J. zyxwvuts Agric. Food Chem. zyxwvut 1992, 40, 1471-1474 1471 Mono- and Polyclonal Antibodies to the Organophosphate Fenitrothion. 2. Antibody Specificity and Assay Performance Amanda S. Hill,+ Helen L. Beasley,+David P. McAdam,l and John H. Skerritt'3t CSIRO Division of Plant Industry, zyxwvu P.O. Box 7, North Ryde, NSW 2113, Australia, and CSIRO Division of Plant Industry, G.P.O. Box 1600, Canberra, ACT 2601, Australia The performance and specificities of antibody-based assays for fenitrothion were studied. Fenitrothion could be detected in grain in assays with either polyclonal or monoclonal antibodies using either immobilized antibody or immobilized hapten-protein conjugates. Most assay formats and antibodies distinguished fenitrothion from structurally similar organophosphates, and variable extents of cross- reaction with metabolites were observed. The concentration of fenitrothion in wheat grain samples was measured in three assay formats with two antibodies, and in each case good correlations were obtained with the results obtained by gas-liquid chromatography. For routine use, a format detecting 3 ng and suitable for analysis over the range 100 ppb-20 ppm in grain was preferred. INTRODUCTION The preceding paper (McAdam et al., 1992) described the development of several enzyme immunoassays for the organophosphate fenitrothion (FN) that were sensitive to low nanogram amounts of this compound. Our purpose for developing these antibodies followed from the wide- spread use of fenitrothion as a grain protectant (Desmarch- elier et al., 1977; Snelson, 1987). With many grain storages being treated annually with FN, there is a need for less expensive high-throughput laboratory methods for de- termination of FN residues. In this paper, several aspects of the performance of these assays are investigated, including (1) cross-reaction with analogs of FN, FN metabolites, other stored-grain pes- ticides, and other organophosphates, (2) effects of solvents required for FN extraction from grain, and (3) effects of coextractives in grain matrices on assay performance. Our results indicate that ELISA assays based on these anti- bodies can accurately quantify fenitrothion in wheat samples at commercially relevant (0.1-20 ppm) levels. MATERIALS AND METHODS Antibodies, Fenitrothion Conjugates, and ELISA Meth- ods. The preparation and properties of the antibodies and FN- HRP and FN-protein conjugates used in this study, as well as the ELISA methods used, are described in the preceding paper (McAdam et al., 1992). Standard curves for each assay are also shown in that paper. Grain Samples. In initial work, samples (500 g) of pesticide- free wheat were spiked with small amounts (0.5-5 mL) of an ether solution of FN to yield approximately3,9, and 14 ppm in the grain. The samples were stored at 20 "C in the dark for 9-12 weeks before analysis. Such "aging" is necessary for the pesticide to achieve constant and reproducible distribution within and between grains and to exhibit extractability properties similar to those of field samples (Sharp et al., 1988). Samples of grain treated commercially at receival with fenitrothion and stored for 3-18 months at the various elevators were obtained from the Australian Wheat Board, Melbourne. Extraction. Previous work has indicated that virtually quantitative extraction of FN from whole wheat could be obtained by standing the grain in 2.5 volumes of methanol for 36-48 h with * Author to whom correspondence should be addressed. t North Ryde. t Canberra. 002 1-856 1 I921 1440-1 47 1$03.00/ 0 intermittent (3 X 5 min at 100 rpm) shaking (Desmarchelier et al., 1977). This result was confirmed by GLC in the current study (not shown), and this extraction method was therefore used. Test Compounds. Pesticide standards were purchased from Chem Service (West Chester, PA). FN metabolites were syn- thesized as follows: Demethylfenitrothion [O-methylO-(3-meth- yl-Cnitrophenyl) phosphorothioate] and fenitrothion S isomer [O,S-dimethyl 0-(3-methy1-4-nitrophenyl) phosphorothioate], were synthesized according to the method of Eto et al. (1968). These products were purified by column chromatography and analyzed by lH NMR. Fenitro-oxon was prepared by reaction of fenitrothion (3.8 mmol) in 15 mL of ethanol-5 mL of water with bromine (18.8 mmol in 10 mL of ethanol) for 1 h at room temperature. The mixture was extracted in ethyl acetate, and the organic layer was washed (water, 0.1 M Na&05, HzO, brine), dried over MgS04, and then concentratedto an oil. Silica column chromatography (eluted with chloroform) gave the product as an oil (650 mg, 69% yield). Proton NMR (200 MHz): 6 8.04 (d, zy JH~,~ = 9.5 Hz, H5), 7.21 (m, 2 H, Hz,~), 3.91 (d, JP,H = 11.5 Hz, OCH3), 2.63 (8, CH,); infrared (1280 cm-l) P=O. 3-Methyl-4- nitrophenol was obtained from Aldrich (Milwaulkee, WI). RESULTS Solvent Effects on Assays. The functional limit of detection for fenitrothion in grain matrices will also depend on the maximum amount of solvent that the assay will tolerate, which in turn affects the dilution factor for the grain extracts. Fenitrothion may be extracted from grain using either nonpolar water-immiscible solvents (e.g., hex- ane) or water-miscible solvents such as acetone, acetoni- trile, ethanol, or methanol (Sharp et al., 1988). A solvent was considered suitable at 10% final concentration if it did not significantly change the ELISA control absor- bance or alter inhibition by free FN. With assays using anti-YNF hound to the solid phase, solvents such as acetone (10%) or methanol (5-20% ) did not reduce ELISA control values but did reduce the inhibitory potency of FN. Others, such as 20% 2-propanol, affected both parameters. Previous studies indicated that methanol was an efficient extractant, especially if grain samples were exposed to the solvent for 36-48 h (Desmarchelier et al., 1977). Immunoassays were less affected by methanol than by other solvents (Table I). Therefore, 10% methanol was routinely used for subsequent work. Use of methanol had the advantage that for confirmatory gas chromato- graphic analyses the same extract may be injected directly into the instrument. 0 1992 American Chemical Society