Host plant-guided engineering of beneficial soil microbiomes that protect against insect pests

Microbial communities are essential for plant health and protection. Recently, it has been suggested that the natural mechanisms plants employ to manage relationships with microbes (i.e., fight pathogens, promote mutualisms) can be used to selectively engineer whole microbial communities with beneficial or protective properties that enhance key plant traits (e.g., growth, yield, stress tolerance). This approach, known as host-guided selection, has been attempted in model species but has yet to be developed for crop plants. We conducted an experiment testing host-guided selection of insect-suppressive soil microbiomes in tomato (Solanum pimpinellifolium). We compared the efficacy of this approach between a phloem-feeder (potato aphids, Macrosiphum euphorbiae) and a chewing insect (tobacco hornworm, Manduca sexta). Aphid abundance and caterpillar growth were used to select the soil that would be transferred to the next generation. Treatments consisted of an insect-suppressive microbiome, an insect-enhancing microbiome, and a randomly selected microbiome. We found that aphid performance was significantly lower in insect-suppressive soils compared with insect-enhancing soils in four out of ten total generations. Caterpillar performance remained the same regardless of treatment. Our results suggest that host-guided selection on soil microbial communities can result in temporary impacts on insect performance, but the ecological stability of engineered communities may be challenging to achieve. Moreover, the focal trait and type of insect herbivore may determine the success of plant-guided microbiome engineering. Our results have important implications for host-guided selection that can be used to develop more effective control measures for insect pests in agricultural systems.