Specific Noncompetitive Immunoassay for HT-2 Mycotoxin Detection

Jan 19, 2016 - (34) Selection of the anti-IC antibodies was made using an automated magnetic bead processor (King Fisher Duo, Thermo Scientific)...
0 downloads 0 Views 656KB Size
Subscriber access provided by UNIV OSNABRUECK

Article

A Specific Non-competitive Immunoassay for HT-2 Mycotoxin Detection Henri Olavi Arola, Antti Tullila, Harri Kiljunen, Katrina Campbell, Harri Siitari, and Tarja K. Nevanen Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.5b04591 • Publication Date (Web): 19 Jan 2016 Downloaded from http://pubs.acs.org on January 27, 2016

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

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

Page 1 of 10

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

Analytical Chemistry

A Specific Non-competitive Immunoassay for HT-2 Mycotoxin Detection Henri O. Arola*,†, Antti Tullila†, Harri Kiljunen‡, Katrina Campbell§, Harri Siitari# and Tarja K. Nevanen† †

VTT Technical Research Centre of Finland, Tietotie 2, FI-02150 Espoo, Finland



VERIFIN, Finnish Institute for Chemical Weapons Convention, A. I. Virtasen aukio 1, Department of Chemistry, FI00014 University of Helsinki, Finland

§

Institute for Global Food Security, Queen's University Belfast, School of Biological Sciences, 8 Cloreen Park, Malone Road, Belfast, United Kingdom #

University of Helsinki, Faculty of Pharmacy, Viikinkaari 5 E, FI-00014 University of Helsinki, Finland

KEYWORDS HT-2 toxin, hapten, mycotoxin, recombinant antibody, Fab, immune complex assay, non-competitive, TR-FRET * E-mail: [email protected]. Phone: +358 40 189 5970 ABSTRACT: Here we demonstrate a novel homogenous one-step immunoassay, utilizing a pair of recombinant antibody Fab fragments, that is specific for HT-2 toxin and has a positive read-out. Advantages over the conventional competitive immunoassay formats such as ELISA are the specificity, speed and simplicity of the assay. Recombinant antibody HT2-10 Fab recognizing both HT-2 and T-2 toxins was developed from a phage display antibody library containing 6 x 107 different antibody clones. Specificity of the immunoassay was introduced by an anti-immune complex (IC) antibody binding the primary antibody - HT-2 toxin complex. When the non-competitive immune complex assay was compared to the traditional competitive assay, an over 10-fold improvement in sensitivity was observed. Although the HT2-10 antibody has 100 % cross-reactivity for HT-2 and T-2 toxins, the immune complex assay is highly specific for HT-2 alone. The assay performance with real samples was evaluated using naturally contaminated wheat reference material. The EC50 value of the TR-FRET assay was 9.6 ng/mL and the limit of detection (LOD) 0.38 ng/mL (19 µg/kg). The labelled antibodies can be pre-dried to the assay vials, e.g. microtiter plate wells, and readout is ready in 10 minutes after the sample application.

The trichothecenes are a class of mycotoxins that are produced as secondary metabolites by Fusarium spp. fungi. HT-2 and T-2 toxins belong to the Type-A trichothecenes, and typically occur in different grains such as oat, barley, wheat, rye, corn, rice and soybeans and products thereof.1,2 Structurally HT2 toxin and T-2 toxin differ only at the C4 position, where T-2 toxin has an acetate and HT-2 toxin a hydroxyl group (Figure S-1). T-2 toxin is rapidly metabolized to HT-2 when consumed.3 HT-2 and T-2 toxins are stable during the processing of contaminated grain and they can even be enriched in some cereal by-products, e.g. bran. HT-2 and T2 toxins are also relatively heat stable, withstanding e.g. baking and cooking.1 Recently, HT-2 toxin was found to be the most prevalent mycotoxin along with deoxynivalenol (DON) in beer in a study including 14 different mycotoxins.4 T-2/HT-2 toxin contamination can cause several healthrelated issues when consumed, such as hemorrhage, vomiting, impaired immune function, fever and headache.2 The current EU recommendations for screening tech-

niques are set for the sum of HT-2 and T-2 toxins. According to the recommendations the limit of detection (LOD) for the sum of these toxins should not exceed 25 μg/kg.5 There are no specific screening methods available for HT2 and T-2 toxins separately. The reported antibodies and immunoassays for T-2 toxin or for T-2 and HT-2 toxins are typically cross-reactive.6,7,8,9,10,11,12,13 Chemically based analytical methods, especially techniques that combine liquid or gas chromatographic separation with mass spectrometry, are well established for mycotoxin determination.14 LC-MS and GC-MS methods are very sensitive and accurate but often they are rather laborious and time-consuming. Despite the vast and still growing utilization of LC-MS and GC-MS methods in mycotoxin analysis, the farmers, customs and small industrial producers need rapid, easy, high-throughput and preferably portable screening methods and devices. For them, owning expensive instruments needing skilled laboratory personnel is not an option.15

ACS Paragon Plus Environment

Analytical Chemistry

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

The immunoassays available today for T-2 and HT-2 toxins are qualitative, semi-quantitative or quantitative competitive assays. Assay formats include more traditional techniques such as lateral flow tests and ELISA11, but a few other techniques have also been developed based on e.g. fluorescence polarization8, surface plasmon resonance9 or magnetic beads16. The cross-reactivity is a potential problem in current immunoassays, lowering their accuracy2 and leading to overestimation of the target analyte due to the cross-reacting molecules in the sample15. Especially modified mycotoxins pose a potential risk of biased results15,17. Theoretically, the sensitivity, kinetics and linear range of the competitive immunoassays are inferior to those of the non-competitive assays.18 For small molecules such as mycotoxins the competitive assay has been the most common option since haptens do not have multiple epitopes for the simultaneous binding of two antibodies. To enable a non-competitive assay for small molecules, some other assay formats have been developed. So-called open-sandwich immunoassays utilize only one antibody and the signal is amplified with a conjugated enzyme19,20. Since only one antibody is involved, it also faces the same problem of cross-reactivity as the competitive assay. Another non-competitive assay format developed by Li et al. (2013) uses quenching of the intrinsic fluorescence of antibodies due to the binding of a small analyte.21 This kind of assay has been developed for a few mycotoxins, namely aflatoxin, ochratoxin A (OTA) and zearalenone. In this case too, only one antibody is used and so the assay is as specific as the antibody used and cross-reactivity is a potential problem. Additionally, a FRET phenomenon between the antibody and the analyte has been utilized, with OTA as an example22. The utilization of antibodies that recognize the immune complex formed by the primary antibody and the small analyte is so far the only way to enable a sandwich type of assay for haptens. The first anti-immune complex antibodies were monoclonal antibodies developed by immunization23,24,25. The success of the immunization depends on the stability of the immune complex. Antibody library and phage display techniques enable the selection of the antibodies for immune complexes in vitro as reported here. Also other immune complex binders than antibodies have been reported. Gonzales-Techera et al. (2007) have developed a phage anti-immune complex assay by using peptide libraries26 and later anti-immune complex peptides inserted to a streptavidin core have been used by Vanrell et al. (2013) to develop lateral flow and ELISA assays27. In this work, we developed a specific, sensitive and rapid one-step homogenous immunoassay for HT-2 toxin detection. The non-competitive assay for HT-2 toxin was implemented using a pair of recombinant Fab fragments. The primary Fab recognizes HT-2 toxin and the secondary Fab binds specifically to the complex formed by the primary antibody and HT-2 toxin (Figure 1). The same principle has been applied previously for rapid diagnostics of drugs of abuse28,29. The immune complex (IC) assay re-

Page 2 of 10

duces the amount of work and the time needed to perform the analysis by simplifying the assay procedure. In its simplest form the labeled antibodies are pre-dried in a vial. No blocking or washing steps are needed and after addition of the sample the readout is ready in 10 minutes. The IC assay can also be applied for a large number of samples in high-throughput format. The homogeneous assay presented in this work is based on the utilization of time-resolved fluorescence resonance energy transfer (TR-FRET)30. Briefly, the energy from a donor fluorophore is excited to a higher energy state and transferred to an acceptor fluorophore via intermolecular dipole-dipole coupling. This phenomenon can occur only if the donor and acceptor are close enough to each other (1-10 nm) and the fluorescence spectra of the donor and the acceptor are matched. The size of a Fab antibody fragment is ideal for setting up a FRET assay. By labelling the Fab antibody pair with suitable donor and acceptor labels, the energy transfer due to the binding of the antiimmune complex Fab can be detected as the acceptor fluorescence (Figure 1). The simplicity of the assay makes it suitable for different rapid on-site tests for challenging samples. This is the first demonstration of an immune complex assay for mycotoxins.

Figure 1. Principle of the HT-2 toxin FRET assay. The anti-immune complex (IC) Fab binds to the primary antibody with HT-2 toxin. The FRET can occur due to the short distance between the two fluorophores.

MATERIALS AND METHODS Chemicals and Reagents. HT-2 and T-2 toxins and the carrier proteins human serum albumin (HSA) and alkaline phosphatase (AP) were purchased from Sigma Aldrich. Mollusk hemocyanin (Blue CarrierTM) was purchased from Thermo Scientific. HT-2, T-2, T-2-triol, T-2tetraol, diacetoxyscirpenol, 15-acetyldeoxynivalenol, 3acetyldeoxy-nivalenol, deoxynivalenol, deoxynivalenol-3glucoside, nivalenol and neosolaniol standard solutions for cross-reactivity studies were from Romer labs (Aus-

ACS Paragon Plus Environment

Page 3 of 10

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

Analytical Chemistry

tria) and were provided by Queen’s University Belfast. For the FRET assay Europium label (Loisto615) was purchased from Kaivogen (Finland) and Alexa Fluor 647 label from Molecular Probes Inc. The blank wheat sample (confirmed not to include either HT-2 or T-2 by LC-MS31) was obtained from Alexis Nathanail (Evira, Finland) and naturally contaminated reference matrix material was purchased from Aokin (Germany) with known concentrations of HT-2 (455 ± 89 µg/kg) and T-2 toxin (1769 ± 355 µg/kg) certified by an HPLC and kinetic fluorescence polarization method with a satisfactory range of ± 20 %. Escherichia coli XL-1 blue electroporation-competent cells were purchased from Agilent Technologies Inc. (TX) and the E. coli RV308 (ATCC 31608) strain used for Fab production was from American Type Culture Collection. Superb Broth (SB, 2 % yeast extract, 3 % tryptone, 1 % MOPS, pH 7) was used as a cultivation medium and Luria Broth (LB, 0,5 % yeast extract, 1 % tryptone, 1 % NaCl) supplemented with 100 µg/mL ampicillin (LB-amp) for preparing agar plates.

Construction of the Antibody Gene Library. HT-2 toxin - Blue protein conjugate was used for immunizing mice in Freund’s adjuvant. The mouse with the best immune response for HT-2 toxin was selected for construction of the antibody gene library. The mouse spleen was homogenized using a sterilized Kinematica AG Polytron PT 1200-homogenizer. A commercial RNeasy® Midi/Maxikit (Qiagen Inc., Germany) was used according to manufacturer’s instructions for the extraction of total RNA from the homogenized spleen. The totRNA was used as a template to synthesize the complementary DNA (cDNA) using a Phusion® RT-PCR kit (Thermo Scientific) with oligo-dT priming. The mouse IgG light chain and heavy chain were amplified by PCR using specific primers32 for heavy and light variable chains and constant regions and the cDNA as a template. An antibody fragment (Fab) phage display library was constructed and displayed on phages as described in Pulli et al (2005)28. The functional size of the final library was determined by sequencing 20 Fab clones (GATC Biotech, Germany).

Toxin Conjugation to a Carrier Protein. HT-2 toxin (5 mg, 1.18 x 10-5 mol) was dissolved in 200 µL of dry 1,4dioxane (Merck). Dry pyridine (Riedel-de Haën) (1.9 µL, 2.36 x 10-5 mol) was added and the reaction mixture was mixed by a vortex mixer. Diphosgene (Fluka) (total amount of 2.5 µL) was added in three parts, with mixing for 10 minutes between each addition. The reaction mixture was allowed to react for three hours with mixing for 1 minute every 30 minutes. After completion of the reaction, remaining solvents and pyridine were evaporated in an argon stream and the drying was completed with reduced pressure (0.5 mmHg) for 5 minutes.

Development of Anti-HT-2 toxin Antibody. HT-2 alkaline phosphatase protein conjugate was used for coating the magnetic beads (Invitrogen, M-270 Epoxy Dynabeads) according to the manufacturer’s instructions. The amount of the coated conjugate on the beads was determined using a Pierce® BCA Protein assay kit (Thermo Scientific). Before use, the magnetic beads were blocked for one hour at room temperature with 1 % milk-PBS solution, washed and resuspended in PBS. A KingFisherTM magnetic bead processor (Thermo Fisher Scientific) was used in the selection of the anti-HT-2 toxin antibodies. 200 μl of phage displayed antibody library was added to the magnetic beads coated with HT-2 toxin-AP conjugate together with 20 μg soluble AP-protein and 100 μl PBST (PBS, 0.05 % Tween 20) and incubated overnight in rotation. Soluble AP was used to deplete possible AP binders from the antibody library. The beads were then washed 56 times for 20 seconds with PBST and the bound phages were eluted with triethylamine (Sigma-Aldrich) buffer, pH 11.75 or 0.1 M Glycine-HCl, pH 2.2 and neutralized with 6.9 μl of 1M HCl or 10 µL of 1M Tris pH 9.6 before infecting 900 μl bacterial XL-1 blue cells. Four rounds of selection were completed with a similar protocol. The selection pressure was increased by decreasing the amount of the beads by 1 μl and increasing the washing time in each selection round. The DNA coding the Fabfragments from all selection output DNA pools was cloned to a pKKTac expression vector33 and plated on LBampicillin plates. The Fab pools were then produced on 96-well plates and screened against HT-2-HSA conjugate using ELISA assay to visualize the possible enrichment of specific antibodies. From every positive selection round, 96 single colonies were picked from LB-amp plates and transferred to 96-well plates containing growth media (100 μl of SB including 100 μg/mL ampicillin, 10 µg/mL tetracycline and + 1 % glucose / well) using a Genetix QPix robotic colony picker. After overnight incubation with shaking at 37 °C and 80 % humidity, 9 μl of fresh cell culture was inoculated on induction plates containing 100

Fresh dry 1,4-dioxane (100 µL) was added into dried HT2 toxin linker product. A suspension of toxin-linker was produced by vortexing for one minute. Linker-activated toxin solution was added dropwise into a solution of human serum albumin (5 mg/mL), mollusk hemocyanin (5 mg/mL), and alkaline phosphatase (AP) (5 mg/mL) in phosphate buffered saline (PBS; 15 mM sodium phosphate pH 7.3, 150 mM NaCl). The reaction mixture was allowed to stand overnight. The toxin-albumin complex was purified using Econopac® columns (Bio-Rad) or Amico Ultra® 10K MWCO centrifugal filters (Millipore). Analysis of free toxin was carried out by LC-MS to confirm the absence of free toxins in the samples. Verification of the HT-2-protein Conjugates. The HT-2- toxin protein conjugates were analyzed with ELISA using commercial anti-HT-2 monoclonal antibodies (Hytest, Finland) for HSA, AP and Blue Protein conjugates. Additionally, matrix assisted laser desorption ionization – time of flight mass spectrometry (MALDI-TOF MS) was used to determine the ratio of conjugation of the HT-2 toxin to HSA and AP conjugates. Sinapic acid dissolved in a 50:50 mixture of 0.1% trifluoroacetic acid and acetonitrile was used as a matrix. Hapten-protein conjugates in water were mixed 1 + 1 with the matrix, and 1 µL of the mixture was spotted on the target plate and dried for 10 minutes. The analysis was made using a Bruker Autoflex II mass spectrometer (Germany).

ACS Paragon Plus Environment

Analytical Chemistry

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

μl of SB (100 μg/mL ampicillin, 10 μg/mL tetracyclin, 0.1 % glucose and 1 mM isopropyl-beta-D-thiogalactopyranoside (IPTG, Sigma-Aldrich)/well). The plates were incubated overnight at 34 °C in 80 % humidity with shaking. For the screening of positive clones, a robotic station (Beckman Coulter) was used, allowing automation of the ELISA test. MaxiSorp plates (NUNC) were coated overnight at 4 °C with 100 ng of HT2-toxin-Blue protein conjugates per well in 100 μl of 0.1 M Na-bicarbonate buffer (pH 9.6). The blocking was performed using Superblock Blocking Buffer (Thermo Scientific) with 0.05 % Tween 20 (Sigma Life Sciences) for 30 min. 100 μl aliquots of culture supernatants were applied from induction plates to each well on ELISA plates and incubated for one hour. The plates were washed with 3 x 200 μl of PBS and 100 μl of the detection antibody (Goat anti-mouse-kappa IgGAlkaline phosphatase, Southern Biotech) diluted 1:2000 in 50 % Super Block-solution was added to each well. After 30 min of incubation with the detection antibody, detection solution consisting of 2 mg of para-nitrophenyl phosphate (pNPP, Sigma Aldrich) powder in 1 mL of diethanolamine-MgCl2-buffer (Reagena) was added. Absorbance was measured at 405 nm after 30 minutes. The best clones from the primary screen were analyzed with competitive ELISA after 4 mL production. The supernatant dilutions of 1:1, 1:4, 1:5 and 1:20 were pre-incubated with 6 concentrations of HT-2 and T-2 toxins ranging from 100 pM to 1 μM for 1 hour. Otherwise the ELISA was performed as described previously. The clone showing the best inhibition with free toxins was selected for further characterization. The selected clone HT2-10 Fab was produced in the RV308 E. coli strain in pKKtac plasmid and purified by metal affinity chromatography according to the manufacturer’s instructions (GE Healthcare). Development of the Anti-IC Antibody. The VTT Human naïve scFv phage display library was used as a source for the selection of the anti-immune complex antibodies.28 The naïve library was displayed on the surface of multivalent M13 Hyperphages (Progen, Germany). Iminodiacetic acid (IDA) modified magnetic beads (Bioclon Inc., CA) charged with cobalt (Co+2) were prepared according to the instructions from the manufacturer. Purified HT2-10 Fab fragments were immobilized (3.7 mg Fab/30 mg beads) via the six- histidine tag in an oriented manner and finally the beads were oxidized with H2O2 to make the bond “exchange inert”34. Selection of the anti-IC antibodies was made using an automated magnetic bead processor (King FisherTM Duo, Thermo Scientific). The magnetic beads coated with HT2-10 were incubated for 60 min with a 25 μM solution of free HT-2 toxin in 1 % DMSO-PBST. After a quick wash with PBST for 2 seconds, the beads were incubated for 1 hour with the naïve antibody library. The beads were washed 3 times (PBST, 30 sec) and the bound antibodies were eluted with 100 µL of Glycine-HCl pH 1.5 (30 min in RT). The pH of the eluted phages was neutralized with 15 μl of 1 M Tris before the infection for the next selection round. The output phages from the first round were diluted to a concentration of 2.5

Page 4 of 10

x 1010 phages/400 μl PBST for the following three rounds of selection. The free toxin concentration was kept constant through every selection round. The phage display format was changed from multivalent to monovalent for the last round using VCS M13 (Stratagene) as the helper phage. For the second round a soluble anti-hapten Fab of mouse origin was added to deplete the binders directed to the constant region of the Fab fragments. The last two rounds 3 and 4 were depleted with the HT2-10 Fab beads without the free HT-2 toxin. On selection rounds 2-4, after the incubation with the library, the beads were washed three times (2 min) and eluted with 100 µL of Glycine-HCl, pH 1.5 for 30 min. Screening of the anti-IC clones was performed by phage ELISA. Biotinylated primary antibody HT2-10 was immobilized onto a streptavidin plate (Kaivogen Kaisa96, Finland) at 500 ng/well. The phages displaying the single antibody clones were incubated with HT-2 toxin (25 µM) for 1 h in a well. After washing 3 times with PBST the detection was performed using anti-M13-HRP antibodies (GE Healthcare) diluted 1:2500 in Superblock with 0.05 % Tween 20. The antibodies were incubated for 1 h, after which the plates were washed 3 times with PBST and detected with ABTS® Peroxidase Substrate (KPL, MD) according to the manufacturer’s instructions. The clones with the best A405 ratio between the HT-2 toxin immune complex and the primary antibody alone were sequenced. The best scFv clone anti-IC-HT2-10 (H5) was synthesized as a Fab fragment (DNA 2.0, CA) and produced in E. coli strain RV308 in a pJExpress expression vector (DNA2.0, CA) and purified by metal affinity chromatography according to the manufacturer’s instructions (GE Healthcare). Characterization of the Binding Properties of the Primary Anti-HT-2/T-2 Antibody HT2-10. The crossreactivities of the primary antibody for HT-2 and T-2 toxins were determined as IC50 values using a Biacore Q instrument as described in Meneely et al. 2010.9 Briefly, the HT-2 toxin was immobilized onto a Biacore CM5 chip (GE Healthcare) and final concentrations of 1-1000 ng/mL of HT-2 toxin were applied on the chip after mixing 50:50 with HT2-10 Fab (3.6 µg/mL). The cross-reactivity (%CR) was calculated: IC50 HT-2 / IC50 T-2 x 100 %. The crossreactivity of the HT2-10 Fab was further studied with a panel of other closely related molecules to HT-2, namely T-2-triol, T-2-tetraol, diacetoxyscirpenol, 15-acetyldeoxynivalenol, 3-acetyldeoxy-nivalenol, deoxynivalenol, deoxynivalenol-3-glucoside, nivalenol and neosolaniol. The assay was carried out as described above, except that only one concentration (200 ng/mL) was used for each molecule instead of full inhibition curves. Immune Complex Biacore Assay. The primary antibody HT2-10 was immobilized onto a Biacore Chip CM5 (GE Healthcare) according to the manufacturer’s instructions with an immobilization level of 7694 RU. HT-2 toxin dilutions in running buffer HBS-EP (10 mM HEPES pH 7.4, 150 mM NaCl, 3 mM EDTA, 0.005% v/v Surfactant P20, GE Healthcare) were mixed 50:50 with the anti-IC antibody H5 (10 µg/mL in PBS) having final HT-2 concen-

ACS Paragon Plus Environment

Page 5 of 10

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

Analytical Chemistry

trations in the range of 1-1000 ng/mL and applied on the chip. The used flow rate was 20 µL/min and sample contact time 4 min. The chip was regenerated with 10 mM Glycine-HCl (pH 2) for 30 seconds. Single-step TR-FRET Assay. For the FRET assay the primary antibody HT2-10 Fab was labeled with Europium by dissolving 0.1 mg isothiocyanate (ITC) activated europium chelate in ddH2O and mixing it with 0.3 mg of HT210 Fab and rotating overnight at 4 °C. NAP5 columns (GE Healthcare) were used for buffer exchange to 50 mM Tris pH 7.8; 0.9 % NaCl. The labeling degree was determined according to instructions from the manufacturer. Alexa Fluor 647 label was used for the anti-IC antibody (H5) following the manufacturer’s instructions. 0.5 µg of Eulabeled HT2-10 and 1.5 µg of Alexa 647 labeled H5 in 10 µL of PBS were dispensed on the bottom of a black 96-well plate (Nunc™ F96 MicroWell™, Thermo Scientific) before applying 80 µL of the sample. The plate was shaken for 10 min and measured using a Victor V fluorimeter (Perkin Elmer) with excitation at 340 nm and detection at 665 nm. Specificity Determination. The specificity of the FRET assay was evaluated by spiking 1-100 ng/mL concentrations of HT-2 and T-2 toxins to wheat blank sample extraction. The effect of excess T-2 toxin on the assay was evaluated by mixing 5 ng/mL of HT-2 toxin with 10 ng/mL, 20 ng/mL and 50 ng/mL of T-2 toxin in 7 % MeOH-water. Samples with no toxins or only T-2 toxin with a concentration of 50 ng/mL were used as controls. Sample Preparation. 5 g wheat samples were weighed and extracted with 25 mL of methanol-water 70:30. The mixture was rotated for 30 min in RT and then centrifuged 6000 g for 10 min. The supernatant was filtered (Whatman GF/CTM), diluted 1:10 with PBS and centrifuged at 13 000 g for 5 min to remove the insolubles. Sample Analysis and Matrix Effect. In order to evaluate the effect of the methanol the assay performance was compared in PBS and 7 % MeOH-PBS, which is a methanol concentration equivalent to 1:10 dilution of sample extract. HT-2 toxin was spiked to both buffers with concentrations ranging from 5 to 100 ng/mL. The assay performance with real samples was evaluated using wheat matrix. Six samples (0, 25, 50, 100, 250 and 400 µg/kg) were prepared by weighing and mixing the reference sample with the known blank sample. For generating a standard curve, equivalent final concentrations of HT-2 toxin (0, 0.5, 1, 5, 8 and 10 ng/mL, respectively) were spiked to the blank wheat sample extract. The limit of detection (LOD) and the limit of quantification (LOQ) of the FRET assay were calculated as follows: LOD = blank average + [3 x blank standard deviation] and LOQ = blank average + [10 x blank standard deviation]. Safety Precautions. Mycotoxin stock solutions were handled with safety gloves and using protective clothing. The toxin waste including the liquids and plastic laboratory equipment were disposed appropriately.

RESULTS AND DISCUSSION Antibody Library Construction. The functional size of the Fab antibody library was 6 x 107 /µg DNA, which is adequate due to the strong immune response to HT-2 toxin after immunization and is comparable to previously constructed immunized libraries. Screening of Primary HT-2 Specific Antibodies. 576 single clones were screened from the selection rounds 2, 4 and 5, and from both low and high pH elution, that showed enrichment of specific antibodies in the selection output pool ELISA. The best clones from each round were chosen and characterized by determining their affinities and specificities. From the 16 characterized clones the best clone HT2-10 was selected for anti-immune complex antibody development. The robotic station offered a fast and easy way to perform large sets of primary and competitive ELISA assays, which would be very laborious manual work. The six 96 well ELISA plates in the primary screen and 12 full plates of competition ELISA were both completed during one day with constant incubation times and comparable results for each plate. Characterization of the Binding Properties of the Primary Anti-HT-2/T-2 Antibody HT2-10. The clone HT2-10 showed similar inhibition for both HT-2 and T-2 toxins in competitive ELISA and Biacore Q assays. The IC50 values of the primary antibody HT2-10 were 134 ng/mL for both HT-2 toxin and T2- toxin, and thus the cross reactivity with T-2 toxin was 100 % (Figure 2). No detectable cross-reactivity with the other tested molecules T-2-triol, T-2-tetraol, diacetoxyscirpenol, 15acetyldeoxynivalenol, 3-acetyldeoxynivalenol, deoxynivalenol, deoxynivalenol-3-glucoside, nivalenol or neosolaniol was observed (Table S-1).

Figure 2. Comparison of IC50 for primary antibody HT210 and EC50 of the immune complex assay using the Biacore Q instrument. Results are an average of two replicates. The EC50 and IC50 values were calculated using OriginPro 2015 software (logistic fit).

ACS Paragon Plus Environment

Analytical Chemistry

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

Page 6 of 10

Development of the Specific Anti-IC Antibody for HT2-10-HT-2 toxin Immune Complex. The aim was to develop an HT-2 toxin specific immunoassay and for this purpose anti-IC antibodies were developed using an optimized selection method. The primary antibody HT2-10 was immobilized on the surface of the cobalt-activated magnetic beads in a site-specific manner via the sixhistidine tag at the carboxy terminus of the Fab fragment in order to ensure the optimal orientation of the Fab and HT-2 toxin complex. The cobalt beads were oxidized to stabilize the bond between the Fab and the bead and to minimize non-specific binding of the phages during the selection.34 In total 16 primary positive anti-IC antibodies on the basis of phage ELISA screening were obtained and 3 unique clones were found in sequence analysis. The clone H5 had the best A405 response ratio between the HT-2 toxin immune complex and the primary antibody alone, and it was selected for non-competitive IC assay development. Sensitivity of the Non-competitive IC Assay. The sensitivity of the non-competitive HT-2 toxin assay was compared to the competitive format using the Biacore Q instrument. The EC50 value for the IC assay was 10.5 ng/mL, making it 13 times more sensitive (Figure 2). One–step Homogenous TR-FRET Assay. The homogeneous assay can be performed in solution. This brings an advantage over traditional ELISA assays, in which immobilization and washing steps are needed. The antiimmune complex antibody makes the assay specific to HT-2 toxin without cross-reactivity with T-2 toxin. The specificity of the FRET assay was studied with a spiked wheat blank sample in order also to take into account the possible effect of the sample matrix. No detectable crossreactivity with T-2 toxin was observed over the concentration range 1-100 ng/mL (Figure 3). The EC50 value for HT2 toxin calculated by OriginPro 2015 software was 9.6 ng/mL. In naturally contaminated samples HT-2 and T-2 toxin occur together with different ratios. Typically the amount of HT-2 toxin in contaminated food matrices is double that of T-2 toxin1. Since the primary antibody HT2-10 is highly cross-reactive with both toxins we studied to what extent high T-2 toxin concentrations may interfere with the FRET assay. The test showed that even a T-2 concentration exceeding the HT-2 concentration by a factor of 10 had no effect on the HT-2 response in 7 % MeOH-water (Figure S-1).

Figure 3. Specific recognition of the HT-2 toxin in the TR-FRET assay.

Matrix Effect. A challenge for competitive assays for small analytes is the matrix effect. Some components in the sample may influence the test result. Here we first studied the influence of methanol, a commonly used solvent, in sample preparation for HT-2 toxin analysis. The effect of the methanol concentration on the assay response was examined using PBS as a control. The presence of methanol decreased the fluorescent response. A final concentration of 7% methanol was selected to maintain sufficient sensitivity of the assay, thus allowing 1:10 dilution of the 70% methanol extract (optimization data not shown). The wheat extracts lowered the assay response even more, necessitating a matrix-based standard curve. TR-FRET Assay With Real Samples. The assay performance with real samples was evaluated using wheat matrix, which is one of the grains known often to have Type-A trichothecene contaminations. The standard curve for determining the assay sensitivity for wheat is presented in Figure 4. The LOD and LOQ of the FRET assay were 0.38 ng/mL (equivalent of 19 µg/kg) and 1.1 ng/mL (55 µg/kg), respectively and the linear range was – 25-400 µg/kg. The sensitivity of the FRET assay is comparable to those of previously reported immunoassays for HT-2 and T-2 toxin. Meneely et al. (2011) compared ELISA, enzyme linked immunomagnetic electrochemical array (ELIME) and SPR methods which had LODs of