Identification and expression analysis of microRNAs at the grain filling stage in rice(Oryza sativa L.)via deep sequencing

Abstract

MicroRNAs (miRNAs) have been shown to play crucial roles in the regulation of plant development. In this study, high-throughput RNA-sequencing technology was used to identify novel miRNAs, and to reveal miRNAs expression patterns at different developmental stages during rice (Oryza sativa L.) grain filling. A total of 434 known miRNAs (380, 402, 390 and 392 at 5, 7, 12 and 17 days after fertilization, respectively.) were obtained from rice grain. The expression profiles of these identified miRNAs were analyzed and the results showed that 161 known miRNAs were differentially expressed during grain development, a high proportion of which were up-regulated from 5 to 7 days after fertilization. In addition, sixty novel miRNAs were identified, and five of these were further validated experimentally. Additional analysis showed that the predicted targets of the differentially expressed miRNAs may participate in signal transduction, carbohydrate and nitrogen metabolism, the response to stimuli and epigenetic regulation. In this study, differences were revealed in the composition and expression profiles of miRNAs among individual developmental stages during the rice grain filling process, and miRNA editing events were also observed, analyzed and validated during this process. The results provide novel insight into the dynamic profiles of miRNAs in developing rice grain and contribute to the understanding of the regulatory roles of miRNAs in grain filling.

Figure 1. Small RNAs in each of the four stages of rice grain development.

(A) Length distribution of small RNAs at different grain development stages. (B-G) Summary of common and specific total (B, C and D) and unique (E, F and G) small RNA sequences between different libraries. (H) Known miRNAs among different libraries.

Figure 2. Predicted novel miRNAs identified in this study.

(A) Predicted stem-loop structures of novel miRNA precursors. The precursor structures of four newly identified rice miRNAs (Osa-2, Osa-14, Osa-35 and Osa-46) were predicted via the MFOLD pipeline. Mature miRNA and miRNA* sequences are highlighted in red and blue, respectively. The numbers along the structure indicate nucleotide sites from the 5′ end of the pre-miRNAs sequence. (B) Stem-loop RT-PCR analysis if the identified novel miRNAs. Five novel miRNAs were confirmed via stem-loop RT-PCR. The sizes of the obtained PCR products were approximately ∼60 bp. M indicates a 20 bp DNA Ladder Marker (Takara, Japan). The arrow indicates 60 bp.

Figure 3. Differential expression analysis of known miRNAs.

(A) Heatmap for clustering analysis of the differentially expressed known miRNAs. The bar represents the scale of the expression levels of the miRNAs (log 2). (B) Validation via quantitative real-time RT-PCR of differentially expressed miRNAs obtained from deep sequencing. U6 snRNA was used as a reference, and the expression levels of each of the miRNAs were then compared with the expression at 5 DAF, which was set to 1.0. Error bars indicate the standard deviation (±SD) of three replicates.

Figure 4. Validation and expression of selected miRNA target genes.

(A) 5′-RLM-RACE analysis of the cleavage of target mRNAs by corresponding miRNAs. The arrows indicate the cleavage sites, and the numbers represent the frequency of the sequenced clones. (B) Expression profiling analysis of several target genes and their corresponding miRNAs in rice grain on different days after fertilization. Actin was used as a reference, and the expression levels of each of the target mRNAs were then compared with their expression at 5 DAF or 12 DAF, which was set to 1.0. Error bars indicate the standard deviation (±SD) of three replicates. Os04g47870 (PINHEAD) and Os12g41680 (no apical meristem protein) were confirmed to be targets of Osa-miR168 and Osa-miR164, respectively, in previous work . Os01g63290 (transporter) was predicted to be target of osa-miR167d/f-h/j using psRNA Target (data not shown).

Figure 5. miRNA editing analysis.

(A) Summary of the nucleotide substitution types observed in each library. (B) Summary of nucleotide substitution positions among miRNAs. (C) Validation of the editing sites inferred from deep sequencing via Sanger sequencing. Sequencing chromatogram traces from four miRNA sequences are shown. The edited positions are highlighted with yellow shading. The top trace is genomic DNA (gDNA), and the bottom trace is cDNA.