Anti-idiotypic nanobody-alkaline phosphatase fusion proteins:Development of a one-step competitive enzyme immunoassay forfumonisin B1 detection in cereal
Mei Shu a, b, Yang Xu a, b, *, Xing Liu a, d, Yanping Li b, Qinghua He b, Zhui Tu a, Jinheng Fu b,Shirley J. Gee c, Bruce D. Hammock c
a State Key Laboratory of Food Science and Technology, Nanchang University, No. 235 Nanjing East Road, Nanchang 330047, People's Republic of Chinab Jiangxi-OAI Joint Research Institute, Nanchang University, No. 235 Nanjing East Road, Nanchang 330047, People's Republic of Chinac Department of Entomology and UCD Comprehensive Cancer Center, University of California, Davis, CA 95616, USAd College of Food Science and Technology, Hainan University, No. 58 Renmin Avenue, Haikou 570228, China
h i g h l i g h t s
* Corresponding author. State Key Laboratory of FNanchang University, No. 235 Nanjing East Road, Napublic of China.
A rapid and sensitive one-step competitive enzyme immunoassay for the detection of FB1 was developed.The anti-idiotypic nanobodyealkaline phosphatase (Ab2b�Nb�AP) was validated by the AP enzymeactivity and the properties of bounding to anti-FB1-mAb (3F11) through colorimetric and chem-iluminescence analyses. The 50% inhibitory concentration and the detection limit (LOD) of colorimetricenzyme-linked immunosorbent assay (ELISA) for FB1 were 2.69 and 0.35 ng mL�1, respectively, with alinear range of 0.93e7.73 ng mL�1. The LOD of the chemiluminescence ELISA (CLIA) was 0.12 ng mL�1,and the IC50 was 0.89 ± 0.09 ng mL�1 with a linear range of 0.29e2.68 ng mL�1. Compared with LC-MS/MS, the results of this assay indicated the reliability of the Ab2b�Nb�AP fusion protein based one-stepcompetitive immunoassay for monitoring FB1 contamination in cereals. The Ab2b�Nb�AP fusion pro-teins have the potential to replace chemically-coupled probes in competitive enzyme immunoassaysystems.
ood Science and Technology,nchang 330047, People's Re-
1. Introduction
Fumonisin B1 (FB1) is the most abundant and toxic metabolite offumonisins and is associated with human cancer [1,2]. FB1 has beenclassified by the International Agency for Research on Cancer as apossible group 2B human carcinogen [3]. Concerned about thepotential detrimental health effects, the United States Food and
M. Shu et al. / Analytica Chimica Acta 924 (2016) 53e5954
Drug Administration recommends that the maximum humanconsumption of residue limits of total fumonisins (FB1þFB2þFB3) incorn and corn products is 2 ppme4 ppm [4]. The European Unionhas set the maximum permissible levels for the sum of FB1 and FB2,which vary from 0.2 ppm to 2 ppm [5].
To efficient control of Fumonisin B1 in cereal and cereal prod-ucts, accurate and easily performed analytical methods, includinginstrumental techniques [6,7] and immunoassays [8e10] will berequired. The instrumental techniques are precise and sensitive,but most require sophisticated equipment and complex purifica-tion steps. Immunoassays are sensitive, specific and cost effectiveencourage themselves to point of use tools for FB1 determination.However, most of the previously reported immunoassays for my-cotoxins use secondary antibodies as heterogeneous probes thatare chemically linked with fluorescent [11] or enzymes, such asalkaline phosphatase (AP) [12] and horseradish peroxidase (HRP)[13,14]. The chemical conjugation of fluorenscent or enzymes toantibodies may result in unstable and randomly cross-linkedmolecules, which often reduce assay sensitivity [15]. Therefore, itis almost impossible to prepare homogeneous probes. With therapid development of genetic engineering techniques, antibodiesare fused to reporter proteins as homogeneous probes. These fusionproteins are constructed into a one-step immunoassay that avoidsthe use of secondary antibodies [16,17]. Many studies have reportedthe detection of low-molecular-weight compounds using thefusion of AP to single-chain fragments of the variable antibodyregion (scFv) or Nb, such as 11-deoxycortisol [18], ochratoxin A [19],and tetrabromobisphenol A [20].
In 1993, Hamers-Casterman Discovered sera of camelids containa novel subclass of IgG antibodies that have unique functionalheavy chain antibodies (HCAbs) and completely devoid of lightchains [21]. nanobodies providemany unique features of antibodiesthat make them more useful compared with the conventional an-tibodies, such as ease of expression in various expression, ther-mostability, biophysical and easy of genetic manipulation [22,23].Due to these advantages, many reports about environmental andbiomedical applications of nanobodies are related to small mole-cules [24,25]. Recently, some studies on the detection of mycotoxinusing nanobodies have been reported, such as ochratoxin A andAflatoxin [26,27].
FB1 as small-molecule hapten is not large enough for directdetection and to illicit an immune response, therefore, it requiresconjugation with a large carrier, such as a protein, as antigen con-jugates (coating antigen or competing antigen) [28]. In the syn-thesized procedures, large amounts of FB1 standard and organicsolvents which are expensive and may pose a threat to humanhealth are involved. In order to reduce hazard condition, biologi-cally derived antigen conjugates would be a preferred reagent. Forthe nominal antigen, the anti-idiotypic antibody (Ab2) is the sec-ond antibody specific to idiotopes, which can bind to the paratopeof the primary antibody (Ab1) [29,30]. When the idiotope recog-nized by Ab2 is part of the paratope of Ab1, it is said to be an Ab2b.An important feature of Ab2b is that it can represent an internalimage of antigenic determinants (epitopes) and compete with theoriginal antigen when binding the primary antibody [31]. Thus, theAb2b nanobody works as a substitute to the original antigen andfused to reporter molecule to be used as an immunoassay label.
In our previous work, we isolated an Ab2b nanobody [32] from anaïve alpaca nanobody phage display library. In the present study,we expand a strategy for determining small molecules that areimmunochemically categorized as haptens by employing anAb2b�Nb�AP fusion protein, which is equivalently immunoreac-tive as the target hapten and bio-reporter. The fusion protein isused to develop a rapid, green, and sensitive one-step competitiveenzyme immunoassay for the detection of FB1 in cereal samples.
2. Materials and methods
2.1. Materials and reagents
T4 DNA ligase and restriction enzyme SfiІ were purchased fromNew England Biolabs, Inc. (Beverly, MA). Mycotoxin fumonisin B1(FB1), fumonisin B2 (FB2), deoxynivalenol (DON), ochratoxin A(OTA), zearalenone (ZEA), aflatoxin B1 (AFB1), and p-nitrophenylphosphate (pNPP) substrate were purchased from Sigma (St. Louis,MO). Disodium 3-(4-methoxyspiro {1, 2-dioxetane-3, 20-(50-chloro)-tricyclo [3.3.1.13, 7] decan}-4-yl) phenyl phosphate (CSPD)was purchased from Roche Applied Science (Basel, CH). The vectorpecan45 [33] was a generous gift from Dr. Jinny L. Liu and Dr. EllenR. Goldman (Naval Research Laboratory, Center for Bio/MolecularScience and Engineering, Washington DC).
2.2. Cloning, expression, and purification of Ab2b�Nb�AP fusionproteins
The Ab2b nanobody specific to anti-FB1-mAb was obtainedpreviously [32]. A recombinant plasmid encoded the Nb�AP fusionprotein with a 6X His tag at its C-terminal end, and thepecan�VHH�AP was constructed as shown (Fig. 1A). Positiveclones were sent to Invitrogen (Shanghai, China) for sequencing.The pecan�VHH�AP plasmid was used into Escherichia coli BL21(DE3) plysS cells. A single colony of the transformants was grown,and protein expression was induced in auto-induction media [34].Crude extracts of the cells were lysed by ultrasonic cell disruption,and the crude protein of Ab2b�Nb�AP was centrifuged at 8000 gfor 10 min. The target proteins, containing a 6X His tag were pu-rified by affinity chromatography using Ni�NTA affinity columnsand were analyzed by 12% SDS-PAGE.
2.3. Measurement of the AP activity of the Ab2b�Nb�AP
2.3.1. Colorimetric analysisSerially diluted Ab2b�Nb�AP fusion protein (50 mL) was added
into a 96-well microplate, followed by the addition of 100 mL of APbuffer (pH 10.4) containing 3.8 mM pNPP, 0.01 M glycine, 50 mMMgCl2, and 50 mM ZnCl2. The plate was mixed and incubated at37 �C for 30 min. The reaction was terminated by the addition of2 M NaOH (50 mL). The well absorbance at 405 nm was thenmeasured on a microtiter plate reader.
Science) that contains disodium 3-(4-methoxyspiro {1, 2-dioxetane-3, 20-(50-chloro)-tricyclo [3.3.1.13, 7] decan}-4-yl)phenyl phosphate (CSPD) (150 mL) was added to microwells inwhite microplates, in which purified Ab2b�Nb�AP diluted in PBS(10 mL) was placed. The plate was mixed and incubated at roomtemperature for 10 min. Luminescence at 475 nm was measuredusing the Thermo Fluoroskan Ascent FL (Thermo Scientific).
2.4. One-step competitive immunoassay using the Ab2b�Nb�AP
One-step competitive immunoassays were performed as fol-lows: Anti-FB1 mAb (3F11) (2 mg mL�1, 100 mL/well) was coated inmicroplate overnight at 4 �C and blocked with 3% skim milk in PBS(10 mM, pH 7.4) for 1 h at 37 �C. After washing with PBST (10 mM,pH 7.4, 0.5% Tween 20), a diluted Ab2b�Nb�AP (periplasmicextract) (50 mL/well) and various concentrations of FB1 standard(from 0.1 to 100 ng mL�1, 50 mL/well), both diluted with PBS(10 mM, pH 7.4) were added; the mixture in the plate was incu-bated at 37 �C for 30 min. After washing three times with PBST, the
Fig. 1. (A) Schematic of the construction and expression of Ab2b�Nb�AP. (B) SDS-PAGE analysis of the expression of the Ab2b�Nb�AP fusion protein. The marker resides in lane M;lane 1, whole cell extract under auto-induction; lane 2, the soluble fraction; lane 3, the liquid flow from the Ni�NTA purification; lane 4, the wash buffer from the Ni�NTA pu-rification, lanes 5 and 6, the elution of the Ab2b�Nb�AP fusion protein indicated by a red arrow. (For interpretation of the references to colour in this figure legend, the reader isreferred to the web version of this article.)
M. Shu et al. / Analytica Chimica Acta 924 (2016) 53e59 55
enzyme activity in the wells was determined with the colorimetricand chemiluminometric assays described above.
For comparison, two-step competitive ELISA was also per-formed. Microwells were coated with 100 mL/well, 2 mg mL�1 anti-FB1mAb (3F11) and then incubated overnight at 4 �C. After blockingwith 3% skim milk in PBS and washing with PBST, each serialconcentration of FB1 (from 0.1 to 100 ng mL�1, 50 mL/well) orsample extract equally with Ab2b nanobodies (periplasmic extract)was added to thewells and incubated at 37 �C for 30min. Followingwashing three times with PBST, 1:4000 dilution of HRP-conjugatedhis-antibody (100 mL) was incubated in the wells at 37 �C for30 min. Finally, 3, 30, 5, 50- tetramethylbenzidine (TMB) (100 mL)substrate was added to the wells. The optical density at 450 nmwasdetermined on a microplate reader (Thermo Scientific).
2.5. Sample analysis and validation
The one-step competitive immunoassay for FB1 was validatedthrough the evaluation of cross-reactivity with other mycotoxinsand through the determination of the spiked samples and incurredsamples [35].
2.5.1. Cross-reactivity evaluation for one-step competitiveimmunoassay
The selectivity of the Ab2b�Nb�AP fusion protein based assay
was determined by its cross-reactivity (CR) with a group of myco-toxins (FB2, AFB1, OTA, ZEN and DON) in the range of0.1e1000 ng mL�1. The cross-reactivity was calculated using thefollowing equation: % CR ¼ 100 � IC50 (FB1)/IC50 (cross-reactingcompound).
2.5.2. Sample preparation for one-step competitive immunoassaySamples were obtained from local markets in Nanchang City,
China. For the recovery of FB1, corn samples, which were found tobe less than 1 mg kg�1of FB1 contamination by LC-MS/MS(LOQ ¼ 1 mg kg�1), were spiked with a known concentration ofFB1 from 10 to 1000 mg kg�1. Incurred samples (corn, rice, andfeedstuff) were prepared and extracted as follows: 5 g of finelyground sample was weighed and mixed with 10 mL of 60% meth-anol in water (v/v) and subjected to ultrasonic extraction for20 min. Thereafter, the supernatant was obtained by centrifugation(10,000 g, 4 �C, 10 min) and filtered through a 0.22 mm celluloseacetate membrane. The supernatant was diluted to a final con-centration of 5% methanol, 10 mM PBS (pH 7.4) for immunoassayand for liquid chromatography tandem mass spectrometry(LCeMS/MS) analysis the supernatant was diluted 2X.
For the LC-MS/MSmethod, the analytical columnwas a ZORBAXSB�C18 analytical column (2.1 mm � 50 mm, 1.8 Micron, Agilent,USA). The mobile phase A consisted of 0.1% (v/v) formic acid inwater, and the mobile phase B is methanol. The following gradient
Fig. 2. (A) Enzyme activity of the Ab2b�Nb�AP fusion protein evaluated by colori-metric and chemiluminometric assays. Error bars indicate the standard derivation(n ¼ 3). (B) Binding kinetic measurement for the Ab2b�Nb�AP and anti-FB1 mAb. (C)Binding of the Ab2b�Nb�AP fusion protein (crude extract) to the 3F11 antibody coatedon the microplates in the absence and presence of 50 ng FB1 by colorimetric assay.
M. Shu et al. / Analytica Chimica Acta 924 (2016) 53e5956
program was applied briefly: 0e1 min, 35% B; 1e3 min, 60% B;5e8 min, 75% B; 8e9 min, 90% B; 9e13 min, 90% B; 13e13.1 min,35% B, and equilibrated for 2.9 min. The flow rate was0.25 mL min�1, and the injection volume was 10 mL [36]. The ef-fluents of the liquid chromatograph was interfaced to a QQQ (6430Triple Quadrupole, Agilent, USA) mass spectrometer equipped withan electrochemical electrospray ionization source (ESI) and oper-ated in positive mode [37]. The following instrument parameterswere used as follows: capillary voltage: 3500 V, gas temperature:325 �C, nebulizer gas pressure: 30 pound per square inch gauge(psig), gas flow: 10 Lmin�1, fragmentor voltage: 135 V, and collisionenergy: 25 V. FB1 was quantitatively analyzed in the multiple re-action monitoring (MRM) mode, in which two correspondingproduct ions (333.9 and 352.3) were selected as precursor ions forFB1 (722.3) [38]. Data were gathered on Agilent MassHunterWorkstation Data Acquisition software and quantified with theQQQ quantitative analysis software [39].
3. Results and discussion
3.1. Expression, purification, and characterization of theAb2b�Nb�AP fusion protein
The positive recombinant plasmid confirmed was transformedinto E. coli BL21 (DE3) plysS cells. The target proteins were culturedin an auto-induction medium at 30 �C for 12 h and were extractedthrough the ultrasonic cell disruption method. The Ab2b�Nb�APwith a 6X His tag at its C-terminal end was purified by the Ni�NTAaffinity columns. Protein size and integrity were verified by 12%SDSePAGE (Fig. 1B). The gels of selected purified Ab2b�Nb�APfusions protein showed the expected band of approximately 65 kDafor the 1:1 fusion of Ab2b�Nb and AP.
3.2. Enzyme activity and anti-idiotype reactivity of Ab2b�Nb�AP
The enzyme activity of Ab2b�Nb�AP was assessed with color-imetric and chemiluminometric assays (Fig. 2A). As expected, thechemiluminometric assay allowed for more sensitive detectionthan the colorimetric assay. The limit of detections (LOD) of theseassays, defined as the amount (per assay) that provided signalswith two standard deviations (SD) greater than the average (n¼ 10)of the signal at zero concentration, were 72.4 amol for the colori-metric assay and 18.1 amol for the chemiluminometric assay. Thelowest amount of Ab2b�Nb�AP detected by the chemilumino-metric assay was approximately four times lower than that bycolorimetric analysis, indicating the greater sensitivity of thechemiluminometric assay than the colorimetric assay. Binding ki-netics for Ab2b�Nb�AP was examined using a label-free biosensor(Thermo Scientific, USA). The binding kinetics for Ab2b�Nb�APhad ka ¼ 6.57E þ 4/Ms and kd ¼ 4.31E � 3/s. The equilibriumdissociation constant (KD ¼ kd/ka) measured for Ab2b�Nb�AP:anti-FB1 mAb was 65.5 mM, which was twice lower than thatmeasured for Ab2b Nb: anti-FB1 mAb (KD ¼ 164.6 nM) [32], rep-resenting a 2-fold increase in affinity for the Ab2b Nb. TheAb2b�Nb�AP fusion protein was bound to the relevant primaryanti-FB1 mAb (3F11), and this binding was inhibited by addition ofFB1, which was the original antigen for 3F11 (Fig. 2C). Reports showthat the fusion of AP�Nb can broaden the application of enhancingthe binding affinity compared to that of unfused Nb [40]. Thus,Ab2b�Nb�AP retained high binding activity to the 3F11 paratope.
3.3. Immunoassay for FB1 using Ab2b�Nb�AP as probe
One-step competitive immunoassay was performed withcolorimetric and chemiluminometric detection to optimize the
concentrations of the anti-FB1 mAb (2 mg mL�1 in 10 mM PBS pH7.4) and Ab2b�Nb�AP (1 mgmL�1 in 10mM PBS pH 7.4) in advance.The bound Ab2b�Nb�AP was detected with a common colori-metric method as described above (Fig. 3A). The IC50 of the assaywas 2.69 ± 0.25 ng mL�1. The LOD of the immunoassay wascalculated to be 0.35 ng mL�1, and the linear range, calculated by20%e80% inhibition of Nb�ELISA, was 0.93e7.73 ng mL�1. Forcomparison, one-step competitive chemiluminometric ELISA (CLIA)was also performedwith chemiluminometric detection. The LOD ofthe assay was 0.12 ng mL�1, and the IC50 was 0.89 ± 0.09 ng mL�1
with a linear range of 0.29e2.68 ng mL�1. An obvious improvementin assay sensitivity was observed in the CLIA. A two-step ELISA wasalso performed using the secondary antibody under optimal
Fig. 3. (A) Standard one-step immunoassay (ELISA and CLIA) and two-step immuno-assay. (B) Cross reactivity of one-step CLIA with other common mycotoxins. Error barsindicate the standard derivation (n ¼ 3).
M. Shu et al. / Analytica Chimica Acta 924 (2016) 53e59 57
conditions. The sensitivity (IC50) of the one-step competitive CLIAwas improved by nine times compared to the two-step competitiveELISA (IC50 ¼ 7.22 ± 0.52 ng mL�1). The cross reactivity (CR) againstFB2, AFB1, ZEN, DON, and OTA was tested, and the results demon-strated CR toward FB2 (5.11%) and negligible CR toward other my-cotoxins (Fig. 3B).
Fig. 4. Effects of ionic strength (A), pH (B), and MeOH (C) on the performance of one-step CLIA. Error bars indicate the standard derivation (n ¼ 3).
3.4. Optimizing the one-step competitive CLIA
Reports show that pH value, ionic strength, and water misciblesolvents can greatly influence immune-reactions [41]. The differentconcentrations of buffer ionic strength was investigated from 0.5Xto 5X PBS buffer (pH 7.4) (Fig. 4A). According to the result, the IC50
values were almost equal in 0.5X and 1X PBS buffer, but the value ofoptical intensity significantly decreased when ionic strength valueswere 2.5X and 5X PBS buffer. The influence of the assay buffer withthe pH values on the one-step competitive CLIAwas evaluated to bebetween 5.0 and 9.0 (Fig. 4B). The IC50 and optical intensity valueswere almost equal at pH 5.0 to pH 8.0, whereas the lowest IC50 andlowest optical intensity values were observed at pH 9.0. Methanol iscommonly used to extract fumonisins from samples and can in-fluence the interaction between antigen and antibody. In moni-toring the methanol effect, the different concentrations of 2.5%, 5%,10%, and 20% methanol�PBS were examined in terms of
performance. There were no significant difference between the IC50obtained from 2.5% to 5%; however, 20% methanol interfered withthe shape of inhibition curves strongly (Fig. 4C). 5% methanol waschosen because of less dilution would give higher final concentra-tion for the detection and appropriate the maximum absorbance.So the best performance buffer containing a final concentration of5% methanol and 10 mM PBS in pH 7.4 was chosen for furtheroptimization.
3.5. Validation study
The samples were validated by analyzing the spiked corn sam-ples using Ab2b�Nb�AP based one-step competitive ELISA and
Table 1Recoveries of FB1 added to corn samples in determinations performed by the one-step competitive immunoassay.
FB1 added (mg/kg) ELISAa (mg/kg) Average recovery (%) CV (%) CLIAb (mg/kg) Average recovery (%) CV (%)
a One-step competitive ELISA.b One-step competitive CLIA.c Each assay was conducted in triplicates on the same day.d Each assay was performed on five different days.
Table 2Detection results of the incurred samples by the one-step competitive immunoassay and LC-MS/MS.
a One-step competitive ELISA.b One-step competitive CLIA.c Each assay was conducted in triplicates on the same day.d Not detectable.
M. Shu et al. / Analytica Chimica Acta 924 (2016) 53e5958
CLIA. These samples spiked with a series of known concentrationsof FB1 (10e1000 mg kg�1) exhibited recovery rates from 71.20% to114.03%, with relative standard deviations (RSD) rates from 2.43% to12.50% by one-step competitive ELISA and recovery of 83.50%e112.68% with RSD of 2.47%e14.86% by one-step competitive CLIA(Table 1). These results indicated that reliable accuracy and repro-ducibility of Ab2b�Nb�AP based one�step competitive immuno-assay in analyzing samples.
In addition, a total of 40 samples of cereals were blindlyanalyzed using Ab2b�Nb�AP based one-step competitive immu-noassay and LC-MS/MS. The one-step competitive immunoassaycorrelated well with the LC-MS/MS method (R2 ¼ 0.97). Among the40 samples, 15 of 20 corn samples and 8 of 15 feedstuffs were FB1positive, and 5 rice samples were all FB1 negative, as detected bythe one-step immunoassay (Table 2). The contents of FB1 in the fourcorn samples were beyond the maximum limit for FB1 in cereal
(1000 mg kg�1). The results demonstrated that using Ab2b�Nb�APbased one-step competitive immunoassay is an acceptable methodfor the detection of FB1 in cereal samples.
4. Conclusion
In this study, a rapid, green and sensitive one-step competitiveenzyme immunoassay (ELISA and CLIA) for detecting FB1 wasdeveloped using an Ab2b�Nb�AP, an anti-idiotypic moleculederived from a naïve camelid library and fused to a enzyme-labelled antigen. The Ab2b�Nb�AP fusion protein is a novel,“clonable” homogeneous probe suitable for various purposes in FB1analysis. Furthermore, through the integration of the one-stepcompetitive immunoassay, rapid and green mycotoxin detectionwith reproducibility, enhanced sensitivity, reduced analysis time,lower reagent volumes, avoiding synthesizing competing antigen
M. Shu et al. / Analytica Chimica Acta 924 (2016) 53e59 59
and chemically-coupled probes can be achieved. The sensitivities(IC50) of the one-step competitive ELISA and CLIAwere improved bythree and nine times, respectively, compared with that of the two-step competitive ELISA. These results indicate that the developedone-step competitive immunoassay would be a tool for small-molecule immunoassays.
Author contributions
The paper was written through contributions of all authors. M.Shu completed major part of the experiments and wrote the paper.X. Liu, Q. He, Z. Tu and J. Fu completed the else part of the experi-ments. Y. Xu and Y. Li supported ideas and funding. S.J. Gee and B.DHammock supported plasmid pecan45 and technique guidance atnanobody paining. All the authors have given approval to the dis-cussion and revision of the paper.
Acknowledgments
This study was financially supported by the National BasicResearch Program of China (Grant No. 2013CB127804), the MajorProgram of Natural Science Foundation of Jiangxi, China (Grant No.20152ACB20005), the National Natural Science Foundation of China(Grant Nos. NSFC-31471648, NSFC-31360386, NSFC-31201360, andNSFC-31171696), and the Education Department of Jiangxi Province(Grant No. GJJ13095).
References
[1] W.F. Marasas, in: L.S. Jackson, J.M. DeVries (Eds.), Fumonisins in Food, SpringerScience and Business Media, New York, 1996, pp. 1e17.
[2] S. Marín, N. Magan, A.J. Ramos, V. Sanchis, Fumonisin-producing strains ofFusarium: a review of their ecophysiology, J. Food Prot. 67 (2004) 1792e1805.
[3] International Agency for Research on Cancer, IARC Monographs on the Eval-uation of Carcinogenic Risks to Humans, vol. 82, IARC, Lyon, 2002, pp.301e345.
[4] US Food and Drug Administration (FDA), Draft guidance for industry: Fumo-nisin levels in human food and animal feeds; final guidance. (http://www.fda.gov/food/guidanceregulation/guidancedocumentsregulatoryinformation/ucm109231.htm).
[5] European Commission (EC), Commission Regulation No. 1126/2007 of 28September 2007, Off. J. EC L225 (2007) 14e17.
[6] E.N. Ediage, J.D. Di Mavungu, S. Song, A. Wu, C. Van Peteghem, S. De Saeger,A direct assessment of mycotoxin biomarkers in human urine samples byliquid chromatography tandem mass spectrometry, Anal. Chim. Acta 741(2012) 58e69.
[7] B. Warth, S. Michael, K. Rudolf, LC-MS/MS-based multibiomarker approachesfor the assessment of human exposure to mycotoxins, Anal. Bioanal. Chem.405 (2013) 5687e5695.
[8] S. Ling, J. Pang, J. Yu, R. Wang, L. Liu, Y. Ma, S. Wang, Preparation and iden-tification of monoclonal antibody against fumonisin B1 and development ofdetection by Ic-ELISA, Toxicon 80 (2014) 64e72.
[9] Z. Hu, H. Li, P. Wu, Y. Li, Z. Zhou, J. Zhang, Y. Liao, An affinity improved single-chain antibody from phage display of a library derived from monoclonal an-tibodies detects fumonisins by immunoassay, Anal. Chim. Acta 867 (2015)74e82.
[10] X. Liu, Y. Xu, Q. He, Z. He, Z. Xiong, Application of mimotope peptides offumonisin B1 in Peptide ELISA, J. Agric. Food Chem. 61 (2013) 4765e4770.
[11] M. Ren, H. Xu, X. Huang, M. Kuang, Y. Xiong, H. Xu, A. Wang, Immunochro-matographic assay for ultrasensitive detection of aflatoxin B1 in maize byhighly luminescent quantum dot beads, Appl. Mater. Inter 6 (2014)14215e14222.
[12] N.A. Burmistrova, T.Y. Rusanova, N.A. Yurasov, I.Y. Goryacheva, S. De Saeger,Multi-detection of mycotoxins by membrane based flow-through immuno-assay, Food control. 46 (2014) 462e469.
[13] Y. Qiu, Q. He, Y. Xu, A.K. Bhunia, Z. Tu, B. Chen, Y.Y. Liu, Deoxynivalenol-mimicnanobody isolated from a naïve phage display nanobody library and itsapplication in immunoassay, Anal. Chim. Acta 887 (2015) 201e208.
[14] Y. Xu, L. Xiong, Y. Li, Y. Xiong, Z. Tu, J. Fu, X. Tang, Citrinin detection usingphage-displayed anti-idiotypic single-domain antibody for antigen mimicry,Food Chem. 177 (2015) 97e101.
[15] S. Daunert, B.R. Payne, L.G. Bachas, Pyruvate carboxylase as a model for oli-gosubstituted enzyme-ligand conjugates in homogeneous enzyme immuno-assays, Anal. Chem. 61 (1989) 2160e2164.
[16] L. Zou, Y. Xu, Y. Li, Q. He, B. Chen, D. Wang, Development of a single-chainvariable fragment antibody-based enzyme-linked immunosorbent assay for
determination of fumonisin B1 in corn samples, J. Sci. Food Agric. 94 (2014)1865e1871.
[17] Z. Xu, J. Dong, H. Wang, Z. Li, R. Beier, Y.M. Jiang, Y.M. Sun, Production andcharacterization of a single-chain variable fragment linked alkaline phos-phatase fusion protein for detection of O, O-diethyl organophosphorus pes-ticides in a one-step enzyme-linked immunosorbent assay, J. Agric. FoodChem. 60 (2012) 5076e5083.
[18] H. Oyama, E. Tanaka, T. Kawanaka, I. Morita, T. Niwa, N. Kobayashi, Anti-id-iotype scFveenzyme fusion proteins: a clonable analyte-mimicking probe forstandardized immunoassays targeting small biomarkers, Anal. Chem. 85(2013) 11553e11559.
[19] X. Liu, Y. Xu, D.B. Wan, Y.H. Xiong, Z.Y. He, X.X. Wang, B.D. Hammock,Development of a nanobodyealkaline phosphatase fusion protein and itsapplication in a highly sensitive direct competitive fluorescence enzymeimmunoassay for detection of ochratoxin A in cereal, Anal. Chem. 87 (2015)1387e1394.
[20] J. Wang, Z. Majkova, C.R. Bever, J. Yang, S.J. Gee, J. Li, B.D. Hammock, One-stepimmunoassay for tetrabromobisphenol A using a camelid single domainantibodyealkaline phosphatase fusion protein, Anal. Chem. 87 (2015)4741e4748.
[21] C. Hamers-Casterman, T. Atarhouch, S. Muyldermans, G. Robinson,C. Hammers, E.B. Songa, R. Hammers, Naturally occurring antibodies devoid oflight chains, Nature 363 (1993) 446e448.
[23] S. Muyldermans, T.N. Baral, V.C. Retamozzo, P. De Baetselier, E. De Genst,J. Kinne, U. Rothbauer, Camelid immunoglobulins and nanobody technology,Vet. Immunol. Immunopathol. 128 (2009) 178e183.
[24] C.R. Bever, Z. Majkova, R. Radhakrishnan, I. Suni, M. McCoy, Y. Wang,B.D. Hammock, Development and utilization of camelid VHH antibodies fromalpaca for 2, 20 , 4, 40-tetrabrominated diphenyl ether detection, Anal. Chem.86 (2014) 7875e7882.
[25] J. Wang, C.R. Bever, Z. Majkova, J.E. Dechant, J. Yang, S.J. Gee, B.D. Hammock,Heterologous antigen selection of camelid heavy chain single domain anti-bodies against tetrabromobisphenol A, Anal. Chem. 86 (2014) 8296e8302.
[26] X. Liu, Y. Xu, Y. Xiong, Z. Tu, Y. Li, Z. He, B.D. Hammock, VHH phage-basedcompetitive real-time immuno-polymerase chain reaction for ultrasensitivedetection of ochratoxin A in cereal, Anal. Chem. 86 (2014) 7471e7477.
[27] T. He, Y. Wang, P. Li, Q. Zhang, J. Lei, Z. Zhang, W. Zhang, Nanobody-basedenzyme immunoassay for aflatoxin in agro-products with high tolerance tocosolvent methanol, Anal. Chem. 86 (2014) 8873e8880.
[28] L. Zou, Y. Xu, Y.P. Li, Q.H. He, B. Chen, D. Wang, Development of a single-chainvariable fragment antibody-based enzyme-linked immunosorbent assay fordetermination of fumonisin B1 in corn samples, J. Sci. Food Agric. 94 (2014)1865e1871.
[29] N.K. Jerne, The immune system, Sci. Am. 229 (1973) 52e60.[30] N.K. Jerne, Towards a network theory of the immune system, Ann. Immunol.
125 (1974) 373e389.[31] N.K. Jerne, J. Roland, P.A. Cazenave, Recurrent idiotopes and internal images,
EMBO J. 1 (1982) 243e247.[32] M. Shu, Y. Xu, D. Wang, X. Liu, Y. Li, Q. He, X. Wang, Anti-idiotypic nanobody:
A strategy for development of sensitive and green immunoassay for Fumo-nisin B1, Talanta 143 (2015) 388e393.
[33] M.D. Swain, G.P. Anderson, J. Serrano-Gonz�alez, J.L. Liu, D. Zabetakis,E.R. Goldman, Immunodiagnostic reagents using llama single domain anti-bodyealkaline phosphatase fusion proteins, Anal. Biochem. 417 (2011)188e194.
[34] F.W. Studier, Protein production by auto-induction in high-density shakingcultures, Protein Expr. Purif. 41 (2005) 207e234.
[35] Y.K. Wang, Y.C. Wang, H.A. Wang, W.H. Ji, J.H. Sun, Y.X. Yan, Animmunomagnetic-bead-based enzyme-linked immunosorbent assay for sen-sitive quantification of fumonisin B1, Food Control. 40 (2014) 41e45.
[36] M. Devreese, S. De Baere, P. De Backer, S. Croubels, Quantitative determinationof several toxicological important mycotoxins in pig plasma using multi-mycotoxin and analyte-specific high performance liquid chromatogra-phyetandem mass spectrometric methods, J. Chromatogr. A 1257 (2012)74e80.
[37] H. Belhassen, I. Jim�enez-Díaz, R. Ghali, H. Ghorbel, J.M. Molina-Molina, N. Olea,A. Hedili, Validation of a UHPLCeMS/MS method for quantification of zear-alenone, a-zearalenol, b-zearalenol, a-zearalanol, b-zearalanol and zear-alanone in human urine, J. Chromatogr. B 962 (2014) 68e74.
[38] F. Soleimany, S. Jinap, F. Abas, Determination of mycotoxins in cereals byliquid chromatography tandem mass spectrometry, Food Chem. 130 (2012)1055e1060.
[39] X. Wang, Q. He, Y. Xu, X. Liu, M. Shu, Z. Tu, D. Cao, Anti-idiotypic VHH phagedisplay-mediated immuno-PCR for ultrasensitive determination of mycotoxinzearalenone in cereals, Talanta 147 (2016) 410e415.
[40] J.L. Liu, D. Zabetakis, A.B. Lee, E.R. Goldman, G.P. Anderson, Single domainantibodyealkaline phosphatase fusion proteins for antigen detectiondanal-ysis of affinity and thermal stability of single domain antibody, J. Immunol.Methods 393 (2013) 1e7.
[41] J. Lei, P. Li, Q. Zhang, Y. Wang, Z. Zhang, X. Ding, W. Zang, Anti-IdiotypicNanobody-Phage Based Real-Time Immuno-PCR for Detection of Hep-atocarcinogen Aflatoxin in Grains and Feedstuffs, Anal. Chem. 86 (2014)10841e10846.
