Plant-herbivore interactions: experimental demonstration of genetic variability in plant-plant signaling

• Plant-herbivore interactions mediated by plant-plant signaling were documented in different species. Here, we tested if herbivore foraging activity on plants was influenced by plant’s prior contact with a damaged plant and if the effect of such plant-plant signaling was variable between genotypes of the same plant species. • We built an innovative design where a snail was filmed during one hour on two plants differing only in a prior contact with a damaged plant or not. We replicated eight times the experiment on 113 natural genotypes of the model species Arabidopsis thaliana. We recorded snails’ first choice, and measured its first duration on a plant, the proportion of time spent on both plants and leaf consumption index. We performed genome-wide association studies (GWAS) on the four response variable. • On average, snails spent more time on plants that experienced a prior contact with a damaged plant, and consumed them more. However, plant-plant signaling effect of snail behavior was highly variable between plant genotypes, which was explained by a small number of single nucleotid polymorphisms (SNPs) related to stress coping ability and jasmonate pathway. • Plant-plant signaling modified the foraging activity of herbivores in A. thaliana. Depending on the plant genotype, specifically on genes involved in plant-herbivores interactions, plant-plant signaling made undamaged plants more repulsive or attractive to snails. This finding questions the theoretical basement of the evolution of plant-herbivore interactions mediated by plant-plant signaling.