Plant EV small RNAs as programmable cross-kingdom signals: Mechanisms, evidence standards, and crop protection

Plant extracellular vesicles (EVs) have emerged as candidate carriers of small RNAs that connect intracellular RNA regulation with apoplastic defense, intercellular signaling, and cross-kingdom communication. The field is advancing rapidly, yet its central mechanistic claims remain unevenly resolved because RNA detection, vesicle association, selective loading, recipient-cell delivery, and target-dependent function are often treated as equivalent observations. The novelty of this review is a mechanistic and evidentiary framework for evaluating when an extracellular RNA is biologically functional, rather than a simple restatement of plant EV-mediated cross-kingdom RNA communication. This review develops a macromolecule-centered framework for plant EV small RNAs in which biological activity is determined by a linked sequence of control points: generation of the donor RNA pool, cargo selection by RNA-binding proteins and subcellular routing, membrane enclosure or surface association, extracellular stability, uptake by recipient cells, endosomal or cytosolic release, and decoding through RNA interference or other RNA-dependent pathways. Direct evidence from plant-fungal interactions is integrated with plant mobile-silencing research, general EV RNA-sorting principles, parasitic and microbial cross-kingdom systems, and emerging RNA-delivery technologies. The review distinguishes established plant mechanisms from comparative inference, examines methodological confounders that have driven overinterpretation, and proposes an evidence ladder for assigning functional delivery. Finally, it assesses native plant EVs, plant-derived nanovesicles, synthetic vesicles, and hybrid nanocarriers as platforms for sequence-programmable crop protection. The resulting framework shifts the field from descriptive cargo catalogues toward quantitative design rules for RNA dose, carrier identity, recipient specificity, and agronomic performance.