- Email: [email protected]
Production and characterization of a biotinylated single-chain variable fragment antibody for detection of parathion-methyl
Protein Expression and Purification 126 (2016) 1e8
Contents lists available at ScienceDirect
Protein Expression and Purification journal homepage: www.elsevier.com/locate/yprep
Review article
Production and characterization of a biotinylated single-chain variable fragment antibody for detection of parathion-methyl Huimin Wang 1, Fengchun Zhao 1, Xiao Han, Zhengyou Yang* Department of Microbiology, College of Life Science, Key Laboratory for Agriculture Microbiology, Shandong Agricultural University, Taian 271018, China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 24 April 2016 Received in revised form 8 May 2016 Accepted 9 May 2016 Available online 12 May 2016
In this article, we reported the development of a biotinylated single-chain variable fragment (scFv) antibody based indirect competitive enzyme-linked immunosorbent assay (IC-ELISA) for parathionmethyl (PM) detection. Firstly, a phage display library was generated using a pre-immunized BALB/C mouse against a specific hapten of PM. After four rounds of panning, the scFv gene fragments were transferred into a secreted expression vector. Then, the scFv antibodies were secreted expressed and screened by IC-ELISA against PM. The selected scFv antibody was fused with a biotin acceptor domain (BAD) and inserted into pET-28a(þ) vector for high-level expression in Escherichia coli BL2 (DE3). After optimizing expression conditions, the scFv-BAD antibody was expressed as a soluble protein and biotinylated in vitro by the E. coli biotin ligase (BirA). Subsequently, the biotinylated scFv-BAD antibody was purified with a high yield of 59.2 ± 3.7 mg/L of culture, and was characterized by SDS-PAGE and western blotting. Finally, based on the biotinylated scFv-BAD, a sensitive IC-ELISA for detection of PM was developed, and the 50% inhibition value (IC50) of PM was determined as 14.5 ng/mL, with a limit of detection (LOD, IC10) of 0.9 ng/mL. Cross-reactivity (CR) studies revealed that the scFv antibody showed desirable specificity for PM. © 2016 Elsevier Inc. All rights reserved.
Keywords: Parathion-methyl Phage display Biotinylation scFv antibody IC-ELISA
Contents 1. 2.
3.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Materials and method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2.1. Materials and instruments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2.2. Mice immunization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2.3. Construction and panning of phage-display library . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2.4. Selection of the scFv antibodies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2.5. High-level expression of scFv antibody and condition optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 2.6. In vitro biotinylation and purification of scFv antibody . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 2.7. Development of biotinylated scFv based IC-ELISA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Results and discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 3.1. Phage-display library construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 3.2. Panning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 3.3. Selection of scFv antibodies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 3.4. High-level expression of scFv antibody and condition optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 3.5. In vitro biotinylation, purification and characterization of scFv antibody . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 3.6. Optimization of scFv-based IC-ELISA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 3.7. Cross-reactivity study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
* Corresponding author. E-mail address: [email protected] (Z. Yang). 1 Authors made equal contributions to this work. http://dx.doi.org/10.1016/j.pep.2016.05.005 1046-5928/© 2016 Elsevier Inc. All rights reserved.
2
H. Wang et al. / Protein Expression and Purification 126 (2016) 1e8
4.
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Supplementary data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1. Introduction
2. Materials and method
Parathion-methyl (O,O-dimethyl-O-4-nitrophenyl phosphorothioate, PM), which is widely used as an insecticide and acaricide to control many insect pests of agricultural products, is a highly toxic insecticide in EPA (United States Environmental Protection Agency) toxicity class I. It could inhibit acetylcholinesterase activity in erythrocytes and brain even at a low concentration [1]. Due to its high toxicity and extensive application, PM has been classified as a restricted pesticide with a strict maximum residue limited (MRLs) standard in many countries [2]. For analysis of PM, some traditional methods, such as gas chromatography (GC) and high-performance liquid chromatography (HPLC), have been successfully developed due to their high sensitivity and reliability [3,4]. However, these classical methods require high cost, skilled analysts, and time-consuming sample preparation steps. Electroanalytical sensors exhibited high sensitivity but showed poor selectivity in determination of practical samples, which limited the application and promotion of electroanalytical methods [5]. Compared with these analysis strategies, immunoassay technology is gaining acceptability as a simple, costeffective screening method for pesticides analysis in many samples [6,7]. Several immunoassay methods based on polyclonal antibodies (PAbs) and monoclonal antibodies (MAbs) for PM detection have been reported [8e10]. But PAbs sometimes react nonspecifically while the preparation of MAbs is time consuming and costly. Recently, the production of recombinant antibodies (RAbs), such as single-chain variable fragment (scFv), has been regarded as an alternative way to obtain low-cost antibodies with desirable affinity and specificity [11e13]. An increasing numbers of scFv antibodies against small molecule contaminants, such as pesticides, fungal toxins, clenbuterol and some other food pollutants, have been produced [14e16]. The low affinity of the scFv antibody always hindered its application [17]. But for the pesticide residue analysis, an alternative approach is improving the affinity of the detection tag of the scFv antibody. Biotin which could be detected by biotinestreptavidin system with extremely strong affinity will be an appropriate tag. For biotinylation of scFv antibody, compared with the random modification of chemical methods, Escherichia coli biotin ligase (BirA) which could catalyze the site-specifically attachment of biotin to a 15 amino acid acceptor peptide (known as Avi-tag) is preferable [18]. In this paper, we selected a scFv antibody against PM from a preimmunized phage display library. The scFv was fused with a biotin acceptor domain (BAD) and high-level expressed in E. coli BL21 (DE3) as soluble protein. Then, the fused protein was biotinylated in vitro by BirA before purification. The purified biotinylated scFvBAD antibody was characterized by sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDSPAGE) and western blotting. Based on the biotinylated scFv-BAD, a sensitive indirect competitive enzyme-linked immunosorbent assay (IC-ELISA) for detection of PM was developed.
2.1. Materials and instruments Pesticides were purchased from National Standards. (China). E. coli BL21 (DE3) and plasmid pET-28a(þ) were preserved in our laboratory. Phagemid vector pIT2, E. coli TG1 and M13KO7 helper phage were purchased from Amersham Biosciences (Germany). Horseradish Peroxidase-labeled goat anti-mouse immunoglobulin (IgG-HRP), horseradish peroxidase-labeled streptavidin (SA-HRP), complete Freund’s adjuvants, incomplete Freund’s adjuvants, isopropyl-b-D-thiogalactoside (IPTG), 3,30 -diaminobenzidine (DAB) and tetramethylbenzidine (TMB) were purchased from Sigma Chemical Co. (St. Louis, USA). RNAiso Plus Kit, Prime ScriptII 1st Strand cDNA Synthesis Kit, dNTP, Prime STAR®GXL DNA Polymerase, T4 DNA ligase and restriction enzymes were purchased from Takara Biotechnology Co. Ltd. (Dalian, China). Ni-IDA agarose was purchased from GenScript. (Nanjing, China). All chemical reagents were analytical grade. Polystyrene 96-well microtiter plates were purchased from Costar. (Corning, MA, USA). ELISA plates were washed with the KHB ST-36W plate washer (Shanghai, China), and well absorbance was determined with the Model 680 plate reader (Bio-Rad, USA). 2.2. Mice immunization The specific hapten of PM was synthesized and characterized as the previous study [8]. The structures of hapten and PM are shown in Fig. 1. The hapten was covalently attached to bovine serum albumin (BSA) or ovalbumin (OVA) to use as immunogen (haptenBSA) or coating antigen (hapten-OVA) by active ester method [19]. Five BALB/C female mice (6e8 weeks) were immunized intraperitoneally with 1 mg/mL of immunogen in 100 mL phosphatebuffered saline (PBS, 10 mM, pH 7.4) mixed with 100 mL complete Freund’s adjuvant. Immunizations were repeated two times with incomplete Freund’s adjuvant at two-week intervals. One week after the last boost, the titer of each mouse was determined by noncompetitive ELISA, the spleen cells of the mouse which showed the highest titer were harvested and used for total RNA preparation by RNAiso Plus Kit according to the manufacturer’s instructions. 2.3. Construction and panning of phage-display library The cDNA was synthesized by PrimeScript II 1st Strand cDNA Synthesis Kit and used as template for PCR amplification of variable heavy (VH) and light (VL) chain genes. The primers and PCR procedures are the same as previously described [20]. The VL and VH genes were then assembled by SOC-PCR to obtain the scFv genes. The scFv gene fragments were purified by gel extraction kit, digested by Sfi I, inserted into pIT2-P phagemid and electrotransformed into E. coli TG1 to obtain the antibody library. The phagemid pIT2-P (Fig. 2B), which contain a tetracycline resistance gene (Tet) and two Sfi I sites between the pelB and full length gIII gene was transformed from pIT2 (Fig. 2A). The antibody library was
H. Wang et al. / Protein Expression and Purification 126 (2016) 1e8
3
Fig. 1. The structures of the hapten and parathion-methyl.
rescued by a helper phage KM13 and scFv antibody fragments were displayed on the surface of phage particles. The procedures of rescue and phage particles purification are the same as previous study [21]. Panning of the phage scFv library was performed as follows. A well of the 12-well plate was coated with 1 mL of coating antigen in PBS and incubated overnight at 4 C. After 3 times washing with 1 PBS (10 mM, pH 7.4), the well was blocked with 4% (w/v) skim milk powder in 1 PBS at 37 C for 1 h. Then the initial phage library was added and incubated at 37 C for 1 h. Unbound phages were removed by washing 10 times with PBST (PBS containing 0.1% Tween-20) and followed by washing 10 times with PBS. Bound phages were competitively eluted by 500 mL parathion-methyl at 25 C for 1 h. The eluent was then added to 2.5 mL E. coli TG1 (OD600 nm ¼ 0.5) cells and incubated at 37 C for 30 min before spreading onto a TYE plate (tryptone 10 g/L, yeast extract 5 g/L and NaCl 8 g/L). The sub-library was collected, rescued and used for the next round of panning. A total of four rounds of panning were performed according to the above procedure. The concentrations of coating antigen and PM used for the panning were shown in Table 1. 2.4. Selection of the scFv antibodies A phagemid pIT2-E was constructed and used for secretory expression and selection of scFv antibodies (see Fig. 2C). For the construction of pIT2-E, a biotin acceptor domain (BAD) [17] which encoding the IgA hinge, Avi-tag, His-tag and a stop codon (TAA) was inserted into pIT2-P phagemid between the EcoR I and Nde I. The phagemids of the fourth round sub-library were digested with Sfi I and inserted into pIT2-E. The ligation product was transformed into E. coli TG1 (BirA) to produce soluble scFv antibodies library and used for scFv antibody selection. A pBirAcm plasmid (encoding the biotin ligase and the chloramphenicol resistance gene) is contained in the E. coli TG1 (BirA), hence, scFv antibodies could be expressed and biotinylated in vivo simultaneously. 60 single clones from soluble scFv antibodies library were inoculated in 1 mL 2 YT medium (tryptone 16 g/L, yeast extract 10 g/L, and NaCl 5 g/L) with 100 mg/mL ampicillin and 10 mg/mL chloramphenicol and shaken at 37 C until OD600 nm reached 0.9. Then 100 mL of 2 TY containing 1 mM IPTG and 500 mM D-Biotin was added to each tube to induce the expression of both the scFv antibody and biotin ligase and shaken for 20 h at 25 C (200 rpm). The cells were harvested by centrifugation, re-suspended in 1 mL PBS and disrupted by ultrasound. The supernatants were collected by centrifugation and screened by IC-ELISA. The IC-ELISA procedure was as follows. ELISA plate was coated with 100 mL coating antigen (1 mg/mL) per well in 1 PBS and incubated at 37 C for 1 h. Then the plate was washed three times with 1 PBS, blocked with 1% skim milk (200 mL per well) and incubated at 37 C for 1 h. After washing three times, 50 mL of PM (200 ng/mL) in 1 PBS and 50 mL of biotinylated scFv supernatant were added to the blocked plate. After 1 h incubation and three times washing, 100 mL per well of diluted SA-HRP (1/5000) was added. The mixture was incubated for 1 h, and followed by six times
Fig. 2. Phagemids used in this study. The phagemid pIT2 (A), pIT2-P (B) was used for phage display and pIT2-E (C) was used for selection of scFv antibodies.
4
H. Wang et al. / Protein Expression and Purification 126 (2016) 1e8
Table 1 Overview of panning. Round
Concentration of coating antigen (mg/mL)
Concentration of PM in eluent solution (ng/mL)
Phages input
1 2 3 4
100 40 10 5
2000 1000 500 200
1.0 1.0 1.0 1.0
1013 1013 1013 1013
Phage output 8.1 3.4 4.2 4.5
106 107 108 108
Recovery (%) 8.1 3.4 4.2 4.5
105 104 103 103
Recovery (%) ¼ (number of phages output)/(number of phages input) 100.
30 and 37 C), the concentration of IPTG (0.1, 0.25, 0.5, 1.0, 2.5, 5.0 mM) and time (2, 4, 6, 8 and 10 h) were optimized to improve the expression level of soluble protein. The cells expressed in different conditions were collected and analyzed by SDS-PAGE. 2.6. In vitro biotinylation and purification of scFv antibody
Fig. 3. Agarose gel electrophoresis of purified VL, VH and scFv gene fragments obtained from the mouse 4. Lane 1, light chains (VL); lane 2, heavy chains (VH); lane 3, scFv gene fragments assembled by SOE-PCR.
washing. Finally, 100 mL per well of TMB solution was added. After incubation for 15 min, the reaction was stopped by 50 mL of 2 M H2SO4, The absorbance was read at 450 nm by a microplate reader. Clones which have high affinity and sensitivity were selected and the phasmids were extracted for sequencing. 2.5. High-level expression of scFv antibody and condition optimization The plasmid of a scFv antibody which showed high affinity and sensitivity for PM was digested with Nco I and Hind III to obtain the scFv-BAD fragment and inserted into the pET-28a(þ) expression vector. The plasmid was transformed into E. coli BL21 (DE3) cells for high-level expression. The expression conditions, such as induced temperature (20, 25,
The primers used for amplification the BirA coding gene were as follows: BirA-F (BamH I) 50 -CGGGATCCATGAAGGAT-AACACCG-30 and BirA-R (Hind III) 50 -CCAAGC TTTTATTTTTCTGCACTACGC-30 . A stop codon (TAA) was added at the end of the gene to guarantee the fusion protein expressed without His-tag. After digesting with BamH I and Hind III, the BirA gene was inserted into pET-28a(þ) and transformed into E. coli BL21 (DE3). The fusion protein was expressed in 10 mL LB broth and induced by 0.5 mM IPTG at 37 C for 4 h (200 rpm). The cells were harvested, suspended in 1 mL BirA buffer A (10 mM Tris-HCl, 20 mM NaCl, pH 8.0), sonicated, and centrifuged at 12,000 rpm for 15 min at 4 C. The supernatant which contains the soluble BirA protein was used for in vitro biotinylation directly. The selected scFv antibody was performed for high-level expression with the optimum conditions in 200 mL LB broth. The cells were harvested with wet weight of 750 mg, suspended in 35 mL LE buffer (50 mM NaH2PO4, 300 mM NaCl, pH 8.0) and broken by ultrasonication. Then the solution was centrifuged at 12,000 rpm for 15 min and the supernatant was transferred into a new 50-mL tube. Subsequently, 4 mL of 10 BirA buffer B (1 mM DBiotin, 100 mM ATP-Na2, pH 8.0) and 1 mL of the unpurified BirA enzyme were added and mixed thoroughly and stored at 4 C for 24 h to complete the biotinylation reaction. Finally, the mixture was purified by Ni-IDA chromatography column (contains 2 mL Ni-IDA resin) according to the manufacturer’s instructions to obtain the biotinylated scFv antibody. The purified scFv antibody was dialyzed against 1 PBS before storage at 4 C. The protein concentration was determined using the Bradford assay with BSA as standard
Fig. 4. Amino acid sequence of PM-30 (scFv-BAD) for parathion-methyl. The complementarity-determining regions of the VH (VH/CDR) and VL (VL/CDR) domains are underlined. The linker fragment (Gly4Ser)4 is shown in italic type. The BAD fragment is shown in gray background. The three different amino acids between PM-30 and PM-25 scFv antibodies are shown in red and italic type (The amino acids M, R and F in PM-30 are replaced by R, P and S, respectively). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
H. Wang et al. / Protein Expression and Purification 126 (2016) 1e8
5
Fig. 6. SDS-PAGE (A) and western blotting (B) analysis of the scFv antibody against PM. (A) Lane M, protein molecular weight standards; lane 1, scFv antibody without biotinylation; lane 2, biotinylated scFv antibody. (B) Western blotting of the biotinylated scFv antibody detected with SA-HRP.
protein and determined by K5600 ultra micro spectrophotometer under the Bradford model. The biotinylated scFv antibody was confirmed by SDS-PAGE and western blotting. 2.7. Development of biotinylated scFv based IC-ELISA The procedure of the developed IC-ELISA based on the biotinylated scFv antibody is the same as the process of scFv antibodies selection. Optimal IC-ELISA conditions, such as the ionic strength, pH and concentration of skim milk and methanol, were determined. The main criterion used to evaluate immunoassay performance was the ratio of maximal absorbance (Amax) and the 50% inhibition value (IC50). The condition which acquired the highest ratio of Amax/IC50 was selected and used as the optimized condition. The specificity of the biotinylated scFv antibody based IC-ELISA was evaluated by determining the cross-reactivity (CR) of 12 organophosphorus pesticides (OPs) which have a similar structure with PM. CR was calculated by the following equation: CR ¼ (IC50 of parathion-methyl/IC50 of other compound) 100
3. Results and discussion 3.1. Phage-display library construction Fig. 5. Effect of induction temperature (A), IPTG concentration (B) and the induction time (C) for expression of the soluble scFv antibody. Lane M, protein molecular weight standards; (A) Lane 1e4, soluble protein of E. coli BL21(DE3) cells induced with 0.5 mM IPTG at 20 C, 25 C, 30 C, 37 C for 10 h, 8 h, 5 h and 3 h, respectively; (B) Lane 1e6, soluble protein induced with 0.1, 0.25, 0.5, 1.0, 2.5, 5.0 mM IPTG for 8 h at 25 C, respectively; (C) Lane 1e5, soluble protein induced for 2, 4, 6, 8 and 10 h with 0.5 mM IPTG at 25 C, respectively. Different conditions were performed with similar yields of cells.
Spleen cells of mouse 4 which showed the highest titer for coating antigen (1/32,000) was collected and used for preparation of total RNA, followed by reverse transcription to obtain the cDNA. The VH and VL genes were cloned from the cDNA and assembled with a 20-amino acid linker (Gly4Ser)4 to obtain the full-length scFv genes (Fig. 3). The scFv genes were gel purified and digested by Sfi I and then inserted into pIT2-P. Finally, the plasmids were electroporated into E. coli TG1 and a phage-display library including
6
H. Wang et al. / Protein Expression and Purification 126 (2016) 1e8
Fig. 8. Standard inhibition curve of parathion-methyl determined by the optimum biotinylated scFv antibody based IC-ELISA. The optimum working dilutions for purified scFv antibody (1.2 mg/mL) and coating antigen (2.0 mg/mL) were at a ratio of 1:1000 and 1:2000, respectively. Each point represents the mean of three replicates.
results were shown in Table 1. The concentrations of coating antigen and PM concentration used for elution were successively reduced from the first to the fourth round of panning, but the phage recovery rates increased sustainably. The results suggested that the scFv-phages against PM were successfully enriched. Moreover, the recovery rate of the fourth round showed no significant increasing compared with the third, indicating that the specific phages were enriched enough and no more panning was needed. 3.3. Selection of scFv antibodies An expression vector pIT2-E was constructed and used in the scFv antibodies selection. The vector which contain the pelB signal peptide and the biotin acceptor domain make sure that the scFv antibodies could be solubly expressed as well as in vivo biotinylated in the TG1 (BirA) strain. 60 clones were expressed by small scale and screened by biotin-streptavidin system based IC-ELISA. A total of 44 clones were positive and 6 clones which provide better features were analyzed by DNA sequencing. The sequence results indicated that only two different sequences were obtained (named PM-25 and PM-30). The deduced amino acid sequences of the two scFv-BAD antibodies are shown in Fig. 4. Both of the two antibodies contain 306 amino acids including VL, VH, linker and BAD. Only three amino acids are different between the two antibodies (Fig. 4) and the PM-30 which showed a little higher sensitivity for PM was selected for high-level expression. 3.4. High-level expression of scFv antibody and condition optimization
Fig. 7. Influence of the ionic strength (A), pH (B), concentration of methanol (C) and concentration of skim milk (D) on the biotinylated scFv antibody based IC-ELISA. Each point represents the mean of three replicates.
1.8 108 independent clones was obtained. 3.2. Panning Four rounds of panning operations were performed and the
pET-28a(þ) vector was employed for high-level expression of scFv antibody. The scFv-BAD gene from phasmid of PM-30 was inserted into the pET-28a(þ) and expressed in E. coli BL21 (DE3). Pre-experiment indicated that the scFv-BAD fusion protein could be partially solubly expressed. Hence, the induction conditions including temperature, concentration of IPTG and induced time were optimized. As shown in Fig. 5A, the expression level of scFv antibody reached the highest level at 25 C. The concentrations of IPTG have no significant effect on production of the soluble scFvBAD and 0.1 mM was selected for inducing soluble protein expression (Fig. 5B). For inducing time optimization, the production
H. Wang et al. / Protein Expression and Purification 126 (2016) 1e8
7
Table 2 Cross-reactivity of the scFv antibody for organophosphorus pesticides. No.
Analytes
1
Parathion-methyl
2
Structure
IC50 (ng mL1)
CR (%)
14.5
100.0
Parathion
206.4
7.0
3
Fenitrothion
256.7
5.7
4
Paraoxon-methyl
323.7
4.5
5
Diazinon
>5000
<0.3
6
Fenthion
>5000
<0.3
7
Pirimiphos-methyl
>5000
<0.3
8
Chlorpyrifos
>5000
<0.3
9
Chlorpyrifos-methyl
>5000
<0.3
10
Paraoxon-ethyl
>5000
<0.3
11
Azinphos-methyl
>5000
<0.3
12
Bromophosc
>5000
<0.3
CR (%) ¼ (IC50 of parathion-methyl/IC50 of other compound) 100.
of target protein was increased with the extension of induction time from 2 h to 8 h, but for 10 h induction, the production did not increase significantly. Hence, the optimum induced time was 8 h (Fig. 5C).
3.5. In vitro biotinylation, purification and characterization of scFv antibody BirA was efficiently expressed and confirmed by SDS-PAGE (Fig. S1). The result indicated that the protein was partially soluble and partially in form of inclusion body. The soluble BirA was employed for in vitro biotinylation of scFv antibody. The scFv-BAD antibody was expressed at a high level under the optimum condition. Then the unpurified antibody was biotinylated in vitro by the unpurified BirA. After the biotinylation reaction, the scFv-BAD antibody was purified by the Ni-IDA agarose column chromatography with 6 His-tag at the end of BAD. The purified soluble scFv
antibody was obtained with yield of 59.2 ± 3.7 mg/L of culture (15.8 mg/g of wet weight cell pellet and 10.9% of the total lysate). The purified biotinylated scFv antibody and scFv antibody without biotinylation were compared by SDS-PAGE (Fig. 6A), and the results indicated the molecular mass of scFv antibody without biotinylation scFv antibody was 33 kDa as expected (Fig. 6A, lane 1) and the molecular mass of the biotinylated scFv-BAD was a little bigger because of the addition of biotin molecule (Fig. 6A, lane 2). For the biotinylation of scFv antibody, the amount of BirA enzyme, D-biotin and ATP were excess to make sure the complete of the reaction. The biotinylated scFv-BAD antibody was also confirmed by western blotting using SA-HRP as a detector, and the result evidenced the successful biotinylation (Fig. 6B).
3.6. Optimization of scFv-based IC-ELISA Ionic strengths, pH values and organic solvents always influence
8
H. Wang et al. / Protein Expression and Purification 126 (2016) 1e8
the ELISA performance [22,23]. The ionic strength, pH and methanol concentration of ELISA performance were optimized in this study. The effect of NaCl concentration on ELISA was shown in Fig. 7A and 2.0 M of NaCl which led to the highest Amax/IC50 value was selected as the optimum ionic strength. Similarly, pH 6.0 and 2.5% methanol which reached the highest values of Amax/IC50 were selected as the optimum conditions (Fig. 7B and Fig. 7C). During the developing process of IC-ELISA, we found that PBS containing a certain amount of skim milk (MPBS) could significantly improve the sensitivity of scFv antibody and the optimized concentration of skim milk was determined as 0.25% (Fig. 7D). Consequently, the optimal assay solution was MPBS (pH 6.0) containing 0.25% skim milk, 2.0 M NaCl and 2.5% methanol. The standard sigmoid inhibition curve of PM determined by the optimum ELISA was shown in Fig. 8. The 50% inhibition value (IC50) of PM was 14.5 ng/mL and the limit of detection (IC10, LOD) was 0.9 ng/mL. 3.7. Cross-reactivity study The hapten used in this study was designed to produce high specific antibody for PM. Table 2 showed the IC50 values and CR of PM and 11 OPs which have similar structures with PM. Although most OPs possess the same moieties: O,O-dimethyl phosphorothioate or benzene ring, the scFv antibody only showed low crossreactivity for parathion, fenitrothion and paraoxon-methyl and the CR of the three pesticides was 7.0%, 5.7% and 4.5%, respectively. For the rest pesticides, cross-reactivities were all below 0.3%. The negligible cross-reactivity proved that the scFv antibody was specific to PM. 4. Conclusions Preparation of high quality antibodies is still a bottleneck problem when establishing immunoassay methods for small molecule contaminants. In this study, we developed an efficient method to select and produce large scale of biotinylated scFv antibodies. A sensitive and specific scFv antibody against PM was selected from a pre-immunized phage display library and expressed as soluble protein in high-level. Then the scFv antibody was simply biotinylated in vitro with BirA. Based on the biotinylated scFv antibody, a sensitive IC-ELISA for detecting PM was developed. All these results indicate that the biotinylation of scFv antibody is an efficient strategy in the development of immunoassay. Acknowledgments This research was supported by National Natural Science Foundation of China (grant numbers 30972050, 31271873). Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.pep.2016.05.005.
References [1] A.Y. Kolosovaa, J.H. Parka, S.A. Ereminb, S.J. Parka, S.J. Kanga, W.B. Shima, H.S. Leec, Y.T. Leed, D.H. Chunga, Comparative study of three immunoassays based on monoclonal antibodies for detection of the pesticide parathionmethyl in real samples, Anal. Chim. Acta 511 (2004) 323e331. [2] F.L. Edwards, P.B. Tchounwou, Environmental toxicology and health effects associated with methyl parathion exposureea scientific review, Int. J. Environ. Res. Public Health 2 (2005) 430e441. [3] C. Ferrer, M.J. Martínez-Bueno, A. Lozano, Pesticide residue analysis of fruit juices by LCeMS/MS direct injection, Talanta 83 (2011) 1552e1561. [4] N. Yoshioka, Y. Akiyama, T. Matsuoka, Rapid determination of five postharvest fungicides and metabolite in citrus fruits by liquid chromatography/ time-of-flight mass spectrometry with atmospheric pressure photoionization, Food Control 21 (2010) 212e216. [5] X.D. Xue, Q. Wei, D. Wu, H. Li, Y. Zhang, R. Feng, B. Du, Determination of methyl parathion by a molecularly imprinted sensor based on nitrogen doped graphene sheets, Electrochim. Acta 116 (2014) 366e371. [6] A. Shanker, C. Sood, V. Kumar, S.D. Ravindranath, A modified extraction and clean-up procedure for the detection and determination of parathion-methyl and chlorpyrifos residues in tea, Pest. Manag. Sci. 57 (2001) 458e462. [7] K. Li, Q.X. Li, Development of an enzyme linked immunosorbent assay for the insecticide imidacloprid, J. Agric. Food Chem. 48 (2000) 3378e3382. [8] M.J. Kim, H.S. Lee, D.H. Chung, Y.T. Lee, Synthesis of haptens of organophosphorus pesticides and development of enzyme-linked immunosorbent assays for parathion-methyl, Anal. Chim. Acta 493 (2003) 47e62. [9] A.Y. Kolosova, J.H. Park, S.A. Eremin, Comparative study of three immunoassays based on monoclonal antibodies for detection of the pesticide parathionmethyl in real samples, Anal. Chim. Acta 511 (2004) 323e331. [10] M.J. Moreno, E. Plana, J.J. Manclús, A. Montoya, Comparative study of monoclonal and recombinant antibody-based immunoassays for fungicide analysis in fruit juices, Food Anal. Method 7 (2014) 481e489. [11] C.P. Chan, Y.C. Cheung, R. Renneberg, M. Seydack, New trends in immunoassays, Adv. Biochem. Eng. Biotechnol. 109 (2008) 123e154. [12] K.Y. Yau, H. Lee, J.C. Hall, Emerging trends in the synthesis and improvement of hapten-specific recombinant antibodies, Biotechnol. Adv. 21 (2003) 599e637. [13] V. Markus, L. Janne, L. Urpo, Directed antibody-engineering techniques and their applications in food immunoassays, TrAC-Trend. Anal. Chem. 30 (2011) 219e226. [14] J. He, Y. Liang, M.T. Fan, Preparation of Anti-idiotype antibodies of O,ODimethyl organophosphorus pesticides by phage display technology, Chin. J. Anal. Chem. 39 (2011) 178e182. [15] B. Lauer, I. Ottleben, H.J. Jacobsen, T. Reinard, Production of a single-chain variable fragment antibody against fumonisin B1, J. Agric. Food Chem. 53 (2005) 899e904. [16] H. Wang, X.X. Liu, Y.S. He, Expression and purification of an anti-clenbuterol single chain Fv antibody in Escherichia coli, Protein Expr. Purif. 72 (2010) 26e31. [17] S.M. Cloutier, S. Couty, A. Terskikh, Streptabody, a high avidity molecule made by tetramerization of in vivo biotinylated, phage display-selected scFv fragments on streptavidin, Mol. Immunol. 37 (2000) 1067e1077. [18] Y. Li, R. Sousa, Expression and purification of E. coli BirA biotin ligase for in vitro biotinylation, Protein Expr. Purif. 82 (2012) 162e167. [19] D.P. McAdam, A.S. Hill, H.L. Beasley, J.H. Skerritt, Mono-and polyclonal antibodies to the organophosphate fenitrothion. 1. Approaches to hapten-protein conjugation, J. Agric. Food Chem. 40 (1992) 1466e1470. [20] A. Krebber, S. Bornhauser, J. Burmester, Reliable cloning of functional antibody variable domains from hybridomas and spleen cell repertoires employing a reengineered phage display system, J. Immunol. Method 201 (1997) 35e55. Gonz [21] E. Garet, A.G. Cabado, J.M. Vieites, A. alez-Fern andez, Rapid isolation of single-chain antibodies by phage display technology directed against one of the most potent marine toxins: palytoxin, Toxicon 55 (2010) 1519e1526. [22] X. Jiang, H. Shi, N. Wu, M. Wang, Development of an enzyme-linked immunosorbent assay for diniconazole in agricultural samples, Food Chem. 125 (2011) 1385e1389. [23] Y. Cui, K. Liu, C. Xu, F. Liu, Q.X. Li, S. Liu, B. Wang, Development of a sensitive monoclonal antibody-based indirect competitive enzyme-linked immunosorbent assay for analyzing chlorantraniliprole residues, Food Chem. 143 (2014) 293e299.
Contents lists available at ScienceDirect
Protein Expression and Purification journal homepage: www.elsevier.com/locate/yprep
Review article
Production and characterization of a biotinylated single-chain variable fragment antibody for detection of parathion-methyl Huimin Wang 1, Fengchun Zhao 1, Xiao Han, Zhengyou Yang* Department of Microbiology, College of Life Science, Key Laboratory for Agriculture Microbiology, Shandong Agricultural University, Taian 271018, China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 24 April 2016 Received in revised form 8 May 2016 Accepted 9 May 2016 Available online 12 May 2016
In this article, we reported the development of a biotinylated single-chain variable fragment (scFv) antibody based indirect competitive enzyme-linked immunosorbent assay (IC-ELISA) for parathionmethyl (PM) detection. Firstly, a phage display library was generated using a pre-immunized BALB/C mouse against a specific hapten of PM. After four rounds of panning, the scFv gene fragments were transferred into a secreted expression vector. Then, the scFv antibodies were secreted expressed and screened by IC-ELISA against PM. The selected scFv antibody was fused with a biotin acceptor domain (BAD) and inserted into pET-28a(þ) vector for high-level expression in Escherichia coli BL2 (DE3). After optimizing expression conditions, the scFv-BAD antibody was expressed as a soluble protein and biotinylated in vitro by the E. coli biotin ligase (BirA). Subsequently, the biotinylated scFv-BAD antibody was purified with a high yield of 59.2 ± 3.7 mg/L of culture, and was characterized by SDS-PAGE and western blotting. Finally, based on the biotinylated scFv-BAD, a sensitive IC-ELISA for detection of PM was developed, and the 50% inhibition value (IC50) of PM was determined as 14.5 ng/mL, with a limit of detection (LOD, IC10) of 0.9 ng/mL. Cross-reactivity (CR) studies revealed that the scFv antibody showed desirable specificity for PM. © 2016 Elsevier Inc. All rights reserved.
Keywords: Parathion-methyl Phage display Biotinylation scFv antibody IC-ELISA
Contents 1. 2.
3.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Materials and method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2.1. Materials and instruments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2.2. Mice immunization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2.3. Construction and panning of phage-display library . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2.4. Selection of the scFv antibodies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2.5. High-level expression of scFv antibody and condition optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 2.6. In vitro biotinylation and purification of scFv antibody . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 2.7. Development of biotinylated scFv based IC-ELISA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Results and discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 3.1. Phage-display library construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 3.2. Panning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 3.3. Selection of scFv antibodies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 3.4. High-level expression of scFv antibody and condition optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 3.5. In vitro biotinylation, purification and characterization of scFv antibody . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 3.6. Optimization of scFv-based IC-ELISA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 3.7. Cross-reactivity study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
* Corresponding author. E-mail address: [email protected] (Z. Yang). 1 Authors made equal contributions to this work. http://dx.doi.org/10.1016/j.pep.2016.05.005 1046-5928/© 2016 Elsevier Inc. All rights reserved.
2
H. Wang et al. / Protein Expression and Purification 126 (2016) 1e8
4.
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Supplementary data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1. Introduction
2. Materials and method
Parathion-methyl (O,O-dimethyl-O-4-nitrophenyl phosphorothioate, PM), which is widely used as an insecticide and acaricide to control many insect pests of agricultural products, is a highly toxic insecticide in EPA (United States Environmental Protection Agency) toxicity class I. It could inhibit acetylcholinesterase activity in erythrocytes and brain even at a low concentration [1]. Due to its high toxicity and extensive application, PM has been classified as a restricted pesticide with a strict maximum residue limited (MRLs) standard in many countries [2]. For analysis of PM, some traditional methods, such as gas chromatography (GC) and high-performance liquid chromatography (HPLC), have been successfully developed due to their high sensitivity and reliability [3,4]. However, these classical methods require high cost, skilled analysts, and time-consuming sample preparation steps. Electroanalytical sensors exhibited high sensitivity but showed poor selectivity in determination of practical samples, which limited the application and promotion of electroanalytical methods [5]. Compared with these analysis strategies, immunoassay technology is gaining acceptability as a simple, costeffective screening method for pesticides analysis in many samples [6,7]. Several immunoassay methods based on polyclonal antibodies (PAbs) and monoclonal antibodies (MAbs) for PM detection have been reported [8e10]. But PAbs sometimes react nonspecifically while the preparation of MAbs is time consuming and costly. Recently, the production of recombinant antibodies (RAbs), such as single-chain variable fragment (scFv), has been regarded as an alternative way to obtain low-cost antibodies with desirable affinity and specificity [11e13]. An increasing numbers of scFv antibodies against small molecule contaminants, such as pesticides, fungal toxins, clenbuterol and some other food pollutants, have been produced [14e16]. The low affinity of the scFv antibody always hindered its application [17]. But for the pesticide residue analysis, an alternative approach is improving the affinity of the detection tag of the scFv antibody. Biotin which could be detected by biotinestreptavidin system with extremely strong affinity will be an appropriate tag. For biotinylation of scFv antibody, compared with the random modification of chemical methods, Escherichia coli biotin ligase (BirA) which could catalyze the site-specifically attachment of biotin to a 15 amino acid acceptor peptide (known as Avi-tag) is preferable [18]. In this paper, we selected a scFv antibody against PM from a preimmunized phage display library. The scFv was fused with a biotin acceptor domain (BAD) and high-level expressed in E. coli BL21 (DE3) as soluble protein. Then, the fused protein was biotinylated in vitro by BirA before purification. The purified biotinylated scFvBAD antibody was characterized by sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDSPAGE) and western blotting. Based on the biotinylated scFv-BAD, a sensitive indirect competitive enzyme-linked immunosorbent assay (IC-ELISA) for detection of PM was developed.
2.1. Materials and instruments Pesticides were purchased from National Standards. (China). E. coli BL21 (DE3) and plasmid pET-28a(þ) were preserved in our laboratory. Phagemid vector pIT2, E. coli TG1 and M13KO7 helper phage were purchased from Amersham Biosciences (Germany). Horseradish Peroxidase-labeled goat anti-mouse immunoglobulin (IgG-HRP), horseradish peroxidase-labeled streptavidin (SA-HRP), complete Freund’s adjuvants, incomplete Freund’s adjuvants, isopropyl-b-D-thiogalactoside (IPTG), 3,30 -diaminobenzidine (DAB) and tetramethylbenzidine (TMB) were purchased from Sigma Chemical Co. (St. Louis, USA). RNAiso Plus Kit, Prime ScriptII 1st Strand cDNA Synthesis Kit, dNTP, Prime STAR®GXL DNA Polymerase, T4 DNA ligase and restriction enzymes were purchased from Takara Biotechnology Co. Ltd. (Dalian, China). Ni-IDA agarose was purchased from GenScript. (Nanjing, China). All chemical reagents were analytical grade. Polystyrene 96-well microtiter plates were purchased from Costar. (Corning, MA, USA). ELISA plates were washed with the KHB ST-36W plate washer (Shanghai, China), and well absorbance was determined with the Model 680 plate reader (Bio-Rad, USA). 2.2. Mice immunization The specific hapten of PM was synthesized and characterized as the previous study [8]. The structures of hapten and PM are shown in Fig. 1. The hapten was covalently attached to bovine serum albumin (BSA) or ovalbumin (OVA) to use as immunogen (haptenBSA) or coating antigen (hapten-OVA) by active ester method [19]. Five BALB/C female mice (6e8 weeks) were immunized intraperitoneally with 1 mg/mL of immunogen in 100 mL phosphatebuffered saline (PBS, 10 mM, pH 7.4) mixed with 100 mL complete Freund’s adjuvant. Immunizations were repeated two times with incomplete Freund’s adjuvant at two-week intervals. One week after the last boost, the titer of each mouse was determined by noncompetitive ELISA, the spleen cells of the mouse which showed the highest titer were harvested and used for total RNA preparation by RNAiso Plus Kit according to the manufacturer’s instructions. 2.3. Construction and panning of phage-display library The cDNA was synthesized by PrimeScript II 1st Strand cDNA Synthesis Kit and used as template for PCR amplification of variable heavy (VH) and light (VL) chain genes. The primers and PCR procedures are the same as previously described [20]. The VL and VH genes were then assembled by SOC-PCR to obtain the scFv genes. The scFv gene fragments were purified by gel extraction kit, digested by Sfi I, inserted into pIT2-P phagemid and electrotransformed into E. coli TG1 to obtain the antibody library. The phagemid pIT2-P (Fig. 2B), which contain a tetracycline resistance gene (Tet) and two Sfi I sites between the pelB and full length gIII gene was transformed from pIT2 (Fig. 2A). The antibody library was
H. Wang et al. / Protein Expression and Purification 126 (2016) 1e8
3
Fig. 1. The structures of the hapten and parathion-methyl.
rescued by a helper phage KM13 and scFv antibody fragments were displayed on the surface of phage particles. The procedures of rescue and phage particles purification are the same as previous study [21]. Panning of the phage scFv library was performed as follows. A well of the 12-well plate was coated with 1 mL of coating antigen in PBS and incubated overnight at 4 C. After 3 times washing with 1 PBS (10 mM, pH 7.4), the well was blocked with 4% (w/v) skim milk powder in 1 PBS at 37 C for 1 h. Then the initial phage library was added and incubated at 37 C for 1 h. Unbound phages were removed by washing 10 times with PBST (PBS containing 0.1% Tween-20) and followed by washing 10 times with PBS. Bound phages were competitively eluted by 500 mL parathion-methyl at 25 C for 1 h. The eluent was then added to 2.5 mL E. coli TG1 (OD600 nm ¼ 0.5) cells and incubated at 37 C for 30 min before spreading onto a TYE plate (tryptone 10 g/L, yeast extract 5 g/L and NaCl 8 g/L). The sub-library was collected, rescued and used for the next round of panning. A total of four rounds of panning were performed according to the above procedure. The concentrations of coating antigen and PM used for the panning were shown in Table 1. 2.4. Selection of the scFv antibodies A phagemid pIT2-E was constructed and used for secretory expression and selection of scFv antibodies (see Fig. 2C). For the construction of pIT2-E, a biotin acceptor domain (BAD) [17] which encoding the IgA hinge, Avi-tag, His-tag and a stop codon (TAA) was inserted into pIT2-P phagemid between the EcoR I and Nde I. The phagemids of the fourth round sub-library were digested with Sfi I and inserted into pIT2-E. The ligation product was transformed into E. coli TG1 (BirA) to produce soluble scFv antibodies library and used for scFv antibody selection. A pBirAcm plasmid (encoding the biotin ligase and the chloramphenicol resistance gene) is contained in the E. coli TG1 (BirA), hence, scFv antibodies could be expressed and biotinylated in vivo simultaneously. 60 single clones from soluble scFv antibodies library were inoculated in 1 mL 2 YT medium (tryptone 16 g/L, yeast extract 10 g/L, and NaCl 5 g/L) with 100 mg/mL ampicillin and 10 mg/mL chloramphenicol and shaken at 37 C until OD600 nm reached 0.9. Then 100 mL of 2 TY containing 1 mM IPTG and 500 mM D-Biotin was added to each tube to induce the expression of both the scFv antibody and biotin ligase and shaken for 20 h at 25 C (200 rpm). The cells were harvested by centrifugation, re-suspended in 1 mL PBS and disrupted by ultrasound. The supernatants were collected by centrifugation and screened by IC-ELISA. The IC-ELISA procedure was as follows. ELISA plate was coated with 100 mL coating antigen (1 mg/mL) per well in 1 PBS and incubated at 37 C for 1 h. Then the plate was washed three times with 1 PBS, blocked with 1% skim milk (200 mL per well) and incubated at 37 C for 1 h. After washing three times, 50 mL of PM (200 ng/mL) in 1 PBS and 50 mL of biotinylated scFv supernatant were added to the blocked plate. After 1 h incubation and three times washing, 100 mL per well of diluted SA-HRP (1/5000) was added. The mixture was incubated for 1 h, and followed by six times
Fig. 2. Phagemids used in this study. The phagemid pIT2 (A), pIT2-P (B) was used for phage display and pIT2-E (C) was used for selection of scFv antibodies.
4
H. Wang et al. / Protein Expression and Purification 126 (2016) 1e8
Table 1 Overview of panning. Round
Concentration of coating antigen (mg/mL)
Concentration of PM in eluent solution (ng/mL)
Phages input
1 2 3 4
100 40 10 5
2000 1000 500 200
1.0 1.0 1.0 1.0
1013 1013 1013 1013
Phage output 8.1 3.4 4.2 4.5
106 107 108 108
Recovery (%) 8.1 3.4 4.2 4.5
105 104 103 103
Recovery (%) ¼ (number of phages output)/(number of phages input) 100.
30 and 37 C), the concentration of IPTG (0.1, 0.25, 0.5, 1.0, 2.5, 5.0 mM) and time (2, 4, 6, 8 and 10 h) were optimized to improve the expression level of soluble protein. The cells expressed in different conditions were collected and analyzed by SDS-PAGE. 2.6. In vitro biotinylation and purification of scFv antibody
Fig. 3. Agarose gel electrophoresis of purified VL, VH and scFv gene fragments obtained from the mouse 4. Lane 1, light chains (VL); lane 2, heavy chains (VH); lane 3, scFv gene fragments assembled by SOE-PCR.
washing. Finally, 100 mL per well of TMB solution was added. After incubation for 15 min, the reaction was stopped by 50 mL of 2 M H2SO4, The absorbance was read at 450 nm by a microplate reader. Clones which have high affinity and sensitivity were selected and the phasmids were extracted for sequencing. 2.5. High-level expression of scFv antibody and condition optimization The plasmid of a scFv antibody which showed high affinity and sensitivity for PM was digested with Nco I and Hind III to obtain the scFv-BAD fragment and inserted into the pET-28a(þ) expression vector. The plasmid was transformed into E. coli BL21 (DE3) cells for high-level expression. The expression conditions, such as induced temperature (20, 25,
The primers used for amplification the BirA coding gene were as follows: BirA-F (BamH I) 50 -CGGGATCCATGAAGGAT-AACACCG-30 and BirA-R (Hind III) 50 -CCAAGC TTTTATTTTTCTGCACTACGC-30 . A stop codon (TAA) was added at the end of the gene to guarantee the fusion protein expressed without His-tag. After digesting with BamH I and Hind III, the BirA gene was inserted into pET-28a(þ) and transformed into E. coli BL21 (DE3). The fusion protein was expressed in 10 mL LB broth and induced by 0.5 mM IPTG at 37 C for 4 h (200 rpm). The cells were harvested, suspended in 1 mL BirA buffer A (10 mM Tris-HCl, 20 mM NaCl, pH 8.0), sonicated, and centrifuged at 12,000 rpm for 15 min at 4 C. The supernatant which contains the soluble BirA protein was used for in vitro biotinylation directly. The selected scFv antibody was performed for high-level expression with the optimum conditions in 200 mL LB broth. The cells were harvested with wet weight of 750 mg, suspended in 35 mL LE buffer (50 mM NaH2PO4, 300 mM NaCl, pH 8.0) and broken by ultrasonication. Then the solution was centrifuged at 12,000 rpm for 15 min and the supernatant was transferred into a new 50-mL tube. Subsequently, 4 mL of 10 BirA buffer B (1 mM DBiotin, 100 mM ATP-Na2, pH 8.0) and 1 mL of the unpurified BirA enzyme were added and mixed thoroughly and stored at 4 C for 24 h to complete the biotinylation reaction. Finally, the mixture was purified by Ni-IDA chromatography column (contains 2 mL Ni-IDA resin) according to the manufacturer’s instructions to obtain the biotinylated scFv antibody. The purified scFv antibody was dialyzed against 1 PBS before storage at 4 C. The protein concentration was determined using the Bradford assay with BSA as standard
Fig. 4. Amino acid sequence of PM-30 (scFv-BAD) for parathion-methyl. The complementarity-determining regions of the VH (VH/CDR) and VL (VL/CDR) domains are underlined. The linker fragment (Gly4Ser)4 is shown in italic type. The BAD fragment is shown in gray background. The three different amino acids between PM-30 and PM-25 scFv antibodies are shown in red and italic type (The amino acids M, R and F in PM-30 are replaced by R, P and S, respectively). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
H. Wang et al. / Protein Expression and Purification 126 (2016) 1e8
5
Fig. 6. SDS-PAGE (A) and western blotting (B) analysis of the scFv antibody against PM. (A) Lane M, protein molecular weight standards; lane 1, scFv antibody without biotinylation; lane 2, biotinylated scFv antibody. (B) Western blotting of the biotinylated scFv antibody detected with SA-HRP.
protein and determined by K5600 ultra micro spectrophotometer under the Bradford model. The biotinylated scFv antibody was confirmed by SDS-PAGE and western blotting. 2.7. Development of biotinylated scFv based IC-ELISA The procedure of the developed IC-ELISA based on the biotinylated scFv antibody is the same as the process of scFv antibodies selection. Optimal IC-ELISA conditions, such as the ionic strength, pH and concentration of skim milk and methanol, were determined. The main criterion used to evaluate immunoassay performance was the ratio of maximal absorbance (Amax) and the 50% inhibition value (IC50). The condition which acquired the highest ratio of Amax/IC50 was selected and used as the optimized condition. The specificity of the biotinylated scFv antibody based IC-ELISA was evaluated by determining the cross-reactivity (CR) of 12 organophosphorus pesticides (OPs) which have a similar structure with PM. CR was calculated by the following equation: CR ¼ (IC50 of parathion-methyl/IC50 of other compound) 100
3. Results and discussion 3.1. Phage-display library construction Fig. 5. Effect of induction temperature (A), IPTG concentration (B) and the induction time (C) for expression of the soluble scFv antibody. Lane M, protein molecular weight standards; (A) Lane 1e4, soluble protein of E. coli BL21(DE3) cells induced with 0.5 mM IPTG at 20 C, 25 C, 30 C, 37 C for 10 h, 8 h, 5 h and 3 h, respectively; (B) Lane 1e6, soluble protein induced with 0.1, 0.25, 0.5, 1.0, 2.5, 5.0 mM IPTG for 8 h at 25 C, respectively; (C) Lane 1e5, soluble protein induced for 2, 4, 6, 8 and 10 h with 0.5 mM IPTG at 25 C, respectively. Different conditions were performed with similar yields of cells.
Spleen cells of mouse 4 which showed the highest titer for coating antigen (1/32,000) was collected and used for preparation of total RNA, followed by reverse transcription to obtain the cDNA. The VH and VL genes were cloned from the cDNA and assembled with a 20-amino acid linker (Gly4Ser)4 to obtain the full-length scFv genes (Fig. 3). The scFv genes were gel purified and digested by Sfi I and then inserted into pIT2-P. Finally, the plasmids were electroporated into E. coli TG1 and a phage-display library including
6
H. Wang et al. / Protein Expression and Purification 126 (2016) 1e8
Fig. 8. Standard inhibition curve of parathion-methyl determined by the optimum biotinylated scFv antibody based IC-ELISA. The optimum working dilutions for purified scFv antibody (1.2 mg/mL) and coating antigen (2.0 mg/mL) were at a ratio of 1:1000 and 1:2000, respectively. Each point represents the mean of three replicates.
results were shown in Table 1. The concentrations of coating antigen and PM concentration used for elution were successively reduced from the first to the fourth round of panning, but the phage recovery rates increased sustainably. The results suggested that the scFv-phages against PM were successfully enriched. Moreover, the recovery rate of the fourth round showed no significant increasing compared with the third, indicating that the specific phages were enriched enough and no more panning was needed. 3.3. Selection of scFv antibodies An expression vector pIT2-E was constructed and used in the scFv antibodies selection. The vector which contain the pelB signal peptide and the biotin acceptor domain make sure that the scFv antibodies could be solubly expressed as well as in vivo biotinylated in the TG1 (BirA) strain. 60 clones were expressed by small scale and screened by biotin-streptavidin system based IC-ELISA. A total of 44 clones were positive and 6 clones which provide better features were analyzed by DNA sequencing. The sequence results indicated that only two different sequences were obtained (named PM-25 and PM-30). The deduced amino acid sequences of the two scFv-BAD antibodies are shown in Fig. 4. Both of the two antibodies contain 306 amino acids including VL, VH, linker and BAD. Only three amino acids are different between the two antibodies (Fig. 4) and the PM-30 which showed a little higher sensitivity for PM was selected for high-level expression. 3.4. High-level expression of scFv antibody and condition optimization
Fig. 7. Influence of the ionic strength (A), pH (B), concentration of methanol (C) and concentration of skim milk (D) on the biotinylated scFv antibody based IC-ELISA. Each point represents the mean of three replicates.
1.8 108 independent clones was obtained. 3.2. Panning Four rounds of panning operations were performed and the
pET-28a(þ) vector was employed for high-level expression of scFv antibody. The scFv-BAD gene from phasmid of PM-30 was inserted into the pET-28a(þ) and expressed in E. coli BL21 (DE3). Pre-experiment indicated that the scFv-BAD fusion protein could be partially solubly expressed. Hence, the induction conditions including temperature, concentration of IPTG and induced time were optimized. As shown in Fig. 5A, the expression level of scFv antibody reached the highest level at 25 C. The concentrations of IPTG have no significant effect on production of the soluble scFvBAD and 0.1 mM was selected for inducing soluble protein expression (Fig. 5B). For inducing time optimization, the production
H. Wang et al. / Protein Expression and Purification 126 (2016) 1e8
7
Table 2 Cross-reactivity of the scFv antibody for organophosphorus pesticides. No.
Analytes
1
Parathion-methyl
2
Structure
IC50 (ng mL1)
CR (%)
14.5
100.0
Parathion
206.4
7.0
3
Fenitrothion
256.7
5.7
4
Paraoxon-methyl
323.7
4.5
5
Diazinon
>5000
<0.3
6
Fenthion
>5000
<0.3
7
Pirimiphos-methyl
>5000
<0.3
8
Chlorpyrifos
>5000
<0.3
9
Chlorpyrifos-methyl
>5000
<0.3
10
Paraoxon-ethyl
>5000
<0.3
11
Azinphos-methyl
>5000
<0.3
12
Bromophosc
>5000
<0.3
CR (%) ¼ (IC50 of parathion-methyl/IC50 of other compound) 100.
of target protein was increased with the extension of induction time from 2 h to 8 h, but for 10 h induction, the production did not increase significantly. Hence, the optimum induced time was 8 h (Fig. 5C).
3.5. In vitro biotinylation, purification and characterization of scFv antibody BirA was efficiently expressed and confirmed by SDS-PAGE (Fig. S1). The result indicated that the protein was partially soluble and partially in form of inclusion body. The soluble BirA was employed for in vitro biotinylation of scFv antibody. The scFv-BAD antibody was expressed at a high level under the optimum condition. Then the unpurified antibody was biotinylated in vitro by the unpurified BirA. After the biotinylation reaction, the scFv-BAD antibody was purified by the Ni-IDA agarose column chromatography with 6 His-tag at the end of BAD. The purified soluble scFv
antibody was obtained with yield of 59.2 ± 3.7 mg/L of culture (15.8 mg/g of wet weight cell pellet and 10.9% of the total lysate). The purified biotinylated scFv antibody and scFv antibody without biotinylation were compared by SDS-PAGE (Fig. 6A), and the results indicated the molecular mass of scFv antibody without biotinylation scFv antibody was 33 kDa as expected (Fig. 6A, lane 1) and the molecular mass of the biotinylated scFv-BAD was a little bigger because of the addition of biotin molecule (Fig. 6A, lane 2). For the biotinylation of scFv antibody, the amount of BirA enzyme, D-biotin and ATP were excess to make sure the complete of the reaction. The biotinylated scFv-BAD antibody was also confirmed by western blotting using SA-HRP as a detector, and the result evidenced the successful biotinylation (Fig. 6B).
3.6. Optimization of scFv-based IC-ELISA Ionic strengths, pH values and organic solvents always influence
8
H. Wang et al. / Protein Expression and Purification 126 (2016) 1e8
the ELISA performance [22,23]. The ionic strength, pH and methanol concentration of ELISA performance were optimized in this study. The effect of NaCl concentration on ELISA was shown in Fig. 7A and 2.0 M of NaCl which led to the highest Amax/IC50 value was selected as the optimum ionic strength. Similarly, pH 6.0 and 2.5% methanol which reached the highest values of Amax/IC50 were selected as the optimum conditions (Fig. 7B and Fig. 7C). During the developing process of IC-ELISA, we found that PBS containing a certain amount of skim milk (MPBS) could significantly improve the sensitivity of scFv antibody and the optimized concentration of skim milk was determined as 0.25% (Fig. 7D). Consequently, the optimal assay solution was MPBS (pH 6.0) containing 0.25% skim milk, 2.0 M NaCl and 2.5% methanol. The standard sigmoid inhibition curve of PM determined by the optimum ELISA was shown in Fig. 8. The 50% inhibition value (IC50) of PM was 14.5 ng/mL and the limit of detection (IC10, LOD) was 0.9 ng/mL. 3.7. Cross-reactivity study The hapten used in this study was designed to produce high specific antibody for PM. Table 2 showed the IC50 values and CR of PM and 11 OPs which have similar structures with PM. Although most OPs possess the same moieties: O,O-dimethyl phosphorothioate or benzene ring, the scFv antibody only showed low crossreactivity for parathion, fenitrothion and paraoxon-methyl and the CR of the three pesticides was 7.0%, 5.7% and 4.5%, respectively. For the rest pesticides, cross-reactivities were all below 0.3%. The negligible cross-reactivity proved that the scFv antibody was specific to PM. 4. Conclusions Preparation of high quality antibodies is still a bottleneck problem when establishing immunoassay methods for small molecule contaminants. In this study, we developed an efficient method to select and produce large scale of biotinylated scFv antibodies. A sensitive and specific scFv antibody against PM was selected from a pre-immunized phage display library and expressed as soluble protein in high-level. Then the scFv antibody was simply biotinylated in vitro with BirA. Based on the biotinylated scFv antibody, a sensitive IC-ELISA for detecting PM was developed. All these results indicate that the biotinylation of scFv antibody is an efficient strategy in the development of immunoassay. Acknowledgments This research was supported by National Natural Science Foundation of China (grant numbers 30972050, 31271873). Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.pep.2016.05.005.
References [1] A.Y. Kolosovaa, J.H. Parka, S.A. Ereminb, S.J. Parka, S.J. Kanga, W.B. Shima, H.S. Leec, Y.T. Leed, D.H. Chunga, Comparative study of three immunoassays based on monoclonal antibodies for detection of the pesticide parathionmethyl in real samples, Anal. Chim. Acta 511 (2004) 323e331. [2] F.L. Edwards, P.B. Tchounwou, Environmental toxicology and health effects associated with methyl parathion exposureea scientific review, Int. J. Environ. Res. Public Health 2 (2005) 430e441. [3] C. Ferrer, M.J. Martínez-Bueno, A. Lozano, Pesticide residue analysis of fruit juices by LCeMS/MS direct injection, Talanta 83 (2011) 1552e1561. [4] N. Yoshioka, Y. Akiyama, T. Matsuoka, Rapid determination of five postharvest fungicides and metabolite in citrus fruits by liquid chromatography/ time-of-flight mass spectrometry with atmospheric pressure photoionization, Food Control 21 (2010) 212e216. [5] X.D. Xue, Q. Wei, D. Wu, H. Li, Y. Zhang, R. Feng, B. Du, Determination of methyl parathion by a molecularly imprinted sensor based on nitrogen doped graphene sheets, Electrochim. Acta 116 (2014) 366e371. [6] A. Shanker, C. Sood, V. Kumar, S.D. Ravindranath, A modified extraction and clean-up procedure for the detection and determination of parathion-methyl and chlorpyrifos residues in tea, Pest. Manag. Sci. 57 (2001) 458e462. [7] K. Li, Q.X. Li, Development of an enzyme linked immunosorbent assay for the insecticide imidacloprid, J. Agric. Food Chem. 48 (2000) 3378e3382. [8] M.J. Kim, H.S. Lee, D.H. Chung, Y.T. Lee, Synthesis of haptens of organophosphorus pesticides and development of enzyme-linked immunosorbent assays for parathion-methyl, Anal. Chim. Acta 493 (2003) 47e62. [9] A.Y. Kolosova, J.H. Park, S.A. Eremin, Comparative study of three immunoassays based on monoclonal antibodies for detection of the pesticide parathionmethyl in real samples, Anal. Chim. Acta 511 (2004) 323e331. [10] M.J. Moreno, E. Plana, J.J. Manclús, A. Montoya, Comparative study of monoclonal and recombinant antibody-based immunoassays for fungicide analysis in fruit juices, Food Anal. Method 7 (2014) 481e489. [11] C.P. Chan, Y.C. Cheung, R. Renneberg, M. Seydack, New trends in immunoassays, Adv. Biochem. Eng. Biotechnol. 109 (2008) 123e154. [12] K.Y. Yau, H. Lee, J.C. Hall, Emerging trends in the synthesis and improvement of hapten-specific recombinant antibodies, Biotechnol. Adv. 21 (2003) 599e637. [13] V. Markus, L. Janne, L. Urpo, Directed antibody-engineering techniques and their applications in food immunoassays, TrAC-Trend. Anal. Chem. 30 (2011) 219e226. [14] J. He, Y. Liang, M.T. Fan, Preparation of Anti-idiotype antibodies of O,ODimethyl organophosphorus pesticides by phage display technology, Chin. J. Anal. Chem. 39 (2011) 178e182. [15] B. Lauer, I. Ottleben, H.J. Jacobsen, T. Reinard, Production of a single-chain variable fragment antibody against fumonisin B1, J. Agric. Food Chem. 53 (2005) 899e904. [16] H. Wang, X.X. Liu, Y.S. He, Expression and purification of an anti-clenbuterol single chain Fv antibody in Escherichia coli, Protein Expr. Purif. 72 (2010) 26e31. [17] S.M. Cloutier, S. Couty, A. Terskikh, Streptabody, a high avidity molecule made by tetramerization of in vivo biotinylated, phage display-selected scFv fragments on streptavidin, Mol. Immunol. 37 (2000) 1067e1077. [18] Y. Li, R. Sousa, Expression and purification of E. coli BirA biotin ligase for in vitro biotinylation, Protein Expr. Purif. 82 (2012) 162e167. [19] D.P. McAdam, A.S. Hill, H.L. Beasley, J.H. Skerritt, Mono-and polyclonal antibodies to the organophosphate fenitrothion. 1. Approaches to hapten-protein conjugation, J. Agric. Food Chem. 40 (1992) 1466e1470. [20] A. Krebber, S. Bornhauser, J. Burmester, Reliable cloning of functional antibody variable domains from hybridomas and spleen cell repertoires employing a reengineered phage display system, J. Immunol. Method 201 (1997) 35e55. Gonz [21] E. Garet, A.G. Cabado, J.M. Vieites, A. alez-Fern andez, Rapid isolation of single-chain antibodies by phage display technology directed against one of the most potent marine toxins: palytoxin, Toxicon 55 (2010) 1519e1526. [22] X. Jiang, H. Shi, N. Wu, M. Wang, Development of an enzyme-linked immunosorbent assay for diniconazole in agricultural samples, Food Chem. 125 (2011) 1385e1389. [23] Y. Cui, K. Liu, C. Xu, F. Liu, Q.X. Li, S. Liu, B. Wang, Development of a sensitive monoclonal antibody-based indirect competitive enzyme-linked immunosorbent assay for analyzing chlorantraniliprole residues, Food Chem. 143 (2014) 293e299.