Quantifying glyphosate and its AMPA metabolite from environmental samples in South Africa

Abstract

Glyphosate-based herbicides (GBHs) are integral to modern agriculture, but concerns about their persistence in the environment and potential impacts on non-target organisms have prompted investigations into their environmental concentrations.It is globally considered to be the most used herbicide. The study aimed to quantify the levels of glyphosate (GLY) and its metabolite, aminomethylphosphonic acid (AMPA), in soil, sediment and water samples from selected agricultural and non-agricultural regions in South Africa and to assess the impact of these compounds on microbial communities and public health risks. To contextualise herbicide application, GBH-use in South Africa was geospatially mapped to understand the distribution of GLY use in key South African crops: maize, soybeans, wheat, and sunflower. Water, sediment, and soil samples were collected from the Vaalharts irrigation scheme in the Northern Cape, which uses canal irrigation from the Vaal and Harts rivers and from rain-fed crop fields in the catchments of the Klein Olifants and Vaal rivers in Mpumalanga. Samples were also collected from nonagricultural areas in the North West and Gauteng provinces. Water, soil, and sediment samples were collected during four events. Analytical methods involved the use of liquid chromatographytandem mass spectrometry for the quantification of GLY and AMPA in environmental matrices. The microbial diversity was evaluated by sequencing the 16S rRNA gene to evaluate whether the presence of GLY affected the structure of the microbial community. Lastly, a Quantitative Microbial Risk Assessment (QMRA) was conducted to evaluate the probable risk of infection posed by GLY-altered microbial populations selecting for environmental pathogens. Unfortunately, the GLY and AMPA concentrations were below the limit of detection in all the samples. This caused the focus to shift to assessing microbial diversity in response to agricultural chemicals in general at sites expected to present microbial diversity. The study pivoted to attempting to explain the absence of GLY and AMPA by investigating the role of UV-radiation, cation-exchange capacity (CEC), pH and GLY's adsorption coefficient. Throughout the study sites, the UV-B indices ranged from high to extremely high (6.7-8.7), which may have contributed to the deterioration of GLY. The CEC mean for the soils ranged from 1.6-2.2 meq/100 g, with some sites having values >6 meq/100 g. The calculated GLY sorption coefficients (mean Kd = 70 L/kg) using the CEC, clay, and pH indicated that GLY was mobile, except for 5/46 sites. Beta diversity communities were classified by matrix (soil vs. water) rather than by location, and microbial sequencing revealed that alpha diversity was higher in Mpumalanga. The QMRA showed that inhalation was the primary exposure route with Legionella and Mycobacterium, with probable infection rates of 1.0 (100%) at 10/20 and 19/20 sites, respectively. Together, these findings create a paradox: heavy reported GBH use with no quantifiable levels of GLY or AMPA. At the same time, the diverse microbial communities and probable pathogen infection risks suggest rapid GLY breakdown conditions, thereby helping to explain the non-detection of GLY and AMPA.