Analysis of dioxins in contaminated soils with the calux and caflux bioassays, an immunoassay, and gas chromatography/high-resolution mass spectrometry

Abstract

The chemically activated luciferase expression assay, the chemically activated fluorescence expression assay, and the enzyme-linked immunosorbent assay (ELISA) are all bioanalytical methods that have been used for the detection and quantification of polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans (PCDD/Fs). However, no comparisons of the results obtained by these three methods have been published analyzing identical replicates of purified sample extracts. Therefore, we have evaluated the performance of each of these methods for analyzing PCDD/Fs in aliquots of extracts from aged-contaminated soil samples and compared the results with those obtained by gas chromatography/high-resolution mass spectrometry (GC/HRMS). The quantitative performance was assessed and the effects of sample purification and data interpretation on the quality of the bioassay results were investigated. Results from the bioanalytical techniques were, in principle, not significantly different from each other or from the GC/HRMS data (p = 0.05). Furthermore, properly used, all of the bioanalytical techniques examined were found to be sufficiently sensitive, selective, and accurate to be used in connection with soil remediation activities when aiming at the remediation goal recommended by the U.S. Environmental Protection Agency (i.e., <1000 pg toxic equivalency/g). However, a site-specific correction factor should be applied with the use of the ELISA to account for differences between the toxic equivalency factors and the ELISA cross-reactivities of the various PCDD/F congeners, which otherwise might significantly underestimate the PCDD/F content.

Fig. 1

(a) Chemically activated luciferase expression (CALUX), (b) chemically activated fluorescence expression (CAFLUX), and (c) enzyme-linked immunosorbent assay (ELISA) dose–response curves for the standard 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) or 2,3,7-trichloro-8-methyl-dibenzo-p-dioxin (TMDD) and the wood 1 soil extract, illustrating the slight deviations from parallelism between the CA(F)LUX standard and sample extract curves and an incomplete ELISA sample extract curve. All data have been corrected for signals from a blank sample (wells treated with dimethyl sulfoxide).

Fig. 2

Comparisons of results of gas chromatography/high-resolution mass spectrometry (GC/HRMS) World Health Organization toxic equivalency (WHO-TEQ) reference values (pg/g dry wt) and chemically activated luciferase expression (CALUX)–TEQ (pg/g dry wt), chemically activated fluorescence expression (CAFLUX)–TEQ (pg/g dry wt), and enzyme-linked immunosorbent assay (ELISA) analysis of 2,3,7-trichloro-8-methyl-dibenzo-p-dioxin (TMDD) equivalents (pg/g dry wt). The GC/HRMS results were calculated with TEQ factors from Van den Berg et al. [7], including only dioxins. For comparison purposes, the ±30% uncertainty in the accredited GC/HRMS procedure used was included in the graph.

Fig. 3

Bioassay toxic equivalency (TEQ) concentration ranges for soil samples analyzed by (a) chemically activated fluorescence expression (CAFLUX) and (b) chemically activated luciferase expression (CALUX). The uncertainties in the determinations because of deviations from parallelism between the sample and standard curves are indicated by the size of the range. Data points outside the bars were derived by extrapolation beyond the empirical data. W = wood; C = chlor.