Ultrasensitive sensing platform for platelet-derived growth factor BB detection based on layered molybdenum selenide–graphene composites and Exonuclease III assisted signal amplification

Biosensors and Bioelectronics 77 (2016) 69–75

Contents lists available at ScienceDirect

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

Ultrasensitive sensing platform for platelet-derived growth factor BB detection based on layered molybdenum selenide–graphene composites and Exonuclease III assisted signal amplification Ke-Jing Huang n, Hong-Lei Shuai, Ji-Zong Zhang College of Chemistry and Chemical Engineering, Xinyang Normal University, Xinyang 464000, China

art ic l e i nf o

a b s t r a c t

Article history: Received 22 July 2015 Received in revised form 1 September 2015 Accepted 11 September 2015 Available online 12 September 2015

A highly sensitive and ultrasensitive electrochemical aptasensor for platelet-derived growth factor BB (PDGF-BB) detection is fabricated based on layered molybdenum selenide–graphene (MoSe2–Gr) composites and Exonuclease III (Exo III)-aided signal amplification. MoSe2–Gr is prepared by a simple hydrothermal method and used as a promising sensing platform. Exo III has a specifical exo-deoxyribonuclease activity for duplex DNAs in the direction from 3′ to 5′ terminus, however its activity is limited on the duplex DNAs with more than 4 mismatched terminal bases at 3′ ends. Herein, aptamer and complementary DNA (cDNA) sequences are designed with four thymine bases on 3′ ends. In the presence of target protein, the aptamer associates with it and facilitates the formation of duplex DNA between cDNA and signal DNA. The duplex DNA then is digested by Exo III and releases cDNA, which hybridizes with signal DNA to perform a new cleavage process. Nevertheless, in the absence of target protein, the aptamer hybridizes with cDNA will inhibit the Exo III-assisted nucleotides cleavage. The signal DNA then hybridizes with capture DNA on the electrode. Subsequently, horse radish peroxidase is fixed on electrode by avidin–biotin reaction and then catalyzes hydrogen peroxide and hydroquinone to produce electrochemical response. Therefore, a bridge can be established between the concentration of target protein and the degree of the attenuation of the obtained signal, providing a quantitative measure of target protein with a broad detection range of 0.0001–1 nM and a detection limit of 20 fM. & 2015 Elsevier B.V. All rights reserved.

Keywords: Molybdenum selenide–graphene Exonuclease III Signal amplification Protein

1. Introduction In recent years, two-dimensional (2D) layered materials are very fascinating in batteries, field effect transistors, photocatalysis, electrocatalysis and biosensing in terms of good electronic and electrochemical properties (Chakrabarti et al., 2013; Brown et al., 2014; Zhang et al., 2015a, 2015b). They possess weak van der Waals forces between molecular layers and strong chemical bonding within the layers. Different from bulk materials, 2D layered materials have shortened path lengths for ions diffusion and large exposed surface areas, which play important roles in enhancing the electrochemical performance (Liu et al., 2015). In particular, the electrochemical activity of 2D MoS2 and other transition metal dichalcogenides has received considerable attention (Huang et al., 2014a; Yang et al., 2015). For example, Li et al. successfully synthesized Ni nanoparticle-MoS2 nanosheet hybrid by chemically reducing Ni nanoparticles on MoS2 nanosheet for n

Corresponding author. E-mail address: [email protected] (K.-J. Huang).

http://dx.doi.org/10.1016/j.bios.2015.09.026 0956-5663/& 2015 Elsevier B.V. All rights reserved.

the non-enzymatic detection of glucose. The as-synthesized hybrid exhibits a low detection limit of 0.31 mM towards glucose detection (Huang et al., 2014b). Thin-layer MoS2 nanosheets were prepared by Jiao's group via a simple ultrasound exfoliation method from bulk MoS2. When evaluated as an electrochemical sensing platform for DNA, the thin-layer MoS2 nanosheets exhibited wide linear range (1.0
10  16–1.0
10  10 M) and low detection limit of 1.9
10  17 M (Wang et al., 2015b). With the same layered structure, it can be concluded that MoSe2 might be a good choice for electrochemical application. Interestingly, compared to the intensive research on MoS2, MoSe2, the metal selenide analogues in the same graphene-type family is rarely studied for use as electrochemical biosensing. It has been reported MoSe2 has higher intrinsic electrical conductivity than MoS2 due to the more metallic nature of Se. And, the unsaturated Se-edges in MoSe2 are found to be electrocatalytically active and beneficial for the electrochemical sensing (Kong et al., 2013; Wang et al., 2013a). However, as similar with the most transition metal oxides, MoSe2 shows the intrinsic low electronic conductivity. Also, its big surface energy between nanosheets make it easily

70

K.-J. Huang et al. / Biosensors and Bioelectronics 77 (2016) 69–75

aggregate, which limits its further application. To address these issues, a feasible way is to combine MoSe2 with other conductive matrixes to keep the sheet structure and improve electrical conductivity of MoSe2. Graphene (Gr) has been widely used in electrochemical sensor construction due to its unique electronic, optical, thermal, and mechanical properties (Huang et al., 2013b; Li et al., 2015; Kumar et al., 2015). Especially with its ultrathin 2D planar structures, extremely high surface areas and good electrical conductivity, it is expected that the novel MoSe2–Gr hybrids can exhibit superior electrochemical performance. Recently, in order to improve the sensitivity of assay, strategies based on enzyme-aided signal amplification have been employed because they can overcome the inherent limitation of target-tosignal ratio of 1:1 in the traditional hybridization assay, such as the hybridization chain reaction (Song et al., 2015), the strand displacement amplification (Li et al., 2015), rolling circle amplification (Cheng et al., 2014) and the catalysis by nucleases (Gao et al., 2013). Target recycling achieved by exonucleases is a new method for promoting signal amplification. As an important DNA-modifying enzyme, Exonuclease III (Exo III) catalyzes stepwise removal of mononucleotides from blunt or recessed 30-hydroxyl termini of duplex DNA, whereas its activity on single-strand DNA or 30protruding termini of double-stranded DNA is inhibited (Bao et al., 2015). It is worth mentioning that Exo III is sequence independent and does not require any specific recognition sequence to function (Lin et al., 2014). Thus, it can provide a more universal platform for sensitive detection of DNAs, RNAs and proteins. Aptamers are artificial single-stranded DNA or RNA oligonucleotieds selected from random-sequence nucleic acid libraries by in vitro evolution process (Famulok et al., 2007). Important characteristics for the success of analytical assays based on aptamers are the affinity and the specificity of the aptamer that provides molecular recognition. Aptamers can bind to their targets with high affinity and they can discriminate between closely related targets. This is because of adaptive recognition: aptamers, unstructured in solution, fold upon associating with their molecular targets into molecular architectures in which the ligand becomes an intrinsic part of the nucleic acid structure (Jayasena, 1999). Compared with antibody, aptamers exhibit several considerable advantages, such as high specificity, good stability, desirable biocompatibility and significant chemical simplicity (Wu et al., 2011). Therefore, they offer a powerful alternative to antibody as recognition molecules. The above viewpoints have inspired the present work, where layered molybdenum selenide–graphene (MoSe2–Gr) composite was prepared by a simple hydrothermal method and used as supporting substrate for a novel electrochemical aptamer sensor construction coupled with Exo III for signal-amplification detection of platelet-derived growth factor BB (PDGF-BB). Encouragingly, the as-developed method exhibited an outstanding analytical performance, i.e. wide dynamic range, low detection limit and good selectivity, and could be used as a universal protocol to determine different targets, such as protein, DNAs, drug and even cells.

2. Experimental 2.1. Reagents and materials Graphite powder, Se powder, sodium molybdate and chloroauric acid (HAuCl4  3H2O) were obtained from Aladdin Chemicals Co. Ltd. (Shanghai, China). PDGF-BB, IgG, thrombin (Hb), carcinoembryonic antigen (CEA), bovine serum albumin (BSA), Exonuclease III (Exo III), Avidin labeled horseradish peroxidase (AvidinHRP), hydroquinone (HQ) and hydrogen peroxide (H2O2) were

obtained from Shanghai Xueman Biotechnology Co. Ltd. (China). 0.1 M phosphate buffer solution (PBS, pH 7.0) was prepared with 0.1 M Na2HPO4 and NaH2PO4 and adjusted by 0.1 M H3PO4 or 0.1 M NaOH solutions. The water used in the experiments was obtained from a water purification system ( Z18 MΩ, Milli-Q, Millipore). The aptamer sequence and other DNA sequences were synthesized by Shanghai Sangon Biological Engineering Technology Co. Ltd. (Shanghai, China) and their sequences were as follows: Aptamer: 5′-CAGGCTACGGCACGTAGAGCATCACCATGATCCTGTTTT3′; Immobilized DNA (iDNA): 5′-SH-TTTTTTTTTTGATCATGGTGATGCTCAT-3′; Complementary DNA (cDNA): 5′-GATCATGGTGATGCTCTATTTT3′; Auxiliary DNA (aDNA): 5′-Biotin-TAGAGCATCACCATGATC-3′ 2.2. Apparatus Cyclic voltammetry (CV) and differential pulse voltammetry (DPV) measurements were performed on a CHI 660E Electrochemical Workstation (Shanghai CH Instruments, China) with a conventional three-electrode system composed of saturated calomel electrode (SCE) as reference, platinum wire as auxiliary and modified glassy carbon electrode (GCE) as working electrode. A Hitachi S-4800 scanning electron microscope (SEM, Tokyo, Japan) and a JEM 2100 transmission electron microscope (TEM, JEOL, Tokyo, Japan) were used for the morphological characterizations. A model D/max-rA diffractometer (Rigaku, Japan) was applied to record the X-ray diffraction (XRD) pattern and a Renishaw Raman system model 1000 spectrometer (Gloucestershire, UK) was used to record the Raman spectra. 2.3. Preparation of Au nanoparticles and MoSe2–graphene composites The AuNPs were prepared according to the previous method (Jiang et al., 2008). 100 mL HAuCl4 solution (0.01%) was boiled with vigorous stirring, and then 2.7 mL trisodium citrate solution (1%) was quickly added and stirred for 10 min at the boiling point. The obtained colloid AuNPs were stored in brown glass bottles at 4 °C. Graphene oxide was prepared from natural graphite powder (Huang et al., 2013a). In short, 5 g graphite powder was slowly added in the solution containing 87.5 mL concentrated H2SO4 and 45 mL fuming HNO3. After that, 55 g KClO3 was added into above mixture and kept stirring. After 96 h, the mixture was washed with water and then filtered. Finally, the Graphene oxide was obtained after exfoliating the as-prepared graphite oxide in water with ultrasonic treatment. For MoSe2–Gr composites synthesis, 0.158 g Se powder was first dissolved in 5 mL hydrazine hydrate under continuous vibration until homogeneous dark red brown solution was obtained. Then the mixture was laid aside for 24 h to obtain hydrazine hydrate-Se. During this process, the color of solution remained unchanged. Meanwhile, 0.242 g sodium molybdate was dissolved in 50 mL water, and then the as-prepared hydrazine hydrate-Se solution was dropwise added in sodium molybdate solution. After that, the as-obtained graphene oxide was dispersed sufficiently in above mixture. Then, the mixture was transferred into a 100-mL Teflon-lined autoclave and heated at 180 °C. After 48 h, the autoclave was cooled down in the air. The black MoSe2–Gr composites were collected after washed with water and dried in vacuum at 60 °C for 12 h.

K.-J. Huang et al. / Biosensors and Bioelectronics 77 (2016) 69–75

2.4. Preparation of biosensor

3. Results and discussion

Prior to modification, GCE was firstly treated with 0.3 and 0.05 μm alumina powder, followed by washing with ethanol and water. 1.0 mg MoSe2–Gr composites were dispersed in 1.0 mL water to obtain a homogeneous suspension (1 mg mL  1). Then, 10 mL MoSe2–Gr suspension was applied on GCE surface and dried in the air to obtain MoSe2–Gr/GCE. Subsequently, the electrode was exposed to AuNPs solution for 12 h following with rinsed with water and then dried to achieve AuNPs/MoSe2–Gr/GCE/GCE. After that, the modified electrode was submerged into the capture DNA solution for 3 h at room temperature and then incubated with blocking solution (BSA, w/w, 0.25%) for 2 h to eliminate non-specific binding effects and block the remaining active groups. The obtained modified electrodes were stored at 4 °C before use. 2 mL aptamer (0.5 mM) and 5 mL PDGF-BB were mixed and incubated at 30 °C for 2 h. Then, 2 mL signal DNA (0.5 mM) and 2 mL cDNA (0.5 mM) were added and incubated at 30 °C for 1 h. Subsequently, 16 U Exo III was added and incubated at 37 °C for 30 min. After that, the mixture was heated at 70 °C for 10 min to terminate the digestion reaction. The obtained mixture was then applied on the modified electrode and incubated at 30 °C for 60 min. The resultant electrode was then immersed in Avidin-HRP solution (20 mg mL  1) for 20 min to load HRP on the electrode.

3.1. Design of the biosensor

2.5. Gel electrophoresis analysis The samples of cDNA, aDNA/cDNA and PDGF-BB/aptamer/ cDNA/aDNA digested by Exo III were prepared. 4% (w/v) agarose gels were dissolved in buffer containing 40 mM Tris-HAc and 1 mM EDTA (pH 8.0). 10 mL of each samples and 2 mL DNA marker were mixed with 1.5 mL GelDye TM Super Buffer Mix. Electrophoresis was carried out at 110 V constant voltage for 30 min.

71

Herein, the proposed system mainly consists of Exo III, a singlestranded signal probe (aDNA), a PDGF-BB aptamer, an immobilized DNA (iDNA) and a complementary DNA (cDNA), MoSe2–Gr and AuNPs. The thiolated DNA is immobilized on the AuNPs/MoSe2– Gr/GCE and its uncomplementary sequence is designed to hybridize with signal DNA. It has been reported that double-stranded DNA would be hydrolyzed by Exo III when it has three mismatched terminal bases at far recessed 3′ ends while that of more than 4 bases could not be hydrolyzed. Herein, the aptamer and the cDNAs are specially designed and both of them have four thymine nucleotides on their 3′ ends to resist the cleavage of Exo III. Biotintagged at 5′ terminus of signal DNA is designed to hybridize with cDNA. As illustrated in Scheme 1, the proposed aptasensor demonstrated a signal-off architecture in response to the target protein. In the absence of PDGF-BB, the aptamer hybridized with the cDNA, which prevent the Exo III from cleaving the singlestrand signal DNA. Then duplex DNAs are formed between the iDNA and aDNA. Subsequently, avidin-HRP is introduced to the electrode by avidin–biotin, which can result in a strongly current response by the catalysis of H2O2 þHQ. In the presence of PDGFBB, the aptamers associates with the target protein, which lead to the formation of duplex DNAs between the cDNAs and aDNA. The Exo III therefore can digest the duplex DNAs from 3′ blunt terminus of signal probes, liberating the cDNA is liberated. The released cDNAs then hybridized with aDNA to undergo a new cleavage reaction. Through such a cyclic hybridization-hydrolysis process, a PDGF-BB molecule can trigger cleavage of a large number of aDNA, resulting in a low current response. The strategy used to construct the aptasensor is outlined in Scheme 1.

Scheme 1. Schematic illustration of the biosensor construction and used for PGDF-BB detection with Exonuclease III assisted signal amplification.

72

K.-J. Huang et al. / Biosensors and Bioelectronics 77 (2016) 69–75

Fig. 1. SEM images of MoSe2–Gr at two scales (A, B); TEM (C) and HRTEM (D) image of MoSe2; EDS spectrum of MoSe2–Gr (E).

3.2. Characterization of as-prepared materials The morphology of the MoSe2–Gr composites obtained by the hydrothermal synthesis was characterized by SEM and TEM. As shown in Fig. 1A and B, the SEM images clearly illustrated the typical morphology of MoSe2–Gr with an average agglomerate size of 80–120 nm. It was clearly observed that there were many finger-like MoSe2 bars aggregating together with the wrinkled Gr nanosheets coating on the surface (Fig. 2A). Many micropores uniformly distributed on the composites surface and the walls of pores were composed by abundant Gr nanosheets. The Gr

lamellars exhibited an interlaced situation, astricting the growth orientation of MoSe2 by stretching the thin layers. It was no doubt that the porous layered morphology of the products processed a large specific surface area and would accelerate the transportation of ions and electrons, therefore leaded to an enhanced electrochemical performance. The pure MoSe2 prepared by the hydrothermal method was also characterized by TEM. Fig. 1C revealed the typical two-dimensional structure of the MoSe2 nanosheets. From the TEM image in Fig. 1D, the lattice fringes of MoSe2 with a D-spacing of 0.62 nm could be assigned to the (002) lattice plane of hexagonal MoSe2, which comprised only 4–10 layers of MoSe2

Fig. 2. XRD patterns (A) and Raman spectra (B) of MoSe2–Gr composites.

K.-J. Huang et al. / Biosensors and Bioelectronics 77 (2016) 69–75

nanosheets. The elemental composition of the as-prepared MoSe2–Gr composite was identified with energy-dispersive spectroscopy (EDS), which showed intense peaks of Mo, Se and C (Fig. 1E). The molar ratio of Mo/Se was calculated as approximately 1:2 from the peak intensities, verifying the desired stoichiometry of the samples. Besides, the EDS data also suggested the existence of elements C and O, which can be assigned to Gr in the composites. The XRD patterns of the as-prepared MoSe2–Gr composite were shown in Fig. 2A. All the diffraction peaks in the XRD pattern were indexed to the porous layered MoSe2–Gr nanostructure (JCPDS No.00-041-1008 (7:1). The broad peak centered at about 25.7° belonging to carbon suggested the containing the few-layer and intact-structure graphene in the composite (Huang and Wang, 2011). The sharp peaks of MoSe2 suggest the polycrystalline nature of the as-prepared material. Importantly, the as-prepared products are almost pure and without the XRD peaks corresponding to Na2MoO4, MoO2 and other selenide phases. Raman spectroscopy was used to analyze the structure of the MoSe2–Gr composite, as shown in Fig. 2B. The MoSe2 exhibited 1 Raman modes located at 243 and characteristic A1g and E2g 1 291 cm , respectively, indicating a red-shift compared with the MoSe2 bulk and films materials reported previously (Kong et al., 2013; Sekine et al., 1980). The Raman peak corresponding to the out-of plane Mo–Se phonon mode (A1g) was preferentially excited for the edge-terminated film, which results in the much higher A1g 1 peak intensity than the E2g peak intensity. The peak positions of MoSe2 demonstrated a minor dependence on the thickness of layer. Two bands located at 1351 cm  1 and 1592 cm  1 were corresponding to the D and G band of Gr in the composite. The D band was related to the defects of edge areas and G band was related to the in-phase vibration of the Gr lattice.

73

3.3. Eectrochemical characteristics Electrochemical impedance spectroscopy (EIS) was applied to test the impedance changes of surface-transfer electrodes. The micircle diameter equaled to charge transfer resistance (Ret) in Nyquist diagram. Fig. 3A showed the characteristic EIS corresponding to the stepwise modification processes. The bare GCE displayed a Ret value of about 502 Ω (curve a). When MoSe2–Gr composite was applied on the GCE, the Ret value decreased to 405 Ω (curve b), which indicated good electronic conductivity. After AuNPs were further modified on MoSe2–Gr/GCE, the Ret greatly decreased and exhibited almost a line (curve c), which was contributed to the excellent conductivity. However, the Ret value obviously increased after iDNA was applied to the electrode surface decreased (642 Ω) due to its obstructing the electron transfer (curve d). When BSA, aDNA and Avidin-HRP reacted with the electrode, the Ret obviously increased to 780 Ω, 1020 Ω and 1370 Ω (curve e, f and g), respectively, which may originate from the hydrophobic protein and negative charges DNA sequences blocking the electrode surface. The different electrodes were also studied by CV. As shown in Fig. 3B, the bare GCE displayed a couple redox peaks in [Fe(CN)6]3  /4  solution (curve a). The peak currents greatly enhanced when MoSe2–Gr composite modified on the GCE because the as-prepared material could improve the electron transfer rate (curve b). The current respond reached the maximum when AuNPs were further modified on the electrode (curve c). However, the current responses decreased greatly when iDNA was applied on the electrode to block the reactive sites (curve d). The redox currents gradually decreased after the electrode step by step reacted with BSA, aDNA and Avidin-HRP (curves e, f, g) due to the fact that the protein and negative charges of DNA insulated the conductive support. The CV results suggested the successful construction of the biosensor.

Fig. 3. EIS (A) and CVs (B) of bare GCE (a), MoSe2–Gr/GCE (b), AuNPs/MoSe2–Gr/GCE/GCE (c), iDNA/AuNPs/MoSe2–Gr/GCE/GCE (d), iDNA/BSA/AuNPs/MoSe2–Gr/GCE/GCE (e), aDNA/iDNA/BSA/AuNPs/MoSe2–Gr/GCE/GCE (f), Avidin-HRP/aDNA/iDNA/BSA/AuNPs/MoSe2–Gr/GCE/GCE (g). CV experiments were conducted in 0.1 M PBS solution containing 1.0 mM [Fe(CN)6]3  /4  and 0.1 M KCl; (C) Plot of Q–t curves of the bare GCE (a) and AuNPs/MoSe2–Gr/GCE (b) in 0.1 mM K3[Fe(CN)6] containing 1.0 M KCl. Inset: plot of Q–t1/2 curves on GCE (a) and AuNPs/MoSe2–Gr/GCE (b); (D) DPVs of modified electrode with (b) and without (a) Exonuclease III assisted signal amplification in 0.1 M PBS (pH 7.0) containing 1.8 mM H2O2 þ 2 mM HQ.

74

K.-J. Huang et al. / Biosensors and Bioelectronics 77 (2016) 69–75

To confirm the formation of aptamer/PDGF-BB complex and aDNA/cDNA duplex, agarose gel electrophoresis analysis was carried out. As shown in Fig. S2, a bright band at lane 1 indicated the existence of cDNA. Lane 2 displayed lower mobility than Lanes 1 and 3 due to the formation of aDNA/cDNA duplex. A bright fluorescent band was also observed in lane 3, which was attributed to the enzymatic reaction. In the presence of PDGF-BB, the Exo III could digest the cDNA and therefore liberated the fluorophores, which made the brighter bands in Lane 3 than that of Lane 2. These results demonstrated the formation of aptamer/ PDGF-BB complex and aDNA/cDNA duplex in the samples. In this work, the MoSe2–Gr was prepared and used as supporting substrate for the construction of biosensor due to their excellent electronic conductivity and large specific surface area. For confirming this, the effective surface areas (A) of GCE and AuNPs/MoSe2–Gr/GCE were tested by chronocoulumetry according to following equation:

Q = 2nFAcD1/2t1/2/π1/2 + Q dl + Q ads

(1)

where c, Qdl and Qads are the substrate concentration, double layer charge and Faradaic charge, respectively. Other symbols possess the usual meanings. As shown in Fig. 3C, the effective surface areas of GCE and AuNPs/MoSe2–Gr/GCE were calculated to be 0.015 and 0.062 cm2, respectively. The effective surface areas value of AuNPs/MoSe2–Gr was about 4.13 times of GCE, suggesting AuNPs/MoSe2–Gr film greatly increased the effective surface area of electrode. Fig. 3D showed the signal-amplification effect of the Exo III. Current signal greatly decreased when the Exo III was employed (curve a). The corresponding signal gain in the presence of Exo III amplification was about 31.6% for that in the absence of Exo III (curve b), indicating the Exo III was helpful to enhance detection sensitivity. 3.4. Optimization of conditions The catalytic activity of the resultant HRP modified electrode was evaluated. As shown in Fig. S1A, no electrochemical signal of the proposed aptasensor was obtained when H2O2 and HQ were absent in electrolyte. The same result was observed when only H2O2 was added in the electrolyte. When both H2O2 and HQ were added, the remarkable signal response was obtained, suggesting HRP could catalyze the mixed solution of H2O2 and HQ to produce prominent electrochemical response. The effects of H2O2 (Fig. S1B) and HQ concentrations (Fig. S1C) in the detection system were also investigated. The results showed that the maximum Differential pulse voltammetry (DPV) response was obtained when H2O2 and

HQ concentrations were 1.8 mM and 2 mM, respectively, in the range of 0–3 mM. Thus, 1.8 mM H2O2 and 2 mM HQ were used in the subsequent experiments. The reaction time of Avidin-HRP with modified electrode was evaluated. The results showed that the highest current signal was obtained when the reaction time was 20 min (Fig. S1D). Therefore, 20 min was used. The effect of reaction time of the formation of aptamer-PDGF-BB complex (Fig. S1E) and aDNA-cDNA duplex (Fig. S1F) on the DPV responses were also studied. It showed that the optimized conditions were obtained when the reaction times of the formation of aptamer/PDGFBB complex and aDNA/cDNA duplex were 2 h and 60 min, respectively. 3.5. Performance of the as-prepared aptasensor Analytical calibrations for the PDGF-BB were recorded by DPV under optimal conditions. Fig. 4A showed that the peak currents of PDGF-BB decreased with increased concentrations of PDGF-BB. As could be seen in the inset of Fig. 4A, there was a good linear relationship between the peak currents and the logarithm of the concentrations of PDGF-BB in the range from 0.0001 to 1 nM. The peak current values were obtained from the mean value with three independent experiments. The linear regression equation was ipa (mA)¼  20.03–4.42 log (c/M) with the correlation coefficient R of 0.993, and the detection limit was estimated to be 20 fM based on the 3s rule. Thus, the proposed electrochemical aptasensor had a good analytical performance for the detection of PDGF-BB. The analytical performance of proposed aptasensor has been compared to those reported previously in Table 1. It was clear that our aptasensor showed a better performance than other assays in the analytical range and detection limit. The reason should be taken into account as the employment of the excellent electrical conductivity and biocompatibility of MoSe2–Gr composites and Exo III-aided dual signal amplification which exhibited satisfactory performance as expected. In order to evaluate the specificity of the proposed aptasensor, BSA, IgG, Hb and CEA were used as control proteins. As shown in Fig. 4B, the peak current with aspecific proteins (1.0 nM Hb, 100 nM BSA, 100 nM IgG, 1.0 nM CEA) were much higher than that with 5.0 pM PDGF-BB, which were quite close to that obtained in blank solution. So the comparisons had clearly shown that the control proteins had nearly no effect on the detection by using proposed aptasensor, which ascribed to the highly selective recognition and binding between aptamer and target protein. The reproducibility of the proposed aptasensors for PDGF-BB was investigated by replicating measurements of the same PDGFBB solution (0.1 nM) with one aptasensor. The variation coefficient of five measurements was 4.9%. The fabrication reproducibility

Fig. 4. (A) DPVs of various PDGF-BB concentrations (from a to i): 0, 1.0
10  13, 5.0
10  13, 5.0
10  12, 1.0
10  11, 5.0
10  11, 1.0
10  10, 1.0
10  9 mol/L, respectively. Inset: the calibration plots of the DPV current versus the logarithm concentration of PDGF-BB; (B) the amperometric responses of as-prepared biosensor reacted with blank, 5.0 pM PDGF, 1.0 nM Hb, 100 nM BSA, 100 nM IgG, 1.0 nM CEA and admixture of BSA (100 nM), CEA (1.0 nM), Hb (1.0 nM), IgG (100 nM), and PDGF-BB (5.0 pM). Error bars show the standard deviations of measurements taken from three independent experiments.

K.-J. Huang et al. / Biosensors and Bioelectronics 77 (2016) 69–75

75

Table 1 Comparison of the proposed assay with other methods for PDGF-BB detection. Analytical method

Linear range (nM)

Detection limit (pM)

Ref.

Cyclic voltammetry Differential pulse voltammetry Ac voltammetry Differential pulse voltammetry Differential pulse voltammetry Differential pulse voltammetry Colorimetric Fluorescence Photoinduced electron transfer Chemiluminescent Differential pulse voltammogram

0.005–60 0.42–8.3 0.017–1660 0.84–83 0.002–40 0.001–35 10–100 0.011–16.7 0.0005–10 0.06–6 0.0001–1

1.7 0.167 8.3 0.42 0.6 8 6 6.8 0.1 60 0.02

Deng et al. (2013) Wang et al. (2013a, 2013b, 2013c) Zhang et al. (2015a, 2015b) Qu et al. (2011) Han et al. (2013) Bai et al. (2012) Chang et al. (2013)

was evaluated by determining the PDGF-BB with five aptasensors made at the same electrode independently. The variation coefficient of 7.9% was obtained for the peak current, showing an acceptable reproducibility. Thus, the proposed electrochemical strategy could provide a simple method for reproducible preparation of the aptasensor in batches.

Wang et al. (2015a) Wang et al. (2013b) This work

Appendix A. Supplementary materials Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.bios.2015.09.026.

References 3.6. Application of proposed biosensor in human serum samples In order to evaluate the application potential of the developed aptasensor, five human serum samples obtained from healthy people were analyzed with the developed assay and the results were listed in Table S1. 0.1 mL sample was diluted with PBS (pH 7.0) to 10 mL and then detected with the aptasensor. In order to verify the accuracy, the same serum samples were detected with ELISA method. From Table S1, it could be seen that the PDGF-BB contents obtained by this assay were in good agreement with the ELISA method. The relative deviations were less than 7.9%, indicating that the developed method was promising for determining PDGF-BB in real biological samples.

4. Conclusion In this work, we have developed a novel highly sensitive assay to quantify PDGF-BB as low as 20 fM by integrating layered MoSe2 –Gr composites and Exo III-aided signal amplification. This strategy had several excellent features. First, the large surface active sites and good conductivity of MoSe2–Gr composites provided a promising sensing platform. Second, the proposed assay had a wide dynamic range with a span of five orders of magnitude. Third, the application of Exo III-aided signal amplification leaded to a remarkable low detection limit (20 fM). Finally, the concept could easily be extended to various targets by simply changing the DNA sequences. To sum up, the experimental results demonstrated that this facile method had the potential to be used as a useful tool for ultrasensitive quantitative analysis of PDGF-BB or other biomolecules in complex samples and to supply valuable information for biomedical research and clinical diagnosis.

Acknowledgments This work was supported by the National Natural Science Foundation of China (U1304214, 21475115) and Program for University Innovative Research Team of Henan (15IRTSTHN001).

Bai, L.J., Yuan, R., Chai, Y.Q., Zhuo, Y., Yuan, Y.L., Wang, Y., 2012. Biomaterials 33, 1090–1096. Bao, T., Shu, H.W., Wen, W., Zhang, X.H., Wang, S.F., 2015. Anal. Chim. Acta 862, 64–69. Brown, B., Swain, B., Hiltwine, J., Brooks, D.B., Zhou, Z.G., 2014. J. Power Sources 272, 979–986. Chakrabarti, M.H., Low, C.T.J., Brandon, N.P., Yufit, V., Hashim, M.A., Irfan, M.F., Akhtar, J., Ruiz-Trejo, E., Hussain, M.A., 2013. Electrochim. Acta 107, 425–440. Chang, C.C., Wei, S.C., Wu, T.H., Lee, C.H., Lin, C.W., 2013. Biosens. Bioelectron. 42, 119–123. Cheng, W., Zhang, W., an, Y.R., Shen, Y.B., Zhu, D., Lei, P.H., Ding, S.J., 2014. Biosens. Bioelectron. 62, 274–279. Deng, K., Xiang, Y., Zhang, L.Q., Chen, Q.H., Fu, Q.W.L., 2013. Anal. Chim. Acta 759, 61–65. Famulok, M., Hartig, J.S., Mayer, G., 2007. Chem. Rev. 107, 3715–3743. Gao, T., Ning, L.M., Li, C., Wang, H.Y., Li, G.X., 2013. Anal. Chim. Acta 788, 171–176. Han, J., Zhuo, Y., Chai, Y.Q., Yuan, R., Xiang, Y., Zhu, Q., Liao, N., 2013. Biosens. Bioelectron. 46, 74–79. Huang, H.J., Wang, X., 2011. Nanoscale 3, 3185–3191. Huang, J.W., He, Y.Q., Jin, J., Li, Y.R., Dong, Z.P., Li, R., 2014b. Electrochim. Acta 136, 41–46. Huang, K.J., Liu, Y.J., Wang, H.B., Wang, Y.Y., Liu, Y.M., 2014a. Biosens. Bioelectron. 55, 195–202. Huang, K.J., Wang, L., Liu, Y.J., Liu, Y.M., Wang, H.B., Gan, T., Wang, L.L., 2013a. Int. J. Hydrog. Energy 38, 14027–14034. Huang, Z.D., Zhang, H.Y., Chen, Y.M., Wang, W.G., Chen, Y.T., Zhong, Y.B., 2013b. Electrochim. Acta 108, 421–428. Jayasena, S.D., 1999. Clin. Chem. 45, 1628–1650. Jiang, Y., Zhao, H., Zhu, N.N., Lin, Y.Q., Yu, P., Mao, L.Q., 2008. Angew. Chem. Int. Ed. 47, 8601–8604. Kong, D.S., Wang, H.T., Judy, J.C., Mauro, P., Kristie, J.K., Yao, J., Cui, Y., 2013. Nano Lett. 13, 1341–1347. Kumar, S., Ahlawat, W., Kumar, R., Dilbaghi, N., 2015. Biosens. Bioelectron. 70, 498–503. Li, B., Pan, G.H., Avent, N.D., Lowry, R.B., Madgett, T.E., Waines, P.L., 2015. Biosens. Bioelectron. 72, 313–319. Lin, C.S., Wu, Y.F., Luo, F., Chen, D.M., Chen, X., 2014. Biosens. Bioelectron. 59, 365–369. Liu, Y., Zhu, M.Q., Chen, D., 2015. J. Mater. Chem. A 3, 11857–11862. Qu, F.L., Lu, H.M., Yang, M.H., Deng, C.Y., 2011. Biosens. Bioelectron. 26, 4810–4814. Sekine, T., Izumi, M., Nakashizu, T., Uchinokura, K., Matsuura, E., 1980. J. Phys. Soc. Jpn. 49, 1069–1077. Song, W.L., Xie, X.X., Sun, W.B., Zhang, N.B., Li, C.X., 2015. Anal. Chim. Acta 860, 77–82. Wang, C.C., Ding, L., Qu, F.L., 2013a. Measurement 46, 279–283. Wang, G.F., Zhu, Y.H., Chen, L., Zhang, X.J., 2015a. Biosens. Bioelectron. 63, 552–557. Wang, H., Kong, D., Johanes, P., Cha, J.J., Zheng, G., Yan, K., Liu, N., Cui, Y., 2013b. Nano Lett. 13, 3426–3433. Wang, P., Song, Y.H., Zhao, Y.J., Fan, A.P., 2013c. Talanta 103, 392–397. Wang, X.X., Nan, F.X., Zhao, J.L., Yang, T., Ge, T., Jiao, K., 2015b. Biosens. Bioelectron. 64, 386–391. Wu, Z., Tang, L.J., Zhang, X.B., Jiang, J.H., Tan, W., 2011. ACS Nano 5, 7696–7699. Yang, Z.L., Gao, D.Q., Zhang, J., Xu, Q., Shi, S.P., Tao, K., Xue, D.S., 2015. Nanoscale 7, 650–658. Zhang, L., Yang, C.L., Zhao, G.Y., Mu, J.S., Wang, Y., 2015a. Sensor. Actuators B 210, 190–196. Zhang, S.B., Hu, X., Yang, X.H., Sun, Q.L., Xu, X.L., Liu, X.W., Shen, G.Y., Lu, J.L., Shen, G.L., Yu, R.Q., 2015b. Biosens. Bioelectron. 66, 363–369.