Environ. Sci. Technol. 2005, 39, 2786-2794
Evaluation of a Novel Malathion Immunoassay for Groundwater and Surface Water Analysis E V A M . B R U N , M A R T A G A R C EÄ S - G A R C IÄ A , M A . J O S EÄ B A N ˜ U L S , J O S EÄ A . G A B A L D O Ä N, ROSA PUCHADES,* AND AÄ N G E L M A Q U I E I R A * Departamento de Quı´mica, Universidad Polite´cnica de Valencia, Camino de Vera s/n, 46022 Valencia, Spain
Malathion is an organophosphorus insecticide commonly used in crops and indoor applications. Negative effects of malathion on human health and ecosystems are of growing concern. In this work, novel malathion haptens are synthesized to develop an ELISA screening method. The immunoassay is based on a conjugate-coated format and shows a limit of detection of 0.11 ng/mL, an IC50 of 1.58 ng/mL, a dynamic range between 0.23 and 10.94 ng/ mL, and a cross-reactivity of 5000 BSA-MA 5 >5000
paring fragmentary haptens MA-2 and MA-4, which contain different parts of the malathion structure, the first one provides more sensitive antibodies. This indicates that the dithiophosphate group is the most antigenic determinant in the molecule. Nevertheless, hapten MA-5, which has a dithiophosphate group, a P-NH-C bond instead of P-S-C, and a spacer arm, rendered the most sensitive sera. The best assay was achieved using hapten MA-4 as the coating antigen, probably due to the heterology. It may result in an appropriate equilibrium between coating antigen and target compound recognition. On this point, haptens MA-2 and MA-5 were chosen as the best for immunization purposes, and their conjugation to cBSA and KLH was assayed to obtain better antibodies since these proteins are supposed to be more immunogenic. All of the sera obtained using cBSA showed low titers and only when OVA-MA-1 or OVA-MA-3 were used as the coating antigens. No competitive assays were achieved in any format. Precipitation problems happened when MA-2 was conjugated to KLH; thus, an immunogen could not be obtained. Sera produced with KLH-MA-5 were tested against OVAhapten and BSA-hapten in the conjugate-coated format. As can be seen in Table 1, sera from the KLH conjugate showed a higher titer than those obtained with BSA. Besides, the titer was higher when BSA was used instead of OVA as the coating protein. In the antibody-coated ELISA format only the conjugate HRP-MA-5 with serum KMLT-5 II showed a high titer. A low titer was obtained with serum KMLT-5 I. All the combinations that showed specific recognition were used to carry out competitive assays. No competitive
curves were obtained in the antibody-coated format, although it is known (27) that the format has a strong influence on the performances (sensitivity, selectivity, assay time) of the ELISA. In this case, high recognition between the tracer and sera may be the main reason for no competition. The best results were achieved with sera raised from the KLH conjugate immunogen (Table 2). The KMLT-5 I (1/1000)/ BSA-MA-2 (30 ng/mL) combination was the most sensitive and was selected to carry out the optimization assays. ELISA Optimization. The influence of several parameters (pH, ionic strength, surfactant concentration, and incubation time) on the ELISA performance was examined. The effect of pH in the range 5.0-7.0 was evaluated, as malathion hydrolyzes out of this interval. It was observed that the immunoassay for malathion is less sensitive (higher IC50) at pH 7.0 (Figure 3a) than at pH 5.0. However, pH 5.0 is very close to the degrading pH. When the assay is carried out between pH 5.5 and pH 6.5, sensitivity is good and the signal (A0) increases with pH, so pH 6.5 was selected for this immunoassay. The effect of the ionic strength on the assay performance is also shown (Figure 3b). The results indicate that the maximum ability to recognize the coating conjugate (maximum signal, A0) was achieved with 5 mM PBS, decreasing gradually with the buffer concentration. However, the lowest IC50 was obtained when the salt concentration was the highest. In conclusion, 20 mM was chosen as a compromise between signal and sensitivity. Using the established pH and buffer concentration, the effect of Tween 20 as surfactant on the signal and sensitivity was assayed. In this study (Figure 3c), a negative influence (lower signal and lower sensitivity) was observed if the surfactant concentration increased. According to these results, no Tween 20 was added in the competition buffer. Finally, the study of the incubation time on the competition step (Figure 3d) indicated that 15 min gave the most sensitive assay with a good signal. Figure 4 depicts the calibration graph of the competitive optimized ELISA obtained for malathion. The assay shows an IC50 of 1.58 ng/mL, a working range between 0.23 and 10.94 ng/mL, and an LOD of 0.11 ng/mL. In this sense, it is worth mentioning the better sensitivity achieved in comparison with that of the immunoassay reported by Nishi et al (15), though different coating haptens and polyclonal antibodies were used. Since the use of organic solvents is unavoidable for the extraction of the pesticide during the preparation of environmental samples employing reference methods (9), it is desirable to assess the effect of these media on ELISA performance. The effect of different percentages of acetone, acetonitrile, 2-propanol, methanol, and methyl sulfoxide in the assay buffer was studied. 2-Propanol and methanol were tolerated up to 10% with IC50 at the nanogram per milliliter level, while acetone, acetonitrile, and methyl sulfoxide were tolerated up to 4%. These are interesting results since when SPE is applied, analyte elution with an organic solvent is mandatory. Cross-Reactivity Studies. Assay selectivity was evaluated by determining the cross-reactivity with a variety of structurally related compounds. As is shown in Table 3, the interferences observed were negligible. Thus, the developed immunoassay for malathion is specific (maximum CR < 10%) against different organophosphorus insecticides and the main malathion metabolitesmalaoxonsfor which the highest interference was obtained (1.3%). Analysis of Water Samples. Nine groundwater samples were spiked with malathion and analyzed by optimized ELISA and GC-MS techniques as blind samples. Nonspiked samples were below the LOD of malathion for both methods. VOL. 39, NO. 8, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2791
FIGURE 3. Influence of pH (a), buffer concentration (PBS) (b), Tween 20 (c), and incubation time (d) on the competition step of malathion ELISA.
FIGURE 4. Calibration curve for the optimized malathion immunoassay. Mean values ( standard deviation (n ) 3). Then, different groundwater samples spiked at six levels of malathion were quantified by ELISA, and a mean recovery of 98% was obtained. Also, good correlation with GC-MS values (99% mean recovery) was reached (Table 4). These results show that the developed assay is promising as a screening procedure for malathion residues, as well as a sensitive quantitative method. 2792
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 8, 2005
Jucar river water samples (see the Experimental Section) were studied to know the performance of the ELISA for malathion. Different properties were measured in the waters sampled at 10 selected points. Among them, the electric conductivity varied from 815 µS/cm at the point least influenced by human activities to 1200 µS/cm, which is maintained at the middle points, and went to 18190 µS/cm (608 mg of Cl-/L) in the estuarine zone, due to the marine intrusion. Dissolved organic matter initially was 1.3 mg of O2/L, remained at about 2.5 mg of O2/L at almost all the studied points, and finally reached 4.3 mg of O2/L in the proximity of the Mediterranean Sea. All the samples were also analyzed in parallel for malathion residues by immunochemical and GC-MS methods, being under the LOD for both techniques. The same waters were spiked with malathion and quantified as blind samples. As shown in Table 5, mean recovery values of 101% for the immunoassay and 103% for GC-MS were obtained, which confirms the huge potential of the malathion immunoassay for the analysis of environmental waters. The analysis of waters for residues of this organophosphorus through the optimized immunoassay was carried out in less than 2 h without cleanup or pretreatment. The effect of dissolved organic matter on the ELISA method was studied, because it is well-known that the organic matter (dissolved and suspended) content in surface waters may have a negative effect on the pesticide determination. For this, 100 mL of each water sample was mixed, obtaining a composed sample. This water (chemical composition shown in the Table 5 footnote) was spiked with humic acid sodium salt at concentrations ranging between 1 and 50 mg/ L. Then, the pH was adjusted to 6.5. All subsamples were increased till 5 ng/mL malathion and analyzed by ELISA.
TABLE 3. Cross-Reactivity of Malathion-Related Compounds compound
IC50 (ng/mL)
CR (%)
malathion malaoxon acephate chlorpyriphos diazinon dichlorvos fenthion fenitrothion
1.58 118 >5 × 104 >5 × 103 >2.5 × 104 >5 × 104 200 2 × 103
100 1.3 5 × 104 >5 × 104
