Agrochemicals related to maize crops in South Africa and associated biological effects : an in vitro study

Abstract

Although agrochemicals reduce pest-associated crop losses, these chemicals end up in non-target environments following rainfall and irrigation. When fractions of agrochemical mixtures are bioavailable, they pose a threat to human and environmental health. The aim of this study was to determine whether the water-soluble fraction of agricultural soils in South Africa elicit selected biological effects in vitro, as well as to determine which pesticides are likely causing the observed effects. Composite soil samples were collected in two maize growing regions—the Mpumalanga province and Vaalharts Valley—known for their extensive herbicide application. Water-soluble compounds were extracted from the soil using deionised water to mimic environmental conditions and obtain the bioavailable fraction. The H4IIE-luc reporter gene bioassay was performed to establish whether the soil-extracts contained aryl hydrocarbon receptor (AhR) ligands. This was followed by assessing (anti-)androgenic and glucocorticoid activity using the human breast carcinoma cell line MDA-kb2. The T47D-KBluc cell line was used to screen the water-soluble agrochemical residues for (anti-)oestrogenicity by evaluating binding to the oestrogen receptor. Oxidative stress responses (reactive oxygen species production, superoxide dismutase content, catalase activity, and lipid peroxidation), and non-neuronal acetylcholinesterase activity was also evaluated in vitro using the rat hepatoma H4IIE-luc and human duodenum HuTu-80 cell lines. Lastly, ultra-high performance liquid chromatography coupled to quadrupole time-of-flight mass spectrometry was used to quantify the following current-use pesticides in the soil: 2,4-dichlorophenoxy acetic acid (2,4-D), atrazine, dicamba, and imidacloprid. Results from the in vitro bioassays indicated that compounds present in the samples did not activate the AhR or androgen receptor (AR). However, at the concentrations evaluated in this study the soil-extracts from some sampling locations caused AR antagonism and (anti-)oestrogenicity, indicating endocrine disruptive effects. Although similar responses were observed in the HuTu-80 and H4IIE-luc cells following exposure to the bioavailable fraction for 24 hours (83 mg/mL), the rat hepatoma cell line was able to detoxify the xenobiotics present in the samples better. Furthermore, all four the target pesticides were quantifiable in at least one of the soil samples. Atrazine (89%) had the highest detection frequency, followed by dicamba (84%), 2,4-D (74%), and imidacloprid (32%). The combined use of in vitro bioassays and instrumental chemical analysis provided a holistic overview of the impact of agriculture on non-target organisms in the South African environment. Moreover, the findings of the present study contribute to the current understanding of the biological effects associated with the water-soluble fraction of agricultural soils, as well as the identity of the current-use pesticides likely causing these effects.