Detection of aflatoxins,trichothecenes, ochratoxin a and zearalenone by test strip enzyme immunoassay: A rapid method for screening cereals for mycotoxins

The development of test strip enzyme immunoassays (EIA) for the rapid detection of a number of mycotoxins is described. Monoclonal or polyclonal antibodies against aflatoxin B₁ (AFB₁), aflatoxin M₁ (AFM₁), ochratoxin A (OA), T‐2 toxin, diacetoxyscirpenol (DAS), 3‐acetyldeoxynivalenol (3‐AcDON), roridin A (RA) and zearalenone (ZEA) were immobilized on a nylon membrane. Using the corresponding toxin‐horseradish peroxidase conjugate in a direct competitive assay, dot colour development of toxinpositive test strips was visually discernible from that of the negative control. The results of the visual evaluation were compared with that of instrumental reflectance measurements of the test strips. The visual detection limits of the test strip assays for mycotoxins were at 0.6 ng ml^?1 (AFB₁) 0.2 ng ml^?1 (AFM₁), 2.0 ng ml^?1 (OA), 1.0 ng ml^?1(T‐2 toxin), 0.2 ng ml^?1 (DAS), 10.0 ng ml^?1 (3‐AcDON), 15.0 ng ml^?1 (RA), and 5.0 ppb (ZEA) respectively. Utilizing a simple extraction procedure, AFB₁, OA, T‐2 toxin, and ZEA in spiked corn samples were detected by test strip EIA at levels of 15 ng g^?1, 100 ng g^?1, 20 ng g^?1 and 80 ng g^?1 respectively.     

Enzyme immunoassays for picloram detection: A comparison of assay formats

Two direct enzyme immunoassays for picloram (4‐amino‐3,5,6‐trichloro‐2‐pyridinecarboxylic acid) detection were developed. The assay using IgG isolated from a rabbit antipicloram antiserum had a working range from 5 to 5000 ng ml^?1 with a mean I₅₀ value of 180 ng mh^?1 and a limit of quantification of 10 ng ml^?1. The assay using a monoclonal anti‐picloram antibody had a working range from 1 to 200 ng mh^?1 with a mean I₅₀ value of 18 ng mh^?1 and a limit of quantification of 5 ng ml^?1. The direct assays were compared to an existing monoclonal antibody‐based indirect enzymes immunoassay for accuracy and precision of picloram determinations in fortified ground and surface waters. In both matrices, the monoclonal antibody‐based indirect enzyme immunoassay was shown to be the superior format in terms of greater precision and equal or greater accuracy.     

Improved enzyme immunoassay for group-specific determination of penicillins in milk

Polyclonal antibodies against penicillin antibiotics were raised in rabbits after immunization with an ampicillin-bovine serum albumine conjugate. The specificity and sensitivity of these antibodies were tested in a competitive direct enzyme immunoassay (EIA), using an ampicillin-horseradish peroxidase conjugate as the labelled antigen, and penicillin G as the reference antibiotic. The 50% binding inhibition concentration of penicillin G standard curves was at 2–4 ng ml^?1, with a detection limit (20% binding inhibition) at 0.5–1 ng ml^?1. The antiserum showed strong relative cross-reactivities with all penicillin-type betalactam antibiotics tested, including those regulated by maximum residue limits (MRLs) within the European Union (ampicillin 64%, amoxicillin 24%, oxacillin 100%, cloxacillin 29%, dicloxacillin 27%, and nafcillin 120%). After alkaline hydrolysis of the betalactam ring, no cross-reactivity of the respective penicillins was observed anymore. Cephalosporin-type betalactam antibiotics did not react in this test system. The recovery rates for penicillin G in artificially contaminated milk samples (2–32 ng ml^?1) were at 89–97%. Penicillin G, ampicillin, amoxicillin, oxacillin, cloxacillin, dicloxacillin and nafcillin were all detectable in spiked milk samples at or below the respective MRL level. Comparison analysis of the EIA and a microbiological inhibition assay using violative milk samples obtained from routine milk testing scheme in Bavaria showed excellent agreement between both methods.     

Strip ELISA for detection of staphylococcal enterotoxins in culture supernatants and foods

For the easy and rapid detection of staphylococcal enterotoxins (SE) A, B, C₁, D and E, a sandwich ELISA (enzyme‐linked immunosorbent assay) was developed using an activated nylon membrane with covalently bound antibodies as a solid phase. Strips coated with antibodies against one SE type were incubated with culture supernatants of strains of Staphylococcus aureus as well as with the extracts of food previously contaminated with SE. The assay of bound toxin was performed using horseradish peroxidase labelled antibodies which were partly purified by affinity chromatography. After addition of 3,3’,5,5'‐tetramethylbenzidine, a distinct blue dot developed in the case of positive samples. The detection limit for individual SE is in the range of 0.5 ng ml^‐1 to 1 ng ml^‐1. A semi‐quantitative evaluation is possible without any special reading equipment. The test can be done within 2 h, sample preparation excluded. While milk needed no extraction, in the case of minced meat or noodles, homogenization with 1.5 or 2 volumes of phosphatebuffered saline followed by centrifugation for 20 min at 10 000 × g and 4° C proved to be adequate. Interferences by constituents of the food extracts were not observed. The results showed good agreement with those of the microtiter plate ELISA and also with those of a commercially available SE ELISA kit.     

Multi‐residue enzyme immunoassays for screening for β‐agonists in faeces and feeds

Multi‐analyte screening methods for β‐agonists in faeces and feeds based on enzyme immunoassay (EIA) were developed. Commercially available EIA kits were used, and the results presented. The extraction for both matrices was performed under acid conditions. The addition of an organic solvent to the extradant, resulting in higher recoveries, was only retained for feeds as the background absorption for faeces increased substantially. Clean‐up for faecal extracts was achieved in a one‐step extraction with isobutyl alcohol. Detection limits in faeces were slightly different for the two types of kits used. They ranged from 0.8ng g^?1 for clenbuterol to 10ng g^?1 for terbutaline. In feeds, corresponding detection limits for both compounds were 5 ng g^?1 and 110 ng g^?1. Confirmation of EIA positives was established by liquid chromatography with post‐column diazotation and gas chromatography‐mass spectrometry. In the screening of faeces over 1 year about 70% of the EIA positive samples were confirmed as positive.     

Development of an ELISA for the detection of chlortetracycline,oxytetracycline and tetracycline residues

A ‘generic’ enzyme immunoassay (ETA) has been developed which is capable of detecting chlortetracycline, oxytetracycline and tetracycline with a limit of determination well below the current maximum residual limit of 600 ng g^‐1 kidney^‐1. The EIA requires no prior sample clean‐up or concentration, being applied directly to tissue homogenate, and will also detect tetracyclines in milk up to 15 ng ml^‐1.     

Production and characterization of monoclonal anti‐idiotype antibody mimics for the pyrethroid insecticides and the herbicide paraquat

This paper highlights some of the difficulties encountered in the production and characterization of beta‐type anti‐idiotype antibodies to haptenic molecules. It describes the techniques employed to overcome these problems and illustrates the importance of the screening assay format in selecting the desired antibodies. The anti‐idiotype antibodies produced were used in a variety of assay formats, some innovative, for the quantification of the pesticides paraquat and phenoxybenzoic acid (PBA). The most sensitive of these assays detected 100 pm ml^‐1 paraquat and 100 ng ml^‐1 PBA respectively. The application of these antibodies to multi‐residue tests which overcome the narrow specificity conferred by conventional immunoassay techniques is discussed.     

Production and characterization of group‐specific antibodies against penicillin antibiotics

Polyclonal antibodies against penicillin antibiotics were produced in rabbits after immunization with an ampicillin‐bovine serum albumin conjugate. The antibodies were tested in a competitive direct enzyme immunoassay (EIA) using an ampicillin‐horseradish peroxidase conjugate as the labelled antigen and penicillin G as the reference antibiotic. The 50% inhibition concentration for penicillin G was 12 ng ml^?1. This penicillin EIA showed broad group‐specificity for penicillins having the intact 6‐aminopenicillanic acid (6‐APA) nucleus, and was able to detect 16 other penicillins tested, whereas cephalosporin antibiotics were not detected. Relative cross‐reactivities (penicillin G = 100%) exceeding 10% were found for amoxicillin (45%), ampicillin (45%), 6–8 (12%), azlocillin (110%), carbenicillin (39%), cloxacillin (49%), dicloxacillin (57%), epicillin (23%), metampicillin (130%), methicillin (32%), oxacillin (98%), penicillin V (160%), pheneticillin (68%), piperacillin (26%) and ticarcillin (110%). After alkaline hydrolysis of the ß‐lactam ring, the corresponding penicilloyl derivatives of these antibiotics completely lost cross‐reactivity in the test system. Therefore, the antibodies described here primarily detect the intact 6‐APA nucleus having any carboxylamide side chain.     

Development of high‐sensitive enzyme immunoassays for gliadin quantification using the streptavidin‐biotin amplification system

Optimization of three enzyme immunoassays of very high sensitivity using three antiprolamin monoclonal antibodies (MAbs) (13B4, 11C4 and 12A1) is presented here. These MAbs are specific for those prolamins toxic for coeliac patients, as determined by immunoblotting analysis. Biotinylated MAbs were used in two of the assays. In a competitive ELISA, the binding of each biotinylated MAb to a gliadin‐coated solid phase was inhibited by gliadin in the fluid phase. The best result was obtained using the biotinylated MAbl3B4 (detection limit: 20 ng ml^?1). With regard to capture ELISA, we tested the performance of the three MAbs. In this sandwich ELISA, the MAb used for antigenic capture was the same as that used as secondary biotinylated antibody. The MAbl2Al had the best performance (detection limit: 1 ng ml^?1). The use of biotin‐labelled gliadin in a quantitative immunoassay with a detection limit of 5 ng ml^?1 is also reported. This assay involves an antigenic capture using the MAbl2Al followed by a competition between biotinylated and non‐biotinylated gliadin. We have found the use of the streptavidin‐biotin interaction as signal amplification system to be very useful. This technique, as far as we know, has not been previously reported for gliadin quantification.     

Validation of a monoclonal antibody‐based indirect enzyme immunoassay method for picloram detection in soil and plants

Picloram (4‐amino‐3,5,6‐trichloro‐2‐pyridinecarboxylic acid) residues in authentic soil and plant samples were determined by indirect enzyme immunoassay (EIA) and by gas chromatography (GC). Results obtained by EIA methods correlated well with the GC results with coefficients of determination ranging from 0.778 to 0.891. Picloram levels determined by EIA in soil extracts obtained by a potassium hydroxide extraction method (EIA‐1) underestimated picloram levels determined by GC. Conversely, EIA determinations of soil extracts obtained by a more rigorous acetonitrile extraction method (EIA‐2) yielded picloram levels that overestimated levels determined by GC. First order linear regression models using EIA determinations as the independent variable resulted in dependent variable coefficients of 0.325 and 1.42 for EIA‐1 and EIA‐2 respectively. Picloram levels determined in plant extracts were approximately equivalent by EIA and GC methods. Using a point on the standard curve representing 5 ng ml^?1 picloram as a threshold value, the EIA method effectively identified samples as either positive or negative for picloram residues based on the results obtained by GC analysis.     

Rapid enzyme immunoassays for the detection of three sulphonamides in milk

Two rapid competitive enzyme immunoassays, utilizing test strip and immunofiltration techniques, for the detection of three sulphonamides (sulphadiazine, sulphamethazine and sulphamethoxypyridazine) in milk were developed. In these direct immunoassays, free sulphonamide competed with the relevant sulphonamide—horseradish peroxidase conjugate for the binding sites of membrane‐bound polyclonal antibodies. Test strips were incubated in the sample solution containing the sulphonamide—enzyme conjugate, while, for the immunofiltration assays, sample and enzyme conjugate solutions were added consecutively. After incubation of enzyme substrate/chromogen solution, the colour reaction was evaluated visually by observing the difference to the negative control. The detection limits for each of the three compounds were between 5 and 10 ng ml^‐1 of milk using the test strip assays and between 10 and 20 ng ml^‐1 of milk using the immunofiltration assays.     

Production and characterization of antibodies against neosaxitoxin utilizing a novel immunogen Synthesis procedure

Polyclonal antibodies against neosaxitoxin (neoSTX) were produced in rabbits after immunization with a neoSTX—glucose oxidase conjugate. This method used a novel approach for immunogen synthesis, which required only trace amounts (28 ug) of neoSTX for the coupling reaction. A competitive indirect enzyme immunoassay (EIA), with saxitoxin conjugated to bovine serum albumin as the solid phase antigen, was established using these antibodies. The detection limit for neoSTX was 17.8 pg ml^‐1. Relative cross‐reactivities of the antibodies with gonyautoxin 1/4 (GTX1/4), saxitoxin (STX), gonyautoxin 2/3 (GTX2/3), decarbamoylsaxitoxin (dcSTX) and N‐sulphocarbamoylsaxitoxin (C1/2) were 104, 3.36, 3.35, 0.11 and 0.04% respectively. Thus, the EIA described here was most specific for neoSTX and GTX1/4.     

Application of an immunosensor for the detection of the β-lactam antibiotic,cephalexin

Public concern surrounding antibiotic contamination in food and food products has made it imperative to develop analytical methods for their detection. Polyclonal antibodies were used in the development of a surface plasmon resonance (SPR)-based inhibition immunoassay for cephalexin. A conjugate consisting of cephalexin-bovine serum albumin (BSA) was immobilized on the dextran gel surface of the sensor chip. Binding/regeneration studies of antibody to immobilized cephalexin were studied and dissociation of the antibody from the immobilized cephalexin was easily achieved with 10 mmol l^?1 NaOH. Forty surface regeneration cycles were carried out and found to be reproducible with only a 7.4% decrease in binding over this number of regenerations. Model inhibition immunoassays for cephalexin were developed in PBS and spiked milk samples with detection ranges of 4.88 to 2,500 ng ml^?1 and 244 to 3,906 pg ml^?1, respectively.     

Development of Immunochromatographic Assay for Identification of Organophosphate Pesticides in Environmental Samples

Microtiter plate enzyme linked immunoassay (ELISA) experiments in competitive format were performed utilizing polyclonal antibody for establishing a detection system for organophosphate pesticides. IC₅₀ value of and limit of detection (LOD) value was determined by standard inhibition curve and value obtained were 0.05 μgmL^?1 and 0.001 μgmL^?1, respectively. Specificity of antibody was investigated with different organophosphate pesticides. Immunochromatographic assay (ICA) experiments were also designed in competitive format by making use of immunochromatographic strip which was assembly of three main components: conjugate pad, membrane and adsorbent pad. Membrane was coated with hapten-OVA conjugate (test line) and antirabbit IgG (control line). ICA experiments were performed by employing gold-labeled antibody as a detector reagent which was applied over conjugate pad. Visual detection limit obtained from ICA was 0.5 μgmL^?1. Major advantage of strip assay was rapid result, i.e., less than 10 min. which makes it suitable for onsite applications.     

Detection of aldrin/dieldrin in milk by competitive enzyme‐linked immunosorbent assay (ELISA)

Polyclonal antibodies raised against 6,7‐dihydro‐6‐carboxyaldrin can be used to detect aldrin and dieldrin. These analytes are preferentially fat soluble. An immunoassay is described which provides a method for detecting these pesticides in milk, a fat‐rich matrix. The enzyme‐linked immunosorbent assay (ELISA) can detect aldrin/dieldrin in milk in the range 1 ng ml^‐1‐5 μg ml^‐1 simply and reliably. The detection range differs in skimmed and semi‐skimmed milk and in cream, reflecting the differences in fat content between these samples.     

Enzyme immunoassay for the detection of streptomycin and dihydrostreptomycin in milk

Polyclonal antisera against streptomycin were prepared by using a streptomycin‐oxime derivative coupled to bovine serum albumin for the immunization of rabbits. The specificity and sensitivity of these antibodies were tested in a competitive assay using a streptomycinenzyme conjugate (prepared by coupling a streptomycin‐hydrazone derivative to horseradish peroxidase) in a double antibody solid phase technique. The only detectable cross‐reactivity of the assay system with other aminoglycoside antibiotics and other substances similar in structure was shown to be with dihydrostreptomycin of about 148.7%. The detection limit in buffer solution was 0.6 ng ml^‐1 for streptomycin and 0–4 ng ml^‐1 for dihydrostreptomycin. Employing rapid sample preparation procedures, streptomycin and dihydrostreptomycin were detected in milk at levels as low as 6 and 0.8 ng ml^‐1respectively.     

Bovine β‐lactoglobulin in hypoallergenic and ordinary infant formulas measured by an indirect competitive ELISA using monoclonal and polyclonal antibodies

Whey protein concentrates of cows’ milk are used widely in dairy products including infant formulas. A major component is the potent allergen bovine β‐lactoglobulin (BLG). Two easy and sensitive substrate‐amplified indirect competitive ELISAs for BLG, one using a rabbit polyclonal antibody (PAb) K97 raised against heat‐treated BLG, and another using a monoclonal antibody (MAb) 61B4 specific to the epitope Thr¹²⁵‐Lys¹³⁵ are described. Furthermore, a new whey protein test kit specific for BLG was tested. A new video‐based digital image processing system was used to read, evaluate and calculate data from the microtitre plates. The detection limit was 0.08 ng ml^‐1 for the PAb assay, which was sufficient for the use of this assay as a test method for ‘milk‐free’ products. The detection limit was 3.2 ng ml^‐1 for the MAb assay and 144 ng ml^‐1’ for the commercially available kit. Thirteen commercially available infant formulas on the Danish market, including four hypoallergenic products and cows’ milk, were tested. The BLG content varied from 0.2 to 727 μg ml^‐1, quantified with the PAb assay. The hypoallergenic infant formula with the lowest amount of BLG was Nutramigen containing about 0.1% residual BLG compared with the ordinary infant formula Nidina 1. The knowledge of residual BLG content in hypoallergenic infant formulas is important for a more qualified nutritional recommendation to parents with children genetically susceptible to allergies and for quality assurance of infant formulas claimed to be hypoallergenic.     

Rapid determination of triazine contamination of fresh water supplies by immunoassay

A highly sensitive, competitive, enzyme‐linked immunosorbent assay (ELISA) was developed for the detection of five triazine herbicides in water samples. Low detection limits were achieved without the need to concentrate samples prior to analysis by ELISA. The limit of detection for ametryn, prometryn and prometon was 0–05 ng ml～’ while atrazine and propazine had a detection limit of 0.2 ng ml^‐1. The concentrations of analyte required to reduce zero standard absorbances by 50% (IC₅₀) far ametryn, prometryn, prometon, propazine and atrazine were 0.18, 0.18, 0.26, 0.48 and 0.37 ng ml^‐1 respectively. De‐ethylatrazine and simazine had respective ICsos of 4.0 and > 25.0 ng ml^‐1. Inter‐assay coefficients of variation were generally less than 10% and recoveries of triazines from fortified water samples were in the range 90–115% for all five triazine herbicides. The results demonstrate the ability of an enzyme immunoassay to detect five triazine herbicides at concentrations below or very close to the maximum concentration allowed under EC guidelines in drinking water (0.1 ng ml^‐1) for individual pesticides.     

Development of a dual label time‐resolved fluoroimmunoassay for the detection of (β‐agonists in cattle urine

β‐Agonists are among the most widely abused drugs in veterinary medicine for the illegal promotion of farm animal growth. An array of analytical procedures has been developed to detect the residues of these compounds in many biological materials. As the number of β‐agonist formulations increases, it has become increasingly difficult to devise screening techniques capable of detecting a broad spectrum of these residues in a single test. A dual immunoassay based on time‐resolved fluorescence was developed that incorporated a monoclonal antibody raised to tertiary butyl amines and a polyclonal antibody to biphenolic β‐agonists. This assay was capable of detecting residues of a range of β‐agonists present in bovine urine without the need for sample extraction. The limits of detection of the assay ranged from 1 to 8.5 ng ml^‐1 depending on the cross‐reactivity of individual compounds with the antibodies employed in the procedure.     

A quantitative real-time immuno-PCR approach for detection of staphylococcal enterotoxins

Bacterial intoxications represent a substantial public health concern with enterotoxins produced by Staphylococcus aureus among the most common causes of food poisoning. In addition to their role in the pathogenicity of food poisoning, staphylococcal enterotoxins have profound effects on the immune system as members of the family of pyrogenic toxin superantigens. As the classical diagnostic bioassays as well as the routinely used immunological methods are hampered by several drawbacks regarding sensitivity, specificity, and practicability, there is a need for the timely identification of toxins by highly sensitive and specific methods. To combine the versatility of an enzyme immunoassay (EIA) with the amplification power of the PCR, a quantitative real-time immuno-PCR (qRT-iPCR) was developed for the detection of staphylococcal enterotoxins A and B and compared to a commercially available EIA. A broadly applicable tool for signal amplification of pre-formed immunocomplexes was established by covalent binding of a reporter DNA to secondary detection antibodies. Therefore, the amino-modified reporter DNA was coupled successfully to N-succinimidyl-S-actyl-thioacetate-activated secondary detection antibodies. The qRT-iPCR was able to detect highly reproducibly as low as approximately 0.6 to 6 pg (4 to 40 amol/microl) of staphylococcal enterotoxin B and staphylococcal enterotoxin A, respectively. In conclusion, the qRT-iPCR approach was shown to overcome clearly the sensitivity limit of traditional immunological detection procedures for bacterial toxins, as demonstrated in this study for staphylococcal enterotoxins. The development of a stable antibody-DNA conjugate providing a universal signal amplification offers a versatile as well as a highly sensitive and specific tool for diagnostic and research purposes generally applicable for pre-formed antibody-antigen complexes.