Determining the (anti)androgenic activity of agricultural pesticides in water systems with a luminescence bio-assay

Abstract

South Africa has a large agricultural sector that produces a wide variety of crops on large scale for local use and exportation. This provides some explanation as to why the country has over 500 pesticide active ingredients registered for use. Studies have indicated that several current use pesticides have endocrine disrupting properties that may cause health effects if humans and animals are exposed to such chemicals. Distribution of pesticides into freshwater systems frequently occur which increases the potential of exposure since many South Africans use untreated water sources for domestic and agricultural purposes. This study aimed to determine the (anti)androgenic activity of several current use pesticides used in different agricultural practices in South Africa. Surface and groundwater samples as well as sediment samples were collected from three different study areas in the country. In the Letsitele River catchment mangos, citrus and avocados are produced on large scale. The intensive cultivation of maize, sorghum and sunflower take place in the Renoster and Vals River catchments while sugarcane and citrus are produced in the Mzinti, Lomati and Ngweti River catchments. The samples were chemically screened for the presence of 291 pesticide active ingredients. The ten priority pesticides were selected based on their frequency of occurrence in each study area as well as on their intensity of use in the country. These pesticides were subjected to a chemical quantification analysis. The samples were also screened for their ability to activate the androgen receptor and/or to inhibit androgen binding to the androgen receptor. This was done with the MDA-kb2 reporter gene bio-assay. The assay also detects ligands capable of activating the glucocorticoid receptor. The chemical and biological analyses results were compared on a seasonal basis and the results of the samples from the different matrices were also compared to one another. Several surface water and sediment samples elicited (anti)androgenic responses from the cells as well as glucocorticoid receptor activation. The most androgenic and/or glucocorticoid responses were caused by the sediment samples collected in the Renoster and Vals River catchments. The most anti-androgenic responses were collectively detected in the surface water and sediment samples from the Mzinti, Lomati and Ngweti River catchments. Atrazine and alachlor were the only priority pesticides that have been declared as definite endocrine disruptors while studies have indicated that atrazine, imazalil and propiconazole may have anti-androgenic properties. Comparison of the data showed no pattern between the samples from the different matrices from the same sampling event or between the samples from different sampling events. Interpretation of the results were further complicated since very few regulatory guidelines exist for pesticides in aquatic ecosystems. This study can be used as the basis for future endocrine disrupting studies during which pesticide guidelines can be developed in a South African context.