Development of a sensitive ELISA for the determination of microcystins in algae

Polyclonal antibodies for microcystin-leucine-arginine (MCYST-LR) were generated from rabbits after immunizing the animals with MCYST-LR conjugated with gamma-globulin. A competitive direct enzyme-linked immunosorbent assay (cdELISA) and a competitive indirect ELISA (ciELISA) were used for the characterization of the antibodies and for analysis of the toxin in algal cultures and dietary supplements. The concentrations causing 50% inhibition (IC(50)) of binding of MCYST-horseradish peroxidase (MCYST-HRP) to the solid-phase antibodies by MCYST-LR, MCYST-arginine-arginine variant (MCYST-RR), MCYST-tyrosine-arginine variant (MCYST-YR), and nodularin (NODLN) in the cdELISA were found to be 0.10, 0.12, 0.14, and 0.20 ng/mL, respectively. In the presence of algae matrix, the detection limit is less than 10 ppb. The overall analytical recovery of MCYST-LR (25 to 500 ng/g) added to the algal dietary supplements and then extracted with 0.1 M ammonium bicarbonate in the cdELISA was found to be 83.7%. Analysis of MCYSTs in algal cultures and dietary supplements showed that six of eleven cultures produce MCYSTs, and five of the algal cultures were not MCYST producers. Eight of eleven tested commercial algal dietary supplements contained MCYSTs at a level lower than 100 ppb. The presence of MCYST-LR in the Microcystis aeruginosa culture was confirmed by high-performance liquid chromatography.     

Enzyme-linked immunosorbent assay for T-2 toxin metabolites in urine

A direct competitive enzyme-linked immunosorbent assay (ELISA) for determination of total T-2 toxin metabolites in urine was developed. The assay involves coating anti-3-acetyl-neosolaniol-hemisuccinate-bovine serum albumin conjugate (anti-3-Ac-NEOS-HS-BSA) antibody to the ELISA plate and using 3-Ac-NEOS-HS-peroxidase as the enzyme marker. Competitive ELISA revealed that the antibody had good cross-reactivity with acetyldiacetoxyscirpenol (Ac-DAS), T-2 tetraol tetraacetate, 3'-OH-Ac-T-2, 3-Ac-NEOS, and 3,4,15-triacetyl-12,13-epoxytrichothec-9-en-8-one (Ac-T-2-8-one), but less cross-reactivity with Ac-T-2 toxin and T-2 toxin. All metabolites of T-2 toxin in urine were converted to T-2 tetraol tetraacetate (T-2-4ol-4Ac) by acetylation of the sample extract before ELISA. To test the ELISA accuracy, a radioimmunoassay (RIA) was performed simultaneously. The linear portion of the standard curve of this direct ELISA for T-2-4ol-4Ac was 0.2-2.0 ng/mL, which was 10 times more sensitive than RIA. The minimum detection level for T-2-4ol-4Ac was 0.02 ng/mL (0.4 pg/assay) in the absence of urine sample. The overall analytical recoveries for T-2 toxin, HT-2, T-2-4ol, 3'-OH-HT-2, NEOS, and a mixture of these 5 toxins added to the urine samples in the ELISA at concentrations of 0.05 and 0.2 ng/mL were 87 and 94%, respectively.     

Development of ELISA and immunochromatographic assay for the detection of gentamicin

Competitive direct enzyme-linked immunosorbent assay (ELISA) and the immunochromatographic assay were developed using a monoclonal antibody to detect gentamicin in the animal plasma and milk. No cross-reactivity of the antibody was observed with other aminoglycosides based on competitive direct ELISA, indicating that the antibody is highly specific for gentamicin. On the basis of the standard curves, the detection limits were determined to be 0.9 ng/mL in phosphate-buffered saline (PBS), 1.0 ng/mL in plasma, and 0.5 ng/mL in milk, respectively. Recoveries of gentamicin from spiked plasma and milk at levels of 25-100 ng/mL ranged from 85 to 112%. The concentration of intramuscularly injected gentamicin was successfully monitored in the rabbit plasma through competitive direct ELISA. The detection limits were estimated to be about 6 ng/mL of gentamicin in PBS, plasma, and milk using the colloidal gold-based immunochromatographic assay, which is suitable for the simple screening of gentamicin residues in the veterinary field. Observed positives can be confirmed using a more sensitive laboratory method such as competitive direct ELISA. Therefore, the assays developed in this study could complement each other as well as veterinary field and laboratory findings.     

Monoclonal antibody-based enzyme linked immunosorbent assay of aflatoxin B1, T-2 toxin, and ochratoxin A in barley

Aflatoxin B1 (B1), T-2 toxin (T2), and ochratoxin A (OA) were assayed in a single extract from barley grain by using competitive enzyme linked immunosorbent assays (ELISAs) with monoclonal antibodies. B1 and T2 monoclonal antibodies were conjugated to horseradish peroxidase for direct competitive ELISA while an indirect competitive ELISA was used for OA determination. The competitive ELISA detected 0.1 ng/mL of B1, 10 ng/mL of T2, or 1 ng/mL of OA. Acetonitrile-0.5% KCl-6% H2SO4 (89 + 10 + 1) extracts of barley grain either were diluted 1:10 for direct assay or were subjected to a simple liquid-liquid cleanup procedure to concentrate the extract 10:1 before assay. For cleanup, water was added to the acetonitrile extract to partition water-soluble interfering substances, and then the mycotoxins were re-extracted with chloroform. The chloroform extract was evaporated to dryness and redissolved in Tris HCl buffer for ELISA. The mean recoveries from barley spiked with 4-60 ng/g of B1, 50-5000 ng/g of T2, and 5-500 ng/g of OA were, respectively, 93.8, 80.6, and 95.8%. The mean within-assay, inter-assay, and subsample coefficients of variation by ELISA of barley grain colonized with toxigenic fungi were less than 12% for B1 and OA but as high as 17% for T2.     

Comparison of radioimmunoassay and enzyme-linked immunosorbent assay for determining aflatoxin M1 in milk

Using a highly specific antibody against aflatoxin M1, a radioimmunoassay (RIA) and an enzyme-linked immunosorbent assay (ELISA) were developed for the quantitation of M1 in milk. RIA was sensitive in the range of 5-50 ng per assay but was subject to interference by whole milk. Extraction and cleanup were therefore necessary for the detection of M1 in milk at 0.5 ng/mL. An ELISA procedure was developed by using an aflatoxin M1-carboxymethyl-horseradish peroxidase conjugate as the ligand. Competitive assays revealed that this system was relatively more sensitive for M1 than for B1, and had a much lower degree of cross-reactivity for aflatoxins B2, G1, G2, B2a, and aflatoxicol. As low as 0.25 ng M1/mL in artificially contaminated milk (raw, whole, skim) could be detected by ELISA in 3 h without extraction or cleanup. Because of its simplicity, sensitivity, and specificity, ELISA is the preferred method for monitoring aflatoxin M1 in milk.     

Monoclonal antibody-based ELISA and colloidal gold immunoassay for detecting 19-nortestosterone residue in animal tissues

This article presents the generation of monoclonal antibodies (mAbs) with high specificity against 19-nortestosterone (NT) through cell fusion techniques and the development of a mAb-based indirect competitive ELISA (icELISA) method and colloidal gold-based immuno-chromatographic assay to detect NT residues in beef and pork samples. A modified carbodiimide method was employed to synthesize the artificial antigen, and BALB/c mice were used to produce anti-NT mAbs. On the basis of the checkerboard titration, an indirect competitive ELISA standard curve was established. This assay was sensitive and had a linear range from 0.03 to 38 ng/mL in phosphate buffered saline (PBS), with IC(50) and LOD values of 0.52 ng/mL and 0.01 ng/mL, respectively. Of all the competitive analogues, the produced mAb exhibited a high cross-reactivity to 17α-nortestosterone (83.6%), the main metabolite of NT in animal tissues. Except for moderate cross-reactivities with trenbolone (22.6%) and β-boldenone (13.8%), the other interference to the assay was negligible (<0.05%). In contrast, the strip test had a visual detection limit of 1 ng/mL in PBS, 2 μg/kg in beef, and 2 μg/kg in pork, respectively, and the results can be judged within 10 min. The ELISA and GC-MS results showed close correlation in beef (R2=0.9945) and in pork (R2=0.9977). Therefore, the combination of two immunoassays provides a useful screening method for quantitative or qualitative detection of NT residues in animal-origin products.     

Application of an ELISA to the determination of benalaxyl in red wines

The applicability of an ELISA for detection and quantification of benalaxyl in red wine samples is described. The study of the influence of this matrix on the reliability of the assay indicates that red wine samples require a rapid and simple cleanup step before ELISA assay. Recovery and precision of the method were evaluated by spiking red wine samples with benalaxyl in the 0.5-24 ng/mL range. Benalaxyl can be determined with good accuracy and precision up to 0. 5 ng/mL in starting red wine samples (detection limit of 0.13 ng/mL). No false negative or positive results were obtained. Authentic red wine samples were analyzed by ELISA and by RP-HPLC. The amounts of benalaxyl found by ELISA were in good agreement with RP-HPLC analysis.     

Rapid enzyme-linked immunosorbent assay and colloidal gold immunoassay for kanamycin and tobramycin in Swine tissues

A monoclonal antibody (Mab) was produced by using the kanamycin-glutaraldehyde-bovine serum albumin (Kan-GDA-BSA) conjugate as the immunogen. The anti-Kan Mab exhibited high cross-reactivity with tobramycin (Tob) and slight or negligible cross-reactivity with other aminoglycosides. The specificity and cross-reactivity of this Mab are discussed regarding the three-dimensional, computer-generated molecular models of the aminoglycosides. Using this Mab, a rapid enzyme-linked immunosorbent assay (ELISA) and a colloidal gold-based strip test for Kan and Tob were developed. The rapid ELISA showed a 50% inhibition value (IC 50) of 0.83 ng/mL for Kan and 0.89 ng/mL for Tob with the analysis time less than 40 min, and the recoveries from spiked swine tissues at levels of 25-200 microg/kg ranged from 52% to 96% for Kan and 61% to 86% for Tob. In contrast, the strip test for Kan or Tob had a visual detection limit of 5 ng/mL in PBS, 50 microg/kg in meat or liver, and 100 microg/kg in kidney, and the results can be judged within 5-10 min. Observed positive samples judged by the strip test can be further quantitated by ELISA, hence the two assays can complement each other for rapid detection of residual Kan and Tob in swine tissues. Moreover, physical-chemical factors that affected the ELISA and strip test performance were also investigated. The effect of pH and antibody amount for gold-antibody conjugation on the strip test sensitivity was determined followed by a theoretical explanation of the effects.     

Development of an immunoassay for the residues of the herbicide bensulfuron-methyl

To develop a competitive indirect enzyme-linked immunosorbent assay based on polyclonal antibodies for the detection of the sulfonylurea herbicide bensulfuron-methyl, seven structurally related haptens were synthesized. Four of them mimicking the target analyte were conjugated to keyhole limpet hemocyanin by the N-hydroxysuccinimide activated ester method to use as immunogens, and all of them were conjugated to bovine serum albumin to use as plate-coating antigens. Polyclonal antibodies raised in rabbits and the coating antigens were screened and selected for the assay in simple homologous and heterologous ELISA formats. Three sensitive heterologous ELISAs were selected and optimized, showing the average IC(50) values of bensulfuron-methyl as low as 0.17, 0.09, and 0.09 ng/mL, the detection ranges of 0.04-0.60, 0.01-0.60, and 0.04-0.25 ng/mL, and the lowest detection limits of 0.03, 0.002, and 0.03 ng/mL, respectively. The cross-reactivities of other sulfonylurea herbicides and metabolites of bensulfuron-methyl to the antibodies were less than 15% in the two assays. Recoveries from the analyte-fortified water samples in assay I were in the range of 81-125% by simple dilution. The correlation between the ELISA and HPLC was 0.999 (n = 15) with a slope of 1.37 in the analysis of groundwater samples fortified with bensulfuron-methyl. The results obtained strongly indicate that the ELISA can be a highly sensitive and convenient tool for detecting bensulfuron-methyl residues in agricultural and environmental samples.     

Fluorescence polarization as a means for determination of fumonisins in maize

Fumonisins, mycotoxins produced by certain species of Fusaria, are commonly found worldwide as contaminants in maize. This paper reports the development of a rapid, portable fluorescence polarization-based assay for fumonisins in maize. The assay was based on the competition of unlabeled fumonisin, from a sample, with a fluorescently tagged fumonisin (FB(1)-FL) for a fumonisin-specific monoclonal antibody in solution. The fluorescence polarization (FP) of the tagged fumonisin was increased upon binding with the antibody. In the presence of free toxin, less of the FB(1)-FL was bound and the polarization signal was decreased. The assays were very simple to perform, requiring only mixing of an aqueous extract of maize with the tagged fumonisin and antibody, and required <2 min per sample, excluding extraction time. Two permutations of the assay were tested, one with each sample matrix serving as its own blank, and the other with all of the samples compared relative to a PBS blank with normalization of the data similar to an ELISA. The limit of detection, defined as the toxin content associated with a fluorescence polarization signal 5 standard deviations from that of a fumonisin-free control, was 0.5 microg of FB(1)/g in spiked maize. Recoveries from spiked maize over the range of 0.5-20 ppm averaged 94.3 +/- 13.8%. Forty-eight samples of field-contaminated maize were tested by the FP and an established HPLC method, with a good correlation between the two (r(2) = 0.85-0.88). For these samples, the two variations of the FP assay also compared well to one another (r(2) = 0.97), suggesting the assay principle is very robust. The results, combined with the speed and ease of use for the assay, suggest that this technology has substantial potential as a screening tool for mycotoxins in foods.     

Development and fast screening of salbutamol residues in swine serum by an enzyme-linked immunosorbent assay in Taiwan

The analysis of salbutamol in swine serum is the more practical basis for large scale surveillance programs in Taiwan. Objectives of the study were to develop a new assay and to compare with a commercially available kit in field test screens. A simple and reliable enzyme-linked immunosorbent assay (ELISA) to monitor the presence of beta-agonist, salbutamol, in 1,358 field samples of swine serum that were collected from local meat markets was described. The method proved to be suitable and sensitive for the detection of beta-agonist residues caused by growth promoting dosage. The limit of detection of the developed ELISA directly performed on diluted serum was 0.25 ng/mL. The application and the results of two ELISA kits (homemade and commercially available) for the screening of salbutamol were presented. For further confirmation, all samples that showed to be ELISA positive for salbutamol residues were analyzed by GC-MS. Adopting 1 ng/mL salbutamol as a cutoff value, the commercial beta-agonist ELISA had a sensitivity of 89.2% and a specificity of 86.7% versus GC-MS at a cutoff of 1 ng/mL. The homemade salbutamol ELISA had a sensitivity of 81.1% and a specificity of 98.6% and gave a low proportion of false-positive rate results. The reliability of the developed kit in terms of the percentage of false-positive rate results is evaluated. In conclusion, a sensitive, specific salbutamol ELISA has been developed that could serve as a rapid screening assay, and the detection of positive samples at the place of sampling can result in more effective control of the illegal use of beta-agonists.     

Hapten syntheses and antibody generation for a new herbicide, metamifop

To develop a competitive indirect enzyme-linked immunosorbent assay for metamifop, a new aryloxyphenoxypropionic acid herbicide, three structurally related haptens were synthesized. Hapten conjugates to keyhole limpet hemocyanin and bovine serum albumin were used as immunogens and plate-coating antigens, respectively. Various sets of polyclonal antibodies from rabbits and the coating antigens were screened for the assay in simple homologous and heterologous ELISA formats. A selected heterologous ELISA was optimized to show an average IC50 value as low as 20.1 ng/mL, detection ranges of 1.0-350 ng/mL, and a lowest detection limit of 0.1 ng/mL. The cross-reactivities of other aryloxyphenoxypropionic acid herbicides to the antibodies were less than 0.5% in the assays except fenoxaprop-P and fenoxaprop-P ethyl, having a diaryl ether group identical to that of metamifop. Molecular modeling studies revealed that the physicochemical properties of the diaryl ether group are the most important determinants of sensitivity and selectivity. The results strongly indicate that the selected set of ELISA is a highly sensitive and convenient tool for detecting metamifop.     

Development of a fluorescent latex microparticle immunoassay for the detection of staphylococcal enterotoxin B (SEB)

Staphylococcus aureus enterotoxin B (SEB) is a highly heat resistant enteric toxin with a potential as a biothreat agent. A sensitive method for the detection of staphylococcal enterotoxins is needed for food safety and food defense monitoring. The objectives of this research were to develop a competitive fluorescent immunoassay with detection of SEB below toxic levels of 1 ng/mL and to minimize sample preparation. Anti-SEB was immobilized onto carboxylated polystyrene microparticles, and SEB was labeled with fluorescein isothiocyanate (FITC). The concentrations of these reagents were optimized for the detection of SEB below 1 part per billion (1 ng/mL), and other assay conditions (sample volumes and incubation periods) were optimized. Drinking water and milk samples were spiked with 0.125-10 ng/mL SEB and were equilibrated overnight prior to analysis. The water and milk samples were directly analyzed, but heating the milk samples for 10 min at 90 degrees C improved the assay performance. SEB in samples bound with the anti-SEB linked to the latex followed by the competitive binding of SEB-FITC tracer. The excess, unbound tracer was separated by centrifugation, and the fluorescence density of the supernatant was measured. SEB was detected at levels as low as 0.125 ng/mL in drinking water and 0.5 ng/mL in whole milk. This fluorescent latex particle immunoassay will be utilized for the detection of SEB in various foods matrices.     

检测多杀性巴氏杆菌毒素抗体的单抗竞争ELISA方法的建立

本研究以纯化的多杀性巴氏杆菌毒素基因片段的原核表达产物作为抗原免疫小鼠制备单抗,并利用表达蛋白和多杀性巴氏杆菌毒素单抗酶结合物建立了竞争ELISA方法检测多杀性巴氏杆菌毒素抗体。经过研究确定抗原包被浓度为223ng／mL,待检血清最佳稀释度为1：2,酶标单抗工作浓度为1：3200,血清抑制率大于50%为阳性。应用单抗竞争ELISA和细胞毒性中和试验同时对82份血清进行猪多杀性巴氏杆菌毒素抗体检测,竞争ELISA的检出率为40.2％,细胞毒性中和试验检出率为36.6％,两者符合率达91.5％。试验结果表明,该ELISA方法特异性强,敏感性高,稳定性和重复性好,操作简便。本方法的建立在实验室诊断的标准化、猪群萎缩性鼻炎疫苗免疫效果的评价及流行病学调查方面具有应用价值。     

Development and comparison of three diagnostic immunoassay formats for the detection of azoxystrobin

The currently accepted method of detection for azoxystrobin, a strobilurin fungicide, involves a labor-intensive organic solvent extraction and gas chromatography analysis. Three diagnostic assay formats, i.e., enzyme-linked immunosorbent assay (ELISA), fluorescence polarization (FP), and time-resolved fluorescence (TR-FIA), were developed and compared with regard to detection and quantification of azoxystrobin in grape extract and river, lake, and well water samples. These three assay formats require no initial sample extraction and were not affected by any of the environmental matrices tested, and each had a linear working range of 0-400 pg/mL. The polyclonal antibodies used for each of the immunoassays were specific to azoxystrobin; that is, the highest cross-reactivity to other pesticides observed was 5.7%. The limits of detection of the immunoassays were similar at 3 (ELISA), 46 (FP), and 28 (TR-FIA) pg/mL, as were the respective IC50 values of 306, 252, and 244 pg/mL. Each of the three immunoassays developed was less labor-intensive and approximately 100-fold more sensitive than the gas chromatographic method. While the three formats were comparable in terms of performance, the fluorescence polarization assay was the least labor-intensive and required the least time to perform.     

Indirect enzyme-antibody sandwich enzyme-linked immunosorbent assay for quantification of TAXI and XIP type xylanase inhibitors in wheat and other cereals

To quantify Triticum aestivum xylanase inhibitor (TAXI) and xylanase inhibiting protein (XIP) type proteins in cereals in general and wheat ( T. aestivum) in particular, a robust enzyme-linked immunosorbent assay (ELISA) using an uncommon enzyme-antibody sandwich format was developed. Bacillus subtilis glycoside hydrolase family (GH) 11 and Aspergillus oryzae GH 10 xylanases were selected for coating ELISA plate wells to capture TAXI and XIP, respectively, prior to probing with antibodies. The detection threshold of the developed ELISA was much lower than that of the currently used xylanase inhibitor assay and the recently described Western blot approach. Because of its broad dynamic range (TAXI, 30-600 ng/mL, and XIP, 3-60 ng/mL), one proper standard extract dilution can be used for analyzing different wheat varieties, whereas for the currently used colorimetric assay, often different dilutions need to be analyzed. The TAXI ELISA for wheat was successfully adapted for barley ( Hordeum vulgare) and could also be used for other cereals.     

Development of an immunoassay (ELISA) for the quantification of thiram in lettuce

Two competitive immunoassays, a laboratory assay based on microwell plates and a field test based on the use of polystyrene tubes, have been developed for the quantification of thiram in lettuces. Concerning the laboratory assay, the calibration curve for thiram had a linear range of 11 to 90 ng/mL and a detection limit of 5 ng/mL. Precision of the assay presented coefficient of variation values <9% and the recovery of thiram from lettuce averaged 89% across the range of the immunoassay method using 30 min extraction with water/acetone (50:50, v/v). The tube-based method was developed in order that an extract of lettuce, containing thiram at the MRL (8 ppm), would be found on the linear part of the standard curve. The calibration curve for thiram has a linear range of 100 to 800 ng/mL (1.39 to 11.1 ppm in lettuce) and a detection limit of 40 ng/mL.     

Production and characterization of a monoclonal anti-anti-idiotype antibody against fumonisin B(1).

A monoclonal anti-anti-idiotype antibody (mAb3) against fumonisin B(1) (FmB1) was produced from the hybridoma cell line 7C7F4, which was generated by the fusion of P3/NS-1/1-AG4-1 myeloma cells with spleen cells isolated from a Balb/c mouse that had been immunized with the Fab fragments of affinity-purified anti-idiotype antibodies. The mAb3 belongs to the immunoglobulin M, kappa light chain. A direct competitive enzyme-linked immunosorbent assay (dc-ELISA) and an indirect competitive ELISA (idc-ELISA) were established for antibody characterization and toxin analysis. In an idc-ELISA using FmB1-ovalbumin (OVA) as the coating antigen, the concentrations causing 50% inhibition of binding (IC(50)) of mAb3 to the solid-phase FmB1-OVA by free FmB1, FmB2, and FmB3 were found to be 75, 95, and 450 ng/mL, respectively. In the dc-ELISA, the concentration causing IC(50) of FmB1-horseradish peroxidase to the solid-phase mAb3 by free FmB1 was found to be 233 ng/mL. Analysis of 12 samples naturally contaminated with fumonisins with mAb3-based idc-ELISA and polyclonal-based dc-ELISA showed a good correlation between these two methods with a correlation coefficient of 0.76 at p < 0.02. The linear regression slope was found to be y[polyclonal ELISA] = 0.87x[mAb3 ELISA] - 52 ppb.     

Development of an enzyme-linked immunosorbent assay for the detection of pentachloronitrobenzene residues in environmental samples

A competitive enzyme-linked immunosorbent assay (ELISA) for pentachloronitrobenzene (PCNB), a fungicide and chemical intermediate, was developed using a polyclonal antiserum produced against a hapten-protein conjugate of pentachlorophenoxypropionic acid-bovine serum albumin (BSA). An indirect competitive ELISA of PCNB showed an IC50 of 37 ng/mL and a limit of detection (LOD) of 7 ng/mL. The ELISA can tolerate up to 10% (v/v) methanol, 5% (v/v) acetonitrile, or 5% (v/v) acetone without significant fluctuation of Amax and IC50. The assay sensitivity showed little change in a range of pH from 6 to 8 and concentrations of 0.05-0.2 M NaCl in the assay buffer. Very low cross-reactivities were observed for some structurally related compounds except for hexachlorobenzene (12%). The average recoveries of PCNB from fortified well water, river water, and soil samples were in ranges of 88-94, 80-91, and 70-81%, respectively. The correlations between the gas chromatographic and ELISA results were excellent (r 2 >or= 0.97, slopes from 0.86 to 1.10) for those fortified samples. The ELISA is a good alternative tool for monitoring PCNB residues in environmental samples.     

Enzyme immunoassay for mycophenolic acid in milk and cheese

Mycophenolic acid (MPA) was reacted with N-hydroxysuccinimide and conjugated to keyhole limpet hemocyanin (KLH), and to horseradish peroxidase (HRP), respectively. The MPA-KLH was used to produce anti-MPA antiserum in rabbits. A competitive direct enzyme immunoassay (EIA) for MPA was established with anti-MPA antiserum and MPA-HRP conjugate. The mean 50% inhibition and detection limit of MPA standard curves (n = 103) were 197 +/- 67 and 81 +/- 48 pg/mL, respectively. The EIA was specific for MPA and its synthetic 2-morpholinoethyl ester, mycophenolate mofetil (91% relative cross-reactivity). Raw bulk milk and pasteurized milk, with and without beta-glucuronidase pretreatment, were analyzed by EIA. No MPA was found in milk, at a detection limit of 100 pg/mL (recovery 58-66% at 0.125-2 ng/mL). Blue-veined cheese from the German market (n = 53) was analyzed by EIA, and the detection limit was at 0.5 ng/g (recovery 68-79% at 5-100 ng/g). All but two cheeses contained MPA, although mostly (66%) at levels of <10 ng/g. MPA at 400-1200 ng/g was found in Roquefort cheeses. Highest levels (4-11 microg/g) were found in a German soft cheese preparation. MPA levels in mycelium-rich parts of cheese were 3 times higher than in mycelium-free parts.