- Review
Edible plant-derived extracellular vesicles (EVs) and EV-like nanoparticles have attracted interest as potential carriers of microRNAs (miRNAs) across the gastrointestinal tract. This critical narrative review examines their terminology, isolation and characterization, RNA composition, gastrointestinal stability, cellular uptake, biodistribution, biological effects, and potential applications. Available studies indicate that plant-derived particle preparations can associate with diverse RNAs and interact locally with intestinal epithelial cells, immune cells, stem/progenitor cells, and gut microorganisms. However, particles recovered from homogenized tissues or juices may include intracellular vesicles, membrane fragments, and non-vesicular complexes; detection of miRNAs in particle-enriched fractions therefore does not establish vesicular encapsulation or functional delivery. Evidence for dietary cross-kingdom regulation remains mixed. Although plant extracellular small RNAs can regulate fungal pathogens and ginger-derived particle-associated small RNAs have been linked to bacterial and host responses, reproducible systemic delivery of physiologically effective plant miRNAs after normal dietary intake has not been established. Lipids, proteins, metabolites, and other RNAs may also contribute to reported effects. Future studies should distinguish naturally secreted EVs, EV-like nanoparticles, and engineered nanovectors, and should combine rigorous particle characterization, RNA localization, absolute quantification, physiologically relevant dosing, and loss-of-function and rescue experiments. Such standards are essential for defining biological relevance and therapeutic potential.
11 November 2026




![Establishment of an arrayed CRISPRi platform for multimodal lncRNA functional profiling. (A) Schematic overview of the arrayed CRISPRi screening strategy. Stable HEK293T-dCas9-KRAB-MeCP2 and KELLY-dCas9-KRAB cells were used to repress selected lncRNA targets. For each perturbation, matched molecular and cellular readouts were collected, including targeted 3′-end RNA capture-seq for knockdown assessment, shallow 3′-end RNA-seq for transcriptomic profiling, live-cell IncuCyte imaging for growth analysis, and Cell Painting-based high-content imaging for morphology profiling. Panel A was created with BioRender.com and incorporates a Cell Painting image reproduced from Cimini et al. [31]. (B) Distribution of genomic distances between selected lncRNA transcription start sites and the nearest protein-coding gene transcription start site. Distances are shown as log10-transformed closest TSS distance. (C) Genomic classification of the 68 selected lncRNA targets relative to nearby protein-coding genes. Numbers indicate the number of targets in each genomic class. (D) Expression levels of the selected lncRNAs measured by total RNA-seq in unperturbed HEK293T-dCas9 and KELLY-dCas9 cells, shown as log10(gene TPM + 1). (E) Classification of targeted lncRNAs according to confidence in knockdown assessment. Target coverage was determined from targeted 3′-end RNA capture-seq data in matched non-targeting controls. Targets with a median normalized count of at least 3 were classified as high-confidence and included in the primary knockdown interpretation. Targets below this threshold were classified as low-confidence and should be interpreted cautiously. (F) Remaining expression after CRISPRi repression among high-confidence targets in HEK293T and KELLY cells. Points represent individual lncRNA targets. Boxplots summarize the distribution of remaining expression relative to non-targeting controls. The stacked bars above indicate the number of high-confidence targets classified as showing detectable knockdown or no knockdown in each cell line.](https://mdpi-res.com/ncrna/ncrna-12-00046/article_deploy/html/images/ncrna-12-00046-g001-550.jpg)
