Article pubs.acs.org/ac
Nanobody-Based Enzyme Immunoassay for Aflatoxin in AgroProducts with High Tolerance to Cosolvent Methanol Ting He,†,‡,∥,⊗ Yanru Wang,†,‡,∥,⊗ Peiwu Li,*,†,‡,§,∥,⊥,⊗ Qi Zhang,*,†,§,∥ Jiawen Lei,†,‡,∥ Zhaowei Zhang,†,‡,§,∥ Xiaoxia Ding,†,‡,∥,⊥ Haiyan Zhou,†,∥,⊥ and Wen Zhang†,‡,⊥ †
Oil Crops Research Institute of the Chinese Academy of Agricultural Sciences, Wuhan 430062, People’s Republic of China Key Laboratory of Biology and Genetic Improvement of Oil Crops, Ministry of Agriculture, Wuhan 430062, People’s Republic of China § Key Laboratory of Detection for Mycotoxins, Ministry of Agriculture, Wuhan 430062, People’s Republic of China ∥ Laboratory of Risk Assessment for Oilseeds Products, Wuhan, Ministry of Agriculture, Wuhan 430062, People’s Republic of China ⊥ Quality Inspection and Test Center for Oilseeds Products, Ministry of Agriculture, Wuhan 430062, People’s Republic of China ‡
ABSTRACT: A phage-displayed library of variable domain of heavy chain of the heavy chain antibody (VHH) or nanobody (Nb) was constructed after immunizing an alpaca with aflatoxin B1 (AFB1) conjugated with bovine serum albumin (AFB1−BSA). Two AFB1-specific nanobodies were selected. The obtained nanobodies were compared to an aflatoxinspecific monoclonal antibody B5 with respect to stability under organic solvents and high temperature. The two nanobodies could bind antigen specifically after exposure to temperatures as high as 95 °C. Besides, the nanobodies showed better or similar tolerance to organic solvents. A competitive ELISA with nanobody Nb26 was developed for the analysis of AFB1, exhibiting an IC50 value of 0.754 ng/mL (2.4 μM), linear range from 0.117 to 5.676 ng/mL. Due to the high tolerance to methanol, sample extracts were analyzed by nanobody-based ELISA without dilution. The recovery from spiked peanut, rice, corn and feedstuff ranged from 80 to 115%. In conclusion, the isolated nanobodies are excellent candidates for immunoassay application in aflatoxin determination. flatoxins were first discovered in 1960s due to the outbreak of “turkey X” disease in the UK.1 Aspergillus flavus was found in the groundnuts meal imported from Brazil and its metabolites were proved to be the cause of the disease.2 The aflatoxins are a group of naturally occurring mycotoxin, produced mainly by A. flavus and Aspergillus paraciticus. They can occur in a wide range of products, including grains, food and feedstuff. More than 20 types of aflatoxins have been identified, among which, four aflatoxins (B1, B2, G1 and G2) occur naturally. Aflatoxin B1 is the most toxic, and one of the most potent carcinogens in nature. It has been classified as a group I carcinogen, which is carcinogenic to humans, by the International Agency for Research on Cancer (IARC) of the World Health Organization (WHO) in 1993.3 Aflatoxin M1 (AFM1) is the hydroxylated metabolite of AFB1. It can be found in mammals’ urine and milk if contaminated food or feed were ingested.4 AFM1 contaminants can occur in commercial milk and milk products, thus becoming potential health hazards to humans. Due to the high toxicity and carcinogenicity of aflatoxins, many countries have set up strict maximum permissible limits for AFB1 or total aflatoxins (aflatoxin B1, B2, G1 and G2) in food and agri-products, varying from 2 to 20 μg kg−1.5 Accurate
A
© XXXX American Chemical Society
and sensitive determination of aflatoxins became an important requirement to meet food safety concerns and official legislation. A wide range of analytical methods are available for aflatoxin determination, ranging from accurate analysis (e.g., high performance liquid chromatography tandem mass spectrometry6) to rapid methods (e.g., imunochromatographic assay7). Aflatoxin-specific antibodies are required not only for screening methods like immunoassays8 but also for instrumental detection with immunoaffinity purification,9 which is a standard cleanup procedure. In the past decades, numerous antibodies against aflatoxins have been developed by polyclonal10 and monoclonal11 techniques and applied in various assays. With the development of molecular engineering and surface display technology, antibody production has moved from conventional antibodies to recombinant antibodies. Aflatoxin-specific recombinant antibodies such as Fab and ScFv antibody have been developed and utilized to ELISAs and biosensors.12−14 However, the application of this technique turns out to be problematic due to Received: June 29, 2014 Accepted: July 31, 2014
A
dx.doi.org/10.1021/ac502390c | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
QIAquick Gel Extraction Kit and QIAquick PCR Purification Kit were all from Qiagen. LeukoLOCK total RNA isolation system was obtained from Applied Biosystems (Foster City, CA). The Costar 96-well EIA/RIA plate was purchased from Corning Incorporated (Corning, NY, USA). 0.01 M phosphatebuffered saline (PBS, pH 7.4) was prepared by adding 8 g of NaCl, 2.9 g of Na2HPO4·12H2O, 0.2 g of KH2PO4 and 0.2 g of KCl in 1000 mL deionized water. pComb3X phagemid vector was a generous gift from Dr. Carlos F. Barbas (The Scripps Research Institute, La Jolla, CA). Safety. Pure aflatoxin standards were handled in a hood with extreme caution. All items coming in contact with aflatoxins, phage and bacterial cultures (glassware, vials, tubes, ELISA plates, etc.) were immersed in a 10% bleach solution for 1−2 h before they were discarded or autoclaved. Alpaca Immunization. A 2-year-old male alpaca was immunized subcutaneously with 200 μg AFB1−BSA with same volume of Freund’s incomplete Adjuvant. The following injections were performed every 2 weeks in a 10 week period. Preimmune serum was collected as a negative control. Ten milliliters of blood was collected 1 week after the third, fourth, fifth and sixth injection to test titer and isolate leukocytes. Phage-displayed Library Construction. According to the titer results, the fifth blood was selected to extract RNA and construct the library. The library was constructed according to Kim et al.25 In brief, RNA was extracted from blood and transcripted to cDNA. Nanobody gene fragments from IgG2 and IgG3 were amplified by PCR using two pairs of primers containing two different SfiI sites (underlined), forward primer Vhh-f (CAT GCC ATG ACT GTG GCC CAG GCG GCC CAG KTG CAG CTC GTG GAG TC) targeting the framework 1 region, reverse primers Vhh-r1 (CAT GCC ATG ACT CGC GGC CGG CCT GGC CGT CTT GTG GTT TTG GTG TCT TGG G) and Vhh-r2 (CAT GCC ATG ACT CGC GGC CGG CCT GGC CAT GGG GGT CTT CGC TGT GGT GCG) corresponding to the IgG2 and IgG3 hinge region, respectively. The PCR products and pComb3x phagemid vector were separately digested with SfiI and subsequently ligated to generate pComb3x/nanobody constructs. After transformation into E. coli ER 2738, the library size was estimated by plating on LB ampicillin agar plates. Twenty clones were selected from the LB-amp plate and sequenced to evaluate library diversity. Library Panning. The library was subjected to four rounds of panning on a 96-well microtiter plate. Antigen AFB1−BSA was coated as 1, 0.25, 0.1 and 0.05 μg per well for round 1, 2, 3 and 4, respectively. Three replicates were used in each round of panning. To eliminate nonspecific binding, three replicates of wells were coated by 3% BSA for preabsorption. 150 microliters of library was added to each well coated by BSA and incubated for 1 h at room temperature (RT). Then the supernatant was transferred to the wells coated with antigen and incubated for 2 h at RT, followed by 10-time washing with 0.05% PBST. Aflatoxin standard was used to competitively elute bound phages. 50, 10, 2 and 1 ng of aflatoxin in 100 μL of 10% methanol−PBS was used respectively as elution buffer for each round of panning. The elution was performed under gentle shaking at RT for 30 min. Two microliters of the elution solution in each round was used to test the phage titer, which was called output titer and the remainder was amplified for the following panning. The titer of the reamplified phages used for panning was also tested, which was termed input titer.
the low expression yield and stability of these antibody fragments.15 Camelidae produce a unique subclass of antibodies that naturally lack light chains, referred to as heavy chain antibody.16 The antigen binding site of the heavy chain antibody is formed solely by the variable domain of the heavy chain (VHH).17 Recombinant expression of these VHHs could produce single domain heavy chain antibodies, also termed as VHH antibodies or nanobodies.18 The development of nanobodies is much easier compared with conventional recombinant antibodies. For example, a relative small library is enough to achieve antigen-specific binders, only one pair of primers is sufficient to amplify the entire nanobody genes and the expression yield is higher, up to 30 mg L−1 in medium.19,20 Nanobody fragments have hydrophilic residue substitutions in the framework-two region and thus are more soluble than conventional antibody fragments. Due to their single domain nature, nanobodies are also turn out to be more stable over high temperatures and organic solvents. With these benefits, nanobodies are promising reagents in the next generation of immunoassays. Because of its favorable properties, nanobody technology has been widely used in diagnostic and therapeutic areas. An increasing number of nanobodies have been isolated against small molecules and applied in immunoassay.21−27 However, there is no report on aflatoxin-specific nanobodies, except one Chinese publication evaluating the efficiency of two elution methods in biopanning of a nonimmune phage displayed VHH library toward AFB1−BSA, and another report about antiidiotypic nanobodies of aflatoxins isolated from an antiaflatoxin monoclonal antibody immunized VHH library, these nanobodies recognize the antigenic determinants of the monoclonal antibody and can be used as surrogates for aflatoxin haptens as coating antigens.28 Immune libraries lead more directly to nanobodies with higher affinity. In the present study, our goal is to isolate nanobodies against aflatoxins from an immune phage displayed library. We are also interested in evaluating its physical chemical properties and comparing it with a monoclonal antibody, such as thermostability, and solvent effect.
■
EXPERIMENTAL SECTION Materials and Reagents. All reagents were of analytical grade unless otherwise specified. Antiaflatoxin monoclonal antibody (mAb) B5 was produced in our laboratory. Aflatoxin B1, B2, G1, G2, M1 standard, bovine serum albumin (BSA), polyethylene glycol 8000 (PEG 8000), Freund’s incomplete adjuvant, 3,3′,5,5′-tetramethylbenzidine (TMB) and goat antimouse monoclonal antibody conjugated to horseradish peroxidase (HRP) were obtained from Sigma (St. Louis, MO, USA). Escherichia coli ER2738 competent cells from the ER2736 line of E. coli were purchased from Lucigen Corp. (Middleton, WI, USA), Top 10F′ competent cells was purchased from Life Technologies (Grand Island, NY). Mouse anti-M13 monoclonal antibody conjugated to horseradish peroxidase (HRP) was purchased from GE Healthcare (Piscataway, NJ, USA). HRP conjugated anti HA-tag mouse monoclonal antibody was purchased from ComWin Biotech (Beijing, China). Helper phage M13KO7 and SfiI were obtained from New England Biolabs (Ipswich, MA, USA). Tween 20 was obtained from J & K Scientific (Beijing, China). xTractor buffer for protein extraction and His60 Superflow Resin were purchased from Clontech Laboratories, Inc. (Mountain View, CA, USA). QIAprep Spin MiniPrep Kit, B
dx.doi.org/10.1021/ac502390c | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
Table 1. Panning Schedule of Phage-displayed Nanobody round of panning
coating antigen (AFB1−BSA)
AFB1 concentration (ng/mL)
1st 2nd 3rd 4th
1 μg/well 0.25 μg/well 0.1 μg/well 0.05 μg/well
500 100 20 10
input titer (cfu/mL) 2 2.2 3.8 4
× × × ×
1013 1013 1013 1013
output titer (cfu/mL) 1 4.6 3 2
× × × ×
107 107 108 108
Figure 1. Alignment of amino acid sequences of selected nanobodies Nb26 and Nb28. Only amino acid residues different from the top one (Nb26) at the same position have been indicated. Dots are used to indicate residues identify to the top sequence. Solid-line boxes outline the characteristic amino acid substitutions of nanobody FR2.
After four rounds of panning, 30 clones were selected from the fourth output titer plate and amplified each in 3 mL of SB culture. After overnight cultivation, the culture was centrifuged at 3000g for 20 min and the supernatant was further characterized by phage enzyme-linked immunosorbent assay (ELISA). Clones binding to AFB1−BSA, but not to BSA, were deemed positive and selected for further characterization. Expression and Purification of Soluble Nanobodies. Phagemids from the unique positive clones were transformed into and expressed in nonsuppressor E. coli strain TOP10F′ cells. For expression, 100 mL of SB medium was incubated with an overnight culture of TOP10F′ cell carrying nanobody expression plasmid and incubated at 37 °C with shaking at 250 rpm. When the culture reached an OD600 value of 0.6−0.8, 1 mM isopropyl β-D-1-thiogalactopyranoside (IPTG) was added, followed by continuous shaking overnight. Proteins were isolated using xTractor buffer following manufacturer’s instructions. Nanobodies containing 6xHis tag were purified with Ni−NTA metal affinity chromatography according to manufacturer’s instruction. The purity and size of nanobodies were assessed using 15% reducing SDS-PAGE according to a standard protocol, followed by staining with Coomassie Brilliant Blue. The concentration of nanobodies was determined by the Bradford method. Sample Preparation. Five grams of peanut, rice, corn and feedstuff was weighed respectively and extracted using 70% methanol/PBS containing 2% BSA under ultrasound for 5 min. Then the mixture was filtered twice and centrifuged at 5000g for 10 min; the supernatant was used for sample analysis directly.
Figure 2. Competitive binding curves with specific nanobody clones. Representative curves showing the inhibition of the nanobodies Nb26 (■) and Nb28 (○) binding to AFB1−BSA with increasing amounts of free AFB1. Each point represents the average of three replicates. The concentrations of the coating antigen-purified nanobodies used in the assays were 1−0.1 and 1−0.02 μg/mL for Nb26 and Nb28, respectively. The inset shows the SDS-PAGE analysis of purified nanobodies Nb26 and Nb28 on a 15% gel.
trations of the coating antigen and the free AFB1 for competitive elution in each round (Table 1). To minimize nonspecific binding, vigorous washing steps and preabsorption were also performed. From Table 1, the phage output titer was increased in each round, which indicated specific phage clones were enriched during the panning. Thirty clones were selected randomly from the fourth output titer plate and tested by phage ELISA. Twenty-eight clones showed an inhibition of binding to the coating antigen by free AFB1. The plasmids of positive clones were extracted and the sequencing results showed two unique clones, named Nb26 and Nb28, were selected. From Figure 1, the two colonies are both long-hinge VHHs (IgG2) and the framework regions are highly conserved. Both contain characteristic substitutions in FR2, including F, E, R and F/G in positions 37, 44, 45, and 47, which explains the solubility of
■
RESULTS AND DISCUSSION Selection of Aflatoxin-specific Phage Clones. Ten milliliters of blood was collected after the fifth immunization and used for RNA extraction. Nanobody genes were amplified by PCR and ligated with pComb3X vector. The whole plasmid was then electroporated to E. coli ER 2738 cells for amplification. The constructed phage displayed nanobody library was estimated to have 2.1 × 107 independent colonies. To obtain high specific antiaflatoxin phage clones, we utilized the panning strategy by increasing selection pressure. Four cycles of panning were carried out with decreasing concenC
dx.doi.org/10.1021/ac502390c | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
Figure 3. Cross-reactivity of nanobodies Nb26 (A) and Nb28 (B) based ELISA toward aflatoxin B1 (■), B2 (○), G1 (▲), G2 (▽) and M1 (●). Each result is represented as an average ± standard deviation of three replicates. The 96-well plates were coated with an AFB1 hapten−BSA conjugate at 1 μg/mL. Serial dilutions of AFB1 or tested compounds in PBS were mixed with an equal amounts of nanobodies. A volume of 100 μL of the mixture was added into the wells. The bound nanobody was detected with a 1/10 000 dilution of HRP conjugated anti HA-tag mAb.
regions. This might suggest that the two antibodies are derived from the same B cell and bind aflatoxin in a similar way. Expression and Purification of Soluble Nanobodies. For expression, the two unique nanobody plasmids were transformed into Top10F′, which is a nonsuppressor E. coli strain, allowing soluble nanobody expression without pIII protein. The nanobody expressed with 6×His tag was obtained by purification on a Ni−NTA affinity column. The purity of nanobodies was detected on a 15% SDS-PAGE gel. From Figure 2, the size of the two nanobodies is about 17 kDa, which is consistent with the theoretical value calculated by ProtParam (Expasy). The concentrations of two nanobodies were determined by the Bradford method. Nanobody Nb26 had a yield of 5.2 ± 0.2 mg/L bacterial culture, and Nb28 was 4.2 ± 0.6 mg/L (purification yields are expressed as means ± standard deviation with three independent experiments).
Table 2. Cross-Reactivity (CR) of the AFT Compounds Competitive ELISA Setup with Nanobodies Nb26 and Nb28 Nb26
Nb28
aflatoxins
IC50 (ng/mL)
cross-reactivity (%)a
IC50 (ng/mL)
cross-reactivity (%)
B1 B2 G1 G2 M1
0.762 78.83 NDb 40.45 7.094
100 0.97
1.064 12.971 28.708 11.596 3.255
100 8.21 3.71 9.18 32.7
1.89 10.75
a
Cross-reactivity was calculated as % cross-reactivity = (IC50AFB1/ IC50analyte) × 100. bND, not detectable.
nanobodies.29 The two clones possess the same CDR1 region, and only seven amino acids are unique in the CDR2 and CDR3
Figure 4. Thermostability of AFB1-specific nanobodies Nb26 (●), Nb28 (▲) and mAb B5 (■). Nanobodies and conventional monoclonal antibody were diluted to working concentrations in PBS and were heated at various temperatures, 20, 50, 65, 75 and 95 °C for 5 min (A) or at 85 °C for various times, 0, 5, 15, 25, 35, 45 and 60 min (B). After re-equilibated to RT, their binding to coated AFB1−BSA in microtiter plate was measured. Values are the mean ± standard deviation of three well replicates. D
dx.doi.org/10.1021/ac502390c | Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Article
Figure 5. Effects of MeOH (A), DMSO (B), DMF (C), acetone (D) and acetonitrile (E) on the performance of nanobodies Nb26, Nb28 and mAb B5-based ELISA. PBS buffers containing each organic solvent at different concentrations, 10%, 20%, 40%, 60% and 80%, were used to make serial dilutions of free AFB1. The serial dilutions were mixed with equal volumes of nanobodies Nb26, Nb28 or mAb B5 and then 100 μL of the mixture was added into wells. The bound immunoreagents were detected by adding 100 μL of HRP conjugated anti HA-tag mAb for nanobody (1/10 000 dilution in PBS), and goat antimouse PAb conjugated with HRP for mAb B5 (1/5000 dilution in PBS), respectively. Values are the mean ± standard deviation of three well replicates.
Nanobody ELISA. The purified nanobodies were tested for their performance by indirect competitive ELISA. The concentration of the coating antigen and nanobody were first optimized by checkerboard. From Figure 2, nanobodies Nb26 and Nb28 exhibited a similar property, with an IC50 value of 0.754 ng/mL (2.4 μM) and 1.012 ng/mL (3.24 μM) toward AFB1, respectively. The specificity of the assay was also evaluated using four other aflatoxins (AFB2, AFG1, AFG2 and AFM1) (Figure 3). Insignificant cross-reactivity (
