Global food systems face challenges from population growth, shifting diets and climate change. While decades of plant-centric breeding and high-input agriculture have generated high yielding crops, this strategy has unintentionally reshaped the plant associated microbiomes, often coinciding with a depletion of their functional diversity. We revisit these breeding strategies and propose extending breeding targets beyond the plant genome to include the plant microbiome. In this regard, resistance breeding shows, albeit unintended, that plant genetics shape the microbiome: by narrowing the genetic base, we have selected for highly specialised pathogenic microbiomes. This raises a key question: can we intentionally apply the same principle to select for beneficial microbiomes? To answer this question, a thorough insight into microbial community architecture, hubs and functional redundancy is key. We outline two complementary avenues: (i) rewilding to restore ancestral microbial partners and (ii) microbiome breeding guided by QTL/GWAS mapped host loci that gate microbial recruitment, immune filtering and exudate composition. This approach comprises the integration of trait-based phenotyping, multi-omics, network-informed SynCom design and field testing across environments to capture G × E × M (genotype × environment × microbiome) interactions. Treating the microbiome as a selectable, designable and heritable trait can convert small gains into durable long-lasting crop resilience.
Zutter et al. (Wed,) studied this question.