General Bioluminescence Resonance Energy Transfer Homogeneous

Subscriber access provided by UNIV OSNABRUECK

Article

A general bioluminescence resonance energy transfer homogeneous immunoassay for small molecules based on quantum dots Xuezhi Yu, Kai Wen, Zhanhui Wang, Xiya Zhang, Chenglong Li, Suxia Zhang, and Jianzhong Shen Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.5b03581 • Publication Date (Web): 07 Mar 2016 Downloaded from http://pubs.acs.org on March 9, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

A general bioluminescence resonance energy transfer homogeneous immunoassay for small molecules based on quantum dots Xuezhi Yu1, a, Kai Wen1, a, Zhanhui Wang1, Xiya Zhang1, Chenglong Li1, Suxia Zhang1, 2, Jianzhong Shen1, 2 *

1

Beijing Advanced Innovation Center for Food Nutrition and Human Health, College

of Veterinary Medicine, China Agricultural University, No.2 Yuanmingyuan West Road, Beijing 100193, China. 2

Beijing Laboratory for Food Quality and Safety and Beijing Key Laboratory of

Detection Technology for Animal-Derived Food Safety, No.2 Yuanmingyuan West Road, Beijing 100193, China.

a

These authors contributed equally to this paper

*

Corresponding author: Tel: +86-106-273-2803; Fax: +86-106-273-1032; E-mail:

[email protected]

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract Here we describe a general bioluminescence resonance energy transfer (BRET) homogeneous immunoassay based on quantum dots (QDs) as the acceptor and Renilla luciferase (Rluc) as the donor (QD-BRET) for the determination of small molecules. The ratio of the donor-acceptor that could occur energy transfer varied in the presence of different concentrations of free enrofloxacin (ENR) - an important small molecule in food safety. The calculated Förster distance (R0) was 7.86 nm. Under optimized conditions, the half-maximal inhibitory concentration (IC50) for ENR was less than 1 ng/mL and the linear range covered four orders of magnitude (0.023 to 25.60 ng/mL). The cross-reactivities (CRs) of seven representative fluoroquinolones (FQs) were similar to the data obtained by an enzyme-linked immunosorbent assay (ELISA). The average intra-and inter-assay recoveries from spiked milk of were 79.8 - 118.0%, and the relative standard deviations (RSD) was less than 10%, meeting the requirement of residue detection, which was a satisfactory result. Furthermore, we compared the influence of different luciferase substrates on the performance of the assay. Considering sensitivity and stability, coelenterazine-h was the most appropriate substrate. The results from this study will enable better-informed decisions on the choice of Rluc substrate for QD-BRET systems. For the future, the QD-BRET immunosensor could easily be extended to other small molecules, and thus represents a versatile strategy in food safety, environment, clinical diagnosis and other fields. Keywords: bioluminescence resonance energy transfer; quantum dots; small molecule; homogeneous immunoassay, coelenterazine derivative


Page 2 of 27

Page 3 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


INTRODUCTION Bioluminescence resonance energy transfer (BRET) is a non-radiative form of Förster resonance energy transfer that occurs between a luciferase and an appropriate acceptor when they are in close proximity.1, 2 Unlike fluorescence resonance energy transfer (FRET), which relies on external excitation, BRET is energized internally by the oxidation of a luciferin-type substrate, reducing the non-specific signals caused by external light excitation. Furthermore, in BRET2, the spectral separation between donor and acceptor emissions is 115 nm, which results in an excellent signal to noise (S/N) ratio and a working distance to 11.3 nm.3 This large working distance is a much better match for the distances measured within/between proteins, enabling detection at the femtomolar level.3-5 Furthermore, as an important homogeneous immunoassay method, the separation-free nature of BRET has shortened the assay times and simplified procedures. In addition, detection in solution can facilitate reaction kinetics and reduce nonspecific adsorption.6 Similar to BRET, chemiluminescence resonance energy transfer (CRET) is also occurred by the oxidation of a luminescent substrate.7 While, the most widely used chemiluminescence reaction in CRET is the luminol-H2O2 system, catalyzed by horseradish peroxidase (HRP). However, the weak luminescence in neutral media, poor energy-transfer efficiency and the small number of energy acceptors have limited its application.8, 9 As a result of these advantageous properties, BRET biosensors have been widely used in immunoassays, clinical/diagnostic assays, and biomolecule binding assays.3, 10-13 For small molecule analysis, fluorescent protein-based BRET homogeneous assays have been developed to estimate estrogen-like compounds, maltose and diacetyl.3, 14, 15

However, applications of BRET are still limited by the narrow, fixed excitation

spectrum and low luminescence of the fluorescent protein acceptors.15-17 Therefore, the development of more suitable acceptors would broaden the applicability of BRET. Quantum dots (QDs) are a kind of fluorescent semiconductors that have received

attention in recent years.18 QDs have a larger Stokes shift, and spectroscopic detection can be achieved at a low signal intensity, making the fluorescence detection easier.19 Furthermore, QDs have strong fluorescence intensity, high stability, good resistance to photo-bleaching, long fluorescence lifetime, good biocompatibility, narrow emission spectrum and broad absorption.18,

21,

22

Consequently, QDs have been widely used as the donors in FRET.23, 24 In 2006, it was



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

demonstrated that QDs could replace fluorescent proteins as the BRET acceptor in vivo imaging.25 In addition, Medintz et al. created a BRET-(FRET-FRET)X format, in which QDs could act as the acceptor for BRET and donor for FRET in the same reaction system.26,

27

Therefore, QD-based Förster resonance energy transfer,

especially the QD-based BRET (QD-BRET) has been widely applied in many areas, such as in vivo imaging, protein-protein interactions and detection of proteases.15, 28, 29 However, to the best of our knowledge, systematic studies focusing on the application of QD-BRET in the field of small molecule immunoassay are still scarce. Herein, we have developed a QD-BRET immunoassay for fluoroquinolones (FQs) to explore its potential application in small molecule analysis. Since various analytical methods have been established based on the same antibody, we can objectively evaluate the advantages and disadvantages of this assay. As shown in Scheme 1, when QD-conjugated norfloxacin (NOR, QD-NOR) is recognized by single-chain variable fragment (scFv), Renilla luciferase (Rluc) and the QDs are in close proximity. Under these conditions, energy is released from the substrate and transferred to the QDs via BRET. In the absence of recognition, the distance between the Rluc and QDs is too far to realize the energy transfer. Furthermore, we have calculated the Förster distance (R0) and compared the influence of various coelenterazine derivatives on the performance of QD-BRET and the results provide a reference for choice of Rluc substrate in QD-BRET systems. The results demonstrated that the QD-BRET detection scheme could be applied to the analysis of small molecules and was a general strategy that could be extended to the homogeneous immunoassay of other small molecules in other fields.

EXPERIMENTAL SECTION Materials The Rluc gene and isopropyl β-D-1-thiogalactopyranoside (IPTG) were obtained from Promega (Madison, WI, USA). Coelenterazine derivatives were obtained from Sigma-Aldrich (St. Louis, MO, USA). Escherichia.coli (E. coli) BL21 (DE3) strains were obtained from TransGen Biotech (Beijing, China). The anti-FQs scFv C4A9H1_mut 2 vector was established previously in our laboratory.30 The HisTrap™ HP column was from GE Healthcare (Beijing, China). DAB Horseradish Peroxidase Color Development Kit was purchased from TIANGEN Biotech (Beijing) Co., Ltd. FQs-free skimmed milk was supplied by the National Reference Laboratory for


Page 4 of 27

Page 5 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Veterinary Drug Residues (Beijing, People’s Republic of China). White opaque high binding plates were purchased from Costar (Cambridge, MA, USA). The solutions and buffers were prepared with water purified using a Milli-Q system from EMD Millipore Corporation (Belleria, MA, USA). Luminescence was measured using a SpectraMax M5 microplate reader from Molecular Devices (Downingtown, PA, USA). FQs standard solutions and all media used in this study were prepared as described in the literature.31, 32 All other chemicals and organic solvents were of reagent grade and were obtained from Beijing Chemical Co. (Beijing, China). The QD-NOR conjugate was prepared by conjugating carboxyl QD620 with the secondary amine of the NOR piperazine ring, and this work was conducted by Shanghai Xintong Bio Technology Co., Ltd.

Preparation of anti-FQ scFv-Rluc fusion protein The genes encoding anti-FQ scFv and Rluc were amplified and assembled using splicing by overlap extension-polymerase chain reaction (SOE-PCR) to yield a full-length scFv-Rluc gene that encoded a scFv-linker-Rluc format with a reference amino acid linker (GSTSGSGKPGSGEGSTSG), EcoRI and XhoI restriction sites.33 After restriction digestion and ligation reaction, the pET22b (+) vector containing the scFv-linker-Rluc was transformed into E.coli BL21 (DE3) competent cells. Positive colonies identified by polymerase chain reaction (PCR) and DNA sequencing were preserved in medium containing 30% glycerol at -80 °C. After optimization of the expression conditions, the positive colony was cultivated overnight at 37 °C in 2×YT medium containing 100 µg/mL ampicillin in a shake flask (200 rpm). The overnight cultures were transferred into 100 mL fresh 2×YT medium containing 100 µg/mL ampicillin and cultivated on the shaker at 200 rpm until the OD600nm of the culture reached approximately 0.6-1.0. A final concentration of 0.05 mM IPTG was added to induce the expression of the fusion protein at 16 °C, 200 rpm for 16 h. The cells were harvested by centrifugation (5000 ×g) and the soluble scFv-Rluc protein was extracted and purified using a 5 mL HisTrap™ HP column according to the manufacturer’s instructions. The purified protein was dialyzed against 0.01 M phosphate

buffered

saline

(PBS)

and

characterized

by

sodium

dodecyl

sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and western blotting. The antigen-specific binding ability of the fusion protein and the specific binding capacity of the prepared QD-NOR were assessed by enzyme-linked immunosorbent



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

assay (ELISA) and fluorescence-linked immunosorbent assay (FLISA).

Development of the QD-BRET immunoassay Enrofloxacin (ENR, 50 µL) standard or sample solution, 50 µL QD-NOR (1.25 nM) and 50 µL of scFv-Rluc fusion protein (270 ng) in 0.01 M PBS were added successively to each well and the plate was incubated at room temperature (25 °C) for 30 min. Following incubation, coelenterazine-h was added to a final concentration of 5 µM and luminescence from 400-700 nm was measured immediately with an integration time of 100 ms.

Assessment of assay specificity The specificity of the QD-BRET was evaluated by determining the cross-reactivity (CR) to seven representative FQs and other veterinary drugs that are commonly used in livestock. The half-maximal inhibitory concentration (IC50) values for each analyte were determined based on the standard curve obtained by plotting B/B0 against the logarithm of analyte concentration with the four-parameter logistic equation (version 8.0, Microcal, Northampton, MA, USA). The CR was calculated according to the equation: CR (%) = [IC50 (ENR)] / [IC50 (analyte)] ×100.31

Performance of different coelenterazine derivatives The Rluc substrate, coelenterazine, is unstable and is oxidized immediately in the presence of oxygen.34 To objectively assess the QD-BRET, the Rluc substrate was an important parameter to consider. Firstly, we measured the luminescence of coelenterazine derivatives from 0 - 5 min at intervals of 5 s and determined the time point at which the luminescence reached maximum. Subsequently, following the described procedure, the performance of various coelenterazine derivatives in this homogeneous immunoassay was evaluated by determining IC50 values and cross-reactivity with other FQs.

Preparation of FQs-spiked milk samples Milk sample preparation was similar to a published method used for another homogeneous assay system (fluorescence polarization immunoassay).35 Negative skimmed milk samples (2 mL) were fortified with the appropriate FQs standard


Page 6 of 27

Page 7 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


solution and then mixed with an equal volume of 1.5% trichloroacetic acid. The mixtures were agitated on a shaker for 2 min and then deproteinized by centrifugation for 10 min (8000 × g at 4 °C). The pH of the supernatant was adjusted to 7.4 before measurement in the BRET assay.

RESULTS AND DISCUSSION Characterization of anti-FQs scFv-Rluc fusion protein and the QD-NOR The most frequently used luciferases in bioassays are from the firefly, Photinus pyralis (Fluc) and the sea pansy, Renilla reniformis (Rluc). Fluc is 62 kDa, ATP-dependent, uses D-luciferin as the substrate, and emits light at 560 nm, whereas Rluc is 36 kDa, ATP-independent, uses coelenterazine as the substrate, and emits at 480 nm. Rluc requires only molecular oxygen and coelenterazine for luminescence, and its smaller size compared with Fluc make it more appropriate for application as a bioluminescent tags.25, 36 Until now, it has generally been accepted that replacement of native Rluc with its mutants, Rluc2 or Rluc8, as the luciferase donor improves luminescence intensity and stability of the BRET assay.36 However, the sensitivity of the native Rluc is better than that of Rluc2 or Rluc8 as the luciferase donor.2 We therefore chose native Rluc for the construction of the BRET immunosensor. The theoretical molecular size of Rluc and scFv were 34 kDa and 29 kDa, respectively. Taking the c-myc-tag, 6×His-tag and linker into consideration, the molecular size of the fusion protein was about 70 kDa. After purification, we characterized the scFv-Rluc by SDS-PAGE and western blotting. The scFv-Rluc fusion protein formed a band of approximately 70 kDa as expected. The culture without induction did not show any background expression (Figure S1. in the Supporting Information). The affinity of the prepared anti-FQs scFv-Rluc was measured by the IC50 and CRs to 20 FQs (Table S1 in the Supporting Information), which were similar to the data obtained in our laboratory.30 And it was demonstrated that the affinity of the prepared anti-FQ scFv-Rluc was good and was not influenced by this fusion format. The specific binding capacity of the prepared QD-NOR was measured by FLISA (Figure S2. in the Supporting Information). In the absence of NOR, the fluorescence at 620 nm of QD-NOR was 145.12, while the fluorescence at 620 nm of QD-B was only 4.5, similar to the blank plate (4.3). Furthermore, the fluorescence declined from 145.12 to 4.3 with the increase of free NOR concentration. The results showed that



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

the QD-NOR was successfully prepared and the specific binding between QDs and scFv-Rluc was via NOR. The results showed that the QD-NOR was successfully prepared with good specific binding capacity and had no non-specific adsorption with the plate.

Calculations of Förster distance The Förster distance (R0) (Equation. (1)) is a characteristic of a donor-acceptor pair, and is a constant of nanometers for a fixed donor-acceptor. The R0 depends on factors including refractive index of the surrounding medium (n), the donor quantum yield (ΦD), the relative orientation between donor emission and acceptor absorption dipoles, and the degree of spectral resonance between the two species.37, 38 These latter two parameters are described by the orientation factor (k2), and spectral overlap integral (J), respectively. The value of J can be calculated from Equation (2), where ID refers to the fluorescence intensity of the donor and εA refers to the molar absorptivity of the acceptor, which can be calculated from Equation (3).39 In Equation (3), A is the absorbance, b is the path length (cm) of the radiation beam used for recording the absorption spectrum and C is the molar concentration (mol/L) of the acceptor. In our experiment, b is 0.1 cm (measured by a Nano Drop 2000c spectrophotometer). R0 = 8.79×10-28 mol × (n-4 k2ΦDJ) 4

J = ∫ ID (λ) εA (λ) λ dλ / ∫ ID (λ) dλ εA = A/bC

(1) (2)

(3)

EBRET =1-FDA/FD (4) EBRET = R06 / (R06 + rBRET6) (5)

When the Rluc substrate was coelenterazine-h, the calculated εA and J were 3.66 × 105 mol-1 L cm-1 and 1.786 × 10-20 mol-1 L cm3. The orientation factor takes on a value of k2 = 2/3. The refractive index of 0.01 M PBS was 1.33 and the quantum yield of Rluc, ΦD, was 0.07.39,

40

Finally, the R0 was calculated as 7.86 nm. The energy

transfer efficiency, EBRET, can be calculated using Equation (4), where FDA and FD are the luminescence of the donor in the presence or absence of acceptor. 41 As shown in Figure S3 (Supporting Information), EBRET was 0.32. The distance between donor and acceptor was calculated to be 8.9 nm according to Equation (5).

The influence of QDs on the luminescence of Rluc During our experiments, we observed a decline of Rluc luminescence in the presence


Page 8 of 27

Page 9 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


of both QD-NOR and blank QDs (QD-B) containing no NOR. We selected the same concentrations of QD-NOR and QD-B to compare the influence of QDs on the luminescence of Rluc and the results are shown in Figure 1. With increasing QD-NOR concentration, the luminescence of Rluc (470 nm) decreased and the luminescence of QDs-NOR (620 nm) clearly increased (Figure 1A). With increasing QD-B concentration, the luminescence of the Rluc (470 nm) decreased, but the luminescence of QDs-B (620 nm) was unchanged (Figure 1B). The positions of the donor and acceptor are typically dynamic and can come close proximity purely by chance. Such nonspecific proximity will produce a background Förster resonance energy transfer (RET) signal.42 This is so-called proximity RET signal. Proximity RET occurs when a donor and an acceptor approach each other by chance within distances of about two Förster radii.43 According to Equation (5), EBRET = R06 / (R06 + rBRET6), EBRET was calculated as 1/65=0.015 when rBRET was 2R0. Nevertheless, the proportion of donors and acceptors within two Förster radii by chance was very low. In addition, the co-existence of well-separated acceptors would still scatter emission light when the acceptors are suspension nanoparticles, resulting in scattering interference.44 The diameter of the QD was approximately 5-10 nm, less than the wavelength of the Rluc emission (470 nm). Due to the Tyndall effect (Figure 1B, inset figure), when the emission passed through the QD solution, there was scattering and little reflection of light.45 With increasing concentration of QD, scattering would be enhanced and the probability that donors and acceptors approach each other would increase. However, the proportion was still too low to produce luminescence that could be detected at 620 nm. Therefore, with increasing of QD-B concentration, the intensity at 470 nm decreased. As is well known, energy transfer efficiency can be calculated by the equation: EBRET = R06 / (R06 + rBRET6). The R0 is the intermolecular distance at which 50% of the maximum possible energy transfer is achieved and is a constant of nanometers for a fixed donor-acceptor.13, 46 Since the efficiency of energy transfer declines as the sixth power of the donor-acceptor distance, in order to realize energy transfer, rBRET must be in nanometer.1 When QD-NOR is recognized by scFv-Rluc, the distance between the Rluc and QDs is within several nanometers and energy is released from the substrate and transferred to the QD via BRET. Consequently, luminescence of QD-NOR (620



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

nm) is increased and luminescence of Rluc (470 nm) is decreased. However, QD-B cannot be recognized by the scFv-Rluc and the distance between Rluc and the QD is too great to realize the energy transfer. Therefore, the luminescence of QD-B (620 nm) is unchanged and the luminescence of Rluc (470 nm) is reduced by the scattering effect of QD-B. Thus, the decline of luminescence intensity at 470 nm was due to both energy transfer and the scattering effect of the QDs.

Optimization of the QD-BRET homogeneous immunosensor To explore the full potential of the scFv-Rluc and QD-BRET based homogeneous immunosensor, we optimized a series of factors that determined the performance of the immunoassay. We chose the luminescence ratio, Lratio = (L470-10 - L470-0) / L620-10, as the parameter to evaluate the effect of a factor on performance. In this equation, L470-10 was the luminescence at 470 nm in the presence of 10 ng/mL ENR, L470-0 was the luminescence at 470 nm in the absence of ENR and L620-10 was the luminescence at 620 nm in the presence of 10 ng/mL ENR. As shown in Figure 2A, Lratio significantly increased from 0.42 to a maximum of 1.04 when the pH was varied from 6.0 - 7.4 and declined from 1.04 to 0.55 when the pH was changed from 7.4-9.0. The results indicated that the system was strictly dependent on the solution pH, and that pH 7.4 was optimal. To evaluate the effect of ionic strength, PBS (pH 7.4) solutions of 0, 0.01, 0.02, 0.03, 0.04 and 0.05 M were used as the buffer. The results (Figure 2B) indicated that 0.01 M PBS was suitable for the assay. Another factor was the type of buffer. We chose three commonly used buffers, 0.01M PBS, 0.01 M borate buffer (BB) and 0.01 M Tris buffer to evaluate. As shown in Figure 2C, the values of Lratio were 1.01, 0.78 and 0.89, respectively. Thus, the optimal buffer was determined to be 0.01 M PBS. In addition, we also determined the optimal incubation time for the BRET-based system. As shown in Figure 2D, the luminescence at 470 nm declined with prolonged incubation, but the luminescence at 620 nm was unchanged from 0 - 20 min. From 25 min, the luminescence at 620 nm rose slightly and reached a stable value at 30 min. Further incubation produced no obvious change in luminescence at either 620 nm or 470 nm. In consideration of saving time to accelerate the whole process, 30 min was identified as the most appropriate incubation time.


Page 10 of 27

Page 11 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Performance of the immunoassay with different coelenterazine derivatives Coelenterazine, the substrate for Rluc, consists of a central aromatic imidazopyrazine structure. In the presence of oxygen, Rluc catalyzes an oxidative decarboxylation in which the imidazole ring is opened and carbon dioxide is released (Figure S4 in the Supporting Information). Relaxation of the electronically excited coelenteramide reaction product is accompanied by emission of a photon of blue (~470 nm) light.47 To date, many coelenterazine derivatives have been synthesized for different purposes.48, 49 In this study, the acceptor was QDs with a broad excitation spectrum such that we could evaluate the performance of different coelenterazine derivatives with various emissions to the excitation of quantum dots. We selected native coelenterazine and three representative coelenterazine derivatives - coelenterazine-h, coelenterazine-hcp and coelenterazine 400a - with emission wavelengths of 466, 466, 445 and 395 nm, respectively (Figure 3A).50, 51 The spectral separation between the BRET donor and acceptor emissions were 154 nm, 154 nm, 175 nm and 225 nm. As shown in Figure 3B, the luminescence of coelenterazine 400a decayed rapidly in aqueous solution, so we evaluated its performance using an endpoint assay, while the other three were evaluated with spectral scanning.52 We subsequently compared the influence of these four coelenterazines on the stability and sensitivity of the QD-BRET immunoassay. The IC50 values for ENR were 3.423, 0.782, 0.235 and 1.427 ng/mL when

the

Rluc

substrate

was

native

coelenterazine,

coelenterazines-h,

acoelenterazine-hcp and coelenterazine 400a, respectively. The relative standard deviations (RSD) were 6.7%, 7.5%, 18.6% and 14.6%. The luminescence of coelenterazine-hcp and coelenterazine 400a was time sensitive. Large differences in luminescence were observed at reaction times that differed by only several seconds. For thisreason, the RSDs for coelenterazine-hcp and coelenterazine 400a were larger than native coelenterazine and coelenterazine-h. Comparatively, the sensitivities of coelenterazine-h and coelenterazine-hcp were better than native coelenterazine. The RSDs for native coelenterazine and coelenterazine-h were similar, but the RSD for coelenterazine-hcp was not ideal. However, the sensitivities and RSDs all met corresponding detection requirements. The results allowed us to choose different substrates for different purposes in further experiments.

Analytical performance of the QD-BRET based homogeneous immunoassay Taking sensitivity and stability into consideration, we selected coelenterazine-h as the



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

substrate for Rluc. Under optimized conditions, the performance of the immunoassay was investigated by monitoring changes in luminescence intensity at 470 and 620 nm in the presence of different concentrations of ENR analytical standard (0-1000 ng/mL). As shown in Figure 4A, the changes at 470 and 620 nm were correlated with the concentration of ENR. As the concentration of ENR increased, more scFv-Rluc combined with free ENR and less QD-NOR was recognized, so the proportion of Rluc that could transfer energy to QD-NOR decreased. Therefore, luminescence increased at 470 nm and decreased at 620 nm as the concentration of ENR increased. To evaluate the relationship between luminescence and ENR concentration, we chose Lratio = (L470-x - L470-0) / L620-x as the Y-coordinate and concentration of ENR as the X-coordinate to plot the standard curve (Figure 4B). In this equation, L470-x represents luminescence at 470 nm under the corresponding concentration of ENR, L470-0 represents luminescence at 470 nm in the absence of ENR, and L620-x is the luminescence at 620 nm in the presence of different concentrations of ENR. There was a linear relationship between Lratio and ENR concentration over the range of 0.023-25.60 ng/mL and the correlation coefficient was 0.9979. At the same time, we compared different parameters as the Y-coordinate. When the Y-coordinate was (L470-x - L470-0) or L620-x, the IC50 values for ENR were 1.86 ng/mL and 3.51 ng/mL. Following a general analysis, we chose (L470-x-L470-0)/L620-x as the Y-coordinate. The receptor used in the assay was a broad-specific single-chain variable fragment that could recognize 20 fluoroquinolones. The specificity of the assay was evaluated by determining the CRs to seven representative FQs and other unrelated chemical compounds. The CRs to the seven representative FQs were similar to those obtained by ELISA (Table S1 and Figure S5 in the Supporting Information). The CRs for unrelated chemical compounds were less than 0.01%. The specificity of the assay is reflected in specific recognition of FQs and absence of reaction with unrelated chemical compounds. The limit of detection (LOD) was determined as the mean of the measured content of blank samples (n = 20) plus three standard deviations (mean + 3SD). Blank milk samples, obtained from 20 different individuals, were extracted and analyzed according to the developed method. The calculated ENR concentration was 1.48 ± 0.353 ng/L. Thus, the LOD for ENR was determined as 2.54 ng/L in milk samples.


Page 12 of 27

Page 13 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


The developed immunoassay was a homogeneous assay that did not require coating, blocking and washing steps compared with solid-phase assays, such as ELISA, chemiluminescent enzyme-linked immunological assay (CL-ELISA) and FLISA. Although the IC50 of ENR (0.782 ng/mL) was not ao good as these solid-phase assays (0.31 ng/mL by ELISA and 0.12 ng/mL by CL-ELISA), the LOD was lower than these methods (2.54 ng/L vs. 195 µg/kg).30, 31 It is well known that fluorescence polarization immunoassay (FPIA) is a typical homogeneous technique that allows quantitative analysis of the binding of a small fluorescent ligand to a larger protein using plain-polarized light to detect the change in effective molecular volume.35 Although it can be completed in minutes, its sensitivity is not ideal. To date, only one paper has described a method for the analysis of ENR using FPIA. The study was conducted in our laboratory and was a directly competitive FPIA assay for multiple-FQs determination.35 The scFv used in the present study and the monoclonal antibody

employed

in

the

above

paper

were

derived

from

the

same

hybridoma-C4A9H1. The scFv mutant, scFvC4A9H1_mut 2, exhibited 10-fold greater affinity for sarafloxacin (SAR) and its analogues, while the affinity for other FQs was fully inherited from the parental mAb-C4A9H1.30 We therefore chose scFv-C4A9H1_mut 2 for the construction of the QD-BRET immunosensor and ENR as the model analyte to compare the QD-BRET and FPIA homogeneous immunoassays. The IC50 for ENR by FPIA was 1.92 ng/mL and the linear range was 1.25 - 5.38 ng/mL. While, the IC50 for ENR was 0.782 ng/mL and the linear range was 0.023 - 25.6 ng/mL. The LOD for FPIA was 0.95 ng/mL, and the LOD for QD-BRET was 2.54 ng/L. The sensitivity of the BRET assay, evaluated by IC50 and LOD, was 2.45-fold and 374-fold that of the FPIA assay, and the working range equal to the linear range covered four orders of magnitude, which was much wider than in the FPIA assay.

Determination of ciprofloxacin, ENR, marbofloxacin, danofloxacin and flumequine in milk samples The QD-BRET immunoassay was used to determine ciprofloxacin (CIP), ENR, marbofloxacin (MAR), danofloxacin (DAN) and flumequine (FLU) in spiked milk samples to confirm its performance (Table 2). The average intra-assay recoveries were 79.8% - 118.0% with a RSD ranging from 1.13% - 7.45%, and the inter-assay



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

recoveries were 82.7% - 114.0% with a RSD ranging from 1.21% - 9.55%. Skimmed milk spiked with CIP, ENR, MAR, DAN and FLU showed good agreement between the spiking level and the concentration detected. The RSDs were less than 10%, which indicated acceptable accuracy and precision of the developed immunosensor as applied to FQs detection in milk samples.

CONCLUSION We have developed a QD-BRET homogeneous immunoassay for the analysis of small molecules for the first time. The QD-BRET immunoassay exhibited a LOD of 2.54 ng/L and the linear range covered four orders of magnitude. In comparison with ELISA, CL-ELISA and FPIA, the QD-BRET immunoassay offers significant advantages for the detection of small analytes. The wider excitation of QDs enabled more flexible choices of the donor substrates, giving the QD-BRET wider applicability. In prospects, by replacing the target of interest, the QD-BRET immunosensor could easily be extended to analysis of other small molecules. Using different quantum dots, it could even be applied in multiplex detection and thus represents a versatile strategy for small molecules in food safety, environment, clinical diagnosis and other fields. AUTHOR INFORMATION *Corresponding author: Jianzhong Shen Phone: +86-106-273-2803. Fax: +86-106-273-1032. E-mail: [email protected] a

These authors contributed equally to this paper.

Notes The authors declare no competing financial interest.

ACKNOWLEDGMENT This study was supported by the general program of the National Natural Science Foundation of China (Grant No. 31372475 and 31472236).

SUPPORTING INFORMATION Description of the affinity measurements of prepared anti-FQ scFv-Rluc fusion protein；description of specific binding capacity measurements of the prepared QD-NOR.


Page 14 of 27

Page 15 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Affinity measurements of prepared anti-FQ scFv-Rluc fusion protein (Table S1); Characterization of the anti-FQs scFv-Rluc (Figure S1); Assessment of specific binding capacity of the prepared QD-NOR (Figure S2); Luminescence spectra of the BRET-based immunosensor (Figure S3); Bioluminescent reaction catalyzed by Renilla luciferase (Figure S4); Standard curve of eight FQs under different coelenterazine derivatives (Figure S5).



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

REFERENCES (1) Baiula, M. Methods. Mol. Biol, 2015, 1230, 105-113. (2) Dacres, H.; Michie, M.; Wang, J.; Pfleger, K. D.; Trowell, S. C. Biochem. Bioph. Res.Co, 2012, 425, 625-629. (3) Dacres, H.; Wang, J.; Leitch, V.; Horne, I.; Anderson, A. R.; Trowell, S. C. Biosens. Bioelectron, 2011, 29, 119-124. (4) Kumar, M.; Zhang, D.; Broyles, D.; Deo, S. K. Biosens. Bioelectron, 2011, 30, 133-139. (5) Roda, A.; Guardigli, M. Anal. Bioanal. Chem, 2012, 402, 69-76. (6) Liu, F.; Li, Y.; Song, C.; Dong, B.; Liu, Z.; Zhang, K.; Li, H.; Sun, Y.; Wei, Y.; Yang, A.; Yang, K.; Jin, B. Anal. Chem, 2010, 82, 7758-7765. (7) Niazov, A.; Freeman, R.; Girsh, J.; Willner, I. Sensors, 2011, 11, 10388-10397. (8) Dong, Y. P.; Gao, T. T.; Zhou, Y.; Zhu, J. J. Anal. Chem, 2014, 86, 11373-11379. (9) Lei, J.; Jing, T.; Zhou, T.; Zhou, Y.; Wu, W.; Mei, S.; Zhou, Y. Biosens. Bioelectron, 2014, 54, 72-77. (10) Jang, H. J.; Ahn, J.; Kim, M. G.; Shin, Y. B.; Jeun, M.; Cho, W. J.; Lee, K. H. Biosens. Bioelectron, 2015, 64, 318-323. (11) Cui, B.; Wang, Y.; Song, Y.; Wang, T.; Li, C.; Wei, Y.; Luo, Z. Q.; Shen, X. mBio, 2014, 5, e01050-01014. (12) Kuo, M. S.; Auriau, J.; Pierre-Eugene, C.; Issad, T. PloS. One, 2014, 9, e92737. (13) Dacres, H.; Michie, M.; Trowell, S. C. Anal. Biochem, 2012, 424, 206-210. (14) Michelini, E.; Mirasoli, M.; Karp, M.; Virta, M.; Roda, A. Anal. Chem, 2004, 76, 7069-7076. (15) Le, N. C.; Gel, M.; Zhu, Y.; Dacres, H.; Anderson, A.; Trowell, S. C. Biosens. Bioelectron, 2014, 62, 177-181. (16) Arai, R.; Nakagawa, H.; Tsumoto, K.; Mahoney, W.; Kumagai, I.; Ueda, H.; Nagamune, T. Anal. Biochem, 2001, 289, 77-81. (17) Hsu, C. Y.; Chen, C. W.; Yu, H. P.; Lin, Y. F.; Lai, P. S. Biomaterials, 2013, 34, 1204-1212. (18) Massey, M.; Wu, M.; Conroy, E. M.; Algar, W. R. Curr. Opin.Biotechnol, 2015, 34, 30-40. (19) Wu, P.; Yan, X. P. Chem. Soc. Rev, 2013, 42, 5489-5521. (20) Zhao, M. X.; Zeng, E. Z. Nanoscale. Res. Lett, 2015, 10, 171. (21) Kumar, P.; Sharma, S. C.; Deep, A. Sensors. Actuat. B-Chem, 2014, 202,


Page 16 of 27

Page 17 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


404-409. (22) Feng, L.; Zhu, A.; Wang, H.; Shi, H. Biosens. Bioelectron, 2014, 53, 1-4. (23) Zhang, Y.; Zhang, H.; Hollins, J.; Webb, M. E.; Zhou, D. Phys. Chem. Chem. Phys, 2011, 13, 19427-19436. (24) Long, F.; Gu, C.; Gu, A. Z.; Shi, H. Anal. Chem, 2012, 84, 3646-3653. (25) So, M. K.; Xu, C.; Loening, A. M.; Gambhir, S. S.; Rao, J. Nat. Biothechnol, 2006, 24, 339-343. (26) Dwyer, C. L.; Díaz, S. A.; Walper, S. A.; Samanta, A.; Susumu, K.; Oh, E.; Buckhout-White, S.; Medintz, I. L. Chem Mater, 2015, 27, 6490-6494. (27) Samanta, A.; Walper, S. A.; Susumu, K.; Dwyer, C. L.; Medintz, I. L. Nanoscale, 2015, 7, 7603-7614. (28) Wu, Q.; Chu, M. Int. J. Nanomed, 2012, 7, 3433-3443. (29) Yao, H.; Zhang, Y.; Xiao, F.; Xia, Z.; Rao, J. Angew. Chem. Int. Edit, 2007, 46, 4346-4349. (30) Wen, K.; Nolke, G.; Schillberg, S.; Wang, Z.; Zhang, S.; Wu, C.; Jiang, H.; Meng, H.; Shen, J. Anal. Bioanal. Chem, 2012, 403, 2771-2783. (31) Tao, X.; Chen, M.; Jiang, H.; Shen, J.; Wang, Z.; Wang, X.; Wu, X.; Wen, K. Anal. Bioanal. Chem, 2013, 405, 7477-7484. (32) Yu, X.; Tao, X.; Shen, J.; Zhang, S.; Cao, X.; Chen, M.; Wang, W.; Wang, Z.; Wen, K. Anal. Methods, 2015, 7, 9032-9039. (33) Zhang, H. In Molecular Imaging and Contrast Agent Database (MICAD): Bethesda (MD), 2004. (34) Pellinen, T.; Rytkonen, K.; Ristiniemi, N.; Karp, M.; Lamminmaki, U. Anal. Biochem, 2006, 358, 301-303. (35) Mi, T.; Wang, Z.; Eremin, S. A.; Shen, J.; Zhang, S. J. Agr. Food. Chem,

2013,

61, 9347-9355. (36) Loening, A. M.; Fenn, T. D.; Wu, A. M.; Gambhir, S. S. Protein. Eng. Des. Sel, 2006, 19, 391-400. (37) Kumar, M.; Zhang, D.; Broyles, D.; Deo, S. K. Biosens. Bioelectron, 2011, 30, 133-139. (38) Massey, M.; Algar, W. R.; Krull, U. J. Anal Chim Acta, 2006, 568, 181-189. (39) Yu, W.; Qu, L.; Guo, W.; Peng, X. Chem Mater, 2003, 15, 2854-2860. (40) Wang, Y. J.; Meng, S. W.; Liang, Y. Z.; Li, L. X.; Peng, W. In OFSC, Hefei, China, 2013.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(41) Dosremedios, C.; Moens, P. J Struct Biol, 1995, 115, 175-185. (42) de Souza, N. Nat. Methods, 2014, 11, 477. (43) King, C.; Sarabipour, S.; Byrne, P.; Leahy, D. J.; Hristova, K. Biophys. J, 2014, 106, 1309-1317. (44) Li, Y. R.; Liu, Q.; Hong, Z.; Wang, H. F. Anal. Chem, 2015. DOI: 10.1021/acs.analchem.5b03064 (45) Kraemer, E. O.; Dexter, S. T. J. Phys. Chem. C, 1926, 31, 764-782. (46) Dacres, H.; Wang, J.; Dumancic, M. M.; Trowell, S. C. Anal. Chem, 2010, 82, 432-435. (47) Woo, J.; von Arnim, A. G. Plant. Methods, 2008, 4, 23. (48) Levi, J.; De, A.; Cheng, Z.; Gambhir, S. S. J. Am. Chem. Soc, 2007, 129, 11900-11901. (49) Dubuisson, M. L.; Rees, J. F.; Marchand-Brynaert, J. Drug. Dev. Ind. Pharm, 2005, 31, 827-849. (50) Teranishi, K.; Shimomura, O. Anal. Biochem, 1997, 249, 37-43. (51) Zhao, H.; Doyle, T. C.; Wong, R. J.; Cao, Y.; Stevenson, D. K.; Piwnica-Worms, D.; Contag, C. H. Mol. Imaging, 2004, 3, 43-54. (52) Pflger, K. D.; Seeber, R. M.; Eidne, K. A. Nat Protoc, 2006, 1 (1), 337-345.


Page 18 of 27

Page 19 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48


Table 1. Cross-reactivity and IC50 values of representative FQs and unrelated compounds in QD-BRET with coelenterazine, coelenterazine-h and coelenterazine-hcp as the Rluc substrate

native coelenterazine Compounds

Ciprofloxacin

coelenterazine-h

coelenterazine-hcp

coelenterazine-400a

IC50

RSDa

CRb

IC50

RSD

CR

IC50

RSD

CR

IC50

RSD

CR

(ng/mL)

(%)

(%)

(ng/mL)

(%)

(%)

(ng/mL)

(%)

(%)

(ng/mL)

(%)

(%)

2.901

5.4

118.0

0.701

7.5

111.6

0.204

19.1

115.6

1.213

18.5

117.6

Enrofloxacin

3.423

6.7

100.0

0.782

7.5

100.0

0.235

18.6

100.0

1.427

14.6

100.0

Norfloxacin

3.962

6.2

86.4

0.923

6.3

84.7

0.270

18.9

87.0

1.603

16.3

89.0

Danofloxacin

5.094

4.5

67.2

1.21

5.6

64.6

0.385

11.5

61.0

2.162

17.5

66.0

Marbofloxacin

7.393

3.6

46.3

1.63

5.4

47.98

0.479

18.9

49.1

3.390

21.4

42.1

Flumequine

16.3

4.9

21.0

3.39

6.1

23.07

0.903

19.2

26.2

5.663

19.4

25.2

Difloxacin

97.348

2.7

2.98

29.362

3.5

2.64

7.833

14.6

3.00

44.593

15.7

3.20

Sarafloxacin

77.166

2.8

3.14

30.171

4.1

2.55

9.179

14.9

3.27

44.733

18.2

3.19

Chloramphenicol

>1000

-

1000

-

1000

-

1000

-

1000

-

1000

-

1000

-

1000

-

General Bioluminescence Resonance Energy Transfer Homogeneous

Recommend Documents