- Email: [email protected]
A photoelectrochemical aptasensor based on a 3D flower-like TiO2-MoS2-gold nanoparticle heterostructure for detection of kanamycin
Biosensors and Bioelectronics 112 (2018) 193–201
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
A photoelectrochemical aptasensor based on a 3D flower-like TiO2-MoS2gold nanoparticle heterostructure for detection of kanamycin
T
⁎
Xiaoqiang Liua, , Peipei Liua, Yunfei Tanga, Liwei Yanga, Lele Lia, Zhichong Qia, Deliang Lia, ⁎ Danny K.Y. Wongb, a
Henan Joint International Research Laboratory of environmental pollution control materials, College of Chemistry and Chemical Engineering, Henan University, Kaifeng, Henan Province 475004, PR China b Department of Molecular Sciences, Macquarie University, Sydney, NSW 2109, Australia
A R T I C LE I N FO
A B S T R A C T
Keywords: TiO2-MoS2-AuNP composite Photoelectrochemical aptasensor Kanamycin detection Visible light excitation
In this work, a sensitive photoelectrochemical aptasensor was developed for kanamycin detection using an enhanced photocurrent response strategy, which is based on the surface plasmon resonance effect of gold nanoparticles deposited on a 3D TiO2-MoS2 flower-like heterostructure. A significant aspect of this development lies in the photoelectrochemical and morphological features of the unique ternary composite, which have contributed to the excellent performance of the sensor. To develop an aptasensor, mercapto-group modified aptamers were immobilised on the photoactive composite as a recognition unit for kanamycin. The TiO2-MoS2AuNP composite was demonstrated to accelerate the electron transfer, increase the loading of aptamers and improve the visible light excitation of the sensor. Under optimal conditions, the aptasensor exhibited a dynamic range from 0.2 nM to 450 nM of kanamycin with a detection limit of 0.05 nM. Overall, we have successfully synergised both the electrical and the optical merits from individual components to form a ternary composite, which was then demonstrated as an effective scaffold for the development of PEC biosensors.
1. Introduction As a new emerging analytical technique, photoelectrochemical (PEC) sensors have attracted increasing research interest (Li et al., 2015a; Wang et al., 2013). In this technique, an illumination is used to excite electrons in a photoactive material from the valence band to the conductive band, while a potential is applied to prevent the recombination of the photogenerated holes and electrons. An electroactive analyte is then oxidised by the photogenerated holes to produce a photocurrent (Gao et al., 2015; Li et al., 2015b). Hence, this is a technique with a unique capability of preventing the excitation source from interfering with the photocurrent signal to yield a high signal-tonoise ratio measurement response (Li et al., 2016). Unfortunately, PEC sensing is usually restricted by its poor specificity to the target analyte. This has stimulated studies of incorporating many recognition elements including molecular imprinted polymers, enzymes, antibodies and aptamers to improve its selectivity (Li et al., 2012; Whitcombe et al., 2011; Zhao et al., 2013, 2014a). Among them, aptamers (singlestranded oligonucleotides with specific sequences) are some of the strong competitors to the other biological recognition systems in analytical applications due to their low-cost in vitro synthesis, mass
⁎
production, reduced immunogenic response, good selectivity, and inherent binding affinity (Huang et al., 2015; Xin et al., 2015). In addition, aptamers are known to be less sensitive to temperature and pH, and are therefore less prone to deactivation, compared to enzymes. Most importantly, aptamers can be used to detect both small molecules and biological macromolecules including proteins, nucleic acids, polypeptides, hormones, sugars, viral particles, enzymes, etc, while molecular imprinted polymers and enzymes are usually restricted to detection of small molecules (Ansari and Husain, 2012; Fuchs et al., 2012; Xue et al., 2012). In general, the features of materials immobilised on an electrode surface will significantly affect the performance of the electrochemical and PEC biosensors. For example, TiO2 was very often used in developing electrochemical and PEC sensors due to its strong photocatalytic activity, good stability, controllable morphology and excellent biocompatibility (Liu et al., 2017). Unfortunately, TiO2 is only amenable to excitation by UV light, which will otherwise damage any biomolecules immobilised on a biosensor (Li et al., 2014a). In addition, the fast electron-hole recombination may significantly attenuate the PEC signal, which has also limited the applications of TiO2 (Luo et al., 2012; Yildirim et al., 2012). To circumvent this problem, many narrow
Corresponding authors. E-mail addresses: [email protected] (X. Liu), [email protected] (D.K.Y. Wong).
https://doi.org/10.1016/j.bios.2018.04.041 Received 24 January 2018; Received in revised form 11 April 2018; Accepted 17 April 2018 Available online 19 April 2018 0956-5663/ © 2018 Elsevier B.V. All rights reserved.
Biosensors and Bioelectronics 112 (2018) 193–201
X. Liu et al.
human (Song et al., 2011). In our work, the ternary composite design was developed using multi-step solution chemistry illustrated in Scheme 1. We have firstly synthesised mesoporous TiO2 microspheres with a large and rough surface, which acted as an effective platform for depositing an enhanced quantity of MoS2 nanosheets. AuNPs were then deliberately decorated on the surface of the 3D TiO2-MoS2 heterostructure to produce the ternary composite. In this way, plasmonic AuNPs will directly transform the incident visible light into electrical energy by injecting photogenerated electrons into the conduction band of MoS2, which will thus improve the photoelectric conversion efficiency. As demonstrated below, compared with a previously reported AuNP functionalised TiO2 nanotube array-based PEC aptasensor (Xin et al., 2015), the presence of MoS2 has increased the conductivity and visible light absorption. Finally, SH-functionalised kanamycin aptamers were immobilised on the modified electrode to develop an aptasensor. The specific interaction between the aptamer and kanamycin is expected to yield an increase in photocurrent response owing to the oxidation of kanamycin by photogenerated holes.
bandgap photoactive materials such as g-C3N4 (~2.7 eV) and MoS2 (~1.9 eV) nanomaterials have been hybridised with TiO2 to form a composite with improved photocatalytic or PEC performance (Zhao et al., 2012; Zhou et al., 2013). For example, our group has previously synthesised a TiO2 nanosheet-g-C3N4 heterostructure and applied it as a scaffold on an electrode to develop a PEC glucose sensor, which exhibited a 0.05 – 16 mM linear range and a 0.01 mM glucose detection limit under visible light (Liu et al., 2017). However, the need for exfoliating g-C3N4 by ultrasonication always resulted in micrometer-size sheets, which are insufficiently small to exhibit some of the nanomaterial properties (Zhang et al., 2014). More recently, molybdenum disulfide (MoS2) has received considerable attention due to its narrow bandgap, high thermal stability and special optical properties (Huang et al., 2014; Li et al., 2014b). In particular, the narrow bandgap of MoS2 makes this material more suitable in PEC sensor development. In addition, MoS2 is also regarded as a promising electrode material in battery developments owing to its 2D lamellar structure that is analogous to graphene (Ren et al., 2017). Accordingly, MoS2 possesses many advantages similar to graphene such as large surface area and high conductivity (Sun et al., 2015). As an example, Zhou et al. synthesised a MoS2-TiO2 heterostructure with a few-layer MoS2 nanosheets coated on TiO2 nanobelts (Zhou et al., 2013). The heterostructure was demonstrated to significantly retard the recombination between photogenerated electrons and holes due to the energy band matching between TiO2 and MoS2. Therefore, the photocatalytic activity of the composite with 50 wt% of MoS2 has increased by a factor of 3 under visible light irradiation, relative to TiO2 nanobelts alone. In this paper, we reported the development of a PEC aptasensor consisting of a novel TiO2-MoS2-gold nanoparticle (AuNP) flower-like nanocomposite using kanamycin as a model analyte. Notably, kanamycin has been widely used to treat serious infections caused by Grampositive and Gram-negative bacteria during protein synthesis (Xu et al., 2015b). However, an overdose or residue of this drug in animal-derived foods can cause ototoxicity, nephrotoxicity and antibiotic resistance in
2. Experimental 2.1. Materials and reagents All reagents were of analytical reagent grade and were used without any further purification. Titanium(IV) isopropoxide (TTIP, 97+% purity) was purchased from Alfa Aesar Chemical Co. Ltd Tianjin, China. Sodium molybdite dihydrate, thiourea, hexadecylamine (HDA, 90%), acetic acid, chloramphenicol, erythromycin, streptomycin, doxycycline, ciprofloxacin and ofloxacin were purchased from Aladdin Reagent Co., Ltd., Shanghai, PR China. Gold(Ⅲ) chloride trihydrate, chitosan (85% deacetylation) and bovine serum albumin (BSA) were acquired from Sigma-Aldrich Chemical Co. (St. Louis, MO). Sodium borohydride, sodium citrate, potassium ferricyanide, potassium ferrocyanide and potassium chloride were obtained from Sinopharm Chemical Reagent Co., Ltd., China. Kanamycin sulfate and a mercapto group modified
Scheme 1. Synthesis of TiO2-MoS2-AuNP composite; fabrication steps of a kanamycin aptasensor and its PEC mechanism for detection. 194
Biosensors and Bioelectronics 112 (2018) 193–201
X. Liu et al.
Ltd., China) was initially cleaned using NaOH (1 M) and H2O2 (30%), and then washed successively with acetone, alcohol and ultrapure water under sonification, and dried in air. Meanwhile, 20 mg of a TiO2MoS2-AuNP composite was dispersed in 1 mL chitosan (0.2 wt% in acetic acid) solution, which was sonicated for 20 min to obtain a homogeneous suspension. Next, 25 μL of the suspension was applied on the cleaned ITO and dried at ambient temperature. Subsequently, 25 μL of aptamer (1.5 μM) was immobilised on the TiO2-MoS2-AuNP|ITO, and the electrode was left at 4 °C for 1 h. Finally, the modified electrode was incubated in 25 μL of 3% (w/v) BSA solution for 30 min to block any remaining active sites to minimise nonspecific adsorption. The obtained aptasensor, denoted as BSA|aptamer|TiO2-MoS2-AuNP|ITO, was stored at 4 °C before use. The procedure for the aptasensor fabrication is illustrated in Scheme 1.
kanamycin-binding DNA aptamer with the sequence 5´-HS-(CH2)6TGG-GGG-TTG-AGG-CTA-AGC-CGA-3´ were purchased from Shanghai Sangon Biotechnology Co. Ltd., Shanghai, China. Phosphate buffered saline (PBS, pH 7.4) was obtained by mixing the corresponding stock solutions of 0.1 M KH2PO4, K2HPO4 and KCl. Indium tin oxide (ITO; resistivity 10 Ω cm−2) electrodes were acquired from Wuhan Lattice Solar Energy Technology, Ltd., China. Ultrapure water obtained from a Millipore water purification system (≥18 MΩ cm−1, Milli-Q, Millipore) was used in the assays. 2.2. Material characterisations Surface morphology of prepared nanomaterials was examined by scanning electron microscopy (JSM-7610F, JEOL, Japan) at an accelerating voltage of 15.0 kV. The powder samples were ultrasonically dispersed in ethanol, applied on a silicon wafer and dried. Chemical composition was studied by X-ray diffraction (XRD, Bruker D8 Advance, Germany) with a Cu Kα radiation. X-ray photoelectron spectroscopy (XPS) analysis was collected on an X-ray photoelectron spectrometer (Thermo Scientific Escalab 250Xi, USA). UV–visible diffuse reflectance spectroscopy of samples was performed using a UV–vis spectrophotometer (UV-2600, Kyoto, Japan). Electrochemical impedance spectroscopic measurements were carried out at an IM6ex electrochemical station (EIS, ZAHNER, Germany).
2.5. PEC and impedance measurements All PEC measurements at TiO2|ITO, MoS2|ITO, TiO2-MoS2|ITO, TiO2-MoS2-AuNP|ITO and BSA|aptamer|TiO2-MoS2-AuNP|ITO were performed at a CHI 630D electrochemical workstation (Shanghai, CH Instruments, China) accommodated with a three-electrode system. The modified ITO acted as a working electrode, a platinum wire as a counter electrode and a Ag|AgCl (3.0 M KCl) as a reference electrode. A Xe lamp equipped with a 400 nm cut-off filter (CEL-HXUV 300, Beijing AULTT, China) was used as an irradiation source. The electrolyte was composed of an aqueous solution of 0.1 M KCl and 0.1 M PBS (pH 7.4). The photocurrent at the modified ITO electrode was recorded versus time with alternate light-on and light-off cycles at a fixed bias voltage of 0.6 V. Impedance experiments were performed in a solution containing 5 mM K3[Fe(CN)6] and 5 mM K4[Fe(CN)6] in 0.1 M KCl at a DC potential of 0.20 V superimposed by an alternating voltage of 5 mV peakto-peak amplitude. Impedance was measured over a frequency range of 100 kHz to 0.1 Hz.
2.3. Preparation of TiO2-MoS2-AuNP composites A TiO2-MoS2-AuNP composite was prepared according to a multistep procedure illustrated in Scheme 1. The anatase TiO2 microspheres were firstly prepared using Chen et al.'s procedure (Chen et al., 2010) with minor modifications. In this procedure, 0.88 g of HDA was dissolved in 100 mL of absolute ethanol, followed by the addition of 0.40 mL of KCl solution (0.1 M) and 0.25 mL of ultrapure water. The mixture was stirred at ambient temperature and 2.16 mL of TTIP was added to this solution under vigorous stirring. The milky white bead suspension was kept quiescent for 18 h and then centrifuged, and the amorphous TiO2 beads were washed with ethanol three times and dried in air at 60 °C for 4 h. Finally, anatase TiO2 microspheres were obtained by a solvothermal reaction. In this reaction, 0.50 g amorphous TiO2 beads were dispersed in a solution containing 20 mL absolute ethanol and 10 mL ultrapure water. The mixture was transferred into a 50 mL Teflon-lined autoclave at 160 °C for 90 min. The final product (white powder) was collected by centrifugation, washed with ethanol and dried at room temperature. To synthesise the TiO2-MoS2 heterostructure, 0.08 g anatase TiO2 microspheres were dispersed in a mixture of 10 mL ethanol and 20 mL ultrapure water, and 0.35 g sodium molybdite dihydrate and 0.52 g thiourea were dissolved in the above solution, followed by a solvothermal reaction at 220 °C for 20 h. After being cooled down to room temperature, the sample was rinsed with ultrapure water and dried at 60 °C. MoS2 nanosheets were synthesised using the same procedure without adding anatase TiO2 microspheres. In preparing a TiO2-MoS2-AuNP composite, 2 mL of 0.01 M HAuCl4 and 2 mL of 0.01 M sodium citrate (a capping agent) were added into a uniform suspension of 40 mg TiO2-MoS2 heterostructure in 40 mL ultrapure water with continuous stirring. Next, 2 mL fresh NaBH4 solution (0.1 M, 0–4 °C) was added to the suspension with vigorous stirring for 30 min. The suspension was then stirred for another 30 min at room temperature before it was left undisturbed overnight. The composite was centrifuged and sequentially washed with ultrapure water and ethanol, before it was dried in a 60 °C vacuum overnight.
3. Results and discussion In this paper, a novel ternary composite of TiO2-MoS2-AuNP was synthesised and used as a scaffold for the construction of an aptasensor. The composite has provided a large and biocompatible platform to accommodate an enhanced loading of aptamers, all of which have contributed to the excellent electrochemical and PEC features. In the following sections, each component of the scaffold was carefully characterised before being assembled to develop the scaffold of a kanamycin aptasensor. 3.1. Morphological investigations of the nanocomposite The dimension and morphology of the nanomaterials used in this work were initially investigated using scanning electron microscopy. As displayed in Fig. 1(a), uniform TiO2 microspheres with a smooth surface are clearly visible and their average diameter was estimated to be 400 nm. These TiO2 microspheres were then solvothermally treated to form anatase TiO2 beads. According to the scanning electron micrograph (SEM) in Fig. 1(b), the surface of anatase TiO2 beads appears to be extremely rough. Nonetheless, their average diameter is undistinguishable from that of smooth TiO2 spheres. Meanwhile, the SEM of MoS2 samples in Fig. 1(c) clearly displays numerous curly MoS2 nanosheets aggregated together to form a bulky structure with a rough surface. These MoS2 samples are of irregular shapes with a few micrometers in size. In Fig. 1(d), the SEM of TiO2-MoS2 composite revealed that MoS2 nanosheets grew vertically on anatase TiO2 microspheres during the solvothermal reaction. More specifically, each TiO2 microsphere was uniformly covered with thick MoS2 nanosheets, showing a flower-like structure of the TiO2-MoS2 heterostructure. The average diameter of the composite is approximately 600 nm (standard deviation 20 nm; n = 6). The SEM of TiO2-MoS2-AuNP is displayed in
2.4. Construction of the PEC aptasensor To construct the PEC aptasensor, an ITO electrode (1 cm × 3 cm × 1.1 mm thick; acquired from Wuhan Lattice Solar Energy Technology, 195
Biosensors and Bioelectronics 112 (2018) 193–201
X. Liu et al.
Fig. 1. SEM of (a) amorphous TiO2, (b) anatase TiO2, (c) MoS2, (d) TiO2-MoS2, (e) and (f) TiO2-MoS2-AuNP samples. The arrows in (f) indicate where AuNPs are located.
nanohybrid paper electrode assembled from 3D ionic liquid functionalised graphene framework decorated by gold nanoflowers (Zhang et al., 2017). Therefore, the elemental composition and chemical state evaluated from the above XPS results have indicated successful immobilisation of elemental gold on the TiO2-MoS2 to form a ternary nanocomposite. XRD is another powerful technique to further verify the formation of a TiO2-MoS2-AuNP composite. The existence of the individual components can be evaluated by comparing their XRD patterns obtained with the standard patterns of Joint Committee on Powder Diffraction Standards (JCPDS). The XRD patterns of TiO2, MoS2, TiO2-MoS2, and TiO2-MoS2-AuNP nanocomposite are shown in Fig. 3(A). Firstly, we identified the anatase phase of TiO2 microspheres after comparing its pattern (trace a) to the standard pattern of JCPDS (no. 21-1272) (Xu et al., 2015a). The XRD pattern of MoS2 nanosheets (trace b) presented the characteristic diffraction peaks at 14.2°, 33.3°, 39.5° and 58.6°, unambiguously assigned to the (002), (100), (103), (110) peaks of hexagonal MoS2 (JCPDS no. 37–1492) (Heydari-Bafrooei and Askari, 2017). The XRD pattern of TiO2-MoS2 (trace c) was dominated by MoS2 except for the weak peaks at ~25.3° and 48.0° ascribed to the (101) and (200) diffraction of anatase TiO2 (Zhao et al., 2014b), which demonstrated the co-existence of MoS2 and TiO2 in the samples. The weak peaks corresponding to TiO2 could be attributed to dense coverage of TiO2 microspheres by MoS2 nanosheets. As for TiO2-MoS2-AuNP composites, trace d exhibited several new peaks at 38.6°, 44.8°, 64.9° and 77.8° corresponding to the (110), (200), (220), and (311) reflections of AuNPs (JCPDS no. 01-1172) (Mehta et al., 2016), which indicated that AuNPs were deposited on the TiO2-MoS2 heterostructure to form a ternary composite. Therefore, the XRD and XPS results are in agreement with each other, indicating a successful preparation of the TiO2-MoS2AuNP composite. Furthermore, the crystal forms of TiO2 and MoS2 were shown to be anatase and a hexagonal system, respectively, both of which will offer strong photocurrent conversion efficiency.
Fig. 1(e), where a large number of shiny points is observed on the surface of the ternary composite, demonstrating the successful deposition of AuNP on the TiO2-MoS2 heterostructure. The magnified SEM of the TiO2-MoS2-AuNP composite in Fig. 1(f) shows that many small particles of ~5–10 nm diameters were evenly distributed on the MoS2 nanosheets. In summary, the above results presented a visual indication for the formation of a ternary composite. As discussed below, additional techniques were used to identify the composition, chemical state and crystal form of all individual components in the ternary composite to provide further supporting evidence for the formation of the composite. Fig. 2(a) shows a typical overall XPS survey spectrum that suggests the presence of Ti, O, Mo, S, C and Au on the ternary composite surface. More specifically, the high resolution Ti 2p XPS spectrum in Fig. 2(b) shows two peaks at binding energies of 459.2 eV and 464.98 eV, which correspond to Ti 2p3/2 and Ti 2p1/2 of Ti4+, respectively, after comparing to the two peaks at 457.8 eV and 463.6 eV in Liu et al.’s work involving the fabrication of 3D mesoporous TiO2/MoS2/TiO2 nanosheets for visible-light-driven photocatalysis (Liu et al., 2016). Similarly, as displayed in Fig. 2(c), an O 1s peak at 530.5 eV (compared to 530.2 eV in Liu et al's work (Liu et al., 2015b)) is attributable to the TiO-Ti bond, and the peak at 532.0 eV (compared to 531.2 eV in (Li et al., 2017)) is assigned to adsorbed water. In Fig. 2(d), the peak at 226.8 eV is assigned to S 2s, and the peaks at 229.1 eV and 232.3 eV are attributed to Mo 3d5/2 and Mo 3d3/2, as suggested by Bai et al., who prepared chemically exfoliated metallic MoS2 nanosheets for photocatalytic enhancement (Bai et al., 2015). They also suggested the peak at 236 eV was attributed to the Mo6+ 3d, which was produced by the oxidation of Mo4+ (Ren et al., 2017). In Fig. 2(e), two peaks at 161.7 eV and 162.9 eV are respectively attributable to S 2p3/2 and S 2p1/2 at the state of S2-, as demonstrated by Zheng et al.'s work on preparing hierarchical MoS2 nanosheet@TiO2 nanotube array composites with enhanced photocatalytic performance (Zheng et al., 2016). Finally, Fig. 2(f) depicts the binding energies of Au 4f7/2 at 84.1 eV and Au 4f5/2 at 87.9 eV, which matched well with the standard binding energy of Au (84.0 and 87.6 eV), as shown in Zhang et al.'s work involving preparation of 196
Biosensors and Bioelectronics 112 (2018) 193–201
X. Liu et al.
Fig. 2. XPS analysis of TiO2-MoS2-AuNP composite: (a) survey scans, and high-resolution spectra of (b) Ti 2p, (c) O 1s, (d) Mo 3d, (e) S 2p and (f) Au 4f.
3.2. UV–vis and electrochemical impedance spectroscopic studies
displayed strong absorption in both UV and visible regions ascribed to its narrow band gap (1.2–1.9 eV) (Wang et al., 2016). Therefore, the absorption of the binary composite in the visible region (above 400 nm), as displayed in trace c, was greatly enhanced after TiO2 microspheres were decorated with MoS2. Compared to TiO2-MoS2, the extra absorption band of the ternary TiO2-MoS2-AuNP composite was observed in the visible wavelength of 510–675 nm, corresponding to the plasmon peak of AuNPs (Xin et al., 2015). More specifically, the enhanced absorption was attributed to AuNPs as an excellent electronic
As an effective complementary technique to XRD, UV–vis diffuse reflectance spectrometry was performed to evaluate the optical absorption properties of TiO2, MoS2, TiO2-MoS2 heterostructure and TiO2MoS2-AuNP. As illustrated by trace a in Fig. 3(B), anatase TiO2 microspheres only exhibited a noticeable absorption at wavelength lower than 400 nm due to the considerable energy bandgap (3.2 eV) of anatase TiO2 (Liu et al., 2015a). As shown in trace b, hexagonal MoS2
197
Biosensors and Bioelectronics 112 (2018) 193–201
X. Liu et al.
Fig. 3. (A) XRD patterns of (a) anatase TiO2, (b) MoS2, (c) TiO2-MoS2, and (d) TiO2-MoS2-AuNP composite; (B) UV–vis diffuse reflectance spectra of (a) anatase TiO2, (b) MoS2, (c) TiO2-MoS2 and (d) TiO2-MoS2-AuNP composite; (C) Nyquist plots of [Fe(CN)6]3-/[Fe(CN)6]4- at (a) bare ITO, (b) TiO2|ITO, (c) TiO2-MoS2|ITO, (d) TiO2MoS2-AuNP|ITO, (e) aptamer|TiO2-MoS2- AuNP|ITO and (f) BSA|aptamer|TiO2-MoS2-AuNP|ITO; (D) Time-based photocurrent response of (a) MoS2|ITO, (b) TiO2|ITO, (c) TiO2-MoS2|ITO, (d) TiO2-MoS2-AuNP|ITO and (e) BSA|aptamer|TiO2-MoS2-AuNP|ITO in 0.1 M PBS (pH 7.4) containing 0.1 M KCl at a bias voltage of 0.6 V in the absence of kanamycin.
the deposition of AuNPs on the TiO2-MoS2 further decreased the Ret value by 60% to ~78 Ω (trace d), attributed to the excellent electronic conductivity of AuNPs. However, as shown in trace e and trace f, the Ret increased gradually to ~121 Ω and ~157 Ω after successive modification with the kanamycin aptamer and BSA, respectively, owing to the strong steric hindrance and insulation effect of proteins on [Fe(CN)6]3-/ [Fe(CN)6]4-. Therefore, these results provided supporting evidence that the materials were successively immobilised on the electrode surface.
conductor, which both promoted the transfer of MoS2 photoelectrons and accelerated the separation of photo-carriers, leading to an obvious decreased recombination rate. Accordingly, a AuNP-modified TiO2MoS2 heterostructure will significantly benefit the performance of PEC sensing. Following the characterisation of the prepared nanomaterials, electrochemical impedance spectroscopy was used to probe the stepwise assembly of the PEC biosensor due to its strong capability of studying the charge transfer on an electrode/solution interface. As shown in Fig. 3(C), Nyquist (imaginary impedance (Z”) versus real impedance (Z’)) plots for different nanomaterial modified electrodes were recorded in the sequential fabrication steps over a frequency range from 100 kHz to 0.1 Hz. In these plots, the diameter of a semicircle in the high frequency region corresponds to the electron transfer resistance (Ret) of the redox marker, [Fe(CN)6]3-/[Fe(CN)6]4-, at the corresponding electrode. Only qualitative evaluation of results is conducted here by comparing the relative sizes of all semicircles. For example, in trace a, the bare ITO displayed the smallest semicircle, and hence the smallest Ret (~40 Ω), among all the electrodes. In contrast, the largest Ret was obtained at TiO2 microsphere modified electrode surface as shown in trace b, probably due to the poor electron transfer capacity of TiO2. The Ret (~195 Ω) from the Nyquist plot at the TiO2MoS2 modified electrode surface in trace c is 25% smaller than that at TiO2 (trace b，~259 Ω), most likely because the introduction of MoS2 accelerated electron transfer of the probe on the electrode. As expected,
3.3. PEC characterisation of the nanocomposites Next, both the optical absorption properties of the nanomaterials and assembling process of the PEC sensor in Section 3.2 were further confirmed by the photocurrent responses at different modified electrodes. In all these experiments, a photocurrent was obtained after applying a visible illumination. As depicted by trace a in Fig. 3(D), pristine MoS2 nanosheets show the smallest photocurrent (0.20 μA) among all the modified electrodes probably because MoS2 is an indirect gap semiconductor with poor excitation capability. The TiO2 microsphere modified ITO (trace b) also displayed a small photocurrent (0.83 μA), attributed to the poor visible light response of anatase TiO2 with large band gap. The photocurrent at MoS2-TiO2 heterostructure (4.1 μA) modified ITO (trace c) is ~3.9 times larger than that at TiO2 microsphere modified ITO (trace b), most likely because the intimate interfacial contact between MoS2 and TiO2 in the composite increased 198
Biosensors and Bioelectronics 112 (2018) 193–201
X. Liu et al.
Fig. 4. Influence of (A) aptamer concentration, (B) bias potential on PEC response of BSA|aptamer|TiO2-MoS2-AuNP|ITO to 200 nM kanamycin in 0.1 M PBS (pH 7.4) containing 0.1 M KCl; (C) Photocurrent at BSA|aptamer|TiO2-MoS2-AuNP|ITO in 0.1 M PBS (pH 7.4) containing 0.1 M KCl at a bias voltage of 0.6 V before and after incubation with 0, 0.05, 0.2, 2, 20, 50, 100, 200, 300, 400, 450, 500, 550 nM (from a to m) of kanamycin and (D) calibration based on PEC response of aptasensor and kanamycin concentration from 0.2 to 550 nM.
photocurrent response of the BSA|aptamer|TiO2-MoS2-AuNP|ITO modified electrode monotonically increased with the enhancement of the applied potential from 0.2 V to 0.6 V. However, when the bias potential has exceeded 0.6 V, the photocurrent was increasing at a slower rate, most likely because a weaker separation effect on the photogenerated charges. Accordingly, a bias potential of 0.6 V was adopted as the optimum potential for further experiments to minimise the oxidation reaction at higher potential.
the visible light absorption and enhanced the charge separation. Notably, TiO2-MoS2-AuNP|ITO (trace d) showed the highest photocurrent (8.4 μA) among all the electrodes, which is approximately a 2-fold increment of that observed at the MoS2-TiO2 modified ITO (trace c). The enhanced photocurrent could be attributed to the enlarged charge separation by the strong visible light adsorption owing to the localised surface plasmon resonance effect of AuNPs. After immobilising the kanamycin aptamer and BSA on the TiO2-MoS2-AuNP|ITO (trace e), the photocurrent (6.0 μA) has noticeably decreased by 28%. The steric hindrance experienced by the biological molecules on ITO prohibited photogenerated electrons from moving to ITO and this has thus accelerated their recombination with the holes.
3.5. PEC detection of kanamycin Under optimal conditions of 1.5 μM aptamer concentration and 0.6 V applied potential, detection of kanamycin was conducted at the fabricated PEC aptasensor using visible light illumination. In Fig. 4(C), the time (t)-dependent PEC photocurrent (the I-t) plots obtained at a BSA|aptamer|TiO2-MoS2-AuNP|ITO biosensor was recorded when the kanamycin concentration (C) was increased from 0 to 550 nM. The periodic photocurrent generated by the PEC oxidation of kanamycin rapidly responded to the repeated on-off cycles of irradiation, and the photocurrent remained highly stable at each kanamycin concentration during the PEC process. The inset in Fig. 4(C) shows the photocurrent measured over the 0–2 nM kanamycin concentration range, demonstrating the good performance of the sensor at low concentrations. Based on a background corrected current (ΔI), a linear calibration relationship between 0.2 and 450 nM was obtained, as shown in Fig. 4(D). However, when kanamycin concentration exceeded 450 nM, ΔI began to deviate from the linear range. The regression equation is determined as ΔI / μA = 0.011 ×C / μA + 0.055 with a correlation coefficient of 0.996 (n = 6), which was found to be statistically significant at the 95% confidence level using t-test. In addition, the result of an analysis of
3.4. Optimisation of the aptamer sensor Several important experimental parameters were optimised before the PEC sensor performance was evaluated. As described in Section 3.3, the loading of an aptamer on the electrode exhibited a major effect on the photocurrent of the aptasensor and was therefore optimised. Fig. 4(A) shows a photocurrent against aptamer concentration plot. In this figure, the photocurrent was observed to increase between 0.5 μM and 1.5 μM aptamer as a result of increasing kanamycin molecules being captured on the electrode surface, leading to larger oxidation current. However, when the aptamer concentration was higher than 1.5 μM, the photocurrent decreased, possibly due to excessive aptamers acting as steric hindrance to the transfer of electrons. Therefore, 1.5 μM aptamer was chosen as the optimum concentration. The bias potential applied on the photoelectrode drives the photoelectrons away from the photogenerated holes and therefore affects the recombination rate of the electron-hole pairs. As shown in Fig. 4(B), the 199
Biosensors and Bioelectronics 112 (2018) 193–201
X. Liu et al.
Table 1 Comparison of analytical performance of different techniques for kanamycin detection. Analytical technique
Linear range / nM
Detection limit / nM
References
Electrochemistry Electrochemistry Colorimetry Aptamer-based cantilever Luminescence Fluorescence PEC PEC PEC
10–150 50–9.0 × 103 1–100 1.0 × 105–1.0 × 107 200–1.5 × 105 10–1 × 103 1－230 0.2－200 0.2－450
5.8 9.4 1.49 5 × 104 143 2.7 0.2 0.1 0.05
(Sun et al., 2014) (Li et al., 2012) (Sharma et al., 2014) (Bai et al., 2014) (Leung et al., 2013) (Lin et al., 2014) (Li et al., 2014c) (Xin et al., 2015) This work
respectively, demonstrating the excellent long-term storage stability of the aptasensor. These results support the feasibility of this aptasensor for sensitive kanamycin assay.
variance also supported a significant linear relationship between C and ΔI. Further, using a signal-to-noise ratio of 3, the detection limit was estimated to be 0.05 nM (the response obtained at 0.05 nM is included in the inset in Fig. 4C). For comparison, the results obtained by several previously reported methods for the detection of kanamycin are displayed in Table 1. While many of them in Table 1 have exhibited either a good linear range or a low detection limit, the PEC aptasensor prepared in this work has shown superior analytical performance in both analytical characteristics.
3.7. Practical application of the aptasensor in milk samples The practical application of the aptasensor was assessed by detecting different concentrations of kanamycin in milk samples. The milk obtained from a local supermarket was firstly centrifuged at 15,000 rpm for 20 min and then the supernatant was diluted ten times with PBS. Subsequently, kanamycin standard solutions were spiked into the diluted milk, yielding final concentrations of 2, 20, 50, 100 nM. Finally, the recoveries of four different concentrations of kanamycin in the spiked milk samples were obtained using the PEC aptasensor. As shown in Table 2, the recoveries are in the range from 91.5% to 97.1%, clearly indicating the successful application of the PEC aptasensor for detection of kanamycin in real-life samples.
3.6. Specificity, reproducibility and stability of the PEC biosensor The selectivity of the aptasensor was investigated by comparing the photocurrent response to 200 nM kanamycin against that to several antibiotic interferents including chloramphenicol, erythromycin, streptomycin, doxycycline, ciprofloxacin and ofloxacin at the same concentration. As indicated in Fig. 5, the target kanamycin caused at least a 20-time increase in photocurrent change (ΔI) relative to the other interfering species owing to the specific interaction between the aptamer and kanamycin. The specificity was also evaluated by measuring ΔI in the presence of a mixed sample composed of kanamycin (200 nM), chloramphenicol (100 nM), erythromycin (100 nM), streptomycin (100 nM), doxycycline (100 nM), ciprofloxacin (100 nM) and ofloxacin (100 nM). The histogram labelled “Mixed Sample” in Fig. 5 shows a very similar ΔI obtained in kanamycin alone. Reproducibility was studied by determining kanamycin using six identical aptasensors in the same experimental conditions. A relative standard deviation (RSD) of 5.1% was obtained, suggesting an acceptable reproducibility of the aptasensor. Meanwhile, the stability of the PEC aptasensor was examined after the aptasensors were stored at 4 °C before further measurement. The results showed that 88.7% and 80.6% of its initial photocurrent response remained after 2 and 4 weeks of storage,
4. Conclusion In summary, we have developed a PEC aptasensor for the highly sensitive detection of the model analyte kanamycin using a TiO2-MoS2AuNP ternary composite as the photoactive materials in conjunction with a DNA aptamer recognition method. The TiO2-MoS2-AuNP heterostructure displayed a separated sphere nanomorphology with extremely large surface area, which increased the loading of immobilised aptamer molecules, and excellent electrical conductivity, which improved the electron transfer rate at the electrode surface. In addition, AuNPs on the TiO2-MoS2 heterostructure significantly improved the PEC performance under visible light irradiation because of the localised surface plasmon resonance effect. The PEC aptasensor presented good detection selectivity, a low detection limit of 0.05 nM and a wide linear range of 0.2–450 nM. The results demonstrated feasible applications of the TiO2-MoS2-AuNP composite to the development of aptasensors for detecting many different biological molecules. Acknowledgments This work was financially supported by National Natural Science Foundation of China (Nos. U1504215，21576071), and International Science and Technology Cooperative Project funded by The Department of Science and Technology of Henan Province (172102410042). Table 2 Determination of kanamycin in milk samples using the PEC aptasensor.
Fig. 5. Anti-interference property of BSA|aptamer|TiO2-MoS2-AuNP|ITO in 0.1 M PBS (pH 7.4) containing 0.1 M KCl at a bias voltage of 0.6 V in the presence of 200 nM kanamycin and the tabulated antibiotics. The error bars were derived from the standard deviation of four measurements. 200
Sample
Added / nM
Found / nM
Recovery / %
Standard deviation / nM
1 2 3 4 5
0 2 20 50 100
0 1.83 19.2 46.6 97.2
– 91.5 96.1 93.2 97.1
– 0.087 0.091 0.102 0.163
Biosensors and Bioelectronics 112 (2018) 193–201
X. Liu et al.
References
ChemSusChem 9, 1118–1124. Luo, J., Ma, L., He, T., Ng, C.F., Wang, S., Sun, H., Fan, H.J., 2012. J. Phys. Chem. C. 116, 11956–11963. Mehta, A., Sharma, M., Kumar, A., Basu, S., 2016. Nano 11, 1650104. Ren, W., Zhou, W., Zhang, H., Cheng, C., 2017. ACS Appl. Mater. Interfaces 9, 487–495. Sharma, T.K., Ramanathan, R., Weerathunge, P., Mohammadtaheri, M., Daima, H.K., Shukla, R., Bansal, V., 2014. Chem. Commun. 50, 15856–15859. Song, K.-M., Cho, M., Jo, H., Min, K., Jeon, S.H., Kim, T., Han, M.S., Ku, J.K., Ban, C., 2011. Anal. Biochem. 415, 175–181. Sun, X., Li, F., Shen, G., Huang, J., Wang, X., 2014. Analyst 139, 299–308. Sun, Y., Fang, Z., Wang, C., Ariyawansha, K.R.R.M., Zhou, A., Duan, H., 2015. Nanoscale 7, 7790–7801. Wang, L., Duan, X., Wang, G., Liu, C., Luo, S., Zhang, S., Zeng, Y., Xu, Y., Liu, Y., Duan, X., 2016. Appl. Catal., B 186, 88–96. Wang, Y., Zang, D., Ge, S., Ge, L., Yu, J., Yan, M., 2013. Electrochim. Acta 107, 147–154. Whitcombe, M.J., Chianella, I., Larcombe, L., Piletsky, S.A., Noble, J., Porter, R., Horgan, A., 2011. Chem. Soc. Rev. 40, 1547–1571. Xin, Y., Li, Z., Zhang, Z., 2015. Chem. Commun. 51, 15498–15501. Xu, Q., Yu, J., Zhang, J., Zhang, J., Liu, G., 2015a. Chem. Commun. 51, 7950–7953. Xu, Y., Han, T., Li, X., Sun, L., Zhang, Y., Zhang, Y., 2015b. Anal. Chim. Acta 891, 298–303. Xue, L., Zhou, X., Xing, D., 2012. Anal. Chem. 84, 3507–3513. Yildirim, N., Long, F., Gao, C., He, M., Shi, H.-C., Gu, A.Z., 2012. Environ. Sci. Technol. 46, 3288–3294. Zhang, S., Li, J., Wang, X., Huang, Y., Zeng, M., Xu, J., 2014. ACS Appl. Mater. Interfaces 6, 22116–22125. Zhang, Y., Xiao, J., Lv, Q., Wang, L., Dong, X., Asif, M., Ren, J., He, W., Sun, Y., Xiao, F., Wang, S., 2017. ACS Appl. Mater. Interfaces 9, 38201–38210. Zhao, S., Chen, S., Yu, H., Quan, X., 2012. Sep. Purif. Technol. 99, 50–54. Zhao, W.-W., Shan, S., Ma, Z.-Y., Wan, L.-N., Xu, J.-J., Chen, H.-Y., 2013. Anal. Chem. 85, 11686–11690. Zhao, W.-W., Xu, J.-J., Chen, H.-Y., 2014a. Chem. Rev. 114, 7421–7441. Zhao, W., Zhang, M., Ai, Z., Yang, Y., Xi, H., Shi, Q., Xu, X., Shi, H., 2014b. J. Phys. Chem. C. 118, 23117–23125. Zheng, L., Han, S., Liu, H., Yu, P., Fang, X., 2016. Small 12, 1527–1536. Zhou, W., Yin, Z., Du, Y., Huang, X., Zeng, Z., Fan, Z., Liu, H., Wang, J., Zhang, H., 2013. Small 9, 140–147.
Ansari, S.A., Husain, Q., 2012. Biotechnol. Adv. 30, 512–523. Bai, S., Wang, L., Chen, X., Du, J., Xiong, Y., 2015. Nano Res. 8, 175–183. Bai, X., Hou, H., Zhang, B., Tang, J., 2014. Biosens. Bioelectron. 56, 112–116. Chen, D., Cao, L., Huang, F., Imperia, P., Cheng, Y.-B., Caruso, R.A., 2010. J. Am. Chem. Soc. 132, 4438–4444. Fuchs, Y., Soppera, O., Haupt, K., 2012. Anal. Chim. Acta 717, 7–20. Gao, P., Ma, H., Yang, J., Wu, D., Zhang, Y., Du, B., Fan, D., Wei, Q., 2015. New J. Chem. 39, 1483–1487. Heydari-Bafrooei, E., Askari, S., 2017. Int. J. Hydrog. Energy 42, 2961–2969. Huang, K.-J., Liu, Y.-J., Wang, H.-B., Wang, Y.-Y., Liu, Y.-M., 2014. Biosens. Bioelectron. 55, 195–202. Huang, K.-J., Liu, Y.-J., Zhang, J.-Z., Cao, J.-T., Liu, Y.-M., 2015. Biosens. Bioelectron. 67, 184–191. Leung, K.-H., He, H.-Z., Chan, D.S.-H., Fu, W.-C., Leung, C.-H., Ma, D.-L., 2013. Sens. Actuators, B 177, 487–492. Li, C., Wang, H., Shen, J., Tang, B., 2015a. Anal. Chem. 87, 4283–4291. Li, H., Li, J., Qiao, Y., Fang, H., Fan, D., Wang, W., 2017. Sens. Actuators, B 243, 1027–1033. Li, H., Qiao, Y., Li, J., Fang, H., Fan, D., Wang, W., 2016. Biosens. Bioelectron. 77, 378–384. Li, J., Cushing, S.K., Zheng, P., Senty, T., Meng, F., Bristow, A.D., Manivannan, A., Wu, N., 2014a. J. Am. Chem. Soc. 136, 8438–8449. Li, Q., Zhang, N., Yang, Y., Wang, G., Ng, D.H.L., 2014b. Langmuir 30, 8965–8972. Li, R., Liu, Y., Cheng, L., Yang, C., Zhang, J., 2014c. Anal. Chem. 86, 9372–9375. Li, X., Zheng, Z., Liu, X., Zhao, S., Liu, S., 2015b. Biosens. Bioelectron. 64, 1–5. Li, Y.-J., Ma, M.-J., Zhu, J.-J., 2012. Anal. Chem. 84, 10492–10499. Lin, X., Cui, L., Huang, Y., Lin, Y., Xie, Y., Zhu, Z., Yin, B., Chen, X., Yang, C.J., 2014. Chem. Commun. 50, 7646–7648. Liu, C., Wang, L., Tang, Y., Luo, S., Liu, Y., Zhang, S., Zeng, Y., Xu, Y., 2015a. Appl. Catal., B 164, 1–9. Liu, E., Fan, J., Hu, X., Hu, Y., Li, H., Tang, C., Sun, L., Wan, J., 2015b. J. Mater. Sci. 50, 2298–2305. Liu, P., Huo, X., Tang, Y., Xu, J., Liu, X., Wong, D.K.Y., 2017. Anal. Chim. Acta 984, 86–95. Liu, X., Xing, Z., Zhang, H., Wang, W., Zhang, Y., Li, Z., Wu, X., Yu, X., Zhou, W., 2016.
201
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
A photoelectrochemical aptasensor based on a 3D flower-like TiO2-MoS2gold nanoparticle heterostructure for detection of kanamycin
T
⁎
Xiaoqiang Liua, , Peipei Liua, Yunfei Tanga, Liwei Yanga, Lele Lia, Zhichong Qia, Deliang Lia, ⁎ Danny K.Y. Wongb, a
Henan Joint International Research Laboratory of environmental pollution control materials, College of Chemistry and Chemical Engineering, Henan University, Kaifeng, Henan Province 475004, PR China b Department of Molecular Sciences, Macquarie University, Sydney, NSW 2109, Australia
A R T I C LE I N FO
A B S T R A C T
Keywords: TiO2-MoS2-AuNP composite Photoelectrochemical aptasensor Kanamycin detection Visible light excitation
In this work, a sensitive photoelectrochemical aptasensor was developed for kanamycin detection using an enhanced photocurrent response strategy, which is based on the surface plasmon resonance effect of gold nanoparticles deposited on a 3D TiO2-MoS2 flower-like heterostructure. A significant aspect of this development lies in the photoelectrochemical and morphological features of the unique ternary composite, which have contributed to the excellent performance of the sensor. To develop an aptasensor, mercapto-group modified aptamers were immobilised on the photoactive composite as a recognition unit for kanamycin. The TiO2-MoS2AuNP composite was demonstrated to accelerate the electron transfer, increase the loading of aptamers and improve the visible light excitation of the sensor. Under optimal conditions, the aptasensor exhibited a dynamic range from 0.2 nM to 450 nM of kanamycin with a detection limit of 0.05 nM. Overall, we have successfully synergised both the electrical and the optical merits from individual components to form a ternary composite, which was then demonstrated as an effective scaffold for the development of PEC biosensors.
1. Introduction As a new emerging analytical technique, photoelectrochemical (PEC) sensors have attracted increasing research interest (Li et al., 2015a; Wang et al., 2013). In this technique, an illumination is used to excite electrons in a photoactive material from the valence band to the conductive band, while a potential is applied to prevent the recombination of the photogenerated holes and electrons. An electroactive analyte is then oxidised by the photogenerated holes to produce a photocurrent (Gao et al., 2015; Li et al., 2015b). Hence, this is a technique with a unique capability of preventing the excitation source from interfering with the photocurrent signal to yield a high signal-tonoise ratio measurement response (Li et al., 2016). Unfortunately, PEC sensing is usually restricted by its poor specificity to the target analyte. This has stimulated studies of incorporating many recognition elements including molecular imprinted polymers, enzymes, antibodies and aptamers to improve its selectivity (Li et al., 2012; Whitcombe et al., 2011; Zhao et al., 2013, 2014a). Among them, aptamers (singlestranded oligonucleotides with specific sequences) are some of the strong competitors to the other biological recognition systems in analytical applications due to their low-cost in vitro synthesis, mass
⁎
production, reduced immunogenic response, good selectivity, and inherent binding affinity (Huang et al., 2015; Xin et al., 2015). In addition, aptamers are known to be less sensitive to temperature and pH, and are therefore less prone to deactivation, compared to enzymes. Most importantly, aptamers can be used to detect both small molecules and biological macromolecules including proteins, nucleic acids, polypeptides, hormones, sugars, viral particles, enzymes, etc, while molecular imprinted polymers and enzymes are usually restricted to detection of small molecules (Ansari and Husain, 2012; Fuchs et al., 2012; Xue et al., 2012). In general, the features of materials immobilised on an electrode surface will significantly affect the performance of the electrochemical and PEC biosensors. For example, TiO2 was very often used in developing electrochemical and PEC sensors due to its strong photocatalytic activity, good stability, controllable morphology and excellent biocompatibility (Liu et al., 2017). Unfortunately, TiO2 is only amenable to excitation by UV light, which will otherwise damage any biomolecules immobilised on a biosensor (Li et al., 2014a). In addition, the fast electron-hole recombination may significantly attenuate the PEC signal, which has also limited the applications of TiO2 (Luo et al., 2012; Yildirim et al., 2012). To circumvent this problem, many narrow
Corresponding authors. E-mail addresses: [email protected] (X. Liu), [email protected] (D.K.Y. Wong).
https://doi.org/10.1016/j.bios.2018.04.041 Received 24 January 2018; Received in revised form 11 April 2018; Accepted 17 April 2018 Available online 19 April 2018 0956-5663/ © 2018 Elsevier B.V. All rights reserved.
Biosensors and Bioelectronics 112 (2018) 193–201
X. Liu et al.
human (Song et al., 2011). In our work, the ternary composite design was developed using multi-step solution chemistry illustrated in Scheme 1. We have firstly synthesised mesoporous TiO2 microspheres with a large and rough surface, which acted as an effective platform for depositing an enhanced quantity of MoS2 nanosheets. AuNPs were then deliberately decorated on the surface of the 3D TiO2-MoS2 heterostructure to produce the ternary composite. In this way, plasmonic AuNPs will directly transform the incident visible light into electrical energy by injecting photogenerated electrons into the conduction band of MoS2, which will thus improve the photoelectric conversion efficiency. As demonstrated below, compared with a previously reported AuNP functionalised TiO2 nanotube array-based PEC aptasensor (Xin et al., 2015), the presence of MoS2 has increased the conductivity and visible light absorption. Finally, SH-functionalised kanamycin aptamers were immobilised on the modified electrode to develop an aptasensor. The specific interaction between the aptamer and kanamycin is expected to yield an increase in photocurrent response owing to the oxidation of kanamycin by photogenerated holes.
bandgap photoactive materials such as g-C3N4 (~2.7 eV) and MoS2 (~1.9 eV) nanomaterials have been hybridised with TiO2 to form a composite with improved photocatalytic or PEC performance (Zhao et al., 2012; Zhou et al., 2013). For example, our group has previously synthesised a TiO2 nanosheet-g-C3N4 heterostructure and applied it as a scaffold on an electrode to develop a PEC glucose sensor, which exhibited a 0.05 – 16 mM linear range and a 0.01 mM glucose detection limit under visible light (Liu et al., 2017). However, the need for exfoliating g-C3N4 by ultrasonication always resulted in micrometer-size sheets, which are insufficiently small to exhibit some of the nanomaterial properties (Zhang et al., 2014). More recently, molybdenum disulfide (MoS2) has received considerable attention due to its narrow bandgap, high thermal stability and special optical properties (Huang et al., 2014; Li et al., 2014b). In particular, the narrow bandgap of MoS2 makes this material more suitable in PEC sensor development. In addition, MoS2 is also regarded as a promising electrode material in battery developments owing to its 2D lamellar structure that is analogous to graphene (Ren et al., 2017). Accordingly, MoS2 possesses many advantages similar to graphene such as large surface area and high conductivity (Sun et al., 2015). As an example, Zhou et al. synthesised a MoS2-TiO2 heterostructure with a few-layer MoS2 nanosheets coated on TiO2 nanobelts (Zhou et al., 2013). The heterostructure was demonstrated to significantly retard the recombination between photogenerated electrons and holes due to the energy band matching between TiO2 and MoS2. Therefore, the photocatalytic activity of the composite with 50 wt% of MoS2 has increased by a factor of 3 under visible light irradiation, relative to TiO2 nanobelts alone. In this paper, we reported the development of a PEC aptasensor consisting of a novel TiO2-MoS2-gold nanoparticle (AuNP) flower-like nanocomposite using kanamycin as a model analyte. Notably, kanamycin has been widely used to treat serious infections caused by Grampositive and Gram-negative bacteria during protein synthesis (Xu et al., 2015b). However, an overdose or residue of this drug in animal-derived foods can cause ototoxicity, nephrotoxicity and antibiotic resistance in
2. Experimental 2.1. Materials and reagents All reagents were of analytical reagent grade and were used without any further purification. Titanium(IV) isopropoxide (TTIP, 97+% purity) was purchased from Alfa Aesar Chemical Co. Ltd Tianjin, China. Sodium molybdite dihydrate, thiourea, hexadecylamine (HDA, 90%), acetic acid, chloramphenicol, erythromycin, streptomycin, doxycycline, ciprofloxacin and ofloxacin were purchased from Aladdin Reagent Co., Ltd., Shanghai, PR China. Gold(Ⅲ) chloride trihydrate, chitosan (85% deacetylation) and bovine serum albumin (BSA) were acquired from Sigma-Aldrich Chemical Co. (St. Louis, MO). Sodium borohydride, sodium citrate, potassium ferricyanide, potassium ferrocyanide and potassium chloride were obtained from Sinopharm Chemical Reagent Co., Ltd., China. Kanamycin sulfate and a mercapto group modified
Scheme 1. Synthesis of TiO2-MoS2-AuNP composite; fabrication steps of a kanamycin aptasensor and its PEC mechanism for detection. 194
Biosensors and Bioelectronics 112 (2018) 193–201
X. Liu et al.
Ltd., China) was initially cleaned using NaOH (1 M) and H2O2 (30%), and then washed successively with acetone, alcohol and ultrapure water under sonification, and dried in air. Meanwhile, 20 mg of a TiO2MoS2-AuNP composite was dispersed in 1 mL chitosan (0.2 wt% in acetic acid) solution, which was sonicated for 20 min to obtain a homogeneous suspension. Next, 25 μL of the suspension was applied on the cleaned ITO and dried at ambient temperature. Subsequently, 25 μL of aptamer (1.5 μM) was immobilised on the TiO2-MoS2-AuNP|ITO, and the electrode was left at 4 °C for 1 h. Finally, the modified electrode was incubated in 25 μL of 3% (w/v) BSA solution for 30 min to block any remaining active sites to minimise nonspecific adsorption. The obtained aptasensor, denoted as BSA|aptamer|TiO2-MoS2-AuNP|ITO, was stored at 4 °C before use. The procedure for the aptasensor fabrication is illustrated in Scheme 1.
kanamycin-binding DNA aptamer with the sequence 5´-HS-(CH2)6TGG-GGG-TTG-AGG-CTA-AGC-CGA-3´ were purchased from Shanghai Sangon Biotechnology Co. Ltd., Shanghai, China. Phosphate buffered saline (PBS, pH 7.4) was obtained by mixing the corresponding stock solutions of 0.1 M KH2PO4, K2HPO4 and KCl. Indium tin oxide (ITO; resistivity 10 Ω cm−2) electrodes were acquired from Wuhan Lattice Solar Energy Technology, Ltd., China. Ultrapure water obtained from a Millipore water purification system (≥18 MΩ cm−1, Milli-Q, Millipore) was used in the assays. 2.2. Material characterisations Surface morphology of prepared nanomaterials was examined by scanning electron microscopy (JSM-7610F, JEOL, Japan) at an accelerating voltage of 15.0 kV. The powder samples were ultrasonically dispersed in ethanol, applied on a silicon wafer and dried. Chemical composition was studied by X-ray diffraction (XRD, Bruker D8 Advance, Germany) with a Cu Kα radiation. X-ray photoelectron spectroscopy (XPS) analysis was collected on an X-ray photoelectron spectrometer (Thermo Scientific Escalab 250Xi, USA). UV–visible diffuse reflectance spectroscopy of samples was performed using a UV–vis spectrophotometer (UV-2600, Kyoto, Japan). Electrochemical impedance spectroscopic measurements were carried out at an IM6ex electrochemical station (EIS, ZAHNER, Germany).
2.5. PEC and impedance measurements All PEC measurements at TiO2|ITO, MoS2|ITO, TiO2-MoS2|ITO, TiO2-MoS2-AuNP|ITO and BSA|aptamer|TiO2-MoS2-AuNP|ITO were performed at a CHI 630D electrochemical workstation (Shanghai, CH Instruments, China) accommodated with a three-electrode system. The modified ITO acted as a working electrode, a platinum wire as a counter electrode and a Ag|AgCl (3.0 M KCl) as a reference electrode. A Xe lamp equipped with a 400 nm cut-off filter (CEL-HXUV 300, Beijing AULTT, China) was used as an irradiation source. The electrolyte was composed of an aqueous solution of 0.1 M KCl and 0.1 M PBS (pH 7.4). The photocurrent at the modified ITO electrode was recorded versus time with alternate light-on and light-off cycles at a fixed bias voltage of 0.6 V. Impedance experiments were performed in a solution containing 5 mM K3[Fe(CN)6] and 5 mM K4[Fe(CN)6] in 0.1 M KCl at a DC potential of 0.20 V superimposed by an alternating voltage of 5 mV peakto-peak amplitude. Impedance was measured over a frequency range of 100 kHz to 0.1 Hz.
2.3. Preparation of TiO2-MoS2-AuNP composites A TiO2-MoS2-AuNP composite was prepared according to a multistep procedure illustrated in Scheme 1. The anatase TiO2 microspheres were firstly prepared using Chen et al.'s procedure (Chen et al., 2010) with minor modifications. In this procedure, 0.88 g of HDA was dissolved in 100 mL of absolute ethanol, followed by the addition of 0.40 mL of KCl solution (0.1 M) and 0.25 mL of ultrapure water. The mixture was stirred at ambient temperature and 2.16 mL of TTIP was added to this solution under vigorous stirring. The milky white bead suspension was kept quiescent for 18 h and then centrifuged, and the amorphous TiO2 beads were washed with ethanol three times and dried in air at 60 °C for 4 h. Finally, anatase TiO2 microspheres were obtained by a solvothermal reaction. In this reaction, 0.50 g amorphous TiO2 beads were dispersed in a solution containing 20 mL absolute ethanol and 10 mL ultrapure water. The mixture was transferred into a 50 mL Teflon-lined autoclave at 160 °C for 90 min. The final product (white powder) was collected by centrifugation, washed with ethanol and dried at room temperature. To synthesise the TiO2-MoS2 heterostructure, 0.08 g anatase TiO2 microspheres were dispersed in a mixture of 10 mL ethanol and 20 mL ultrapure water, and 0.35 g sodium molybdite dihydrate and 0.52 g thiourea were dissolved in the above solution, followed by a solvothermal reaction at 220 °C for 20 h. After being cooled down to room temperature, the sample was rinsed with ultrapure water and dried at 60 °C. MoS2 nanosheets were synthesised using the same procedure without adding anatase TiO2 microspheres. In preparing a TiO2-MoS2-AuNP composite, 2 mL of 0.01 M HAuCl4 and 2 mL of 0.01 M sodium citrate (a capping agent) were added into a uniform suspension of 40 mg TiO2-MoS2 heterostructure in 40 mL ultrapure water with continuous stirring. Next, 2 mL fresh NaBH4 solution (0.1 M, 0–4 °C) was added to the suspension with vigorous stirring for 30 min. The suspension was then stirred for another 30 min at room temperature before it was left undisturbed overnight. The composite was centrifuged and sequentially washed with ultrapure water and ethanol, before it was dried in a 60 °C vacuum overnight.
3. Results and discussion In this paper, a novel ternary composite of TiO2-MoS2-AuNP was synthesised and used as a scaffold for the construction of an aptasensor. The composite has provided a large and biocompatible platform to accommodate an enhanced loading of aptamers, all of which have contributed to the excellent electrochemical and PEC features. In the following sections, each component of the scaffold was carefully characterised before being assembled to develop the scaffold of a kanamycin aptasensor. 3.1. Morphological investigations of the nanocomposite The dimension and morphology of the nanomaterials used in this work were initially investigated using scanning electron microscopy. As displayed in Fig. 1(a), uniform TiO2 microspheres with a smooth surface are clearly visible and their average diameter was estimated to be 400 nm. These TiO2 microspheres were then solvothermally treated to form anatase TiO2 beads. According to the scanning electron micrograph (SEM) in Fig. 1(b), the surface of anatase TiO2 beads appears to be extremely rough. Nonetheless, their average diameter is undistinguishable from that of smooth TiO2 spheres. Meanwhile, the SEM of MoS2 samples in Fig. 1(c) clearly displays numerous curly MoS2 nanosheets aggregated together to form a bulky structure with a rough surface. These MoS2 samples are of irregular shapes with a few micrometers in size. In Fig. 1(d), the SEM of TiO2-MoS2 composite revealed that MoS2 nanosheets grew vertically on anatase TiO2 microspheres during the solvothermal reaction. More specifically, each TiO2 microsphere was uniformly covered with thick MoS2 nanosheets, showing a flower-like structure of the TiO2-MoS2 heterostructure. The average diameter of the composite is approximately 600 nm (standard deviation 20 nm; n = 6). The SEM of TiO2-MoS2-AuNP is displayed in
2.4. Construction of the PEC aptasensor To construct the PEC aptasensor, an ITO electrode (1 cm × 3 cm × 1.1 mm thick; acquired from Wuhan Lattice Solar Energy Technology, 195
Biosensors and Bioelectronics 112 (2018) 193–201
X. Liu et al.
Fig. 1. SEM of (a) amorphous TiO2, (b) anatase TiO2, (c) MoS2, (d) TiO2-MoS2, (e) and (f) TiO2-MoS2-AuNP samples. The arrows in (f) indicate where AuNPs are located.
nanohybrid paper electrode assembled from 3D ionic liquid functionalised graphene framework decorated by gold nanoflowers (Zhang et al., 2017). Therefore, the elemental composition and chemical state evaluated from the above XPS results have indicated successful immobilisation of elemental gold on the TiO2-MoS2 to form a ternary nanocomposite. XRD is another powerful technique to further verify the formation of a TiO2-MoS2-AuNP composite. The existence of the individual components can be evaluated by comparing their XRD patterns obtained with the standard patterns of Joint Committee on Powder Diffraction Standards (JCPDS). The XRD patterns of TiO2, MoS2, TiO2-MoS2, and TiO2-MoS2-AuNP nanocomposite are shown in Fig. 3(A). Firstly, we identified the anatase phase of TiO2 microspheres after comparing its pattern (trace a) to the standard pattern of JCPDS (no. 21-1272) (Xu et al., 2015a). The XRD pattern of MoS2 nanosheets (trace b) presented the characteristic diffraction peaks at 14.2°, 33.3°, 39.5° and 58.6°, unambiguously assigned to the (002), (100), (103), (110) peaks of hexagonal MoS2 (JCPDS no. 37–1492) (Heydari-Bafrooei and Askari, 2017). The XRD pattern of TiO2-MoS2 (trace c) was dominated by MoS2 except for the weak peaks at ~25.3° and 48.0° ascribed to the (101) and (200) diffraction of anatase TiO2 (Zhao et al., 2014b), which demonstrated the co-existence of MoS2 and TiO2 in the samples. The weak peaks corresponding to TiO2 could be attributed to dense coverage of TiO2 microspheres by MoS2 nanosheets. As for TiO2-MoS2-AuNP composites, trace d exhibited several new peaks at 38.6°, 44.8°, 64.9° and 77.8° corresponding to the (110), (200), (220), and (311) reflections of AuNPs (JCPDS no. 01-1172) (Mehta et al., 2016), which indicated that AuNPs were deposited on the TiO2-MoS2 heterostructure to form a ternary composite. Therefore, the XRD and XPS results are in agreement with each other, indicating a successful preparation of the TiO2-MoS2AuNP composite. Furthermore, the crystal forms of TiO2 and MoS2 were shown to be anatase and a hexagonal system, respectively, both of which will offer strong photocurrent conversion efficiency.
Fig. 1(e), where a large number of shiny points is observed on the surface of the ternary composite, demonstrating the successful deposition of AuNP on the TiO2-MoS2 heterostructure. The magnified SEM of the TiO2-MoS2-AuNP composite in Fig. 1(f) shows that many small particles of ~5–10 nm diameters were evenly distributed on the MoS2 nanosheets. In summary, the above results presented a visual indication for the formation of a ternary composite. As discussed below, additional techniques were used to identify the composition, chemical state and crystal form of all individual components in the ternary composite to provide further supporting evidence for the formation of the composite. Fig. 2(a) shows a typical overall XPS survey spectrum that suggests the presence of Ti, O, Mo, S, C and Au on the ternary composite surface. More specifically, the high resolution Ti 2p XPS spectrum in Fig. 2(b) shows two peaks at binding energies of 459.2 eV and 464.98 eV, which correspond to Ti 2p3/2 and Ti 2p1/2 of Ti4+, respectively, after comparing to the two peaks at 457.8 eV and 463.6 eV in Liu et al.’s work involving the fabrication of 3D mesoporous TiO2/MoS2/TiO2 nanosheets for visible-light-driven photocatalysis (Liu et al., 2016). Similarly, as displayed in Fig. 2(c), an O 1s peak at 530.5 eV (compared to 530.2 eV in Liu et al's work (Liu et al., 2015b)) is attributable to the TiO-Ti bond, and the peak at 532.0 eV (compared to 531.2 eV in (Li et al., 2017)) is assigned to adsorbed water. In Fig. 2(d), the peak at 226.8 eV is assigned to S 2s, and the peaks at 229.1 eV and 232.3 eV are attributed to Mo 3d5/2 and Mo 3d3/2, as suggested by Bai et al., who prepared chemically exfoliated metallic MoS2 nanosheets for photocatalytic enhancement (Bai et al., 2015). They also suggested the peak at 236 eV was attributed to the Mo6+ 3d, which was produced by the oxidation of Mo4+ (Ren et al., 2017). In Fig. 2(e), two peaks at 161.7 eV and 162.9 eV are respectively attributable to S 2p3/2 and S 2p1/2 at the state of S2-, as demonstrated by Zheng et al.'s work on preparing hierarchical MoS2 nanosheet@TiO2 nanotube array composites with enhanced photocatalytic performance (Zheng et al., 2016). Finally, Fig. 2(f) depicts the binding energies of Au 4f7/2 at 84.1 eV and Au 4f5/2 at 87.9 eV, which matched well with the standard binding energy of Au (84.0 and 87.6 eV), as shown in Zhang et al.'s work involving preparation of 196
Biosensors and Bioelectronics 112 (2018) 193–201
X. Liu et al.
Fig. 2. XPS analysis of TiO2-MoS2-AuNP composite: (a) survey scans, and high-resolution spectra of (b) Ti 2p, (c) O 1s, (d) Mo 3d, (e) S 2p and (f) Au 4f.
3.2. UV–vis and electrochemical impedance spectroscopic studies
displayed strong absorption in both UV and visible regions ascribed to its narrow band gap (1.2–1.9 eV) (Wang et al., 2016). Therefore, the absorption of the binary composite in the visible region (above 400 nm), as displayed in trace c, was greatly enhanced after TiO2 microspheres were decorated with MoS2. Compared to TiO2-MoS2, the extra absorption band of the ternary TiO2-MoS2-AuNP composite was observed in the visible wavelength of 510–675 nm, corresponding to the plasmon peak of AuNPs (Xin et al., 2015). More specifically, the enhanced absorption was attributed to AuNPs as an excellent electronic
As an effective complementary technique to XRD, UV–vis diffuse reflectance spectrometry was performed to evaluate the optical absorption properties of TiO2, MoS2, TiO2-MoS2 heterostructure and TiO2MoS2-AuNP. As illustrated by trace a in Fig. 3(B), anatase TiO2 microspheres only exhibited a noticeable absorption at wavelength lower than 400 nm due to the considerable energy bandgap (3.2 eV) of anatase TiO2 (Liu et al., 2015a). As shown in trace b, hexagonal MoS2
197
Biosensors and Bioelectronics 112 (2018) 193–201
X. Liu et al.
Fig. 3. (A) XRD patterns of (a) anatase TiO2, (b) MoS2, (c) TiO2-MoS2, and (d) TiO2-MoS2-AuNP composite; (B) UV–vis diffuse reflectance spectra of (a) anatase TiO2, (b) MoS2, (c) TiO2-MoS2 and (d) TiO2-MoS2-AuNP composite; (C) Nyquist plots of [Fe(CN)6]3-/[Fe(CN)6]4- at (a) bare ITO, (b) TiO2|ITO, (c) TiO2-MoS2|ITO, (d) TiO2MoS2-AuNP|ITO, (e) aptamer|TiO2-MoS2- AuNP|ITO and (f) BSA|aptamer|TiO2-MoS2-AuNP|ITO; (D) Time-based photocurrent response of (a) MoS2|ITO, (b) TiO2|ITO, (c) TiO2-MoS2|ITO, (d) TiO2-MoS2-AuNP|ITO and (e) BSA|aptamer|TiO2-MoS2-AuNP|ITO in 0.1 M PBS (pH 7.4) containing 0.1 M KCl at a bias voltage of 0.6 V in the absence of kanamycin.
the deposition of AuNPs on the TiO2-MoS2 further decreased the Ret value by 60% to ~78 Ω (trace d), attributed to the excellent electronic conductivity of AuNPs. However, as shown in trace e and trace f, the Ret increased gradually to ~121 Ω and ~157 Ω after successive modification with the kanamycin aptamer and BSA, respectively, owing to the strong steric hindrance and insulation effect of proteins on [Fe(CN)6]3-/ [Fe(CN)6]4-. Therefore, these results provided supporting evidence that the materials were successively immobilised on the electrode surface.
conductor, which both promoted the transfer of MoS2 photoelectrons and accelerated the separation of photo-carriers, leading to an obvious decreased recombination rate. Accordingly, a AuNP-modified TiO2MoS2 heterostructure will significantly benefit the performance of PEC sensing. Following the characterisation of the prepared nanomaterials, electrochemical impedance spectroscopy was used to probe the stepwise assembly of the PEC biosensor due to its strong capability of studying the charge transfer on an electrode/solution interface. As shown in Fig. 3(C), Nyquist (imaginary impedance (Z”) versus real impedance (Z’)) plots for different nanomaterial modified electrodes were recorded in the sequential fabrication steps over a frequency range from 100 kHz to 0.1 Hz. In these plots, the diameter of a semicircle in the high frequency region corresponds to the electron transfer resistance (Ret) of the redox marker, [Fe(CN)6]3-/[Fe(CN)6]4-, at the corresponding electrode. Only qualitative evaluation of results is conducted here by comparing the relative sizes of all semicircles. For example, in trace a, the bare ITO displayed the smallest semicircle, and hence the smallest Ret (~40 Ω), among all the electrodes. In contrast, the largest Ret was obtained at TiO2 microsphere modified electrode surface as shown in trace b, probably due to the poor electron transfer capacity of TiO2. The Ret (~195 Ω) from the Nyquist plot at the TiO2MoS2 modified electrode surface in trace c is 25% smaller than that at TiO2 (trace b，~259 Ω), most likely because the introduction of MoS2 accelerated electron transfer of the probe on the electrode. As expected,
3.3. PEC characterisation of the nanocomposites Next, both the optical absorption properties of the nanomaterials and assembling process of the PEC sensor in Section 3.2 were further confirmed by the photocurrent responses at different modified electrodes. In all these experiments, a photocurrent was obtained after applying a visible illumination. As depicted by trace a in Fig. 3(D), pristine MoS2 nanosheets show the smallest photocurrent (0.20 μA) among all the modified electrodes probably because MoS2 is an indirect gap semiconductor with poor excitation capability. The TiO2 microsphere modified ITO (trace b) also displayed a small photocurrent (0.83 μA), attributed to the poor visible light response of anatase TiO2 with large band gap. The photocurrent at MoS2-TiO2 heterostructure (4.1 μA) modified ITO (trace c) is ~3.9 times larger than that at TiO2 microsphere modified ITO (trace b), most likely because the intimate interfacial contact between MoS2 and TiO2 in the composite increased 198
Biosensors and Bioelectronics 112 (2018) 193–201
X. Liu et al.
Fig. 4. Influence of (A) aptamer concentration, (B) bias potential on PEC response of BSA|aptamer|TiO2-MoS2-AuNP|ITO to 200 nM kanamycin in 0.1 M PBS (pH 7.4) containing 0.1 M KCl; (C) Photocurrent at BSA|aptamer|TiO2-MoS2-AuNP|ITO in 0.1 M PBS (pH 7.4) containing 0.1 M KCl at a bias voltage of 0.6 V before and after incubation with 0, 0.05, 0.2, 2, 20, 50, 100, 200, 300, 400, 450, 500, 550 nM (from a to m) of kanamycin and (D) calibration based on PEC response of aptasensor and kanamycin concentration from 0.2 to 550 nM.
photocurrent response of the BSA|aptamer|TiO2-MoS2-AuNP|ITO modified electrode monotonically increased with the enhancement of the applied potential from 0.2 V to 0.6 V. However, when the bias potential has exceeded 0.6 V, the photocurrent was increasing at a slower rate, most likely because a weaker separation effect on the photogenerated charges. Accordingly, a bias potential of 0.6 V was adopted as the optimum potential for further experiments to minimise the oxidation reaction at higher potential.
the visible light absorption and enhanced the charge separation. Notably, TiO2-MoS2-AuNP|ITO (trace d) showed the highest photocurrent (8.4 μA) among all the electrodes, which is approximately a 2-fold increment of that observed at the MoS2-TiO2 modified ITO (trace c). The enhanced photocurrent could be attributed to the enlarged charge separation by the strong visible light adsorption owing to the localised surface plasmon resonance effect of AuNPs. After immobilising the kanamycin aptamer and BSA on the TiO2-MoS2-AuNP|ITO (trace e), the photocurrent (6.0 μA) has noticeably decreased by 28%. The steric hindrance experienced by the biological molecules on ITO prohibited photogenerated electrons from moving to ITO and this has thus accelerated their recombination with the holes.
3.5. PEC detection of kanamycin Under optimal conditions of 1.5 μM aptamer concentration and 0.6 V applied potential, detection of kanamycin was conducted at the fabricated PEC aptasensor using visible light illumination. In Fig. 4(C), the time (t)-dependent PEC photocurrent (the I-t) plots obtained at a BSA|aptamer|TiO2-MoS2-AuNP|ITO biosensor was recorded when the kanamycin concentration (C) was increased from 0 to 550 nM. The periodic photocurrent generated by the PEC oxidation of kanamycin rapidly responded to the repeated on-off cycles of irradiation, and the photocurrent remained highly stable at each kanamycin concentration during the PEC process. The inset in Fig. 4(C) shows the photocurrent measured over the 0–2 nM kanamycin concentration range, demonstrating the good performance of the sensor at low concentrations. Based on a background corrected current (ΔI), a linear calibration relationship between 0.2 and 450 nM was obtained, as shown in Fig. 4(D). However, when kanamycin concentration exceeded 450 nM, ΔI began to deviate from the linear range. The regression equation is determined as ΔI / μA = 0.011 ×C / μA + 0.055 with a correlation coefficient of 0.996 (n = 6), which was found to be statistically significant at the 95% confidence level using t-test. In addition, the result of an analysis of
3.4. Optimisation of the aptamer sensor Several important experimental parameters were optimised before the PEC sensor performance was evaluated. As described in Section 3.3, the loading of an aptamer on the electrode exhibited a major effect on the photocurrent of the aptasensor and was therefore optimised. Fig. 4(A) shows a photocurrent against aptamer concentration plot. In this figure, the photocurrent was observed to increase between 0.5 μM and 1.5 μM aptamer as a result of increasing kanamycin molecules being captured on the electrode surface, leading to larger oxidation current. However, when the aptamer concentration was higher than 1.5 μM, the photocurrent decreased, possibly due to excessive aptamers acting as steric hindrance to the transfer of electrons. Therefore, 1.5 μM aptamer was chosen as the optimum concentration. The bias potential applied on the photoelectrode drives the photoelectrons away from the photogenerated holes and therefore affects the recombination rate of the electron-hole pairs. As shown in Fig. 4(B), the 199
Biosensors and Bioelectronics 112 (2018) 193–201
X. Liu et al.
Table 1 Comparison of analytical performance of different techniques for kanamycin detection. Analytical technique
Linear range / nM
Detection limit / nM
References
Electrochemistry Electrochemistry Colorimetry Aptamer-based cantilever Luminescence Fluorescence PEC PEC PEC
10–150 50–9.0 × 103 1–100 1.0 × 105–1.0 × 107 200–1.5 × 105 10–1 × 103 1－230 0.2－200 0.2－450
5.8 9.4 1.49 5 × 104 143 2.7 0.2 0.1 0.05
(Sun et al., 2014) (Li et al., 2012) (Sharma et al., 2014) (Bai et al., 2014) (Leung et al., 2013) (Lin et al., 2014) (Li et al., 2014c) (Xin et al., 2015) This work
respectively, demonstrating the excellent long-term storage stability of the aptasensor. These results support the feasibility of this aptasensor for sensitive kanamycin assay.
variance also supported a significant linear relationship between C and ΔI. Further, using a signal-to-noise ratio of 3, the detection limit was estimated to be 0.05 nM (the response obtained at 0.05 nM is included in the inset in Fig. 4C). For comparison, the results obtained by several previously reported methods for the detection of kanamycin are displayed in Table 1. While many of them in Table 1 have exhibited either a good linear range or a low detection limit, the PEC aptasensor prepared in this work has shown superior analytical performance in both analytical characteristics.
3.7. Practical application of the aptasensor in milk samples The practical application of the aptasensor was assessed by detecting different concentrations of kanamycin in milk samples. The milk obtained from a local supermarket was firstly centrifuged at 15,000 rpm for 20 min and then the supernatant was diluted ten times with PBS. Subsequently, kanamycin standard solutions were spiked into the diluted milk, yielding final concentrations of 2, 20, 50, 100 nM. Finally, the recoveries of four different concentrations of kanamycin in the spiked milk samples were obtained using the PEC aptasensor. As shown in Table 2, the recoveries are in the range from 91.5% to 97.1%, clearly indicating the successful application of the PEC aptasensor for detection of kanamycin in real-life samples.
3.6. Specificity, reproducibility and stability of the PEC biosensor The selectivity of the aptasensor was investigated by comparing the photocurrent response to 200 nM kanamycin against that to several antibiotic interferents including chloramphenicol, erythromycin, streptomycin, doxycycline, ciprofloxacin and ofloxacin at the same concentration. As indicated in Fig. 5, the target kanamycin caused at least a 20-time increase in photocurrent change (ΔI) relative to the other interfering species owing to the specific interaction between the aptamer and kanamycin. The specificity was also evaluated by measuring ΔI in the presence of a mixed sample composed of kanamycin (200 nM), chloramphenicol (100 nM), erythromycin (100 nM), streptomycin (100 nM), doxycycline (100 nM), ciprofloxacin (100 nM) and ofloxacin (100 nM). The histogram labelled “Mixed Sample” in Fig. 5 shows a very similar ΔI obtained in kanamycin alone. Reproducibility was studied by determining kanamycin using six identical aptasensors in the same experimental conditions. A relative standard deviation (RSD) of 5.1% was obtained, suggesting an acceptable reproducibility of the aptasensor. Meanwhile, the stability of the PEC aptasensor was examined after the aptasensors were stored at 4 °C before further measurement. The results showed that 88.7% and 80.6% of its initial photocurrent response remained after 2 and 4 weeks of storage,
4. Conclusion In summary, we have developed a PEC aptasensor for the highly sensitive detection of the model analyte kanamycin using a TiO2-MoS2AuNP ternary composite as the photoactive materials in conjunction with a DNA aptamer recognition method. The TiO2-MoS2-AuNP heterostructure displayed a separated sphere nanomorphology with extremely large surface area, which increased the loading of immobilised aptamer molecules, and excellent electrical conductivity, which improved the electron transfer rate at the electrode surface. In addition, AuNPs on the TiO2-MoS2 heterostructure significantly improved the PEC performance under visible light irradiation because of the localised surface plasmon resonance effect. The PEC aptasensor presented good detection selectivity, a low detection limit of 0.05 nM and a wide linear range of 0.2–450 nM. The results demonstrated feasible applications of the TiO2-MoS2-AuNP composite to the development of aptasensors for detecting many different biological molecules. Acknowledgments This work was financially supported by National Natural Science Foundation of China (Nos. U1504215，21576071), and International Science and Technology Cooperative Project funded by The Department of Science and Technology of Henan Province (172102410042). Table 2 Determination of kanamycin in milk samples using the PEC aptasensor.
Fig. 5. Anti-interference property of BSA|aptamer|TiO2-MoS2-AuNP|ITO in 0.1 M PBS (pH 7.4) containing 0.1 M KCl at a bias voltage of 0.6 V in the presence of 200 nM kanamycin and the tabulated antibiotics. The error bars were derived from the standard deviation of four measurements. 200
Sample
Added / nM
Found / nM
Recovery / %
Standard deviation / nM
1 2 3 4 5
0 2 20 50 100
0 1.83 19.2 46.6 97.2
– 91.5 96.1 93.2 97.1
– 0.087 0.091 0.102 0.163
Biosensors and Bioelectronics 112 (2018) 193–201
X. Liu et al.
References
ChemSusChem 9, 1118–1124. Luo, J., Ma, L., He, T., Ng, C.F., Wang, S., Sun, H., Fan, H.J., 2012. J. Phys. Chem. C. 116, 11956–11963. Mehta, A., Sharma, M., Kumar, A., Basu, S., 2016. Nano 11, 1650104. Ren, W., Zhou, W., Zhang, H., Cheng, C., 2017. ACS Appl. Mater. Interfaces 9, 487–495. Sharma, T.K., Ramanathan, R., Weerathunge, P., Mohammadtaheri, M., Daima, H.K., Shukla, R., Bansal, V., 2014. Chem. Commun. 50, 15856–15859. Song, K.-M., Cho, M., Jo, H., Min, K., Jeon, S.H., Kim, T., Han, M.S., Ku, J.K., Ban, C., 2011. Anal. Biochem. 415, 175–181. Sun, X., Li, F., Shen, G., Huang, J., Wang, X., 2014. Analyst 139, 299–308. Sun, Y., Fang, Z., Wang, C., Ariyawansha, K.R.R.M., Zhou, A., Duan, H., 2015. Nanoscale 7, 7790–7801. Wang, L., Duan, X., Wang, G., Liu, C., Luo, S., Zhang, S., Zeng, Y., Xu, Y., Liu, Y., Duan, X., 2016. Appl. Catal., B 186, 88–96. Wang, Y., Zang, D., Ge, S., Ge, L., Yu, J., Yan, M., 2013. Electrochim. Acta 107, 147–154. Whitcombe, M.J., Chianella, I., Larcombe, L., Piletsky, S.A., Noble, J., Porter, R., Horgan, A., 2011. Chem. Soc. Rev. 40, 1547–1571. Xin, Y., Li, Z., Zhang, Z., 2015. Chem. Commun. 51, 15498–15501. Xu, Q., Yu, J., Zhang, J., Zhang, J., Liu, G., 2015a. Chem. Commun. 51, 7950–7953. Xu, Y., Han, T., Li, X., Sun, L., Zhang, Y., Zhang, Y., 2015b. Anal. Chim. Acta 891, 298–303. Xue, L., Zhou, X., Xing, D., 2012. Anal. Chem. 84, 3507–3513. Yildirim, N., Long, F., Gao, C., He, M., Shi, H.-C., Gu, A.Z., 2012. Environ. Sci. Technol. 46, 3288–3294. Zhang, S., Li, J., Wang, X., Huang, Y., Zeng, M., Xu, J., 2014. ACS Appl. Mater. Interfaces 6, 22116–22125. Zhang, Y., Xiao, J., Lv, Q., Wang, L., Dong, X., Asif, M., Ren, J., He, W., Sun, Y., Xiao, F., Wang, S., 2017. ACS Appl. Mater. Interfaces 9, 38201–38210. Zhao, S., Chen, S., Yu, H., Quan, X., 2012. Sep. Purif. Technol. 99, 50–54. Zhao, W.-W., Shan, S., Ma, Z.-Y., Wan, L.-N., Xu, J.-J., Chen, H.-Y., 2013. Anal. Chem. 85, 11686–11690. Zhao, W.-W., Xu, J.-J., Chen, H.-Y., 2014a. Chem. Rev. 114, 7421–7441. Zhao, W., Zhang, M., Ai, Z., Yang, Y., Xi, H., Shi, Q., Xu, X., Shi, H., 2014b. J. Phys. Chem. C. 118, 23117–23125. Zheng, L., Han, S., Liu, H., Yu, P., Fang, X., 2016. Small 12, 1527–1536. Zhou, W., Yin, Z., Du, Y., Huang, X., Zeng, Z., Fan, Z., Liu, H., Wang, J., Zhang, H., 2013. Small 9, 140–147.
Ansari, S.A., Husain, Q., 2012. Biotechnol. Adv. 30, 512–523. Bai, S., Wang, L., Chen, X., Du, J., Xiong, Y., 2015. Nano Res. 8, 175–183. Bai, X., Hou, H., Zhang, B., Tang, J., 2014. Biosens. Bioelectron. 56, 112–116. Chen, D., Cao, L., Huang, F., Imperia, P., Cheng, Y.-B., Caruso, R.A., 2010. J. Am. Chem. Soc. 132, 4438–4444. Fuchs, Y., Soppera, O., Haupt, K., 2012. Anal. Chim. Acta 717, 7–20. Gao, P., Ma, H., Yang, J., Wu, D., Zhang, Y., Du, B., Fan, D., Wei, Q., 2015. New J. Chem. 39, 1483–1487. Heydari-Bafrooei, E., Askari, S., 2017. Int. J. Hydrog. Energy 42, 2961–2969. Huang, K.-J., Liu, Y.-J., Wang, H.-B., Wang, Y.-Y., Liu, Y.-M., 2014. Biosens. Bioelectron. 55, 195–202. Huang, K.-J., Liu, Y.-J., Zhang, J.-Z., Cao, J.-T., Liu, Y.-M., 2015. Biosens. Bioelectron. 67, 184–191. Leung, K.-H., He, H.-Z., Chan, D.S.-H., Fu, W.-C., Leung, C.-H., Ma, D.-L., 2013. Sens. Actuators, B 177, 487–492. Li, C., Wang, H., Shen, J., Tang, B., 2015a. Anal. Chem. 87, 4283–4291. Li, H., Li, J., Qiao, Y., Fang, H., Fan, D., Wang, W., 2017. Sens. Actuators, B 243, 1027–1033. Li, H., Qiao, Y., Li, J., Fang, H., Fan, D., Wang, W., 2016. Biosens. Bioelectron. 77, 378–384. Li, J., Cushing, S.K., Zheng, P., Senty, T., Meng, F., Bristow, A.D., Manivannan, A., Wu, N., 2014a. J. Am. Chem. Soc. 136, 8438–8449. Li, Q., Zhang, N., Yang, Y., Wang, G., Ng, D.H.L., 2014b. Langmuir 30, 8965–8972. Li, R., Liu, Y., Cheng, L., Yang, C., Zhang, J., 2014c. Anal. Chem. 86, 9372–9375. Li, X., Zheng, Z., Liu, X., Zhao, S., Liu, S., 2015b. Biosens. Bioelectron. 64, 1–5. Li, Y.-J., Ma, M.-J., Zhu, J.-J., 2012. Anal. Chem. 84, 10492–10499. Lin, X., Cui, L., Huang, Y., Lin, Y., Xie, Y., Zhu, Z., Yin, B., Chen, X., Yang, C.J., 2014. Chem. Commun. 50, 7646–7648. Liu, C., Wang, L., Tang, Y., Luo, S., Liu, Y., Zhang, S., Zeng, Y., Xu, Y., 2015a. Appl. Catal., B 164, 1–9. Liu, E., Fan, J., Hu, X., Hu, Y., Li, H., Tang, C., Sun, L., Wan, J., 2015b. J. Mater. Sci. 50, 2298–2305. Liu, P., Huo, X., Tang, Y., Xu, J., Liu, X., Wong, D.K.Y., 2017. Anal. Chim. Acta 984, 86–95. Liu, X., Xing, Z., Zhang, H., Wang, W., Zhang, Y., Li, Z., Wu, X., Yu, X., Zhou, W., 2016.
201