Millar, D.P., 1993. Biochemistry 32, 13852–13860. Fang, X., Tan, W., 2010. Acc. Chem. Res. 43, 48–57. Fu, Y., Zhang, J., Lakowicz, J.R., 2009. Chem. Commun., 313–315. Geddes, C.D., Lakowicz, J.R., 2002. J. Fluoresc. 12, 121–129. Goldman, E.R., Medintz, I.L., Whitley, J.L., Hayhurst, A., Clapp, A.R., Tetsuo Uyeda, H.,
Deschamps, J.R., Lassman, M.E., Mattoussi, H., 2005. J. Am. Chem. Soc. 127, 6744–6751. He, Y., Wang, H.-F., Yan, X.-P., 2008. Anal. Chem. 80, 3832–3837. Hirata, J., Jong, C.F., Van dongen, M.M., Buijs, J., Ariese, F., Irth, H., Gooijer, C., 2004. Anal. Chem. 76, 4292–4298. Hiroshi, Y., Masaki, W., Yukihiro, O., Manabu, T., Noritada, K., Yoshitaka, M., Hirofumi, N., Yoshinobu, B., Shuji, H., 2012. Biomaterials 33, 2177–2186. Ho, Y.P., Kung, M.C., Yang, S., Wang, T.H., 2005. Nano Lett. 5, 1693–1697. Huang, F., Lerner, E., Sato, S., Amir, D., Haas, E., Fersht, A.R., 2009. Biochemistry 48, 3468–3476. Hu, M., He, Y., Song, S.P., Yan, J., Lu, H.-T., Weng, L.-X., Wang., L.-H., Fan, C.H., 2010. Chem. Commun. 46, 6126–6128. Kishore, M., Krishnamoorthy, G., Udgaonkar, J.B., 2013. Biochemistry 52, 9482–9496. Laitala, V., Hemmila1, I., 2005. Anal. Chem. 77, 1483–1487. Lakowicz, J.R., 2001. Anal. Biochem. 298, 1–24. Lakowicz, J.R., 2005. Anal. Biochem. 337, 171–194. Lakowicz, J.R., Ray, K., Chowdhury, M., Szmacinski, H., Fu, Y., Zhang, J., Nowaczyk, K., 2008. Analyst 133, 1308–1346. Liu, J.F., Bao, C.Y., Zhong, X.H., Zhao, C.C., Zhu, L.Y., 2010. Chem. Commun. 46, 2971–2973. Liu, Y.S., Zhou, S.Y., Tu, D.T., Chen, Z., Huang, M.D., Zhu, H.M., Ma, E., Chen, X.Y., 2012. J. Am. Chem. Soc. 134, 15083–15090. Mackowski, S., Wörmke, S., Maier, A.J., Brotosudarmo, T.H.P., Harutyunyan, H., Hartschuh, A., Govorov, A.O., Scheer, H., Bräuchle, C., 2008. Nano Lett. 8, 558–564. Medintz, I.L., Tetsuo Uyeda, H., Goldman, E.R., Mattoussi, H., 2005. Nat. Mater. 4, 435–446. Michalet, X., Pinaud, F.F., Bentolila, L.A., Tsay, J.M., Doose, S., Li, J.J., Sundaresan, G., Wu, A.M., Gambhir, S.S., Weiss, S., 2005. Science 307, 538–544. Muskens, O.L., Giannini, V.J., Sanchez-Gil, A., Gomez Rivas, J., 2007. Nano Lett. 7, 2871–2875. Nieto., S., Santana., A., Herna'ndez., S.P., Lareau., R., Chamberlain., R.T., Castro., M.E., 2004. Proc. SPIE – Int. Soc. Opt. Eng. 5403, 256–260. Norris, D.J., Efros, A.L., Erwin, S.C., 2008. Science 319, 1776–1779. Ozers, M.S., Marks, B.D., Gowda, K., Kupcho, K.R., Ervin, K.M., Rosier, T.D., Qadir, N., Eliason, H.C., Riddle, S.M., Shekhani, M.S., 2007. Biochemistry. 46 683-69. Qin, Y.Q., Ji, X.H., Jing, J., Liu, H., Wu, H.L., Yang, W.S., 2010. Colloids Surf. A: Physicochem. Eng. Asp. 372, 172–176. Sandin, P., Lincoln, P., Albinsson, B., 2008. J. Phys. Chem. C 112, 13089–13094. Seveus, L., Vaisala, M., Syrjanen, S., Sandberg, M., Kuusisto, A., Harju, R., Salo, J., Hemmila, I., Kojola, H., Soini, E., 1992. Cytometry 13, 329–338. Sokolov, K., Chumanov, G., Cotton, T.M., 1998. Anal. Chem. 70, 3898–3905. Suyver, J.F., Wuister, S.F., Kelly, J.J., Meijerink, A., 2001. Nano.Lett. 1, 429–433. Tu., R.Y., Liu., B.H., Wang., Z.Y., Gao., D.M., Wang., F., Fang., Q.L., Zhang, Z.P., 2008. Anal. Chem. 80, 3458–3465. Wang, H.-F., Li, Y., Wu, Y.-Y., He., Y., Yan, X.-P., 2010. Chem. Eur. J. 16, 12988–12994. Wu, P., He, Y., Wang, H.-F., Yan, X.-P., 2010. Anal. Chem. 82, 1427–1433. Wu, P., Yan, X.-P., 2013. Chem. Soc. Rev. 42, 5489–5521. Xie, R.G., Battaglia, D., Peng, X.G., 2007. J. Am. Chem. Soc. 129, 15432–15433. Xie, R.G., Peng, X.G., 2008. Angew. Chem. Int. Ed. 47, 7677–7680. Zhang, C.Y., Yeh, H.C., Kuroki, M.T., Wang, T.H., 2005. Nat. Mater. 4, 826–831. Zhang, J., Fu, Y., Chowdhury, M.H., Lakowicz, J.R., 2007. Nano Lett. 7, 2101–2107. Zhang, J., Fu, Y., Lakowicz, J.R., 2009. J. Phys. Chem. C 113, 19404–19410. Zhelev, Z., Bakalova, R., Ohba, H., Jose, R., Imai, Y., Baba, Y., 2006. Anal. Chem. 78, 321–330. Zhu, A.X., Duda, D.G., Sahani, D.V., Jain, R.K., 2011. HCC and angiogenesis: possible targets and future directions. Nat. Rev. Clin. Oncol. 8, 292–301. Zhu, D., Li, W., Ma, L., Lei, Y., 2014. RSC Adv. 4, 9372–9378. Zhu, D., Li., W., Wen, H.-M., Chen, Q., Ma, L., Hu, Y., 2014. Anal. Methods 6, 7489–7495.
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
Silver nanoparticles-enhanced time-resolved fluorescence sensor for VEGF165 based on Mn-doped ZnS quantum dots Dong Zhu n, Wei Li n, Hong-Mei Wen, Sheng Yu, Zhao-Yi Miao, An Kang, Aihua Zhang School of Pharmacy, Nanjing University of Chinese Medicine, Nanjing 210023, PR China
art ic l e i nf o
a b s t r a c t
Article history: Received 28 May 2015 Received in revised form 4 August 2015 Accepted 5 August 2015 Available online 6 August 2015
A silver nanoparticles (AgNPs)-enhanced time-resolved fluorescence (TR-FL) sensor based on long-lived fluorescent Mn-doped ZnS quantum dots (QDs) is developed for the sensitive detection of vascular endothelial growth factor-165 (VEGF165), a predominant cancer biomarker in cancer angiogenesis. The aptamers bond with the Mn-doped ZnS QDs and the BHQ-2 quencher-labelling strands hybridized in duplex are coupled with streptavidin (SA)-functionalized AgNPs to form the AgNPs-enhanced TR-FL sensor, showing lower fluorescence intensity in the duplex state due to the fluorescence resonance energy transfer (FRET) between the Mn-doped ZnS QDs and quenchers. Upon the addition of VEGF165, the BHQ-2 quencher-labelling strands of the duplex are displaced, leading to the disruption of the FRET. As a result, the fluorescence of the Mn-doped QDs within the proximity of the AgNPs is recovered. The FL signal can be measured free of the interference of short-lived background by setting appropriate delay time and gate time, which offers a signal with high signal-to-noise ratio in photoluminescent biodetection. Compared with the bare TR-FL sensor, the AgNPs-based TR-FL sensor showed a huge improvement in fluorescence based on metal-enhanced fluorescence (MEF) effect, and the sensitivity increased 11-fold with the detection limit of 0.08 nM. In addition, the sensor provided a wide range of linear detection from 0.1 nM to 16 nM. & 2015 Elsevier B.V. All rights reserved.
Keywords: Metal-enhanced fluorescence Time-resolved Fluorescence resonance energy transfer Mn-doped ZnS quantum dots
1. Introduction The exploration of time-resolved fluorescence (TR-FL) biosensing systems is a topic of considerable interest (Eis and Millar, 1993; Sandin et al., 2008; Kishore et al., 2013; Huang et al., 2009; Ozers et al., 2007;) since TR-FL assay brings the advantage of near zero background signal from the time-resolved (TR) technique, provides an excellent solution to eliminate the interference of short-lived autofluorescence from cells, tissues, and assay multiwell plate and thus offers remarkably high sensitivity as compared with the conventional fluorometry (Liu et al., 2012). The probe is key in this technique, and the ideal probe should have long fluorescence life time, large Stoke's shift and narrow emission peaks. Currently, the TR-FL bio-probes are mainly lanthanide chelates and their corresponding nanoparticles (Liu et al., 2012; Laitala and Hemmila1, 2005; Hirata et al., 2004) owing to their long fluorescence lifetime (as long as 20–1000 μs). However, the development of lanthanide probe is limited by their shortcomings such as weak luminescence and photobleaching (Seveus et al., n
Corresponding authors. Fax: þ 86 2585811839. E-mail addresses: [email protected] (D. Zhu), [email protected] (W. Li). http://dx.doi.org/10.1016/j.bios.2015.08.005 0956-5663/& 2015 Elsevier B.V. All rights reserved.
1992), and thus, development of a fluorescent probe with stronger luminescence and better photostability is critical for the further application of TR-FL in biological studies. Quantum dots (QDs) have emerged as the desirable candidates for applications in biomolecular sensing (Zhang et al., 2005; Ho et al., 2005; Zhelev et al., 2006) and cellular imaging (Michalet et al., 2005; Hiroshi et al., 2012), mainly due to their unique luminescent quality such as strong photoluminescence and good photostability. Cadmium chalcogenide QDs are widely explored. However, their fluorescence lifetime is similar to that of the background autofluorescence (Michalet et al., 2005), which obstructs the application of these traditional QDs in the time-resolved fluorometry. Recently, Mn-doped ZnS QDs have attracted increasing attention (Wu and Yan, 2013; He et al., 2008; Zhu et al., 2014; Wu et al., 2010; Wang et al., 2010) because of their charming characteristics such as strong photoluminescence, high resistance to photobleaching and especially long fluorescence life time (as long as 1–2 ms), from the sharp 4T1(4G)-6A1(6S) emission of the Mn2 þ (Norris et al., 2008; Michalet et al., 2005; Xie et al., 2007; Xie and Peng , 2008). Some researchers have successfully employed the Mn-doped ZnS QDs for TR fluorescence detection (He et al., 2008; Zhu et al., 2014; Wu et al., 2010; Wang et al., 2010). However, the fluorescence lifetime of ms scale about Mn2 þ emission is too long for the only a few nanoseconds lifetime from
1054
D. Zhu et al. / Biosensors and Bioelectronics 74 (2015) 1053–1060
background autofluorescence (Michalet et al., 2005) and such a long decay time results in a low photon count rate, which may lead to a relatively low quantum yield and limit the improvement of detection sensitivity (Zhang et al., 2009). Therefore, development of a strategy that can increase the emission intensity and simultaneously shorten the fluorescence lifetime is critical for the further application of TR-FL detection in biological studies. Lately, the metal-enhanced fluorescence (MEF) effect, based on the surface plasmon resonance (SPR) of metallic nanostructures, seems to be very promising for solving the problem (Lakowicz, 2001). The SPR of metallic nanostructures, especially the Ag nanoparticles (NPs), can enhance the local electromagnetic field surrounding them and finally lead to the photoluminescence increase and lifetime decrease of nearby fluorophores (Geddes and Lakowicz, 2002; Sokolov et al., 1998). Most MEF materials developed thus far have been targeted towards enhancing fluorescence from organic fluorophores (Medintz et al., 2005; Mackowski et al., 2008; Muskens et al., 2007; Zhang et al., 2007), while few has been developed for inorganic fluorescent species, such as QDs (Akimov et al., 2007; Fu et al., 2009), and more importantly, there is very limited progress in the development of biomolecular sensing by this MEF approach. In this paper, based on the MEF principle, an AgNPs-enhanced TR-FL biosensor is developed using the Mn-doped ZnS QDs bond with aptamers that are hybridized with quencher BHQ-2 labelling strands for the detection of vascular endothelial growth factor-165 (VEGF165), a predominant cancer biomarker in cancer angiogenesis (Zhu et al., 2011). Aptamers are short, single-stranded oligonucleotides (RNA or DNA o 100 nt) or peptides (o 100-mer) generated from an in vitro method known as SELEX (systematic evolution of ligands by exponential enrichment) (Fang and Tan, 2010), which have high affinity, specificity and stability, therefore, they are regarded as alternative reagents to antibodies (Cho et al., 2009). Furthermore, aptamers change structures from an elongated metastable structure to a uniquely folded stable structure upon interaction with targets (Bock et al., 1992). As illustrated in Scheme 1, the aptamers modified with biotin on the 3′ end and the Mn-doped ZnS QDs on the 5′ end and BHQ-2 quencher-labelling strands which are hybridized in duplex are bond with streptavidin (SA)-functionalized AgNPs to form an AgNPs-enhanced TR-FL sensor. The TR-FL sensor shows lower fluorescence
in the duplex sate because the quencher BHQ-2 would quench the fluorescence of the Mn-doped ZnS QDs. However, the addition of VEGF165 results in the displacement of the quencher-labelling strands (BHQ-2), thereby disrupting the FRET, and the appearance of AgNPs enhanced fluorescence on the Mn-doped ZnS QDs. What's more, the FL signal can be measured free of the interference from short-lived background by setting appropriate delay time and gate time.
2. Experimental section 2.1. Chemicals 3-Mercaptopropionic acid (MPA) was from Fluka. AgNO3, Zn (NO3)2, Na2S, MnCl2, ascorbic acid, trisodium citrate and sodium palmitate were obtained from Shanghai Reagent Company. Platelet derived growth factor (PDGF)-BB, VEGF165 and VEGF121 were purchased from Shanghai Sangon Biotechnology Co. All the synthetic oligonucleotides used in this study were purchased from Shanghai Sangon Biotechnology Co. The sequences of oligonucleotides are as follows: Apt-1(31 bp): 5′-SH-(CH2)6-TTG TCC CGT CTT CCA GAC AAG AGT GCA GGG A Biotin-3′ Apt-1(42 bp):5′-SH-(CH2)6-TTG TCC CGT CTT CCA GAC AAG AGT GCA GGG ATG ACA AAA AAA-Biotin-3′ Apt-1(51 bp): 5′-SH-(CH2)6-TTG TCC CGT CTT CCA GAC AAG AGT GCA GGG ATG ACA AAA AAA AAA AAA AAA-Biotin-3′ BHQ2-strands (12 bp): 5′-AAG ACG GGA CAA-BHQ2-3′. 2.2. Apparatus and characterization TEM images were recorded on a Shimadzu JEM-2010 CX with an accelerating voltage of 100 kV. UV–vis absorption spectra were obtained by using an UV-3600 spectrophotometer (Shimadzu). Fluorescence measurements were performed by using a Shimadzu RF-5301 PC fluorescence spectrometer. The time-resolved fluorescence spectrum was performed on an LS-55 fluorometer (Perkin-Elmer). The fluorescence lifetime of the Mn-doped ZnS QDs was measured with a FLS 920 time-resolved spectroscope (Edinberge). The fluorescence decays were analyzed in terms of the multi-exponential model as the sum of individual single exponential decays: n
I=
∑ αiexp( − i=1
t ) τi
τi are the decay times and αi are the amplitudes. The fractional contribution of each component to the steady-state intensity is described by: fi =
αiτi ∑j αjτj
and the amplitude-weighted lifetime is given by:
τ=
∑ αiτi i
2.3. Preparation of AgNPs
Scheme 1. Schematic illustration of the preparation of the silver nanoparticleenhanced TR-FL sensor based on the Mn-doped ZnS QDs and the determination of vascular endothelial growth factor 165.
AgNPs were prepared according to the literature (Qin et al., 2010). Briefly, a 40 mL aqueous solution containing ascorbic acid (6.0 10 4 M) and trisodium citrate (3.0 10 3 M) was adjusted to pH 9.0 by addition of 0.2 mol/L citrate acid or 0.1 mol/L NaOH solution. The 0.4 mL of 0.1 M aqueous solution of AgNO3 was added under gentle stirring in a 30 °C water bath. The reaction
D. Zhu et al. / Biosensors and Bioelectronics 74 (2015) 1053–1060
solutions were observed to change from being colorless to dark yellow. After 15 min, no further change in color took place, indicating the reactions were complete. The AgNPs suspension was washed three times with a water-and-ethanol mixture (5:4) and centrifuged. And then the prepared AgNPs were dispersed to Tris– HCl buffer (pH 7.4, 50 mM). 2.4. Synthesis of the MPA-capped Mn-doped ZnS QDs The Mn-doped ZnS QDs were prepared according to the literature (Zhu et al. 2014). Briefly, 2.0 g sodium palmitate was added to the mixture of 15 mL water and 5 mL ethanol, and the pellucid solution was obtained. Then 10 mL of aqueous solution containing 0.25 g of zinc nitrate and 0.02 g of manganese chloride, and 5 mL fresh Na2S solution was added in turn to the sodium palmitate solution. The typical molar ratio of Zn2 þ :S2 was 2:1 in our experiments. A microwave synthesis system (CEM Discover) was used for the preparation. The mixture was transferred to an 80 mL cylindrical digestion vessel under agitation. The reaction was maintained at 170 °C and 180 psi for 35 min under microwave irradiation (260 W). The Mn-doped ZnS QDs collected at the bottom of the container were dispersed in 50 mL of chloroform. After centrifugation at 6000 rpm, the transparent upper Mn-doped ZnS QDs solution was collected. The solution was dried in vacuum, and the Mn-doped ZnS QDs powders of about 100 mg were obtained. The Mn-doped ZnS QDs coated with the original alkyl ligands were dissolved in 20 mL of chloroform and treated with 200 μL of MPA. The mixture was shaken for 60 min under sonication. The MPAcoated Mn-doped ZnS QDs precipitate was isolated by centrifugation and decantation. Excessive MPA was further removed by washing the precipitate with chloroform for three times. The final precipitate was dried in vacuum, and then the Mn-doped ZnS QDs powders were obtained. The powders can be dispersed to Tris–HCl buffer (pH 7.4, 50 mM). 2.5. SA-functionalized AgNPs AgNPs were functionalized with SA according to the reported methods with minor modifications (Chen et al., 2010). Briefly, one milliliter of AgNPs was mixed with 200 μL of 0.5 mg/mL SA for 3 h with gentle shaking at 37 °C. The mixture was centrifuged at 15,000 rpm for 15 min, washed, and then re-dispersed and stored in Tris buffer solution (pH 7.4, 50 mM) containing 0.1% BSA. 2.6. Bioconjugation of the Mn-doped ZnS QDs with Apt-1 Immobilization of Apt-1 onto the Mn-doped ZnS QDs was performed via ligand exchange according to the literature (Hu et al., 2010) with slight modification. Briefly, 1.2 nmol Mn-doped ZnS QDs was mixed with 12 nmol Apt-1. An appropriately calculated amount of 10 Tris buffer (0.1 M TrisHCl, 1 M NaCl, pH 7.6) was then added to obtain a final buffer concentration of 1 Tris buffer (10 mM TrisHCl, 0.1 M NaCl, pH 7.6). The solution was mixed and stored at 25 °C overnight in order to allow the complete exchange of the thiol group MPA associated with at the surface of the Mn-doped ZnS QDs with the thiolated oligonucleotides. To reduce nonspecific interaction, the rest active sites were blocked with 1% BSA solution. After reaction overnight, the free nonconjugated Apt-1 were then removed by ultrafiltration. The above mixture was subjected to ultrafiltration using a 50,000 MW filter; after the lower phase was removed, the upper phase containing Mn-doped ZnS QDs bond with Apt-1 conjugations (Mn-QDs/Apt1) was decanted and diluted with Tris buffer (pH 7.4, 50 mM), and the solution was stored at 4 °C.
1055
2.7. Fabrication of TR-FL biosensor One milliliter of SA-functionalized AgNPs was mixed with 200 μL of 1 μM Mn-QDs/Apt-1 with gentle shaking at 37 °C for 1 h. Then 200 μL of 10 μM BHQ-2 strands was added to the above solution, and the mixture was shaken gently for 1 h at 37 °C. Subsequently, the above mixture was subjected to ultrafiltration using a 50,000 MW filter; after the lower phase was removed, the upper phase containing TR-FL biosensor was decanted and diluted with PBS buffer (pH 7.4, 50 mM), and the solution was stored at 4 °C in the dark until use. 2.8. Strategy for detection of vascular endothelial growth factor-165 (VEGF165) Various concentrations of VEGF165 in the range of 0.1–50 nM were added to the TR-FL sensor (50 μL) for 55 min at 37 °C in a micro quartz cuvette. Then, the TR-FL spectra were recorded, and the fluorescence intensities at 585 nm were determined. The timeresolved fluorescence measurements were carried out with delay time of 0.2 ms, gate time of 0.4 ms and cycle time of 20 ms. The experiments for the selectivity test were performed with the VEGF165 analogs PDGF-BB, and VEGF121 under identical conditions.
3. Results and discussion 3.1. Characterization of the SA-functionalized AgNPs (SA-AgNPs) The AgNPs enhanced fluorescence effect reflects the scattering contribution, with relative proportions depending on the size, shape, and composition of the NPs (Lakowicz, 2005). Lakowicz et al. (2008) proved that the diameter of AgNPs is optimal from 50 to 70 nm to provide the strength distribution of the enhanced electromagnetic field of localized surface plasmon resonance and the depression of competitive quenching. The TEM image of Fig. 1 (A) shows generally spherical and proportioned AgNPs with a diameter at 5073 nm, a size that has been shown to be suitable for MEF and the radiating plasmon model. In Fig. 1(B), the AgNPs show an absorption peak at 445 nm in the plasmon resonance band. After SA was attached to the AgNPs, an obvious absorption peak occurs at 280 nm, which is a characteristic of the SA, indicating the successful binding between SA and the AgNPs. The zeta potential of the AgNPs decreased from 23.2 to 24.9 mV (n ¼3) when binding with SA occurred. The decrease of zeta potential is consistent with the binding of oligonucleotides since oligonucleotides are negatively charged. 3.2. Characterization of the Mn-doped ZnS QDs/aptamers (Apt-1). The Mn-doped ZnS QDs shows good monodispersity with particle sizes ranging from 4.5 to 5.5 nm as shown in Fig S1 (see the Supplementary information). In Fig. 2(A), the Mn-doped ZnS QDs show an absorption peak at 315 nm, and the two photoluminescence (PL) peaks, a strong peak around 585 nm from an internal electronic transition of the Mn (4T1–6A1), and a weak blue peak around 420 nm from the defect related emission (Suyver et al., 2001). After Apt-1 was bond onto the Mn-doped ZnS QDs, an obvious absorption peak occurs at 260 nm as shown in Fig. 2(C), which is a characteristic of the DNA strand, indicating the successful binding between Apt-1 and the Mn-doped ZnS QDs. In Fig. 2(B), the shape and intensity of the emission peak around 585 nm is almost the same before and after the modification of the QDs with Apt-1, suggesting good stability of covalent conjugation between the Mn-doped ZnS QDs and Apt-1, while the defect related emission around 420 nm is greatly suppressed. In Fig. 3(D),
1056
D. Zhu et al. / Biosensors and Bioelectronics 74 (2015) 1053–1060
Fig.1. (A) TEM image of the prepared AgNPs with an average diameter of 50 nm. (B) UV absorption spectra of the AgNPs and the AgNPs immobilized with SA.
the peaks at 1553 and 3428 cm 1 correspond to the stretching vibrations of C ¼ O and O–H, respectively, which demonstrates the carboxylic group capped on the Mn-doped ZnS QDs. The characteristic absorption peaks of Apt-1 at 1642, 1081, and 991 cm 1 are preserved after the formation of the Mn-doped ZnS QDs/Apt-1 conjugates, which further proves the successful connection. The
amount of conjugated Apt-1 is also estimated to be 8 Apt-1 per the Mn-doped ZnS QD (see the Supplementary information, Section S1).
Fig. 2. (A) Fluorescence and UV–vis absorption spectrum of the Mn-doped ZnS QDs. (B) Normalized fluorescence spectra of the Mn-doped ZnS QDs and the Mn-doped ZnS QDs/Apt-1. (C) UV–vis absorption spectra of the Mn-doped ZnS QDs and the Mn-doped ZnS QDs/Apt-1. (D) FT-IR spectra of the Mn-doped ZnS QDs and the Mn-doped ZnS QDs/Apt-1.
D. Zhu et al. / Biosensors and Bioelectronics 74 (2015) 1053–1060
1057
Fig. 3. (A) Emission spectrum (a) of donor (the Mn-doped ZnS QDs), the absorption spectrum (b) of acceptor (BHQ-2). (B) Loss of the Mn-doped ZnS fluorescence with increasing number of BHQ-2 quench oligonucleotides. (C) ΔF/F0 at different lengths of Apt-1; The concentration of VEGF165 was 6 nM. (ΔF/F0 ¼fluorescence intensity change at a specified concentration of VEGF165/fluorescence intensity with absence of VEGF165). (D) ΔF/F0 at different dilution ratio of SA-functionalized AgNPs for 55 min at 37 °C. (E) Fluorescence changes of the Mn-doped ZnS QDs at λ¼ 585 nm upon different incubation time in the treatment of the TR-FL sensor with VEGF165, 6 nM. Spectra were recorded at time intervals of 5 min. (F) Gel-shift analysis of the TR-FL probe–VEGF165 complex, as visualized with ethidium bromide. Lane 1: the Apt-1 bond with the Mndoped ZnS QDs; Lane 2: TR-FL sensor based on AgNPs–SA–Apt-1/BHQ-Mn-QDs complex; Lane 3: the TR-FL sensor/VEGF165 complex; Lane 4: DNA marker; Lane 5: BHQ-2 quencher-labelling strands.
3.3. Optimization of the Ag-enhanced TR-FL sensor The emission spectrum of the donor (the Mn-doped Zns QDs) and the absorption spectrum of the acceptor (BHQ-2) are shown in Fig. 3(A). BHQ-2 has a broad absorption band around 580 nm that matches well with the Mn (4T1-6A1) emission centered at 585 nm. FRET response relies on not only the overlap between the donor emission and acceptor absorption spectrum, but also the
donor–acceptor distances (R) and the number of quencher molecules interacting with donor. Fig. 3(B) shows a plot of the relative fluorescence at 585 nm emission versus increasing amounts of added BHQ-2 Apt (reported as BHQ-2 Apt to QD ratio, n) through hybridization with Apt-1/Mn-doped ZnS QDs, which indicates that saturation occurs at BHQ-2 to QD ratio of 10 with an ultimate quenching efficiency of 90%. The hybridization efficiency of BHQ-2 Apt with Apt-1 QDs can be calculated to 89% of the
1058
D. Zhu et al. / Biosensors and Bioelectronics 74 (2015) 1053–1060
Fig. 4. (A) Steady-state fluorescence spectra of BSA (a), VEGF165 (b), Apt-1 (c) a mixture of BSA, the Mn-doped ZnS QDs, Apt-1 and VEGF165 (d), and time-resolved fluorescence spectra of the mixture (e) in PBS solution (pH ¼7.4). (B) Comparison of ΔF/F0 of AgNPs-based TR-FL sensor and bare TR-FL sensor at different concentrations of VEGF165. (C) Fluorescence decay of AgNPs-based TR-FL sensor and bare TR-FL sensor at 8 nM VEGF165.
immobilized BHQ-2-labelling strands by means of measuring the absorbance of the BHQ-2-labelling strands solution before hybridization and that of the collected ultrafiltration solution after hybridization. So, Apt-1-binding sites on the quantum-dot surface occurs at 8 BHQ-2 Apt per dot with an ultimate quenching efficiency of 90%. From the basis of the equations (Clapp et al., 2004.; Goldman et al., 2005)
E=
nR 06 nR 06 + R6
R 0 = 0.211[k 2nd−4 YJ ]1/6 where E is the energy transfer efficiency, n is the number of quencher molecules, R0 is the critical transfer distance, R is actual distance, k is the orientation factor (k2 ¼2/3 for random collisions), nd is the index of refraction (n ¼1.33 in an aqueous medium), Y is the quantum yield of the Mn-doped Zns QDs (Y¼0.086) and the overlap integral J is obtained by a program that overlays the normalized emission of the donor and absorption spectra of the acceptor, J ¼2.58 1015 cm 1 M 1 nm4, the critical transfer distance R0 was calculated as 3.91 nm, and the separation distance (r) between the QD center and the acceptor was estimated to 3.83 nm.
Using the estimate for the donor–acceptor separation distance and the QD finite radii extracted from size measurements (5 nm) using transmission electron microscopy (TEM) and scattering techniques, we infer a distance between the QD surface and the BHQ-2 of 12–14 Å. In such close distance, the Mn-doped ZnS QDs could contact with BHQ-2 that contributes more of quenching mechanism than FRET because electron withdrawing groups and large conjugated groups of BHQ-2 could quench fluorescence of QDs, and such quenching by other similar strong electron transfer agents has been reported (Nieto et al., 2004; Tu. et al., 2008; Liu et al., 2010). The MEF is highly distance-dependent, and the length of the Apt-1 bond with the Mn-doped ZnS QDs, a single stranded DNA, can affect MEF effect. When the BHQ-2-labelling strands are dissociated from the Apt-1/Mn-doped ZnS QDs by VEGF165, the distance between the Mn-doped ZnS QDs and AgNPs is also changed, and this can be controlled by the length of the Apt-1. The Apt-1 of different lengths(31 bp, 42 bp and 51 bp) are evaluated as shown in Fig. 3(C), and AgNPs show better fluorescence enhancement effect on the Mn-doped ZnS QDs with 42 bp Apt-1. The concentration of SA-functionalized AgNPs was also optimized as shown in Fig. 3(D), a significant decrease of the ΔF/F0 ratio is observed with the increase in dilution ratio between 2 and
D. Zhu et al. / Biosensors and Bioelectronics 74 (2015) 1053–1060
8. When the SA-functionalized AgNPs was diluted with 1 mg/mL BSA by 2-fold (the volume ratio of BSA to SA-functionalized was 1:1), the ΔF/F0 ratio was almost equal to that of undiluted SAfunctionalized AgNPs. Therefore, SA-functionalized AgNPs with 2-fold dilution ratio was selected for the next experiments. In addition, the incubation time between TR-FL sensor and VEGF165 was estimated. Fig. 3(E) shows the fluorescence changes upon different incubation time in the treatment of the TR-FL sensor with VEGF165 at a concentration of 6 nM. A time-dependent increase in the fluorescence is observed, consistent with the dissociation of the BHQ-2 quencher-labelling strands from the Mndoped ZnS QDs, due to the formation of the aptamer–VEGF complex. The increases of fluorescence reached a saturation value after ca. 55 min and represent the kinetics of formation of the equilibrium state of the sensor–VEGF165 complex. Therefore, 55 min incubation time was chosen for the next experiments. Gel-shift confirmed the formation of the TR-FL sensor and the sensor–VEGF165 complex (Fig. 3(F)). Compared with the Apt-1 bond with the Mn-doped ZnS QDs (Lane 1) and BHQ-2 quencherlabelling strands (Lane 5), the TR-FL sensor complex band presented a much slower migration rate (Lane 3). After incubation with VEGF165, the sensor–VEGF165 complex also showed slower migration than the sensor (Lane 3). 3.4. Superiority of AgNPs-enhanced TR-FL sensor Fig. 4(A) compares the fluorescence spectra of a mixture of the Mn-doped ZnS QDs, BSA, DNA and VEGF165 in the steady-state PL mode and time-resolved PL technique. In the steady-state fluorescence spectra, the emissions of both BSA and VEGF165 from 400 to 620 nm interfered with the emission peaks at 420 and 585 nm of the Mn-doped ZnS QDs (curves a and b in Fig. 4(A)). The steadystate fluorescence interferences would inevitably lead to a serious limitation on analytical accuracy and testing sensitivity. In sharp contrast, the short-lived fluorescence of both BSA and VEGF165 from 400 to 620 nm disappeared, and only intense emission originating from the Mn-doped ZnS QDs was detected in the TR-FL spectra by setting up appropriate delay time and gate time. The MEF ability of TR-FL sensor was estimated by comparing the fluorescence intensity change (ΔF/F0) of the AgNPs-based TRFL sensor with bare sensor as shown in Fig. 4(B). The ΔF/F0 ratio of the AgNPs-based TR-FL sensor is much higher in the corresponding concentration range of VEGF165, which can be attributed to the fluorescence enhancement effect of the Mn-doped ZnS QDs based
1059
on AgNPs. The fluorescence intensity increases by 3.5 times when the concentration of VEGF165 is 16 nM. The MEF process is further confirmed by fluorescence decay measurements as shown in Fig. 4(C). The intensity decay of the bare TR-FL sensor was fitted with a multiple exponential with three lifetime components of 125 μs (11%), 450 μs (17% ) and 1330 μs (72%) with the total amplitude-weighted lifetime being 1.28 ms. The intensity decay of AgNPs-based TR-FL sensor was also fitted with a multiple exponential with three lifetimes of 67 μs (9%), 319 μs (12%) and 1130 μs (79%). Hence, the amplitudeweighted fluorescence lifetime of the AgNPs-based TR-FL sensor was calculated to be 1.03 ms. The reduction in the fluorescence lifetime of the Mn-doped ZnS QDs despite an increase in the fluorescence intensity is a characteristic of the fluorophore plasmon coupling effect (Zhang et al., 2009; Lakowicz 2001). 3.5. Assay performance on VEGF165 Fig. 5(A) depicts the resulting calibration curve corresponding to the fluorescence change of the Mn-doped ZnS QDs in the presence of different concentrations of VEGF165. The ΔF/F0 ratio increased with the VEGF165 concentration, and a linear relationship (r ¼0.995) between ΔF/F0 and VEGF165 concentration in the range of 0.1–16 nM was observed. Based on the averaged values of experiments, the concentration of VEGF165 was estimated by using the following equation:
ΔF /F0 = 0.1946 + 0.4167CVEGF165 The precision for seven repeated measurements of 4 nM VEGF165 was 3.3% (RSD), and the detection limit (3δ) was 0.08 nM. The specificity of the TR-FL sensor toward VEGF165 was also investigated in the presence of other cancer-related proteins such as platelet-derived growth factor (PDGF)-BB and VEGF121. As shown in Fig. 5(B), the TR-FL sensor produced a significant signal change with VEGF165 at 3 nM, while signal changes from 3 nM PDGF-BB and 3 nM VEGF121 were not significant. The presence of 10-fold human serum (10% serum), 50% serum and 100% serum did not show any interference. Then, VEGF165 in 100% human serum was also tested to examine whether the complex matrix affected the AgNPs-enhanced TR-FL sensor. The fluctuation of 1.8% was weak compared with VEGF165 as shown in Fig. 5(B), and the recovery of 92–106% of spiked VEGF165 in 100% human serum were obtained as shown in Table S1 in Supplementary information.
Fig. 5. (A) Response curve of ΔF/F0 to different VEGF165 concentrations, where ΔF/F0 represents the fluorescence intensity change of the modified Mn-doped ZnS QDs at λ ¼585 nm, in the absence and presence of VEGF165, respectively. (Error bars were evaluated from N ¼3 experiments.). (B) Specificity of the TR-FL for VEGF165. The concentrations of molecules are physiologically relevant values: 3 nM (VEGF165, PDGF-BB and VEGF121), 10% serum, 50% serum and 100% serum (error bars were evaluated from N ¼3 experiments.)
1060
D. Zhu et al. / Biosensors and Bioelectronics 74 (2015) 1053–1060
4. Conclusions A silver nanoparticles-enhanced TR-FL sensor based on the Mndoped ZnS QDs has been successfully developed for the detection of VEGF165. The TR-FL sensor shows good selectivity and sensitivity toward the target species in the presence of other proteins or in the complex matrix. Compared with the bare FL sensor, the AgNPs-enhanced TR-FL sensor clearly shows remarkable advantages in terms of sensitivity. What is more, the TR-FL sensor also offers a signal with high signal-to-noise ratio in luminescent biodetection as compared with conventional FL. The sensitivity was increases 11-fold compared with bare FL sensor with the detection limit of 0.08 nM. In addition, the sensor provided a wide range of linear detection from 0.1 nM to 16 nM. Furthermore, the principle of this TR-FL sensor can be extended to other proteins or new biomarkers for disease diagnosis and fundamental research.
Acknowledgments We greatly appreciate the National Natural Science Foundation of China for the financial support (21205064). This work has also been supported by Fund of State Key Laboratory of Analytical Chemistry for Life Science (SKLACLS1208). We also greatly appreciate “A Project Funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions” (PAPD).
Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.bios.2015.08.005.
References Akimov, A.V., Mukherjee, A., Yu, C.L., Chang, D.E., Zibrov, A.S., Hemmer, P.R., Park, H., Lukin, M.D., 2007. Nature 450, 402–406. Bock, L.C., Griffin, L.C., Latham, J.A., Vermaas, E.H., Toole, J.J., 1992. Nature 355, 564–566. Chen, L.Q., Xiao, S.J., Peng, W.T., Ling, J., Li, Y.F., Huang, C.Z., 2010. J. Phys. Chem. B 114, 3655–3659. Cho, E.J., Lee, J.-W., Ellington, A.D., 2009. Annu. Rev. Anal. Chem. 2, 241–264. Clapp, A.R., Medintz, I.L., Mauro, J.M., Fisher, B., Bawendi, M.G., Mattoussi, H., 2004. J. Am. Chem. Soc. 126, 301–310. Eis, P.S., Millar, D.P., 1993. Biochemistry 32, 13852–13860. Fang, X., Tan, W., 2010. Acc. Chem. Res. 43, 48–57. Fu, Y., Zhang, J., Lakowicz, J.R., 2009. Chem. Commun., 313–315. Geddes, C.D., Lakowicz, J.R., 2002. J. Fluoresc. 12, 121–129. Goldman, E.R., Medintz, I.L., Whitley, J.L., Hayhurst, A., Clapp, A.R., Tetsuo Uyeda, H.,
Deschamps, J.R., Lassman, M.E., Mattoussi, H., 2005. J. Am. Chem. Soc. 127, 6744–6751. He, Y., Wang, H.-F., Yan, X.-P., 2008. Anal. Chem. 80, 3832–3837. Hirata, J., Jong, C.F., Van dongen, M.M., Buijs, J., Ariese, F., Irth, H., Gooijer, C., 2004. Anal. Chem. 76, 4292–4298. Hiroshi, Y., Masaki, W., Yukihiro, O., Manabu, T., Noritada, K., Yoshitaka, M., Hirofumi, N., Yoshinobu, B., Shuji, H., 2012. Biomaterials 33, 2177–2186. Ho, Y.P., Kung, M.C., Yang, S., Wang, T.H., 2005. Nano Lett. 5, 1693–1697. Huang, F., Lerner, E., Sato, S., Amir, D., Haas, E., Fersht, A.R., 2009. Biochemistry 48, 3468–3476. Hu, M., He, Y., Song, S.P., Yan, J., Lu, H.-T., Weng, L.-X., Wang., L.-H., Fan, C.H., 2010. Chem. Commun. 46, 6126–6128. Kishore, M., Krishnamoorthy, G., Udgaonkar, J.B., 2013. Biochemistry 52, 9482–9496. Laitala, V., Hemmila1, I., 2005. Anal. Chem. 77, 1483–1487. Lakowicz, J.R., 2001. Anal. Biochem. 298, 1–24. Lakowicz, J.R., 2005. Anal. Biochem. 337, 171–194. Lakowicz, J.R., Ray, K., Chowdhury, M., Szmacinski, H., Fu, Y., Zhang, J., Nowaczyk, K., 2008. Analyst 133, 1308–1346. Liu, J.F., Bao, C.Y., Zhong, X.H., Zhao, C.C., Zhu, L.Y., 2010. Chem. Commun. 46, 2971–2973. Liu, Y.S., Zhou, S.Y., Tu, D.T., Chen, Z., Huang, M.D., Zhu, H.M., Ma, E., Chen, X.Y., 2012. J. Am. Chem. Soc. 134, 15083–15090. Mackowski, S., Wörmke, S., Maier, A.J., Brotosudarmo, T.H.P., Harutyunyan, H., Hartschuh, A., Govorov, A.O., Scheer, H., Bräuchle, C., 2008. Nano Lett. 8, 558–564. Medintz, I.L., Tetsuo Uyeda, H., Goldman, E.R., Mattoussi, H., 2005. Nat. Mater. 4, 435–446. Michalet, X., Pinaud, F.F., Bentolila, L.A., Tsay, J.M., Doose, S., Li, J.J., Sundaresan, G., Wu, A.M., Gambhir, S.S., Weiss, S., 2005. Science 307, 538–544. Muskens, O.L., Giannini, V.J., Sanchez-Gil, A., Gomez Rivas, J., 2007. Nano Lett. 7, 2871–2875. Nieto., S., Santana., A., Herna'ndez., S.P., Lareau., R., Chamberlain., R.T., Castro., M.E., 2004. Proc. SPIE – Int. Soc. Opt. Eng. 5403, 256–260. Norris, D.J., Efros, A.L., Erwin, S.C., 2008. Science 319, 1776–1779. Ozers, M.S., Marks, B.D., Gowda, K., Kupcho, K.R., Ervin, K.M., Rosier, T.D., Qadir, N., Eliason, H.C., Riddle, S.M., Shekhani, M.S., 2007. Biochemistry. 46 683-69. Qin, Y.Q., Ji, X.H., Jing, J., Liu, H., Wu, H.L., Yang, W.S., 2010. Colloids Surf. A: Physicochem. Eng. Asp. 372, 172–176. Sandin, P., Lincoln, P., Albinsson, B., 2008. J. Phys. Chem. C 112, 13089–13094. Seveus, L., Vaisala, M., Syrjanen, S., Sandberg, M., Kuusisto, A., Harju, R., Salo, J., Hemmila, I., Kojola, H., Soini, E., 1992. Cytometry 13, 329–338. Sokolov, K., Chumanov, G., Cotton, T.M., 1998. Anal. Chem. 70, 3898–3905. Suyver, J.F., Wuister, S.F., Kelly, J.J., Meijerink, A., 2001. Nano.Lett. 1, 429–433. Tu., R.Y., Liu., B.H., Wang., Z.Y., Gao., D.M., Wang., F., Fang., Q.L., Zhang, Z.P., 2008. Anal. Chem. 80, 3458–3465. Wang, H.-F., Li, Y., Wu, Y.-Y., He., Y., Yan, X.-P., 2010. Chem. Eur. J. 16, 12988–12994. Wu, P., He, Y., Wang, H.-F., Yan, X.-P., 2010. Anal. Chem. 82, 1427–1433. Wu, P., Yan, X.-P., 2013. Chem. Soc. Rev. 42, 5489–5521. Xie, R.G., Battaglia, D., Peng, X.G., 2007. J. Am. Chem. Soc. 129, 15432–15433. Xie, R.G., Peng, X.G., 2008. Angew. Chem. Int. Ed. 47, 7677–7680. Zhang, C.Y., Yeh, H.C., Kuroki, M.T., Wang, T.H., 2005. Nat. Mater. 4, 826–831. Zhang, J., Fu, Y., Chowdhury, M.H., Lakowicz, J.R., 2007. Nano Lett. 7, 2101–2107. Zhang, J., Fu, Y., Lakowicz, J.R., 2009. J. Phys. Chem. C 113, 19404–19410. Zhelev, Z., Bakalova, R., Ohba, H., Jose, R., Imai, Y., Baba, Y., 2006. Anal. Chem. 78, 321–330. Zhu, A.X., Duda, D.G., Sahani, D.V., Jain, R.K., 2011. HCC and angiogenesis: possible targets and future directions. Nat. Rev. Clin. Oncol. 8, 292–301. Zhu, D., Li, W., Ma, L., Lei, Y., 2014. RSC Adv. 4, 9372–9378. Zhu, D., Li., W., Wen, H.-M., Chen, Q., Ma, L., Hu, Y., 2014. Anal. Methods 6, 7489–7495.