Analysis of the global transcriptome and miRNAome associated with seed dormancy during seed maturation in rice (Oryza sativa L. cv. Nipponbare)

Abstract

Background: Seed dormancy is a biological mechanism that prevents germination until favorable conditions for the subsequent generation of plants are encountered. Therefore, this mechanism must be effectively established during seed maturation. Studies investigating the transcriptome and miRNAome of rice embryos and endosperms at various maturation stages to evaluate seed dormancy are limited. This study aimed to compare the transcriptome and miRNAome of rice seeds during seed maturation.

Results: Oryza sativa L. cv. Nipponbare seeds were sampled for embryos and endosperms at three maturation stages: 30, 45, and 60 days after heading (DAH). The pre-harvest sprouting (PHS) assay was conducted to assess the level of dormancy in the seeds at each maturation stage. At 60 DAH, the PHS rate was significantly increased compared to those at 30 and 45 DAH, indicating that the dormancy is broken during the later maturation stage (45 DAH to 60 DAH). However, the largest number of differentially expressed genes (DEGs) and differentially expressed miRNAs (DEmiRs) were identified between 30 and 60 DAH in the embryo and endosperm, implying that the gradual changes in genes and miRNAs from 30 to 60 DAH may play a significant role in breaking seed dormancy. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses confirmed that DEGs related to plant hormones were most abundant in the embryo during 45 DAH to 60 DAH and 30 DAH to 60 DAH transitions. Alternatively, most of the DEGs in the endosperm were related to energy and abiotic stress. MapMan analysis and quantitative real-time polymerase chain reaction identified four newly profiled auxin-related genes (OsSAUR6/12/23/25) and one ethylene-related gene (OsERF087), which may be involved in seed dormancy during maturation. Additionally, miRNA target prediction (psRNATarget) and degradome dataset (TarDB) indicated a potential association between osa-miR531b and ethylene biosynthesis gene (OsACO4), along with osa-miR390-5p and the abscisic acid (ABA) exporter-related gene (OsMATE19) as factors involved in seed dormancy.

Conclusions: Analysis of the transcriptome and miRNAome of rice embryos and endosperms during seed maturation provided new insights into seed dormancy, particularly its relationship with plant hormones such as ABA, auxin, and ethylene.

Keywords: Plant hormone; Rice; Seed dormancy; Seed maturation; Transcriptome; miRNAome.

Fig. 1

Pre-harvest sprouting (PHS) assay in the Nipponbare seeds and the sampling scheme. A The PHS phenotypes of the Nipponbare panicles at 30, 45, and 60 days after heading (DAH). The phenotypes, both before and after PHS assay, are presented. Scale bar, 5 cm. B The PHS rates of the Nipponbare panicles at 30, 45, and 60 DAH (N = 9). The significance was determined using the Student’s t-test, ***P < 0.001. C The sampling scheme employed in this study: embryo and endosperm

Fig. 2

Biological function analysis of differentially expressed genes (DEGs) in the embryo and endosperm. A Principal component analysis (PCA) results from transcriptome of the embryo and endosperm at 30, 45, and 60 DAH. Data for each maturation stage of the embryo and endosperm are represented in different colors; biological replicates of samples from the same maturation stage are depicted in the same color. B, C Venn diagram of the relationship based on the number of DEGs during seed maturation stages in the embryo (B) and endosperm (C). 30 vs. 45: DEGs between 30 and 45 DAH. 45 vs. 60: DEGs between 45 and 60 DAH. 30 vs. 60: DEGs between 30 and 60 DAH. D, E, F Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment terms based on DEGs between 30 and 45 DAH (D), 45 and 60 DAH (E), and 30 and 60 DAH (F). The Y-axis represents the enriched GO and KEGG pathway terms. The X-axis represents the amount of fold enrichment of GO and KEGG pathway terms. The top five GO terms associated with “biological process”, “molecular function”, and “cellular component”, and the top five KEGG terms based on fold enrichment > 2 and false discovery rate (FDR) < 0.05 were selected. The enriched KEGG pathway terms were obtained using ShinyGO

Fig. 3

Biological function analysis of the most commonly altered DEGs in UpSet plots. A The UpSet plot shows the number of shared DEGs between the tissues and the maturation stages. The horizontal bar graphs and written numbers on the left of intersection matrix represent the numbers of DEGs between two compared conditions. The X-axis in the upper graph represents the number of DEGs corresponding to the lower filled dots, with counts of fewer than 10 DEGs excluded. The sets of connected filled dots indicate a specific intersection of DEGs among the maturation stages in the embryo and endosperm. Red bars, arrows, and dots indicate the up-regulated DEGs, while the blue bars, arrows, and dots indicate the down-regulated DEGs. B GO terms based on 2,091 DEGs down-regulated from 45 to 60 DAH and 30 to 60 DAH in the embryo. The Y-axis indicates the enriched GO terms, and the X-axis indicates the amount of fold enrichment of the GO terms. The top ten GO terms associated with “biological process”, “molecular function”, and “cellular component”, based on fold enrichment > 2 and FDR < 0.05 were selected. The KEGG pathway terms were not identified via the KEGG enrichment analysis

Fig. 4

Profiling of genes expected to be involved in seed dormancy during maturation. A-C Identification of hormone-related DEGs between 30 and 45 DAH (A), 45 and 60 DAH (B), and 30 and 60 DAH (C) in the embryo and endosperm using MapMan analysis. The “Regulation overview” of the MapMan analysis was employed, and the figures were adapted from Additional file 1: Fig S2 for DEGs related to hormones. 30 vs 45: DEGs between 30 and 45 DAH. 45 vs 60: DEGs between 45 and 60 DAH. 30 vs 60: DEGs between 30 and 60 DAH. D, E The graph displays the relative expression levels of auxin-related genes (D) and an ethylene-related gene (E) using quantitative real-time polymerase chain reaction. Data are represented as the mean ± standard error of mean (SEM; N = 3). The significance was determined using Student’s t-test, **P < 0.01 and ***P < 0.001

Fig. 5

The expression dynamics of microRNAs (miRNAs) after 30 DAH in the embryo and endosperm. A The PCA results showed that the miRNA population exhibit more dynamic changes in embryo than in endosperm during seed maturation. PCA plot was drawn after regularized logarithm transformation of the normalized miRNA counts. B Visualization of the UpSet plot for the numbers of differentially expressed miRNAs (DEmiRs) and intersections between the testing sets. The horizontal bar graphs and written numbers on the left of intersection matrix represent the numbers of DEmiRs between two compared conditions. The vertical bar graphs and written numbers from above intersection matrix represents the size of the intersections. C Heatmap representation of hierarchically clustered DEmiRs during rice seed maturation in the embryo and endosperm. Hierarchical clustering was conducted after log2-transformation of the counts per million (CPM)-normalized miRNA read counts. Boxes colored in red or blue indicate differentially expressed DEmiRs (p.adj < 0.05) between the two sampling timepoints, which are labeled on top of boxes

Fig. 6

Predicted target genes of DEmiRs in the embryo involved in ethylene and ABA signaling-related genes. A, C Schematic representation of the pairing between two DEmiRs and their target genes. B, D Profiled expression levels of the DEmiRs and their predicted target genes in this dataset. TPM: transcript per million mapped, RP10M: reads per ten million mapped