The Plant Microbiome in a Changing World: Ecology, Evolution, Stress Resilience and Sustainable Applications

Plant microbiome research has been dominated by bacterial and fungal components, yet viruses, protists, and other microbial eukaryotes can regulate the abundance, behaviour, and ecological effects of those better-studied groups. This critical narrative review evaluates how these underrepresented components contribute to plant health and stress resilience through direct infection, predation, parasitism, nutrient remineralisation, host filtering, and indirect trophic cascades. Literature published from 1 January 2000 to 15 July 2026 was considered, with older foundational evidence retained where necessary. Evidence is strongest for three mechanisms: bacteriophages restructuring bacterial communities and, in some systems, contributing to pathogen suppression; predatory protists altering bacterial and fungal communities in ways that affect plant growth and disease; and viruses of fungi or oomycetes changing pathogen fitness, sometimes towards hypovirulence but in other cases towards enhanced virulence. Evidence that plant viruses themselves can generate stress-beneficial phenotypes is credible in defined host–virus combinations but remains context dependent rather than a general property of persistent infection. Cross-kingdom network studies increasingly identify protists as connectors and reveal associations spanning plants, bacteria, fungi, archaea, and microbial eukaryotes, but co-occurrence alone cannot establish interaction or causal direction. Major technical blind spots include marker and size-fraction bias, weak reference databases for microbial eukaryotes and viruses, uncertain virus–host assignment, compositional data artefacts, and limited use of absolute abundance, time series, and perturbation experiments. Translation into biocontrol or resilience engineering is therefore most defensible when ecological associations are followed by isolate-based, synthetic-community, and field validation. A plant microbiome framework that explicitly includes viral and eukaryotic trophic layers can improve mechanistic understanding, but its agricultural value will depend on resolving context dependence, non-target effects, evolutionary stability, and performance under realistic combinations of abiotic and biotic stress.