The effect of different sample preparatory protocols on the induction of the aryl hydrocarbon receptor (AhR) in the H4IIE–luc reporter gene bio–assay

Abstract

Concern on a global scale gave rise to the founding of the Stockholm Convention on persistent organic pollutants (POPs) with a view to restrict the use and production of these toxic chemicals. As a signatory, South Africa is legally bound to abide to the Convention’s objectives, including participating in relevant research and monitoring. However, developing countries such as South Africa have limited information concerning POPs, partially because these countries do not have sufficient analytical capabilities, and thus method development and refinement are necessary. One group of POPs consisting of polychlorinated dibenzo-pdioxins (PCDDs), polychlorinated dibenzofurans (PCDFs) and dioxin-like polychlorinated biphenyls (dl-PCBs), collectively referred to as dioxins, are of particular concern due to their high toxicity and persistence. Additionally, the analysis of dioxins is recognised as one of the most analytically challenging of its kind. This study investigated the effect of different preparatory protocols on the semi-quantification of dioxins using the H4IIE-luc-reporter gene assay. The protocols evaluated were either Soxhlet or pressurised liquid extraction (PLE) combined with a manual acid digestion, gel permeation chromatography (GPC) and Florisil fractionation clean-up procedure as well as the automated Total Rapid Prep™(TRP) system which makes use of a PLE combined with a multi-layer silica, alumina and carbon column clean-up procedure. To evaluate the protocols, an eight point matrix matched calibration curve, two soil samples and a certified reference material (CRM) were used. The extracts were semi-quantified by the H4IIE-luc bio-assay. During the course of the assay, the appropriateness of different standards was investigated, and a mixed standard containing all 17 toxic PCDD/Fs was chosen for quantification. During the data review process, higher bioassay equivalent (BEQ) values were obtained from PLE compared to Soxhlet extraction, while no statistically significant difference (Kruskal-Wallis ANOVA: p > 0.05) was found between the assay quantifications for the different preparatory techniques. However, the results of the H4IIE bio-assay were larger than the expected values. The identity of the chemicals that were in all likelihood responsible for the higher response was investigated through instrumental analysis using comprehensive two-dimensional gas chromatography coupled to time of flight mass spectrometry (GCxGC-TOFMS). Instrumental results indicated a high level of PAHs in the extracts, which could lead to super induction of the aryl hydrocarbon receptor (AhR) and therefore, to a positive bias in the results. Instrumental screening proved that all selected preparatory protocols were inadequate at removing interfering compounds and not sufficiently selective for PCDD/Fs, although the TRP was more successful in removing interferences. The high matrix interference hindered peak identification. Additionally, as indicated by instrumental analysis, the weak recovery of PCDD/Fs could be ascribed to high evaporation temperatures. The effect of different reference standards in the H4IIE bio-assay used during semi-quantification needs further investigation; similarly, the optimisation of extraction, evaporation and clean-up protocols and the use of different GCxGC-TOFMS column combinations aimed at more efficient separation needs to be investigated. The assistance of the National Metrology Institute (funded through the Department of Trade and Industry) towards this research is hereby acknowledged.