Skip to main content
NCBI home page
ISME Communications logoLink to ISME Communications
. 2024 Oct 12;4(1):ycae120. doi: 10.1093/ismeco/ycae120

Rhizospheric miRNAs affect the plant microbiota

Harriet Middleton 1,2,#, Jessica Ann Dozois 3,#, Cécile Monard 4, Virginie Daburon 5, Emmanuel Clostres 6, Julien Tremblay 7, Jean-Philippe Combier 8, Étienne Yergeau 9,✉, Abdelhak El Amrani 10,✉
PMCID: PMC11520407  PMID: 39474459

Abstract

Small ribonucleic acids (RNAs) have been shown to play important roles in cross-kingdom communication, notably in plant–pathogen relationships. Plant micro RNAs (miRNAs)—one class of small RNAs—were even shown to regulate gene expression in the gut microbiota. Plant miRNAs could also affect the rhizosphere microbiota. Here we looked for plant miRNAs in the rhizosphere of model plants, and if these miRNAs could affect the rhizosphere microbiota. We first show that plant miRNAs were present in the rhizosphere of Arabidopsis thaliana and Brachypodium distachyon. These plant miRNAs were also found in or on bacteria extracted from the rhizosphere. We then looked at the effect these plants miRNAs could have on two typical rhizosphere bacteria, Variovorax paradoxus and Bacillus mycoides. The two bacteria took up a fluorescent synthetic miRNA but only V. paradoxus shifted its transcriptome when confronted to a mixture of six plant miRNAs. V. paradoxus also changed its transcriptome when it was grown in the rhizosphere of Arabidopsis that overexpressed a miRNA in its roots. As there were differences in the response of the two isolates used, we looked for shifts in the larger microbial community. We observed shifts in the rhizosphere bacterial communities of Arabidopsis mutants that were impaired in their small RNA pathways, or overexpressed specific miRNAs. We also found differences in the growth and community composition of a simplified soil microbial community when exposed in vitro to a mixture of plant miRNAs. Our results support the addition of miRNAs to the plant tools shaping rhizosphere microbial assembly.

Keywords: plant miRNAs, rhizosphere, bacterial communities, Variovorax, transcriptomics

Introduction

Small ribonucleic acids (sRNAs) are thought to play a major role in plant–microbe interactions [1–3]. For example, cotton plants used sRNAs to inhibit the fungal pathogen Verticillium dahliae [4] and, similarly, Arabidopsis thaliana and tomato used miRNAs to inhibit another fungal pathogen, Botrytis cinerea [5–7]. Arabidopsis also delivered sRNAs into the oomycete pathogen Phytophthora capsici to silence its genes [8]. Similarly, wheat used sRNAs to silence the alpha/beta hydrolase gene in Fusarium graminearum [9]. Applying sRNAs targeting Botrytis genes on the surface of plants led to an inhibition of the disease [10]. Conversely, microorganisms also use sRNAs to modulate plant gene expression. B. cinerea delivered sRNAs to plant cells, silencing the host immune response [11, 12]. Rhizobium delivered transfer RNA-derived sRNAs into soybean cells to regulate nodulation [13]. Puccinia striiformis used sRNAs to increase their pathogenicity by suppressing wheat pathogenesis-related genes [14]. Arbuscular mycorrhizal fungi and plants also exchange sRNAs to regulate their mutualistic interactions [15–17]. More recently, rhizosphere microbial sRNAs activated specific pathways in the host plant, playing a role in the soil suppressiveness of Rhizoctonia solani [18]. The role of sRNAs in shaping more generally the plant microbiota is, however, not known.

One class of sRNAs that have the potential to shape the microbiota is micro RNAs (miRNAs). miRNAs are small ~21 nt non-coding RNAs that control target gene expression, through sequence complementarity. Their roles in eukaryotes vary from regulating developmental processes to responding to abiotic and biotic stresses. In the mammalian gut, host miRNAs regulated bacterial gene expression and growth, shaping the gut bacterial community [19]. The oral administration of a single miRNA shifted the gut microbiota, mainly by increasing the abundance of Akkermansia muciniphila [20]. miRNAs from edible plants, conveyed in exosome-like nanoparticles, were preferentially taken up by gut bacteria and regulated bacterial gene expression [21]. The plant miRNA miR159 was taken up by various gut bacteria and it influenced their gene expression and growth, which led to shifts in the gut microbiota of mice that were fed with this miRNA [22]. In view of these two last studies, it is therefore likely that plant miRNAs also affect the plant microbiota.

Arabidopsis roots contain many miRNAs, of which over half are expressed in a tissue-specific manner, with several being enriched at the root tip, in the early meristematic zone [23]. This zone is also a recognized hotspot for plant-driven microbial selection [24]. The current paradigm is that the plant selects the rhizospheric microbiota through active or passive rhizodeposition, including exudates such as sugars, peptides, amino acids, nucleic acids, nucleotides, fatty acids, or secondary metabolites [25]. We had previously hypothesized that miRNAs could play a role in shaping the rhizosphere microbiota [3]. Here, we hypothesize that plant miRNAs can be found in the rhizosphere of plants and that they influence the gene expression of rhizosphere bacteria, thereby shaping the bacterial community. We report several independent experiments designed to test these hypotheses (Fig. 1).

Figure 1.

Figure 1

Open in a new tab

Experimental approach overview. Overview of the experiments leading to the identification of plant miRNAs in the rhizosphere and the confirmation of their effect on microbial communities. Created with BioRender.com

Materials and methods

The full method description is available as supplementary material

Detection of plant micro ribonucleic acids in the rhizosphere, roots, and rhizosphere bacteria

For rhizosphere analyses, triplicate A. thaliana (Col-0) and Brachypodium distachyon (Bd21-3) were grown for a month, alongside unplanted soils. For root analyses, we grew 5 replicated A. thaliana (Col-0) plants for 21 days, with unplanted controls. RNA was extracted from the plant roots and rhizosphere and from the unplanted soils. miRNA library preparation with size selection for miRNAs was performed prior to small RNA sequencing. Sequences between 18 and 27 nucleotides were mapped against Arabidopsis and Brachypodium genomes and were assigned to known miRNAs. This approach ensured that the profiled miRNAs were derived from our model plants. From these miRNAs we constructed an abundance table that was used to compare the rhizosphere and roots to the unplanted controls. Rhizospheric and root miRNAs were defined as the miRNAs that had at least 10 reads for each of the rhizosphere or root samples and a maximum of one read across all the bulk soil samples. Singleton reads were considered to be a sequencing artifact.

We further wanted to show that the plant miRNA had the potential to interact with the rhizosphere bacteria, by showing that they were either on the surface or inside the bacterial cells. We extracted bacterial cells from the rhizosphere soil of 1-month old A. thaliana (Col-0) using a Nycodenz gradient. RNA was extracted from the bacterial pellet [26], sequenced, and processed as described above, but with a lower threshold because of the lower number of plant miRNA sequences retrieved. Only the miRNAs that were represented by at least five reads in each rhizosphere sample and absent in the bulk soil samples were kept. To further confirm that miRNAs found in the rhizospheric bacteria originated from the plant and not the bacteria, miRNA sequences were searched for on the + and − strands of 3837 bacterial genomes [27], of which 1160 were isolated from plants.

Effect of micro ribonucleic acids on bacterial gene expression

Bacterial transcriptomic response to micro ribonucleic acids

Variovorax paradoxus EPS and Bacillus mycoides YL123 were grown until the exponential phase, when they were exposed to a mix of the six most abundant rhizospheric miRNAs: miR159a, miR159b, miR159c, miR161.1, miR158a, and miR165b, or a control mix of six scrambled miRNAs (same nucleotide composition but in a random order). The cultures were sampled after 20 min and 120 min of incubation, pelleted, and the RNA of the cell pellet was extracted and sequenced. The transcripts were mapped on their respective genomes, and differential expression analyses (plant miRNA vs. scrambled miRNA) were done using DESeq2.

In silico analysis of micro ribonucleic acids targets

To predict potential targets of the six miRNA used on the genome of V. paradoxus, we implemented a workflow [28] named “mirnatarget 1.0: miRNA target finder” (https://github.com/jtremblay/MiRNATarget), which was inspired from the plant miRNA target finder, “psRNAtarget” [29].

Confocal microscopy

We then wanted to know if the difference between the response of V. paradoxus EPS and B. mycoides YL123 was due to a barrier blocking entry into the cell. We grew the two bacteria overnight in liquid media. The cultures were incubated for 4 h with 3’-Cy5 fluorescent miRNAs (ath-miR159a or a scrambled control—same nucleic acid content but in different order), or a pCp-Cy5 control at a final concentration of 2 μM. Twenty minutes before visualization, the cultures were also treated with MitoTracker Green FM (Invitrogen), which stains all live bacteria. We visualized the washed and concentrated culture using a confocal microscope (Zeiss LSM780).

Flow cytometry

We prepared the cultures as described in the confocal microscopy section. The cells were fixed with 4% paraformaldehyde and then stained with a deoxyribonucleic acid (DNA) marker (Hoechst 33342). The cells were diluted in PBS and processed by flow cytometry (BD LSRFortessa). We used a variety of controls to ensure that our statistical analyses were conducted on bacteria that were positive for our green (MitoTraker), red (Cy5 tagged miRNAs), and blue (Hoechst) markers.

In vitro micro ribonucleic acid-encoded peptide treatment and transcriptomic experiment

To confirm in planta the transcriptomic results for V. paradoxus, ~50 A. thaliana (Col-0) surface-sterilized seeds were grown axenically in Petri dishes in a growth chamber. After 20 days of growth, the plants were treated twice within 24 h by inoculating miPEP159c, a scrambled micro ribonucleic acid-encoded peptide (miPEP), or water at the plant crown. One hour after the second miPEP treatment, V. paradoxus was inoculated along the roots of the seedlings. Two hours after the bacterial inoculation, the plants and rhizospheres were sampled, and RNA was extracted. From the potential targets of miR159c found above, three were selected for reverse transcriptase real-time quantitative PCR (RT-qPCR) analyses in the rhizosphere—alpha-2-macroglubulin, phosphatidate cytidylyltransferase (CdsA), and LysR. We also quantified the abundance in plant tissues of the primary transcript of mir159c, pri-miR159c.

Effect of micro ribonucleic acids on the bacterial community

Arabidopsis mutant experiment

Five A. thaliana mutants were chosen. RTL1 mutant overexpresses RTL1 protein which results in a suppression of siRNA pathway without affecting miRNAs [30]. RTL1myc overexpresses RTL1 protein flagged with Myc epitope, rendering RTL1 less active, so siRNA pathway is less suppressed than with RTL1 mutant. Ago1-27 mutant has AGO protein function partially impaired and is completely post-transcription gene silencing (PTGS) deficient [31]. Dcl1-2 mutant has total loss of function of DCL1 protein resulting in low levels of miRNA and developmental problems [32]. Hen1-4 mutant is miRNA defective but is also affected in some siRNA–PTGS [33]. HEN1 methylates siRNA and miRNA not only to maintain their levels and size but also to protect them from uridylation and subsequent degradation [34]. The plants were grown for a month, after which the roots and attached rhizosphere were sampled, their DNA extracted, amplified using 16S ribosomal RNA (rRNA) gene primers, and sequenced. The same primers were used in real-time quantitative PCR to quantify bacterial abundance. Amplicon sequencing data was processed with AmpliconTagger [35] and the R package “phyloseq” v 1.32.0 [36].

Micro ribonucleic acid-encoded peptide experiment

We treated A. thaliana Col-0 with 500 μL of water (control condition) or a miPEP solution (20 μM of miPEP159a, miPEP159b, or miPEP159c), applied at the base of the crown, 3 times a week for a total of 10 applications. We then extracted the DNA from the roots attached rhizosphere and sequenced and quantified the 16S rRNA gene as described above.

Simplified soil community laboratory experiment

We created, in vitro, a simplified soil community by inoculating five different growth media with 2 g of agricultural soil. The cultures were normalized to the same optical density, pooled, pelleted, and suspended in PBS. The cells were inoculated in a 96-wells plate containing a mixture of 17 amino acids as nitrogen source. Five wells were treated with a mixture of rhizospheric miRNAs (ath-miR158a-3p, ath-miR158b, ath-miR159a, ath-miR827, and ath-miR5642b) and five wells were treated with a mixture of scrambled miRNAs. These miRNAs were all found in the rhizosphere (but for some below the stringent threshold used above) and were predicted to target bacterial genes associated with nitrogen cycling. We measured bacterial growth every hour for 52 h, after which we sampled the bacteria, extracted the DNA, amplified and sequenced the 16S rRNA gene as described above.

Results

Plant micro ribonucleic acids are present in the rhizosphere

We sequenced small RNA extracted from A. thaliana rhizosphere soil and unplanted soils. One-hundred-eleven ath-miRNAs (mapped on A. thaliana’s genome) were detected, of which 14 were present in the rhizosphere with >10 reads per sample and absent in unplanted soil (Fig. 2A). The most abundant miRNAs in the rhizosphere were ath-miR158a-3p, ath-miR161.1, and various members of the miR159, miR166, and miR165 families (Fig. 2A). We then sequenced the rhizosphere miRNAs of a second model plant, B. distachyon. Out of the 81 bdi-miRNAs (mapped on B. distachyon’s genome) detected in the rhizosphere, 10 were represented by >10 reads per sample and absent in unplanted soil (Fig. 2B). The most abundant miRNAs were bdi-miR159b-3p, bdi-miR156, bdi-miR166, bdi-miR396, and bdi-miR167. Among the rhizospheric miRNAs detected above our threshold, four were common between A. thaliana and B. distachyon: miR159b-3p, miR167, miR166, and miR396.

Figure 2.

Figure 2

Open in a new tab

Plant miRNAs are present in the rhizosphere, roots, and bacteria and absent in unplanted soil. Relative abundance of the plant miRNAs found in the rhizosphere of (A) A. thaliana, (B) B. distachyon, (C) in the roots of A. thaliana, and (D) in or on rhizosphere bacteria while completely absent from unplanted soil.

We then grew more Arabidopsis plants and sequenced their root miRNAs to confirm that the rhizospheric miRNAs could be coming from the plant roots. Eleven miRNAs were represented by at least 10 reads in each root sample. There was a clear dominance of ath-miR165, ath-miR166, ath-miR159a, ath-miR160, and ath-miR159b.3p (Fig. 2C). Among these 11 miRNAs, 7 were common with the miRNAs found in Arabidopsis rhizosphere (ath-miR158a-3p, ath-miR159a, ath-miR159b-3p, ath-miR161.1, ath-miR162, ath-miR165, ath-miR166), which included most of the top 5 most abundant root and rhizosphere miRNAs.

To confirm that plant miRNA could interact with rhizosphere bacteria, bacterial cells were isolated from the rhizosphere of 1-month-old A. thaliana, washed and their RNA content was extracted and sequenced. The miRNAs were mapped against A. thaliana TAIR10.1 genome, identifying a total of 34 ath-miRNAs. Five miRNAs—namely ath-miR158a-3p, ath-miR161.1, miR162, ath-miR159b-3p, and ath-miR159a—were represented by at least five reads per rhizosphere samples and absent in bacteria extracted from the unplanted soil (Fig. 2D). Four out of these five miRNAs were among the five most abundant miRNAs found in the rhizosphere of A. thaliana (Fig. 2A) and were found in a similar rank-abundance order. To ensure that these miRNAs did not come from the bacteria themselves, their gene-encoding sequences were searched for in 3837 soil bacterial genomes, of which 1160 were of bacteria isolated from plants [27]. No matches were found, meaning that the miRNAs detected in the bacteria could not be produced by the bacteria.

Plant micro ribonucleic acids shift rhizosphere bacterial gene expression

We incubated V. paradoxus EPS [37] and B. mycoides [38] with a synthetic mixture of the six most abundant A. thaliana rhizosphere miRNAs: miR159a, miR159b, miR159c, miR161.1, miR158a and miR165b, or a mixture of scrambled miRNAs (same nucleotide composition but in a random order) at the same concentration. Bacillus did not respond to the treatment—no gene was significantly differentially expressed following incubation with the synthetic miRNAs. In contrast, Variovorax showed important changes in response to the miRNA confrontation, as revealed by a differential abundance analysis of plant miRNA vs. scrambled miRNA treatments. After 20 min of incubation, the expression of 79 genes was significantly lower in the plant miRNA-treated cultures and the expression of 44 genes was significantly higher (adjusted P < .05, Fig. 3A and Supplementary Table S1). After 120 min of incubation, the expression of 24 genes was significantly lower in the plant miRNA-treated cultures and the expression of 104 genes was significantly higher (adjusted P < .05, Fig. 3B and Supplementary Table S1). Many genes were repressed after 20 min following the addition of the synthetic plant miRNAs to the bacterial culture, whereas after 120 min, more genes presented an increased expression. Only one gene was differentially expressed at both time points, a gene coding for a methionine synthase (VARPA_RS01000), which was overexpressed in response to the plant miRNA treatment.

Figure 3.

Figure 3

Open in a new tab

Plant miRNAs affect the transcriptome of a rhizosphere bacterium. Gene expression of Variovorax paradoxus after (A) 20 min and (B) 120 min exposure to a mixture of six synthetic miRNAs (plant miRNAs compared to scrambled miRNAs), confocal microscopy images of (C) V. paradoxus and (D) B. mycoides after a 4-h exposure to ath-miR159a tagged with Cy5 and flow cytometry experiment showing (E) the percentage of active bacterial cells positive for Cy5 signal and (F) the median intensity of the Cy5 signal in active bacterial cells. Arrows in (C) and (D) indicate cells containing the fluorescent molecule. *: P < .05.

Using the rules of a plant small RNA target finder, psRNAtarget, we compared the differentially expressed genes from the previous experiment with the predicted target genes for the six miRNAs used. The six miRNAs were predicted to target 237 sequences in the V. paradoxus EPS genome. Among these, 100 targets were positioned too far from any coding sequence (CDS), so they were removed from the following analysis, resulting in 137 potential targets. Among the 123 genes differentially expressed at 20 min, only 2 were predicted as targets in silico: VARPA_RS05960 (targeted by miR165), coding for an L-iditol 2-dehydrogenase, and VARPA_RS26385 (targeted by miR159a, b and c), coding for a CDP-diacylglycerol synthase (CdsA). Only two genes were predicted as targets among the 128 differentially expressed genes at 120 min: VARPA_RS00680 (targeted by miR165), coding for a hypothetical protein, and VARPA_RS22555 (targeted by miR158a-3p and miR159c), coding for a non-ribosomal peptide synthetase.

To test if the differences in sensitivity between V. paradoxus and B. mycoides were due to a barrier blocking the entry into the cell, we exposed them to a Cy5-tagged synthetic ath-miR159a and visualized its localization using confocal microscopy. Images show a clear localization of the miRNA inside many bacterial cells (Fig. 3C and D). Flow cytometry confirmed that an average of 6.51% Variovorax cells contained the Cy5 signal from plant miRNAs (average median fluorescence intensity = 123), compared to 4.95% of Bacillus cells (average median fluorescence intensity = 85.75; Fig. 3E and F). The scrambled miRNA, containing the same nucleotides as ath-miR159a but in a different order, was internalized as efficiently (Supplementary Fig. S1 and Fig. 3E), suggesting a general sequence-independent internalization mechanism for miRNAs. Variovorax, in contrast to Bacillus, internalized more efficiently the tagged plant miRNAs than the tagged single nucleotide. Indeed, a comparable amount of pCp-Cy5 was also internalized by Bacillus (on average 5.04%, average median fluorescence intensity = 75.8), but this was an order of magnitude lower for Variovorax (on average 0.67%, adjusted P = .0200, average median fluorescence intensity = 81.5, adjusted P = .0355; Fig. 3E and F and Supplementary Fig. S1).

To confirm the effect of miRNAs on the bacterial transcriptome in planta, we exposed Arabidopsis plants growing in vitro to the miRNA-encoded peptide (miPEP) miPEP159c and then inoculated them with Variovorax. miPEPs increased the expression of specific plant miRNAs [39, 40]. We selected miR159c because it was among the most abundant miRNAs in the rhizosphere, was in the mixture of miRNAs that modulated the gene expression of Variovorax, and was predicted to target several key genes. The relative expression of the corresponding primary transcript (pri-miR159c) in the Arabidopsis plant tissue increased by a factor 1.58 as compared to the scrambled miPEP control (same amino acid composition as the miPEP, but in different order; t-test: t = 3.33, P = .00929; Fig. 4). We then quantified the expression of three Variovorax genes determined to be potential targets of the miR159c according to our bioinformatic and transcriptomic analyses. One hundred and twenty minutes after the miPEP159c application, the relative expression of the LysR genes decreased by a factor 0.69 (t-test: t = −3.06, P = .0195, Fig. 4), whereas the expression of alpha-2 macroglobulin gene increased by a factor 0.34 (t-test: t = 2.18, P = .0520, Fig. 4) in comparison with the scrambled miPEP control. The expression of the CdsA gene did not differ between the miPEP159c and the scrambled miPEP control (Fig. 4). The relative expressions of the three genes and the pri-miR159c following the application of the miPEP159c were all significantly different from the water control, whereas it was never the case for the scrambled miPEP (Fig. 4).

Figure 4.

Figure 4

Open in a new tab

miPEP 159c shifts the expression of V. paradoxus. Expression of pri-miR159c in A. thaliana roots and V. paradoxus expression of CdsA, alpha-2-macroglobulin and LysR in the rhizosphere after exposure of the plant to miPEP159c vs. a water and a scrambled miPEP controls. ns: P > .05; *: P < .05, **: P < .01.

Plant micro ribonucleic acids influence the rhizosphere bacterial community

To investigate the role of plant small RNAs on the rhizospheric microbial diversity, we grew A. thaliana mutants with disturbed miRNA and/or siRNA biosynthesis pathways and analyzed their rhizospheric microbial communities by 16S rRNA gene amplicons sequencing. The rhizosphere bacterial communities varied across the different genotypes (permutational multivariate analysis of variance [PERMANOVA]: P < .05; Fig. 4A). In principal coordinates analysis (PCoA) ordinations, ago1-27 and RTL1myc mutants’ communities were more like unplanted soil communities than those of wild-type (WT) plants (not shown). In subsequent pairwise PERMANOVA, there were significant differences between the community composition of the following mutant pairs: ago1-27 and hen1-4, ago1-27 and RTL1, hen1-4 and RTL1myc, and RTL1 and RTL1myc. The microbial community of WT plants was nearly significantly different (P < .10) from all the mutants. These differences were mirrored in the community composition at the phylum level, with significant differences between the genotypes for the relative abundance of Acidobacteria (F = 7.8, P = .00856), Actinobacteria (F = 3.2, P = .0363), Chloroflexi (F = 5.7, P = .00391), Planctomycetes (F = 5.3, P = .00534), Spirochaete (F = 3.6, P = .0255), and Verrucomicrobia (F = 6.4, P = .00231; Fig. 5A). For the Acidobacteria, Actinobacteria, and Chloroflexi, most of these differences were due to differences between the WT plants and some or all the mutants (Fig. 5A), according to post-hoc Tukey honestly significant difference (HSD) tests. For the other phyla, the differences were rather between different mutants (Fig. 5A). Bacterial diversity was higher in the rhizosphere of most mutant plants and in the unplanted soil as compared with the rhizosphere of the WT plants, whereas there was no difference between the rhizosphere of the mutants and the unplanted controls (Fig. 5B).

Figure 5.

Figure 5

Open in a new tab

Plant miRNAs affect the bacterial community. (A) Phylum-level bacterial community composition and (B) Shannon diversity for different Arabidopsis mutants with impaired small RNAs pathways. (C) Phylum-level bacterial community composition for Arabidopsis plants inoculated with miPEP159a, miPEP159b, or miPEP159c. (D) Growth curve, (E) community composition at genus level, and (F) relative abundance of significantly affected ASVs for a simplified soil microbial community exposed to a mixture of synthetic plant or scrambled miRNAs. *: P < .05, **: P < .01. Different letters in (B) indicate significant differences (P<.05) in Tukey HSD tests.

We then treated soil-grown Arabidopsis with miPEPs to further test for the effect of the overexpression of specific miRNAs on the rhizospheric bacterial community. The application of three miPEPs (miPEP159a, miPEP159b, and miPEP159c) on the crown of Arabidopsis growing in soil changed the bacterial community in the roots/rhizoplane (PERMANOVA: R2 = 0.160, P = .031). All pairwise comparisons with the water control were significant or nearly significant (pairwise PERMANOVA: miPEP159a R2 = 0.11, adjusted P = .083; miPEP159b R2 = 0.16, adjusted P = .025; miPEP159c R2 = 0.15, adjusted P = .033). At the phylum level, the application of the miPEPs increased the relative abundance of the Proteobacteria (F = 8.9, P = .00029) and decreased the relative abundance of the Planctomycetes (F = 40.8, P = 3.64 × 10−10; Fig. 5C). This was due to significant (adjusted P < .05) differences between the water control and each of the miPEP treatments in post-hoc Tukey HSD tests. The application of the miPEP on the crown of Arabidopsis did not, however, affect the bacterial alpha diversity nor abundance.

Finally, we subjected in vitro a simplified microbial community, that was enriched from an agricultural soil, to a mix of five synthetic plant miRNAs (ath-miR158a-3p, ath-miR158b, ath-miR159a, ath-miR827, and ath-miR5642b), or a mix of their scrambled counterparts. Exposure to the plant miRNAs significantly disturbed the growth of the microbial communities during the log phase (t-test: P < .05, Fig. 5D). The bacterial community contained 20 amplicon sequence variants (ASVs) at the endpoint of the incubation (52 h) and we found significant shifts in the bacterial community composition due to the plant miRNA exposure (PERMANOVA: R2 = 0.39, P = .003, Fig. 5E). At the genus level, the plant miRNAs increased the relative abundance of the Acinetobacter and decreased the relative abundance of the Citrobacter and Enterobacter (Fig. 5E). At the ASV level, plant miRNAs decreased the abundance of three ASVs related to the genera Enterobacter (P = .0208) and Citrobacter (P = .0184) and to the Enterobacteriaceae family (P = .0184) and increased the relative abundance of an ASV related to the genus Acinetobacter (P = .00604; Fig. 5F).

Discussion

Using multiple lines of independent evidence, we confirmed our hypotheses. Plant miRNAs are present in the rhizosphere, and they induce changes in the transcriptome of a rhizosphere bacterium, leading to shifts in the bacterial community. Plants and microorganisms are known to interact using small RNAs [4–14, 18] and the gut microbiota was shown to be shaped by miRNAs, including plant miRNAs [19–22], but this is the first report of this mechanism for plant-bacterial community interactions in the rhizosphere.

The two model plants, A. thaliana and B. distachyon harbored a similar complement of plant miRNAs in their rhizospheres. Although it would require confirmation from more plant species, the presence of similar miRNAs in the rhizosphere of a dicotyledon and a monocotyledon suggests a conserved feature among land plants. All the major miRNAs that we found in the rhizosphere of Arabidopsis were also detected in the roots. This agrees well with previous reports of root miRNAs, where the two most abundant rhizospheric miRNAs found in our study—ath-miR158a and ath-miR161.1—were reported to be highly enriched in the early meristematic zone of the roots [23]. Many root exudates, such as extracellular DNA, soluble compounds, and mucilage, are produced and secreted, by border cells, in this region of the root tip [41]. Even though there was a large overlap between the root and rhizosphere miRNAs in our two experiments, the rank abundance of the miRNAs was not the same, alluding to a potential selection mechanism for the miRNAs that make it to the rhizosphere. Alternatively, this pattern could also be explained by different half-lives in the rhizosphere, preferential uptake by bacteria, the use of entire roots for sequencing, or by the slightly different conditions under which the two experiments were run. Finally, bacteria growing in the rhizosphere of Arabidopsis harbored plant miRNA inside or outside their cells, confirming that the rhizosphere miRNAs reached the rhizosphere bacteria.

In planta, in the same way that some bacteria secrete small RNAs in outer membrane vesicles [42], bacteria may internalize external DNA via vesiduction [43], i.e. membrane fusion of a vesicle containing DNA or RNA. The use of vesicles seems, however, not necessary for plant-microbe miRNA-based interactions, as exposure to naked miRNAs led to transcriptomic and community shifts. The uptake of naked or vesicle-borne miRNAs was already shown for the gut microbiota [19–22] and is now generally accepted. The incorporation of eukaryote miRNAs in bacteria is also consistent with their ability to absorb environmental nucleic acids, such as extracellular DNA, through natural competence [44]. It is, however, difficult to measure the total concentration of miRNA in the rhizosphere, and how that compared to the amounts used for our in vitro experiments. In vitro, we used concentrations previously used for gut microbiota studies [19], but we are not sure if this corresponds to the concentrations usually found in the rhizosphere.

In plants, miRNA induce mRNA cleavage or translation inhibition, through near perfect sequence complementarity [45]. Studies in the human gut also suggested that host miRNAs interact with bacterial mRNA through sequence complementarity [19, 21], so we used the rules for plant miRNA based on sequence homology, to search for targets in bacterial genomes. We also used this model to select the genes targeted by ath-miR159c in the V. paradoxus genome, which expression was quantified by RT-qPCR in Variovorax growing in the rhizosphere of gnotobiotic Arabidopsis treated with miPEP 159c. We found that one out of three targets was indeed inhibited following miPEP 159c application and increased expression of the primary transcript of miR159c. The third gene, encoding for alpha-2-macroglobulin, was, however, overexpressed in the presence of the miPEP. In bacteria, non-coding small RNAs can sometimes induce expression of target mRNAs [19, 42]. Even though this gnotobiotic experiment showed that bioinformatic prediction tools worked relatively well when focusing on a few genes, V. paradoxus differentially expressed only 4 of the 137 predicted targets and differentially expressed another 247 non-predicted genes when exposed in liquid cultures to a mixture of six rhizospheric miRNAs as compared to a control mixture of six corresponding scrambled miRNAs. Such a low level of overlap is comparable to what we would expect from two random set of genes. This suggests that either (i) the bacteria adjusted their transcriptome in response to the shift in the expression of the miRNA-targeted gene, (ii) as it is often the case in plants [46], the miRNAs targeted bacterial transcription factors, such as the ones from the LysR family that were differentially expressed at both time points, (iii) target genes were translationally repressed, which would be undetectable with transcriptomics, though this is less common in plants than mRNA cleavage [45], (iv) miRNAs protected targets from repression, as it was shown for arbuscular mycorrhizal fungi [47], or (v) because many predicted targets of the rhizospheric miRNAs were not in CDS, miRNAs could have affected DNA methylation [48] or interacted with gene promoters [49]. Clearly, the bioinformatic tools used were not appropriate to predict the full complement of genes that were affected by the miRNAs. Several supplementary experiments will be needed to confirm the targets and mechanism of action of plant miRNA in bacteria. Among other approaches [50], reporter gene assays [22, 51], biochemical fishing [52, 53], in vivo RNA proximity ligation [54], 5′-RLM RACE [55] and mutant bacteria with resistant targets could be envisioned in further studies.

In contrast to Variovorax, plant miRNAs did not impact the transcriptome of Bacillus. In our simplified soil community, only 4 out of the 20 ASVs were significantly impacted by plant miRNAs, and, similarly, intestinal miRNAs only impacted the growth of specific bacterial strains [19]. Plant exosomes containing miRNAs are also preferentially taken up by some bacteria, affecting their gene expression and activity [21]. Our experiments were, however, carried out using naked miRNAs, excluding this explanation. After the transcriptomic experiment, our hypothesis was that the different responses between the two bacteria could be related to differences in the cell wall that made miRNA entry impossible for Gram-positive. But our microscopy work disproved that. Both our Gram-positive and Gram-negative model bacteria showed that they could take up the miRNAs. Another possibility is that the mechanism of interaction differs between the two groups of bacteria. In bacteria, Argonaute homologs [56, 57] and chaperone proteins such as RNA-binding Hfq, ProQ, or CrsA proteins protect small RNAs and stabilize their interaction with mRNA, improving the formation of sRNA-mRNA duplexes that lead to gene silencing [58]. This crucial role of chaperones in sRNA-mediated interactions was mostly reported for Gram-negative bacteria and only for a handful of Gram-positive bacteria [59, 60]. Alternatively, competence for DNA uptake depends on environmental conditions, such as stress, nutrient availability, and cell density [44]. The two bacteria tested might have different cues to initiate nucleic acid uptake, and the growing conditions might not have been met during the transcriptomic experiment to trigger this behavior in Bacillus. Another possibility is that these bacteria employ distinct sRNA turnover mechanisms through the action of RNA nucleases such as, RNase III, RNase E, and PNPase, which accelerate sRNA turnover [61, 62]. In any case, the differential transcriptomic response to miRNAs of the two bacteria tested suggests a selective mechanism in the rhizosphere. Any effect on a bacterium could have cascading effects on the rest of the community.

Arabidopsis miRNAs impacted the bacterial community in the root environment. First, we examined the root-associated bacterial community of A. thaliana mutants affected in the biosynthesis of miRNA and/or sRNA. Many of these mutants had disrupted bacterial communities compared to WT plants. One of the most relevant mutants, the dcl1-2 mutant, which is specifically impaired in miRNA production, was severely affected in its community composition at the phylum level and harbored a more diversified community than the WT plants. The microbial community composition and diversity in the roots and rhizosphere of the mutants resembled those of an unplanted soil more than those of WT plants. This suggests that mutations in small RNA related pathways lead to a weaker selective pressure in the roots and rhizosphere of the mutants. The mutations used are, however, pleiotropic, and plants were severely affected in their phenotype, which could have also affected the bacterial community.

Second, the bacterial community associated with soil-grown Arabidopsis responded significantly to miPEP application. As we showed in the in vitro experiment using miPEP159c, miPEPs stimulate the production of their corresponding miRNA in plant tissues. This means that the upregulation of a single miRNA could lead to changes in the bacterial community. Other than the direct effect of the overexpressed miRNA on the bacteria, the shifts observed in bacterial community could, however, also be explained by other factors. For instance, plant miRNAs alter various physiological processes within the plant, such as root development and plant immune response [46].

Third, to exclude most of the unavoidable indirect plant-mediated effects of the mutant and the miPEP experiments, we did an in vitro experiment with a simplified soil-derived bacterial community of twenty ASVs exposed to a mixture of synthetic Arabidopsis miRNAs. The plant miRNAs affected the abundance of four ASVs and the growth of the community during the log phase, as compared to a mixture of scrambled miRNAs. This shows that plant miRNAs directly affect bacterial communities. At the individual level, a bacterium could change in relative abundance because of (i) direct effect of the miRNA on its growth or (ii) changes in the relative abundance of other bacteria with which it interacts. Some species can have a keystone role in interaction networks [63], and a shift in these species would influence the entire community. For instance, the presence of Variovorax in the rhizosphere of Arabidopsis counteracted the root growth inhibition induced by many other members of the community [64]. Plant miPEPs, through their effect on miRNAs, also modulated the interactions between plants and key root symbionts, such as arbuscular mycorrhizal fungi [47] and rhizobia [65]. This effect would profoundly alter the microbial community, even if the miRNAs had only affected a single or a few keystone species.

We showed here for the first time that plant miRNAs are found in the rhizosphere of two model plants and that they affect the transcriptome of a rhizosphere bacterium, and that they modulate soil microbial community composition and growth. The rhizosphere effect is thought to be mainly due to the rhizodeposition of various small organic molecules. Our study shows that miRNAs should be added to the molecules that plants use to interact with their rhizosphere microbiota.

Supplementary Material

figS1_ycae120
figs1_ycae120.jpeg (14.2MB, jpeg)
SupplementaryTableS1_ycae120
supplementarytables1_ycae120.pdf (225.5KB, pdf)
SupplementaryTableS2_ycae120
supplementarytables2_ycae120.docx (18.4KB, docx)
SupplementaryTableS3_ycae120
supplementarytables3_ycae120.docx (219.2KB, docx)
20240510SuppMethods_ycae120
20240510suppmethods_ycae120.docx (64.2KB, docx)

Acknowledgements

We want to thank Hervé Vaucheret and Taline Elmayan for sharing their Arabidopsis mutant seeds with us and Paul M. Orwin for providing the V. paradoxus EPS strain. We also thank Christophe Penno for the helpful discussions and Jessy Tremblay for the expertise and support regarding the confocal microscopy and flow cytometry experiments.

Contributor Information

Harriet Middleton, Écosystèmes, Biodiversité, Évolution (ECOBIO), Unité mixte de recherche (UMR) 6553, Centre national de la recherche scientifique (CNRS) - Université de Rennes, Campus Beaulieu, 263 Avenue du Général Leclerc, Rennes, 35042, France; Institut National de la Recherche Scientifique, Centre Armand-Frappier Santé Biotechnologie, 531 boulevard des Prairies, Laval, Québec, H7V 1B7, Canada.

Jessica Ann Dozois, Institut National de la Recherche Scientifique, Centre Armand-Frappier Santé Biotechnologie, 531 boulevard des Prairies, Laval, Québec, H7V 1B7, Canada.

Cécile Monard, Écosystèmes, Biodiversité, Évolution (ECOBIO), Unité mixte de recherche (UMR) 6553, Centre national de la recherche scientifique (CNRS) - Université de Rennes, Campus Beaulieu, 263 Avenue du Général Leclerc, Rennes, 35042, France.

Virginie Daburon, Écosystèmes, Biodiversité, Évolution (ECOBIO), Unité mixte de recherche (UMR) 6553, Centre national de la recherche scientifique (CNRS) - Université de Rennes, Campus Beaulieu, 263 Avenue du Général Leclerc, Rennes, 35042, France.

Emmanuel Clostres, Écosystèmes, Biodiversité, Évolution (ECOBIO), Unité mixte de recherche (UMR) 6553, Centre national de la recherche scientifique (CNRS) - Université de Rennes, Campus Beaulieu, 263 Avenue du Général Leclerc, Rennes, 35042, France.

Julien Tremblay, Institut National de la Recherche Scientifique, Centre Armand-Frappier Santé Biotechnologie, 531 boulevard des Prairies, Laval, Québec, H7V 1B7, Canada.

Jean-Philippe Combier, Laboratoire de recherche en sciences végétales (LRSV), UMR 5546, Université Paul-Sabatier - CNRS -Institut national polytechnique, 24 chemin de Borde Rouge, Auzeville-Tolosane, 31320, France.

Étienne Yergeau, Institut National de la Recherche Scientifique, Centre Armand-Frappier Santé Biotechnologie, 531 boulevard des Prairies, Laval, Québec, H7V 1B7, Canada.

Abdelhak El Amrani, Écosystèmes, Biodiversité, Évolution (ECOBIO), Unité mixte de recherche (UMR) 6553, Centre national de la recherche scientifique (CNRS) - Université de Rennes, Campus Beaulieu, 263 Avenue du Général Leclerc, Rennes, 35042, France.

Conflicts of interest

The authors report no conflict of interest.

Funding

This work was financially supported by a CNRS MITI (Mission pour les Initiatives Transverses et Interdisciplinaires 2020–2021), by an IEA (International Emerging Actions) grant (France), by the French National program EC2CO (Ecosphère Continentale et Côtière 2024–2025), a CRSNG discovery grant (RGPIN-2020-05723), a FRQNT Samuel-de-Champlain grant (2019-FQ-264131). We acknowledge Compute Canada for access to the Graham compute cluster.

Data availability

We deposited the sequence data generated in this study under NCBI BioProject accessions PRJNA836586 (rhizosphere miRNAs), PRJNA1107220 (root miRNAs), PRJNA1111839 (soil-extracted bacteria miRNAs), PRJNA1111831 (mutant and miPEP 16S rRNA gene amplicons), PRJNA1111829 (simplified soil community 16S rRNA gene amplicons), and PRJNA1111827 (Bacillus and Variovorax transcriptomics). The R code used to manipulate the data and generate the figures is available on GitHub (https://github.com/le-labo-yergeau/Middleton_miRNA) with the accompanying data being available on Zenodo (https://zenodo.org/doi/10.5281/zenodo.11105307).

References

  • 1. Huang  C-Y, Wang  H, Hu  P  et al.  Small RNAs—big players in plant–microbe interactions. Cell Host Microbe  2019;26:173–82. 10.1016/j.chom.2019.07.021 [DOI] [PubMed] [Google Scholar]
  • 2. Regmi  R, Penton  CR, Anderson  J  et al.  Do small RNAs unlock the below ground microbiome-plant interaction mystery?  Front Mol Biosci  2022;9:1017392. 10.3389/fmolb.2022.1017392 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3. Middleton  H, Yergeau  É, Monard  C  et al.  Rhizospheric plant–microbe interactions: miRNAs as a key mediator. Trends Plant Sci  2021;26:132–41. 10.1016/j.tplants.2020.09.005 [DOI] [PubMed] [Google Scholar]
  • 4. Zhang  T, Zhao  Y-L, Zhao  J-H  et al.  Cotton plants export microRNAs to inhibit virulence gene expression in a fungal pathogen. Nat Plants.  2016;2:16153. 10.1038/nplants.2016.153 [DOI] [PubMed] [Google Scholar]
  • 5. Cai  Q, Qiao  L, Wang  M  et al.  Plants send small RNAs in extracellular vesicles to fungal pathogen to silence virulence genes. Science  2018;360:1126–9. 10.1126/science.aar4142 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Qin  S, Veloso  J, Puccetti  G  et al.  Molecular characterization of cross-kingdom RNA interference in Botrytis cinerea by tomato small RNAs. Front Plant Sci  2023;14:1107888. 10.3389/fpls.2023.1107888 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Wu  F, Huang  Y, Jiang  W  et al.  Genome-wide identification and validation of tomato-encoded sRNA as the cross-species antifungal factors targeting the virulence genes of Botrytis cinerea. Front Plant Sci  2023;14:1072181. 10.3389/fpls.2023.1072181 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Hou  Y, Zhai  Y, Feng  L  et al.  A Phytophthora effector suppresses trans-kingdom RNAi to promote disease susceptibility. Cell Host Microbe  2019;25:153–165.e5. 10.1016/j.chom.2018.11.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Jiao  J, Peng  D. Wheat microRNA1023 suppresses invasion of Fusarium graminearum via targeting and silencing FGSG_03101. J Plant Interact  2018;13:514–21. 10.1080/17429145.2018.1528512 [DOI] [Google Scholar]
  • 10. Wang  M, Weiberg  A, Lin  FM  et al.  Bidirectional cross-kingdom RNAi and fungal uptake of external RNAs confer plant protection. Nat Plants  2016;2:16151. 10.1038/nplants.2016.151 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Weiberg  A, Wang  M, Lin  F-M  et al.  Fungal small RNAs suppress plant immunity by hijacking host RNA interference pathways. Science  2013;342:118–23. 10.1126/science.1239705 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Wang  M, Weiberg  A, Dellota  E  Jr  et al.  Botrytis small RNA Bc-siR37 suppresses plant defense genes by cross-kingdom RNAi. RNA Biol  2017;14:421–8. 10.1080/15476286.2017.1291112 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Ren  B, Wang  X, Duan  J  et al.  Rhizobial tRNA-derived small RNAs are signal molecules regulating plant nodulation. Science  2019;365:919–22. 10.1126/science.aav8907 [DOI] [PubMed] [Google Scholar]
  • 14. Wang  B, Sun  Y, Song  N  et al.  Puccinia striiformis f. sp. tritici microRNA-like RNA 1 (Pst-milR1), an important pathogenicity factor of Pst, impairs wheat resistance to Pst by suppressing the wheat pathogenesis-related 2 gene. New Phytol  2017;215:338–50. 10.1111/nph.14577 [DOI] [PubMed] [Google Scholar]
  • 15. Ledford  WC, Silvestri  A, Fiorilli  V  et al.  A journey into the world of small RNAs in the arbuscular mycorrhizal symbiosis. New Phytol  2024;242:1534–44. 10.1111/nph.19394 [DOI] [PubMed] [Google Scholar]
  • 16. Qiao  SA, Gao  ZY, Roth  R. A perspective on cross-kingdom RNA interference in mutualistic symbioses. New Phytol  2023;240:68–79. 10.1111/nph.19122 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Silvestri  A, Fiorilli  V, Miozzi  L  et al.  In silico analysis of fungal small RNA accumulation reveals putative plant mRNA targets in the symbiosis between an arbuscular mycorrhizal fungus and its host plant. BMC Genomics  2019;20:169. 10.1186/s12864-019-5561-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Regmi  R, Panchal  S, Hicks  M  et al.  Multi omics aided small RNA profiling of wheat rhizosphere and their potential targets in contrasting soils for Rhizoctonia solani AG8 suppression. bioRxiv  2024; 2024.06.17.599338. 2024.06.17.599338
  • 19. Liu  S, da Cunha  AP, Rezende  RM  et al.  The host shapes the gut microbiota via fecal MicroRNA. Cell Host Microbe  2016;19:32–43. 10.1016/j.chom.2015.12.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Liu  S, Rezende  RM, Moreira  TG  et al.  Oral administration of miR-30d from feces of MS patients suppresses MS-like symptoms in mice by expanding Akkermansia muciniphila. Cell Host Microbe  2019;26:779–794.e8. 10.1016/j.chom.2019.10.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Teng  Y, Ren  Y, Sayed  M  et al.  Plant-derived exosomal MicroRNAs shape the gut microbiota. Cell Host Microbe  2018;24:637–652.e8. 10.1016/j.chom.2018.10.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Xu  Q, Qin  X, Zhang  Y  et al.  Plant miRNA bol-miR159 regulates gut microbiota composition in mice: in vivo evidence of the crosstalk between plant miRNAs and intestinal microbes. J Agricult Food Chem  2023;71:16160–73. 10.1021/acs.jafc.3c06104 [DOI] [PubMed] [Google Scholar]
  • 23. Breakfield  NW, Corcoran  DL, Petricka  JJ  et al.  High-resolution experimental and computational profiling of tissue-specific known and novel miRNAs in Arabidopsis. Genome Res  2012;22:163–76. 10.1101/gr.123547.111 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. DeAngelis  KM, Brodie  EL, DeSantis  TZ  et al.  Selective progressive response of soil microbial community to wild oat roots. ISME J  2009;3:168–78. 10.1038/ismej.2008.103 [DOI] [PubMed] [Google Scholar]
  • 25. Tian  T, Reverdy  A, She  Q  et al.  The role of rhizodeposits in shaping rhizomicrobiome. Environ Microbiol Rep  2020;12:160–72. 10.1111/1758-2229.12816 [DOI] [PubMed] [Google Scholar]
  • 26. Griffiths  RI, Whiteley  AS, O’Donnell  AG  et al.  Rapid method for coextraction of DNA and RNA from natural environments for analysis of ribosomal DNA- and rRNA-based microbial community composition. Appl Environ Microb  2000;66:5488–91. 10.1128/AEM.66.12.5488-5491.2000 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Levy  A, Salas Gonzalez  I, Mittelviefhaus  M  et al.  Genomic features of bacterial adaptation to plants. Nat Genetics  2017;50:138–50. 10.1038/s41588-017-0012-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Penno  C, Tremblay  J, O'Connell Motherway  M  et al.  Analysis of small non-coding RNAs as signaling intermediates of environmentally integrated responses to abiotic stress. Methods Mol Biol  2023;2642:403–27. 10.1007/978-1-0716-3044-0_22 [DOI] [PubMed] [Google Scholar]
  • 29. Dai  X, Zhuang  Z, Zhao  PX. psRNATarget: a plant small RNA target analysis server (2017 release). Nucl Acids Res  2018;46:W49–54. 10.1093/nar/gky316 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Shamandi  N, Zytnicki  M, Charbonnel  C  et al.  Plants encode a general siRNA suppressor that is induced and suppressed by viruses. PLoS Biol  2015;13:e1002326. 10.1371/journal.pbio.1002326 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Morel  JB, Godon  C, Mourrain  P  et al.  Fertile hypomorphic ARGONAUTE (ago1) mutants impaired in post-transcriptional gene silencing and virus resistance. Plant Cell  2002;14:629–39. 10.1105/tpc.010358 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Xie  Z, Kasschau  KD, Carrington  JC. Negative feedback regulation of dicer-Like1 in Arabidopsis by microRNA-guided mRNA degradation. Curr Biol  2003;13:784–9. 10.1016/S0960-9822(03)00281-1 [DOI] [PubMed] [Google Scholar]
  • 33. Boutet  S, Vazquez  F, Liu  J  et al.  Arabidopsis HEN1: a genetic link between endogenous miRNA controlling development and siRNA controlling transgene silencing and virus resistance. Curr Biol  2003;13:843–8. 10.1016/S0960-9822(03)00293-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Li  J, Yang  Z, Yu  B  et al.  Methylation protects miRNAs and siRNAs from a 3′-end uridylation activity in Arabidopsis. Curr Biol  2005;15:1501–7. 10.1016/j.cub.2005.07.029 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35. Tremblay  J, Yergeau  É. Systematic processing of rRNA gene amplicon sequencing data. GigaScience  2019;8:giz146. 10.1093/gigascience/giz146 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. McMurdie  PJ, Holmes  S. Phyloseq: an R package for reproducible interactive analysis and graphics of microbiome census data. PLoS One  2013;8:e61217. 10.1371/journal.pone.0061217 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Han  JI, Spain  JC, Leadbetter  JR  et al.  Genome of the root-associated plant growth-promoting bacterium Variovorax paradoxus strain EPS. Genome Announc  2013;1:1. 10.1128/genomeA.00843-13 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38. Agoussar  A, Azarbad  H, Tremblay  J  et al.  The resistance of the wheat microbial community to water stress is more influenced by plant compartment than reduced water availability. FEMS Microbiol Ecol  2021;97:fiab149. 10.1093/femsec/fiab149 [DOI] [PubMed] [Google Scholar]
  • 39. Lauressergues  D, Couzigou  J-M, San Clemente  H  et al.  Primary transcripts of microRNAs encode regulatory peptides. Nature  2015;520:90–3. 10.1038/nature14346 [DOI] [PubMed] [Google Scholar]
  • 40. Couzigou  J-M, Lauressergues  D, Bécard  G  et al.  miRNA-encoded peptides (miPEPs): a new tool to analyze the roles of miRNAs in plant biology. RNA Biol  2015;12:1178–80. 10.1080/15476286.2015.1094601 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41. Sasse  J, Martinoia  E, Northen  T. Feed your friends: do plant exudates shape the root microbiome?  Trends Plant Sci  2018;23:25–41. 10.1016/j.tplants.2017.09.003 [DOI] [PubMed] [Google Scholar]
  • 42. Choi  J-W, Um  J-H, Cho  J-H  et al.  Tiny RNAs and their voyage via extracellular vesicles: secretion of bacterial small RNA and eukaryotic microRNA. Exp Biol Med  2017;242:1475–81. 10.1177/1535370217723166 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43. Soler  N, Forterre  P. Vesiduction: the fourth way of HGT. Environ Microbiol  2020;22:2457–60. 10.1111/1462-2920.15056 [DOI] [PubMed] [Google Scholar]
  • 44. Blokesch  M. Natural competence for transformation. Curr Biol  2016;26:R1126–30. 10.1016/j.cub.2016.08.058 [DOI] [PubMed] [Google Scholar]
  • 45. Yu  Y, Jia  T, Chen  X. The ‘how’ and ‘where’ of plant microRNAs. New Phytol  2017;216:1002–17. 10.1111/nph.14834 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46. Tang  J, Chu  C. MicroRNAs in crop improvement: fine-tuners for complex traits. Nat Plants  2017;3:17077. 10.1038/nplants.2017.77 [DOI] [PubMed] [Google Scholar]
  • 47. Couzigou  JM, Lauressergues  D, André  O  et al.  Positive gene regulation by a natural protective miRNA enables arbuscular mycorrhizal symbiosis. Cell Host Microbe  2017;21:106–12. 10.1016/j.chom.2016.12.001 [DOI] [PubMed] [Google Scholar]
  • 48. Wu  L, Zhou  H, Zhang  Q  et al.  DNA methylation mediated by a MicroRNA pathway. Mol Cell  2010;38:465–75. 10.1016/j.molcel.2010.03.008 [DOI] [PubMed] [Google Scholar]
  • 49. Toscano-Garibay  JD, Aquino-Jarquin  G. Transcriptional regulation mechanism mediated by miRNA–DNA•DNA triplex structure stabilized by Argonaute. Biochim Biophys Acta  2014;1839:1079–83. 10.1016/j.bbagrm.2014.07.016 [DOI] [PubMed] [Google Scholar]
  • 50. Vogel  J, Wagner  EGH. Target identification of small noncoding RNAs in bacteria. Curr Opin Microbiol  2007;10:262–70. 10.1016/j.mib.2007.06.001 [DOI] [PubMed] [Google Scholar]
  • 51. Jen  JY, Tang  YA, Lu  YH  et al.  Oct4 transcriptionally regulates the expression of long non-coding RNAs and to promote lung cancer progression. Mol Cancer  2017;16:104. 10.1186/s12943-017-0674-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52. Antal  M, Bordeau  V, Douchin  V  et al.  A small bacterial RNA regulates a putative ABC transporter. J Biol Chem  2005;280:7901–8. 10.1074/jbc.M413071200 [DOI] [PubMed] [Google Scholar]
  • 53. Douchin  V, Bohn  C, Bouloc  P. Down-regulation of porins by a small RNA bypasses the essentiality of the regulated intramembrane proteolysis protease RseP in. J Biol Chem  2006;281:12253–9. 10.1074/jbc.M600819200 [DOI] [PubMed] [Google Scholar]
  • 54. Zhang  YF, Han  K, Chandler  CE  et al.  Probing the sRNA regulatory landscape of P. aeruginosa: post-transcriptional control of determinants of pathogenicity and antibiotic susceptibility. Mol Microbiol  2017;106:919–37. 10.1111/mmi.13857 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55. Llave  C, Xie  ZX, Kasschau  KD  et al.  Cleavage of mRNA targets directed by a class of miRNA. Science  2002;297:2053–6. 10.1126/science.1076311 [DOI] [PubMed] [Google Scholar]
  • 56. Lisitskaya  L, Kropocheva  E, Agapov  A  et al.  Bacterial Argonaute nucleases reveal different modes of DNA targeting in vitro and in vivo. Nucleic Acids Res  2023;51:5106–24. 10.1093/nar/gkad290 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57. Makarova  KS, Wolf  YI, van der Oost  J  et al.  Prokaryotic homologs of Argonaute proteins are predicted to function as key components of a novel system of defense against mobile genetic elements. Biol Direct  2009;4:29. 10.1186/1745-6150-4-29 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58. Boutet  E, Djerroud  S, Perreault  J. Small RNAs beyond model organisms: have we only scratched the surface?  Int J Mol Sci  2022;23:4448. 10.3390/ijms23084448 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59. Nielsen  JS, Lei  LK, Ebersbach  T  et al.  Defining a role for Hfq in gram-positive bacteria: evidence for Hfq-dependent antisense regulation in listeria monocytogenes. Nucl Acids Res  2010;38:907–19. 10.1093/nar/gkp1081 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60. Müller  P, Gimpel  M, Wildenhain  T  et al.  A new role for CsrA: promotion of complex formation between an sRNA and its mRNA target in Bacillus subtilis. RNA Biol  2019;16:972–87. 10.1080/15476286.2019.1605811 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 61. Andrade  JM, Pobre  V, Matos  AM  et al.  The crucial role of PNPase in the degradation of small RNAs that are not associated with Hfq. RNA  2012;18:844–55. 10.1261/rna.029413.111 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 62. Saramago  M, Bárria  C, dos Santos  RF  et al.  The role of RNases in the regulation of small RNAs. Curr Opin Microbiol  2014;18:105–15. 10.1016/j.mib.2014.02.009 [DOI] [PubMed] [Google Scholar]
  • 63. Carlström  CI, Field  CM, Bortfeld-Miller  M  et al.  Synthetic microbiota reveal priority effects and keystone strains in the Arabidopsis phyllosphere. Nat Ecol Evol  2019;3:1445–54. 10.1038/s41559-019-0994-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 64. Finkel  OM, Salas-González  I, Castrillo  G  et al.  A single bacterial genus maintains root growth in a complex microbiome. Nature  2020;587:103–8. 10.1038/s41586-020-2778-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 65. Couzigou  J-M, André  O, Guillotin  B  et al.  Use of microRNA-encoded peptide miPEP172c to stimulate nodulation in soybean. New Phytol  2016;211:379–81. 10.1111/nph.13991 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

figS1_ycae120
figs1_ycae120.jpeg (14.2MB, jpeg)
SupplementaryTableS1_ycae120
supplementarytables1_ycae120.pdf (225.5KB, pdf)
SupplementaryTableS2_ycae120
supplementarytables2_ycae120.docx (18.4KB, docx)
SupplementaryTableS3_ycae120
supplementarytables3_ycae120.docx (219.2KB, docx)
20240510SuppMethods_ycae120
20240510suppmethods_ycae120.docx (64.2KB, docx)

Data Availability Statement

We deposited the sequence data generated in this study under NCBI BioProject accessions PRJNA836586 (rhizosphere miRNAs), PRJNA1107220 (root miRNAs), PRJNA1111839 (soil-extracted bacteria miRNAs), PRJNA1111831 (mutant and miPEP 16S rRNA gene amplicons), PRJNA1111829 (simplified soil community 16S rRNA gene amplicons), and PRJNA1111827 (Bacillus and Variovorax transcriptomics). The R code used to manipulate the data and generate the figures is available on GitHub (https://github.com/le-labo-yergeau/Middleton_miRNA) with the accompanying data being available on Zenodo (https://zenodo.org/doi/10.5281/zenodo.11105307).

Articles from ISME Communications are provided here courtesy of Oxford University Press

RESOURCES