How to Study Plant microRNA

MicroRNA (miRNA) are endogenous, single-stranded non-coding RNA molecules prevalent in eukaryotic organisms, typically comprising approximately 22 nucleotides in length. Since the initial identification of miRNA in plants in 2002, these molecules have garnered significant attention within the domain of plant molecular biology. Extensive research over the past decades has elucidated the critical regulatory roles of miRNA in various physiological and developmental processes, including plant growth and development, signal transduction, response to environmental stress, and the biosynthesis of secondary metabolites. Consequently, research focused on miRNA is of paramount scientific importance and holds substantial practical significance.

Background of miRNA

Biogenesis of miRNA

The biogenesis of miRNA in plants is a complex and sequential series of biochemical reactions, encompassing key stages such as transcription, precursor cleavage, methylation modification, and the formation of the RNA-induced silencing complex (RISC). Initially, within the nucleus, miRNA-coding genes undergo transcription catalyzed by RNA polymerase II, producing primary transcripts (pri-miRNAs) that are several hundred nucleotides in length. These pri-miRNAs subsequently undergo 5' capping and 3' polyadenylation, and fold into a hairpin structure to form the primary miRNA.

Following this, the pri-miRNA is processed by a microprocessor complex consisting of core proteins such as Dicer-like 1 (DCL1), Hyponastic Leaves 1 (HYL1), and Serrate (SE), which cleaves the pri-miRNA into precursor miRNA (pre-miRNA). The pre-miRNA is then further cleaved to generate a miRNA duplex.

The miRNA duplex undergoes methylation at the 3' ends, a modification catalyzed by the methyltransferase HEN1. This methylation is crucial for the stability and function of miRNA, as it prevents degradation. Finally, one strand of the mature miRNA duplex is selectively incorporated into the RISC, facilitating the subsequent gene expression regulation, while the complementary strand is degraded. This intricate series of biochemical processes constitutes the miRNA biogenesis pathway in plants, providing a fundamental molecular basis for the precise regulation of gene expression within the plant system.

Figure 1 illustrates the principal processes of plant miRNA biogenesis (Deng et al., 2022).

Regulatory Mechanisms of miRNA

In plants, miRNA primarily exerts negative regulation on target genes through two mechanisms: mRNA cleavage and translational repression (Figure 2). The mRNA cleavage process is predominantly dependent on the endonuclease activity of Argonaute (AGO) proteins. Mature miRNAs associate with AGO proteins to form the RNA-induced silencing complex (RISC). Upon binding to target mRNAs that are complementary to the miRNA sequence, AGO1 cleaves the phosphodiester bond at the 10th to 11th nucleotide position of the target mRNA. This cleavage results in two mRNA fragments, which are subsequently degraded by exonucleases. It is noteworthy that some of these cleaved mRNA fragments may generate phased small interfering RNAs (phasiRNAs).

The progress in detecting miRNA target proteins at the protein level has been hindered by the slow development of plant protein antibodies. Consequently, research has predominantly focused on miRNA-mediated mRNA cleavage. Initially, it was widely believed that plant miRNAs primarily function through mRNA cleavage. However, accumulating evidence suggests that some plant miRNAs reduce target gene expression at the protein level without significantly affecting mRNA levels, indicating a translational repression mechanism similar to that observed in animals.

In this mechanism, mature miRNAs bind with AGO1 and other associated proteins to form a complex that, with the assistance of endoplasmic reticulum-associated proteins such as ALTERED MERISTEM PROGRAM 1 (AMP1), precisely targets translating mRNAs. This interaction inhibits ribosome assembly and the translation process. Further investigation and understanding of this mechanism remain essential for a comprehensive elucidation of miRNA function in plants.

Figure 2 provides an overview of the operational models of plant miRNAs (Yu et al., 2017).

Basic Research Approaches for miRNA

This section outlines the fundamental research approaches utilized in the study of miRNAs. Prior to initiating miRNA research, it is imperative to establish clear research objectives, specifically identifying the target miRNAs. Researchers typically achieve this by analyzing miRNA-seq data from experimental and control groups to identify differentially expressed miRNAs. Additionally, a review of relevant literature and database resources can aid in the identification of specific miRNAs of interest.

Identification of Target miRNAs

The preliminary task in miRNA research involves defining the research goals, particularly the identification of miRNAs to be studied. Researchers frequently employ miRNA-seq analyses to compare expression levels between experimental and control groups, thus pinpointing miRNAs with significant differential expression. Furthermore, consulting existing literature and database information is essential for selecting specific miRNA candidates.

Functional Validation

Upon identification of the target miRNAs, researchers can proceed with functional validation by constructing overexpression or silencing vectors. These vectors can be employed in either stable or transient transformation experiments to elucidate the biological roles of the miRNAs. Such experimental techniques are pivotal for comprehending the mechanisms through which miRNAs influence biological systems.

Regulatory Mechanisms and Target Gene Identification

To further elucidate the regulatory mechanisms of miRNAs, bioinformatics prediction methods can be utilized alongside experimental approaches such as degradome sequencing, 5' RLM-RACE, quantitative real-time PCR, and western blotting. These methods facilitate the identification of miRNA target genes. In-depth analysis of these target genes allows for a comprehensive understanding of the regulatory relationships between miRNAs, their target genes, and the resulting phenotypes, thereby providing robust support for subsequent research endeavors.

Techniques and Methodologies in miRNA Research

Figure 3 miRNA Research Techniques and Methodologies

miRNA-seq

miRNA-seq constitutes a high-throughput sequencing technology specifically designed for the detection of miRNA expression profiles. Given the brevity of miRNA sequences, adapter ligation can be directly incorporated during library preparation to facilitate reverse transcription amplification. Subsequent to this, PCR amplification and the addition of sequencing adapters are performed, followed by the purification of DNA fragments within a specific size range for sequencing. The integration of data analysis enables comprehensive elucidation of miRNA expression patterns (Figure 4).

Figure 4 depicts the workflow for miRNA sequencing experiments.

Case study

A research article titled "Integrated transcriptome and microRNA sequencing analyses reveal gene responses in poplar leaves infected by the novel pathogen bean common mosaic virus (BCMV)" was published in Frontiers in Plant Science. This study reports on the impact of Bean common mosaic virus (BCMV) on poplar trees and the molecular mechanisms of poplar response to viral infection.

In this study, RNA-seq and miRNA-seq analyses were utilized to unveil differentially expressed genes and miRNAs in diseased poplar leaves. The miRNA-seq data analysis identified 84 differentially expressed miRNAs in short-term diseased leaves (SD, showing newly emerged mild mosaic symptoms) and 89 differentially expressed miRNAs in long-term diseased leaves (LD, exhibiting typical and significant mosaic symptoms). Among these differentially expressed miRNAs, 78 were known miRNAs belonging to 21 miRNA families, with miR156 being the largest family comprising 12 members, followed by miR395 and miR167. Furthermore, results indicated that miRNAs belonging to the same family exhibited consistent expression patterns in both SD and LD leaves.

Subsequently, TargetFinder was employed to predict the target genes of these miRNAs. In SD leaves, 2079 target genes were predicted, whereas 1656 target genes were predicted in LD leaves. KEGG pathway analysis of predicted genes revealed no significantly enriched metabolic pathways in SD leaves, whereas in LD leaves, the peroxisome pathway and isoquinoline alkaloid biosynthesis pathway were significantly enriched (Figure 5).

This study provides insights into the regulatory roles of miRNAs and their target genes in poplar leaves infected by BCMV, shedding light on the molecular mechanisms underlying the host-pathogen interaction.

Figure 5 illustrates the Venn diagrams of differentially expressed genes (DEGs) and differentially expressed miRNAs (DEMs) in Populus leaf tissues affected by leaf spot disease, along with KEGG pathway enrichment analysis diagrams (Wang et al., 2023). Panel (A) shows the Venn diagram of DEGs in diseased Populus leaves. Panels (B) and (C) depict KEGG pathway enrichment analysis diagrams of DEGs in short-day (SD) and long-day (LD) leaf samples, respectively. Panel (D) displays the Venn diagram of DEMs in diseased Populus leaves, while Panels (E) and (F) present KEGG pathway enrichment analysis diagrams of DEMs in SD and LD leaf samples, respectively.

Overexpression of miRNA

To study the overexpression of target miRNAs, it is essential to identify their pre-miRNAs initially. Subsequently, constructs are designed using robust promoters to drive the expression of pre-miRNAs (Figure 6). Following this, stable transformation or transient transfection is conducted according to experimental objectives to investigate the biological functions of miRNAs.

Several websites are recommended for locating miRNA precursor sequences:

(1) miRBase: http://microrna.sanger.ac.uk

(2) PmiRKB: PmiRKB Homepage (zju.edu.cn)

(3) RNAcentral: https://rnacentral.org/search?q=sbi-MIR156d

Figure 6 illustrates a schematic diagram of miRNA overexpression vector construction.

Silencing of miRNA

In addition to overexpressing miRNAs, silencing miRNAs is a commonly employed method for elucidating their biological functions. This section primarily introduces two techniques for miRNA silencing: Short Tandem Target Mimic (STTM) technology and CRISPR/Cas9 technology.

Short Tandem Target Mimic (STTM)

The STTM technology specifically induces the degradation of miRNA within plant cells, thereby reducing endogenous miRNA levels. The principle of STTM involves the artificial synthesis of a short tandem target mimic, which contains a sequence of approximately 48 nucleotides flanked by target miRNA binding sites at both ends. Each miRNA binding site includes three non-complementary bases (CTA), allowing the sequence to form an imperfectly complementary duplex with the target miRNA. This interaction effectively prevents the miRNA from binding to its target mRNA, thereby inhibiting miRNA function.

CRISPR/Cas9

The CRISPR/Cas9 technology can be utilized for the gene editing and functional analysis of certain miRNAs. When editing MIR gene sequences with this technology, single guide RNA (sgRNA) is typically designed to target regions within the mature miRNA or its precursor stem-loop structure, thereby achieving miRNA knockout. However, the presence of appropriate Protospacer Adjacent Motif (PAM) sequences is a limitation, and suitable sgRNA may not always be available in practical research applications (Deng et al., 2022).

In July 2020, Plant Biotechnology Journal published a study titled "Disruption of miRNA sequences by TALENs and CRISPR/Cas9 induces varied lengths of miRNA production." In this study, the authors employed CRISPR/Cas9 technology to edit the MIR160a gene in Arabidopsis thaliana. They designed gRNAs targeting both the miR160a and miR160a* strands, constructed a vector containing these two gRNAs, and performed genetic transformation. This approach successfully generated mutants with large segment deletions (Figure 7).

Figure 7 depicts the gene editing of Arabidopsis MIR160a using CRISPR/Cas9 (Bi et al., 2020). Panel (a) illustrates the design of CRISPR/Cas9 with dual gRNAs. miR160 and miR160a* strands are denoted in bold and underlined, respectively. PAM and double-strand break (DSB) sites are marked. Pairing sequences between gRNA and genomic DNA are highlighted in blue and orange. The resulting 47 or 48 bp deletion fragments are indicated. Panel (b) shows wild-type and mir160aΔ47 plants at two weeks old. Panel (c) presents the floral phenotype of mir160aΔ47 mutants. Panel (d) displays the siliques phenotype of wild-type and mir160aΔ47 mutants at the same developmental stage. Panel (e) depicts seeds of wild-type and mir160aΔ47 mutants, with blue arrows indicating seeds with delayed or arrested development and white arrows indicating unfertilized ovules.

Detection of miRNA

Following the overexpression or silencing of miRNA, it is imperative to assess miRNA levels to determine the experimental outcome. Given that miRNAs are not translatable into proteins, their detection is restricted to the transcriptional level. Current methodologies for evaluating miRNA expression include qPCR, Northern blotting, microarray analysis, and RNA-seq. Among these techniques, qPCR and Northern blotting are most frequently employed.

It is noteworthy that due to the brevity of mature miRNA sequences, specific adaptations are required for accurate quantification via qPCR. Techniques such as stem-loop primers or tailing methods are utilized to extend the mature miRNA sequences, thus facilitating precise quantification.

Prediction and Validation of miRNA Target Genes

Bioinformatics Prediction

The use of bioinformatics for predicting miRNA target genes provides essential insights for subsequent functional analyses. In plants, miRNAs typically bind to their target mRNAs with nearly perfect complementarity, simplifying the prediction process and obviating the need for complex algorithms. Consequently, the prediction of miRNA target genes in plants is relatively straightforward. Several commonly utilized prediction tools include:

psRNATarget: http://plantgrn.noble.org/psRNATarget/

TAPIR: http://bioinformatics.psb.ugent.be/webtools/tapir/

miRDeepFinder: http://www.leonxie.com/DeepFinder.php

psRobot: http://omicslab.genetics.ac.cn/psRobot/index.php

WMD3: http://wmd3.weigelworld.org/cgi-bin/webapp.cgi

It should be noted that target genes predicted through bioinformatics may contain false positives. Whether miRNAs can effectively target and regulate gene expression through cleavage or inhibition ultimately requires validation through biological experiments.

For further information, refer to "How to Predict miRNA Targets?"

Degradome Sequencing Technology

Degradome sequencing, also known as parallel analysis of RNA ends (PARE), employs high-throughput sequencing to analyze cleaved mRNA fragments, thus providing an accurate and efficient method for identifying miRNA target genes (German et al., 2008). This technique is predicated on the principle that miRNA binding to its target mRNA induces cleavage at the 10th or 11th nucleotide of the complementary site, resulting in two fragments. The 5' cleavage fragment retains the 5' cap structure and the 3' hydroxyl group, while the 3' cleavage fragment possesses a free 5' monophosphate and a 3' poly(A) tail.

The specific capture of miRNA cleavage products is achieved using a 5' adapter that selectively binds to 5' monophosphate RNA. The captured 3' cleavage fragments are then reverse-transcribed into cDNA and amplified. Restriction enzymes are used to digest the amplified products, and the resulting fragments are ligated to a 3' double-stranded DNA adapter and subsequently amplified and purified to construct a library for high-throughput sequencing (Figure 9). Analysis of sequencing data reveals a peak at specific mRNA sites, which corresponds to candidate miRNA cleavage sites.

Degradome sequencing significantly enhances the accuracy and reliability of miRNA target gene identification, thereby becoming a powerful tool in miRNA research. However, it should be noted that this technology is limited to detecting miRNA-mediated cleavage and does not identify miRNAs that inhibit target gene expression through translational repression.

For further information, refer to "Degradome Sequencing: Introduction, Features, Workflow, and Applications"

Figure 8 Scheme representing the exploration of degradome libraries. (Julie Leclercq et al., 2020)

Case study

In June 2022, the journal Plant Physiology published a research article titled "Maize miR167-ARF3/30-Polyamine Oxidase 1 Module-Regulated H2O2 Production Confers Resistance to Maize Chlorotic Mottle Virus," elucidating how the Zma-miR167-ZmARF3/30 module regulates the expression of ZmPAO1 to inhibit Maize chlorotic mottle virus (MCMV) infection. In this study, the authors initially employed the online tools WMD3 and TargetMiRna to predict potential targets of Zma-miR167. The results indicated that ZmARF3, ZmARF9, ZmARF16, ZmARF18, ZmARF22, ZmARF30, and ZmARF34 all contained binding sites for Zma-miR167. The focus on ZmARF genes in the bioinformatics predictions was due to previous studies suggesting that miR167 targets ARF genes, thereby modulating various biological processes (Varaud et al., 2011; Kinoshita et al., 2012; Barik et al., 2015; Wang et al., 2015; Na et al., 2019; Song et al., 2019; Yao et al., 2019).

Subsequently, the authors conducted degradome sequencing on total RNA extracted from Zma-miR167 overexpression lines (OE2) and interference lines (CMV-STTM167). The results revealed peaks corresponding to ZmARF3 and ZmARF30, indicating the presence of a cleavage site for Zma-miR167. No significant cleavage peaks were observed for the other predicted ZmARF targets. Additionally, 5'-RNA ligase-mediated rapid amplification of cDNA ends (RLM-RACE) experiments further validated that ZmARF3 and ZmARF30 are directly targeted by Zma-miR167 (Figure 9).

Figure 9 illustrates the cleavage of ZmARF3 and ZmARF30 by Zma-miR167 (Liu et al., 2022). Panel (A) depicts degradation profiling revealing cleavage sites on the targets of Zma-miR167, indicated by red arrows highlighting degradation peaks. Panel (B) shows degradation products of ZmARF3 and ZmARF30 with cleavage sites targeted by Zma-miR167, denoted by black arrows. Nucleotide peak charts correspond to amplified fragments.

RLM-RACE

5' RNA Ligase-Mediated Rapid Amplification of cDNA Ends (5' RLM-RACE) is a frequently utilized technique for identifying cleavage sites of plant miRNAs. The detection mechanism of 5' RLM-RACE is analogous to that of degradome sequencing. In this method, T4 RNA ligase is employed to attach a synthetic RNA adapter to the 5' end of the 3' cleavage fragment of the mRNA. Notably, the intact mRNA cannot ligate with the RNA adapter due to the presence of a 5' cap structure.

Subsequently, reverse transcription is performed using random primers. Nested PCR is then conducted using two sets of primers: an inner and outer anchor primer derived from the adapter sequence and two specific primers targeting downstream sequences of the gene of interest. The amplified products are cloned into a vector for sequencing. By aligning the obtained sequences with the target gene sequence, it is possible to determine whether the miRNA cleaves the target mRNA and to identify the precise cleavage site (Zheng et al., 2021).

Case study

A study published in the Horticultural Plant Journal titled "Identification, Characterization, and Verification of miR399 Target Gene in Grape" elucidated the evolutionary characteristics of the grape miR399 family and validated its cleavage effects on target gene mRNA. To identify the target genes of grape miR399, the authors utilized psRNATarget for initial predictions. The results indicated that members of the miR399 family could target VIT_13s0067g03280 (inorganic phosphate transporter 1-3). Additionally, VIT_12s0035g00200 (phospholipase D delta-like) and VIT_12s0059g02000 (beta-glucuronosyltransferase GlcAT14A) were identified as targets for most miR399 members. Previous studies reported that vvi-miR399b exhibited differential expression in the transcriptomes of 'Kyoho' and 'Fengzao' grapes (Guo et al., 2018). Consequently, vvi-miR399b was selected to verify its cleavage effect on target gene mRNA.

The dual-luciferase reporter assay demonstrated that vvi-miR399b targets the binding sites of inorganic phosphate transporter 1-3, phospholipase D delta-like, and beta-glucuronosyltransferase, and regulates these target genes (Figure 10 A, B). Further validation using 5' RNA Ligase-Mediated Rapid Amplification of cDNA Ends (5' RLM-RACE) revealed that vvi-miR399b cleaves its target genes primarily at the 10th or 11th nucleotide of the complementary region between vvi-miR399b and the target mRNA (Figure 10 C).

Figure 10 illustrates the validation of grapevine miR399 cleavage effects on its target genes (Pei et al., 2023). Panel (A) depicts a schematic of the dual-luciferase reporter vector. Panel (B) shows the Luc/Ren ratio with "WT" indicating the vvi-miR399b target sequence and "MUT" indicating the mutated vvi-miR399b target sequence. Panel (C) displays the results of 5′ RLM-RACE experiments used for target gene validation.

RT-qPCR and Protein Immunoblotting

The outcome of miRNA-mediated negative regulation on target genes typically involves a decrease in mRNA and protein levels of the targets. Therefore, RT-qPCR and protein immunoblotting can be employed to further validate target genes identified through bioinformatics approaches. By assessing changes in mRNA and protein expression levels of target genes in plants before and after miRNA overexpression or inhibition, the regulatory relationship between miRNA and its targets can be established. However, the application of protein immunoblotting is relatively limited in plants due to challenges associated with the preparation of high-quality antibodies.

Case study

In August 2023, Xu et al. published a study titled "Intronic microRNA-directed regulation of mitochondrial reactive oxygen species enhances plant stress tolerance in Arabidopsis" in New Phytologist. This study reported that under cadmium (Cd) stress conditions, intron retention of pre-miR400 results in decreased miR400 levels, leading to the upregulation of its downstream target gene PPR1, reduced accumulation of reactive oxygen species (ROS), and consequently, mild oxidative damage.

In this research, bioinformatics analyses initially identified PPR1 and PPR2 as potential miR400 target genes. To further elucidate the miR400 regulatory network, the authors examined the relative expression levels of PPR1 and PPR2 in miR400 overexpression lines (OxmiR400) and miR400 interference lines (STTM400). The results indicated that the expression levels of PPR1 and PPR2 were reduced in OxmiR400 and elevated in STTM400 (Figure 11), confirming that miR400 targets both PPR1 and PPR2.

Figure 11 illustrates the relative expression levels of PPR1 (panel a) and PPR2 (panel b) in WT, OXmiR400, and STTM400 plants determined by RT-qPCR.

References:

1. Barik S, Kumar A, Sarkar Das S, et al. Coevolution pattern and functional conservation or divergence of miR167s and their targets across diverse plant species. Scientific reports. 2015;5(1):14611.
2. Bi H, Fei Q, Li R, et al. Disruption of miRNA sequences by TALENs and CRISPR/Cas9 induces varied lengths of miRNA production. Plant Biotechnology Journal. 2020;18(7):1526-1536.
3. Deng F, Zeng F, Shen Q, et al. Molecular evolution and functional modification of plant miRNAs with CRISPR. Trends in plant science. 2022;27(9):890-907.
4. German MA, Pillay M, Jeong DH, et al. Global identification of microRNA-target RNA pairs by parallel analysis of RNA ends. Nature biotechnology. 2008;26(8):941-946.
5. Guo DL, Li Q, Lv WQ, et al. MicroRNA profiling analysis of developing berries for 'Kyoho'and its early-ripening mutant during berry ripening. BMC plant biology. 2018;18(1):1-16.
6. Kinoshita N, Wang H, Kasahara H, et al. IAA-Ala Resistant3, an evolutionarily conserved target of miR167, mediates Arabidopsis root architecture changes during high osmotic stress. The Plant Cell. 2012;24(9):3590-3602.
7. Liu X, Liu S, Chen X, et al. Maize miR167-ARF3/30-polyamine oxidase 1 module-regulated H2O2 production confers resistance to maize chlorotic mottle virus. Plant Physiology. 2022;189(2):1065-1082.
8. Na GN, Mu X, Grabowski P, et al. Enhancing micro RNA 167A expression in seed decreases the α‐linolenic acid content and increases seed size in Camelina sativa. The Plant Journal. 2019;98(2):346-358.
9. Pei M, Liu H, Wei T, et al. Identification, characterization, and verification of miR399 target gene in grape. Horticultural Plant Journal. 2023.
10. Song X, Li Y, Cao X, et al. MicroRNAs and their regulatory roles in plant–environment interactions. Annual review of plant biology. 2019;70:489-525.
11. Varaud E, Brioudes F, Szecsi J, et al. AUXIN RESPONSE FACTOR8 regulates Arabidopsis petal growth by interacting with the bHLH transcription factor BIGPETALp. The Plant Cell. 2011;23(3):973-983.
12. Wang Y, Li K, Chen L, et al. MicroRNA167-directed regulation of the auxin response factors GmARF8a and GmARF8b is required for soybean nodulation and lateral root development. Plant physiology. 2015;168(3):984-999.
13. Wang L, Zhang W, Shen W, et al. Integrated transcriptome and microRNA sequencing analyses reveal gene responses in poplar leaves infected by the novel pathogen bean common mosaic virus (BCMV). Frontiers in Plant Science. 2023;14:1163232.
14. Xu WB, Zhao L, Liu P, et al. Intronic microRNA‐directed regulation of mitochondrial reactive oxygen species enhances plant stress tolerance in Arabidopsis. New Phytologist. 2023;240(2):710-726.
15. Yu Y, Jia T, Chen X. The 'how'and 'where'of plant microRNAs. New Phytologist. 2017;216(4):1002-1017.
16. Yao X, Chen J, Zhou J, et al. An essential role for miRNA167 in maternal control of embryonic and seed development. Plant physiology. 2019;180(1):453-464.