Thomas Mitchell-Olds

Baseline levels of glucosinolates—important defensive phytochemicals in brassicaceous plants—are determined by both genotype and environment. However, the ecological causes of glucosinolate plasticity are not well characterized. Fertilization is known to alter glucosinolate content of Brassica crops, but the effect of naturally occurring soil variation on glucosinolate content of wild plants is unknown. Here, we conducted greenhouse experiments using Boechera stricta to ask (i) whether soil variation among natural habitats shapes leaf and root glucosinolate profiles; (ii) whether such changes are caused by abiotic soil properties, soil microbes, or both; and (iii) whether soil-induced glucosinolate plasticity is genetically variable. Total glucosinolate quantity differed up to 2-fold between soils from different natural habitats, while the relative amounts of different compounds were less responsive. This effect was due to physico-chemical soil properties rather than microbial communities. We detected modest genetic variation for glucosinolate plasticity in response to soil. In addition, glucosinolate composition, but not quantity, of field-grown plants could be accurately predicted from measurements from greenhouse-grown plants. In summary, soil alone is sufficient to cause plasticity of baseline glucosinolate levels in natural plant populations, which may have implications for the evolution of this important trait across complex landscapes.