ORIGINAL ABSTRACT

The progamic phase from pollination to fertilisation is a critical stage of plant life cycle that is vulnerable to chemical exposure and requires controlled production of reactive oxygen species (ROS). Herbicide formulations are integral to agriculture, horticulture, and forestry, though surprisingly little is known about pollen exposure to commercial herbicide formulations, especially for non-target (i.e., non-weed) plants. Here, we compared the effects of a synthetic formulation of glyphosate (Roundup) and a natural formulation of pelargonic acid (Herbistop) on the in vitro performance of hazel (Corylus avellana L.) pollen. At equal pH values, Herbistop (half-maximal inhibitory concentration [IC50] 10.7-11.2 mg acid equivalents [a.e.] L-1) restricted pollen tube elongation at concentrations one order of magnitude lower than Roundup (IC50 332.0-410.0 mg a.e. L-1) and in a narrower range. Pollen staining with triphenyl tetrazolium chloride confirmed the toxicity of both formulations, with Herbistop compromising viability more severely. At respective IC50 values, both herbicide formulations curtailed the production of the ROS hydrogen peroxide during pollen tube elongation, impairing viability, and slowing down in vitro germination. Overall, our in vitro evidence shows that the natural formulation, Herbistop, was more hazardous for hazel pollen performance than the synthetic formulation, RoundUp. This highlights the common misconception that usage of natural herbicide formulations, rather than synthetic, is safer for non-target plants, other non-target organisms, and the environment. We propose that pollen performance is an additional toxicological endpoint of all herbicide formulations that requires special surveillance in gardens, woodlands, and agroecosystems.