Single-Step LRET Aptasensor for Rapid Mycotoxin Detection

Dec 6, 2017 - ... Department of Electrical Engineering and Computer Science, and Biophysics Graduate Program, University of California, Berkeley, Cali...
0 downloads 0 Views 2MB Size
Article pubs.acs.org/ac

Cite This: Anal. Chem. 2018, 90, 716−722

Single-Step LRET Aptasensor for Rapid Mycotoxin Detection Eun-Jung Jo,†,⊥ Ju-Young Byun,‡,⊥ Hyoyoung Mun,† Doyeon Bang,§ Jun Ho Son,§ Jae Young Lee,∥ Luke P. Lee,*,§ and Min-Gon Kim*,†

Downloaded via UNIV OF SUSSEX on July 3, 2018 at 17:06:50 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.



Department of Chemistry, School of Physics and Chemistry, and ∥School of Materials Science and Engineering, Gwangju Institute of Science and Engineering, Gwangju Institute of Science and Technology (GIST), 123 Cheomdangwagi-ro, Buk-gu, Gwangju, 61005, Republic of Korea ‡ Hazards Monitoring Bionano Research Center, Korea Research Institute of Bioscience and Biotechnology (KRIBB), Daejeon 34141, Republic of Korea § Department of Bioengineering, Department of Electrical Engineering and Computer Science, and Biophysics Graduate Program, University of California, Berkeley, California 94720, United States S Supporting Information *

ABSTRACT: Contamination of foods by mycotoxins is a common yet serious problem. Owing to the increase in consumption of fresh produce, consumers have become aware of food safety issues caused by mycotoxins. Therefore, rapid and sensitive mycotoxin detection is in great demand in fields such as food safety and public health. Here we report a single-step luminescence resonance energy transfer (LRET) aptasensor for mycotoxin detection. To accomplish the single-step sensor, our sensor was constructed by linking a quencher-labeled aptamer through a linker to the surface of upconversion nanoparticles (UCNPs). Our LRET aptasensor is composed of Mn2+-doped NaYF4:Yb3+,Er3+ UCNPs as the LRET donor, and black hole quencher 3 (BHQ3) as the acceptor. The maximum quenching efficiency is obtained by modulating the linker length, which controls the distance between the quencher and the UCNPs. Our distinctive design of LRET aptasensor allows detection of mycotoxins selectively in colored food samples within 10 min without multiple bioassay steps. We believe our single-step aptasensor has a significant potential for on-site detection of food contaminants, environmental pollutants, and biological metabolites. assays (ELISA),10,11 flow-through immunoassays,12 lateral flow immunoassays,11,13 electrochemical immunosensors,14 and surface plasmon resonance (SPR).15 Although these detection methods are reasonably sensitive for mycotoxin detection, they require laborious washing steps and labeling procedures. Also, implementation of antibody-based immunoassays in point-ofuse test systems remains highly challenging owing to the complexity of reaction steps and the low stabilities of antibodies.16−18 Recently developed bioassays,19 which rely on the use of localized surface plasmon resonance (LSPR),20 chemiluminescence,21 electrochemistry,22 and fluorescence,23 have been explored in the context of methods for mycotoxin detection that are based on aptamer−mycotoxin interactions. However, these methods require multiple steps, which limit their use in on-site screening. Furthermore, the complexity of the tested samples, which are composed of complex biological and colored substances, can cause potential interference with these assays. As a result, in these cases the mycotoxins must be separated from other materials in the test samples in order

T

he presence of mycotoxins in human and animal food supplies is recognized to be a serious safety issue. The most prominent mycotoxins, including ochratoxin, aflatoxins, zearalenone (ZEA), patulin, and deoxynivalenol (DON), cause a potential threat to human and animal health. These substances cause health problems such as mutagenic and teratogenic effects, kidney and liver damage, cancers, and birth defects, which lead to symptoms ranging from skin irritation to immune-suppression, neurotoxicity, and death.1−5 Because of these health issues, mycotoxins are categorized by the International Agency for Research on Cancer (IARC) as possible human carcinogens.6 The health relevance of mycotoxins demands a rapid and accurate bioassay for on-site screening of commodity foods. As a result, several accurate laboratory techniques have been developed for detecting mycotoxins, such as those that utilize high-performance liquid chromatography (HPLC) with fluorescence detection7 and gas (GC-MS)8 and liquid (LC-MS) chromatography with mass spectrometry detection.9 However, these techniques are unsuitable for on-site testing because they require multiple bioassay steps and bulky and expensive instruments. Because of these limitations, a number of immunoassay-based detection methods have been developed, including those that involve enzyme-linked immunosorbent © 2017 American Chemical Society

Received: June 20, 2017 Accepted: December 6, 2017 Published: December 6, 2017 716

DOI: 10.1021/acs.analchem.7b02368 Anal. Chem. 2018, 90, 716−722

Article

Analytical Chemistry

Figure 1. Detection of mycotoxin by using a luminescence resonance energy transfer (LRET)-based aptasensor. (a) A colored food sample absorbs light in the visible region (violet line). On the other hand, the absorption spectrum (gray line) of UCNPs is located in the NIR region. (b) Schematic energy level diagram of Mn2+-doped UCNPs (left) and BHQ3 (middle). Spectral overlap (lower right): the absorption spectrum (navy line) of quencher (BHQ3) and the upconversion luminescence spectrum (red dotted line) of UCNPs. (c) A schematic illustrating the detection of ochratoxin A (OTA) via LRET between quenchers (BHQ3) and UCNPs under NIR irradiation. (d) Strong upconversion luminescence (LRET OFF) is obtained in the absence of OTA. However, weak luminescence (LRET ON) is observed in the presence of OTA that forms a Gquadruplex−OTA complex.

single-probe-type LRET aptasensor, which can be utilized to detect the target mycotoxin, ochratoxin A (OTA). The system uses a quencher-labeled OTA-specific aptamer through a linker to the surface of UCNPs for construction of the single-probetype sensor. Our distinctive design of the LRET aptasensor allows direct detection of mycotoxins selectively in colored food samples within 10 min without multiple bioassay steps.

to perform an accurate assay. Because of the many issues outlined above, we need to develop a single-step biosensor that has high sensitivity and selectivity for on-site detection of mycotoxins. In this paper, we report the single-step LRET aptasensor for mycotoxin detection. Our LRET aptasensor is composed of a Mn2+-doped NaYF4:Yb3+,Er3+ UCNPs as the LRET donor, and black hole quencher 3 (BHQ3) as the acceptor. The principles for operation of LRET are similar to those for FRET (fluorescence resonance energy transfer) in that resonance energy transfer takes place between closely located donors and acceptors when the donor emission overlaps with the acceptor absorption.24 In comparison to FRET, which uses conventional organic dye, LRET has the advantage25−27 that lanthanide cations can be employed as donor fluorophores. UCNPs are lanthanide-doped nanomaterials that convert 980 nm nearinfrared (NIR) radiation into luminescence in the ultraviolet (UV) or visible region. These nanomaterials typically have high chemical stabilities, sharp emission bands, high signal-to-noise ratios, deep light penetration, minimal photobleaching, and low toxicity.28,29 Moreover, the anti-Stokes nature of emission from the UCNPs enables avoidance of coexcitation of the donor and acceptor, and thus possible false positive signals are precluded. Although UCNPs have poor quantum yields and require expensive raw materials, such as rare earth salts,30−32 they are ideal luminescent nanomaterials for the construction of sensors because they enable circumvention of interference caused by contaminants in test samples that absorb UV/visible light.25−29,33 Recently, UCNP-based LRET aptasensors have demonstrated promising applications in detecting mycotoxins. These sensors require additional materials as quenchers, such as magnetic nanobeads,34 gold nanoparticles,35 gold nanorods,36 and graphene oxide,37 to be tagged to receptors for detection of the target. These systems also need an additional step, hybridization of the aptamers with their corresponding complementary oligonucleotides. Here we have employed a



EXPERIMENTAL SECTION Materials and Sampling. 2-(N-Morpholino)ethanesulfonic acid (MES) hydrate, HEPES, Tris, N-hydroxysuccinimide (NHS), ochratoxin A (OTA), ochratoxin B (OTB), zearalenone (ZEA), aflatoxin B1 (AFB1), sodium carbonate (Na2CO3), sodium bicarbonate (NaHCO3), boric acid (H3BO3), sodium tetraborate decahydrate (Na2B4O7· 10H2O), sodium phosphate monobasic (NaH2PO4), sodium phosphate dibasic (Na2HPO4), sodium chloride (NaCl), potassium chloride (KCl), magnesium chloride (MgCl2), and water (DNase and RNase free) were purchased from SigmaAldrich Chemical (St. Louis, MO). Tris(hydroxymethyl)aminomethane buffer (Tris-HCl, pH 8.8) and phosphatebuffered saline (PBS, pH 7.4) were from Biosesang (Seongnam, South Korea). Instrumentation. Upconversion luminescence spectra were recorded with a fluorescence spectrometer (FluoroMate FS-2; SCINCO, Seoul, Korea) that was modified with a 980 nm laser (SSL-LM-980-600-D; Shanghai Sanctity Laser Technology, Shanghai, China) for upconversion excitation. The absorbance was recorded using a microplate reader (TECAN; Mannedorf, Switzerland). LRET-Based Aptasensor for OTA Detection. For preparation of the LRET-based aptasensor, 20 μL of UCNP− DNA oligonucleotide conjugates and 8 μL of cations were added to the upconversion luminescence measurement cell along with 152 μL of buffer. After the addition of each concentration of 20 μL of OTA (final concentration: 1 μg 717

DOI: 10.1021/acs.analchem.7b02368 Anal. Chem. 2018, 90, 716−722

Article

Analytical Chemistry

Figure 2. Quenching efficiency with various adenine-linker lengths. (a−d) A schematic illustrating the distance (R) and radiative decay rate (k) via LRET between BHQ3 and UCNPs in the absence (a) and presence (b) of OTA under NIR irradiation. In the presence of OTA, the distance (R2) between the donor (UCNPs) and acceptor quenchers decreases which causes resonance energy transfer (RET) between the donor and the acceptor. (c) Differential RET efficiency (blue) of UCNP-quencher system and conformational degree of freedom (green) of adenine-linker with various adenine-linker lengths between UCNP region and quencher-labeled aptamer in conjugates. Differential RET efficiency is defined as the difference of RET efficiency before and after the target binding event. (d) Quenching efficiency with various adenine-linker lengths between UCNP region and quencher-labeled aptamer in conjugates. The optimum linker length (5-mer) by measuring the upconversion luminescence in the presence and absence of 100 ng mL−1 OTA was matched with differential RET efficiency with the steric effect calculation.

mL−1, 0.1 μg mL−1, 10 ng mL−1, 1 ng mL−1, and 0.1 ng mL−1), the cell was incubated at room temperature prior to measurement. The upconversion luminescence signal was measured with a fluorescence spectrometer that was modified with a 980 nm cw laser for upconversion excitation. To optimize LRET aptasensor conditions, some important factors, such as the linker length, conjugate concentration, reaction buffer, cations type, and reaction time, were studied. Detection of OTA Spiked in Colored Food Samples. Colored food samples such as red wine, grape juice, and beer were obtained in local markets. To determine OTA in red wine, grape juice, and beer spiked with various OTA concentrations, the colored food samples were diluted ten times with 10 mM borate buffer (pH 8.5) to obtain OTA solution with methanol (final concentration: 10%). The solutions obtained by mixing the spiked samples with 0.002× conjugates and 4 mM KCl were incubated at room temperature for 10 min. Then the upconversion luminescence was measured using a fluorescence spectrometer. Specificity of the LRET-Based Aptasensor System. To determine mycotoxin in red wine, grape juice, and beer spiked with various mycotoxin (ZEA, OTB, AFB1, or OTA), the colored food samples were diluted ten times with 10 mM borate buffer (pH 8.5) to yield mycotoxin solution with methanol (final concentration: 10%). In the presence of each mycotoxin (final concentration: 0.1 μg mL−1), the upconversion luminescence signal was measured as described in Detection of OTA Spiked in Colored Food Samples.



aptasensor is composed of a Mn2+ -doped NaYF4:Yb3+,Er3+ UCNPs and black hole quencher 3 (BHQ3) as the respective LRET donor and acceptor. Electronic energy diagrams for Yb3+, Er3, and Mn2+ ions present in the UNCPs along with a representation of the proposed mechanism for upconversion energy transfer between the Mn2+-doped UCNP and BHQ3 are given in Figure 1b. In the typical case, energy transfer upconversion, which produces light of higher energy than that absorbed, results from sequential energy transfer processes between excited states of ions produced by near-infrared irradiation. Accordingly, red emission from Mn2+-doped UCNP results from a mechanistic pathway involving nonradiative energy transfer between excited Er3+, produced by NIR light absorption, and Mn2+. This event is followed by back-energy transfer to Er3+ that causes a decrease in the rate of radiative transition of Er3+ to the ground state. Moreover, the population density of excited Mn2+ is increased by resonance energy transfer. Finally, back-energy transfer leads to a red emission. In other words, Mn2+ ions resist transition possibilities between green and red emissions of Er3+ ions, resulting in a red emission.38,40 Resonance energy transfer then occurs between the red-emitting UCNPs and BHQ3 because the UCNP emission overlaps with the BHQ3 absorption. Consequently, the resonance energy transfer process enables quenching of upconversion luminescence generated by laser irradiation of 980 nm when the donor and acceptor are in close proximity (Figure 1b). In the mycotoxin assay system, the LRET quencher, BHQ3, is attached to the 3′ end of an aptamer. The BHQ3-linked amine-containing aptamer (Table S1) is then immobilized on carboxylated-PEG coated UCNPs39 (UCNP nanoprobe). Addition of the colored food sample containing a target mycotoxin to a solution of the UCNP nanoprobe results in folding of the BHQ3-labeled aptamer on the UCNP surface owing to the aptamer−target complex formation. The folding process moves BHQ3 closer to the UCNP surface (Figure 1c) where it more efficiently participates in LRET with the UCNPs. As a consequence of this design, BHQ3 linked to surfaces of UCNPs using selected aptamers can be employed as probes for highly sensitive and selective detection of target mycotoxins (Figure 1d). Moreover, a system of this type can be readily implemented by employing a single mixing step.

RESULTS AND DISCUSSION

Principles of Operation of the LRET-Based Aptasensor. The basic principle of the rapid single-step detection of mycotoxins relies on the use of a UCNP-based LRET aptasensor and is schematically displayed in Figure 1. In comparison to detection systems that utilize fluorescing quantum dots and organic dyes, those based on UCNPs can employ NIR radiation. Consequently, interference associated with background autofluorescence and light scattering caused by visible- and UV-light absorbing contaminants present in foods can be avoided (Figure 1a).28,29 Taking advantage of this unique feature of UCNPs, we designed a single-step assay system for on-site detection of mycotoxin that utilizes a LRET-based aptasensor. The LRET 718

DOI: 10.1021/acs.analchem.7b02368 Anal. Chem. 2018, 90, 716−722

Article

Analytical Chemistry

Figure 3. Optimization of LRET-based aptasensor. (a) Quenching efficiency with various concentrations of conjugates (BHQ1-labeled OTA aptamer−UCNPs). (b) Quenching using various reaction buffer (10 mM MES pH 5, 10 mM MES pH 6, 1X PBS pH 7.2, 10 mM PB pH 7.4, 10 mM borate pH 8.5, 10 mM Tris-HCl pH 8.8, and 10 mM carbonate pH 9.5). (c) Quenching efficiency with various cations in 10 mM borate buffer (pH 8.5) in the presence or absence of the target (100 ng mL−1 OTA). (d) Quenching efficiency with various concentrations of KCl in 10 mM borate buffer (pH 8.5).

Quenching Efficiency with Various Adenine-Linker Lengths. LRET depends mainly on the distance between the UCNP energy donor and BHQ3 acceptor quencher (Figure 2a). As a result of this distance dependence, it is expected that the LRET efficiency would be larger when quenchers are located closer to the surface of the UCNPs. To demonstrate that this proposal is correct, we prepared several aptamers (Table S1) that have the BHQ3 moiety linked by chains containing 0, 5, 10, 15, 20, and 25 adenine groups. The BHQ3labeled aptamers were used to construct conjugates with the UCNPs. A detailed description related to the calculations for determining quenching efficiencies, considering factors such as steric effects, has been added to Supporting Information (section 3). The quenching efficiency of the donor (UCNP) and acceptor (BHQ3) aptamer-containing conjugates is, as expected, dependent on the length of the adenine linker (Figure 2c, blue points). Because the contribution of steric factors to quenching efficiencies is not negligible, we evaluated conformational degrees of freedom of the polymer (Figure 2c, green dots).41 Differential quenching efficiency (E′) caused by steric effects is equal to the differential quenching efficiency (E) multiplied by a factor of 1/α, which explains the reduced degrees of freedom of the aptamer linker (Figure 2d, black dots). The optimum linker length was determined by measuring the upconversion luminescence at 660 nm of conjugates in the presence and absence of 0.1 μg mL−1 OTA. Quenching efficiencies are expressed as [(I0 − I)/I0], where I0 is intensity in the absence of OTA and I is intensity in the presence of OTA. Upconversion luminescence spectra for the conjugated UCNPs with different linker lengths are shown in Figure S3b, and the quenching efficiency at 660 nm in the presence of OTA as a function of linker length is plotted in Figure 2d (red points). The results show that as the linker length increases, the quenching efficiency of the BHQ−UCNPs

LRET system decreases. Thus, when UCNPs are in very close proximity to BHQ, LRET dominates and leads to relatively weak upconversion luminescence. The change in quenching efficiency in the presence and absence of 100 ng mL−1 OTA reaches a maximum at a linker length corresponding to five adenines, after which it gradually decreases (Figure 2d). The fact that the luminescence of BHQ3-labeled aptamerconjugated UCNPs without a linker is only weakly quenched compared to that of the UCNP conjugates with a five-adenine linker is possibly a result of steric hindrance in a conformation with a G-quadruplex structure. Optimization of the LRET-Based Aptasensor. The next phase of this effort was aimed at elucidating conditions that would optimize aptamer−target molecule interactions. For this purpose, the conjugate with a five-adenine linker was employed. We found that the luminescence quenching efficiency of red emission at 660 nm reaches a maximum upon the addition of 0.002× UCNP−aptamer conjugates (conjugation between 149.08 pmol of BHQ3-labeled aptamer) and decreased upon addition of higher concentrations of conjugates (Figure S4 and Figure 3a). In Figure 3b are shown the results of buffer tests which demonstrate that borate buffer at pH 8.5 is the most suitable for LRET process. Divalent cation (Mg2+) is typically required to bring about specific OTA binding to aptamers,42,43 and the G-quadruplex complexes formed are normally stable under monovalent cation (K+ and Na+) conditions.43,44 As shown in Figure 3c, the quenching efficiency increases when buffers that contain KCl are used. Finally, to evaluate the optimal cation concentration, KCl concentrations (0−10 mM) in borate buffer were investigated. KCl at 4 mM was shown to be optimal for LRET quenching (Figure 3d). Consequentially, the optimized conditions for operation of the BHQ3-labeled aptamer-conjugated UCNPs for detection of OTA are as follows: five-adenine linker, 0.002× UCNP−aptamer con719

DOI: 10.1021/acs.analchem.7b02368 Anal. Chem. 2018, 90, 716−722

Article

Analytical Chemistry

Figure 4. Confirmation of the utility of the LRET-based aptasensor system. (a) Quenching efficiency with various probe-conjugated UCNPs. (b) Upconversion luminescence imaging (upper) and spectrum (lower) with various incubation times (0−10 min). (c) Plots of quenching efficiency versus incubation time for OTA concentration ranging from 0.5 to 50 ng mL−1.

Figure 5. Detection of ochratoxin A in colored food samples. (a) A schematic illustrating the quantitative LRET measurements under NIR irradiation made after addition of mycotoxin-contaminated foods. (b) Absorption spectra of the colored food samples. The absorption spectrum of UCNPs is located in the NIR region. (c) Standard curve of the LRET aptasensor performed with various concentrations of OTA (0.1−1000 ng mL−1) in real food samples under optimized conditions. The regression coefficients (R2) of buffer, wine, grape juice, and beer are 0.9947, 0.9784, 0.9741, and 0.9810, respectively. (d−f) Specificity of the LRET-based aptasensor. In contrast to target OTA, addition of other mycotoxins (OTB, AFB1, and ZEA) in food samples (0.1−100 ng mL−1) does not lead to emission quenching (d: wine, e: grape juice, f: beer).

To examine if this system can be used for rapid on-site detection, we measured the time course for evolution of the images and spectra of the aptasensor following addition of OTA under the optimized conditions (Figure 4b). Inspection of the figure shows that the degree of quenching of luminescence from the UCNP nanoprobes rises gradually, and it reaches a maximum at 10 min, indicating that the occurrence of energy transfer from UCNPs to BHQ3 is due to the G-quadruplex formation. The quenching efficiency determined by using the upconversion luminescence imaging data closely matches that measured by using upconversion luminescence spectroscopic data (Figure S6). The results also show that the conformation of the G-quadruplex changes rapidly in the presence of OTA. Furthermore, the rate at which emission quenching increases slowly increases as the concentration of OTA increases (Figure 4c and Figure S7) and then reaches saturation. The results of the kinetic studies, in which OTA concentrations are varied in

jugates, 10 mM borate at pH 8.5 for the reaction buffer, and 4 mM KCl. Confirmation of the Utility of the LRET-Based Aptasensor System. To test the mycotoxin specificity of the UCNP nanoprobe, we examined the quenching efficiency using the OTA specific aptamer (without BHQ3) and BHQ3labeled poly A DNA-conjugated UCNPs (Figure 4a and Figure S5). In the presence of OTA, upconversion luminescence of the UCNP probes is strongly quenched through LRET (Figure 4a, red bar; Figure S5a). Luminescence of the OTA specific aptamer-functionalized UCNPs (without BHQ3) is not quenched by the addition of OTA (Figure 4a, green bar; Figure S5b). In addition, the luminescence of BHQ3-labeled poly A DNA-functionalized UCNPs is only weakly quenched by addition of OTA (Figure 4a, blue bar; Figure S5c). These observations suggest that the BHQ3-labeled OTA aptamerfunctionalized UCNPs bind OTA specifically. 720

DOI: 10.1021/acs.analchem.7b02368 Anal. Chem. 2018, 90, 716−722

Article

Analytical Chemistry

additional experiments with spiked food samples demonstrate the specificity of the assay for OTA over other mycotoxins. For this purpose, the efficiency of quenching promoted by OTA addition was compared to those of other mycotoxins including OTB, AFB1, and ZEA in the concentration range from 0.1 to 100 ng mL−1 in real food samples. Inspection of the bar graphs in Figure 5d−f shows that exposure to other mycotoxins does not lead to quenching of emission (