Temporal small RNA transcriptome profiling unraveled partitioned miRNA expression in developing maize endosperms between reciprocal crosses

Abstract

In angiosperms, the endosperm nurtures the embryo and provides nutrients for seed germination. To identify the expression pattern of small interfering RNA in the developing maize endosperm, we have performed high-throughput small RNA transcriptome sequencing of kernels at 0, 3, and 5 days after pollination (DAP) and endosperms at 7, 10, and 15 DAP using B73 and Mo17 reciprocal crosses in previous study. Here, we further explored these small RNA-seq data to investigate the potential roles of miRNAs in regulating the gene expression process. In total, 57 conserved miRNAs and 18 novel miRNAs were observed highly expressed in maize endosperm. Temporal expression profiling indicated that these miRNAs exhibited dynamic and partitioned expression patterns at different developmental stages between maize reciprocal crosses, and quantitative RT-PCR results further confirmed our observation. In addition, we found a subset of distinct tandem miRNAs are generated from a single stem-loop structure in maize that might be conserved in monocots. Furthermore, a SNP variation of Zma-miR408-5p at 11th base position was characterized between B73 and Mo17 which might lead to completely different functions in repressing targets. More interestingly, Zma-miR408-5p exhibited B73-biased expression pattern in the B73 and Mo17 reciprocal hybrid endosperms at 7, 10, and 15 DAP according to the reads abundance with SNPs and CAPS experiment. Together, this study suggests that miRNA plays a crucial role in regulating endosperm development, and exhibited distinct expression patterns in developing endosperm between maize reciprocal crosses.

Keywords: maize endosperm; miRNA profiling; partitioned expression; reciprocal cross; tandem miRNA.

Figure 1

Distribution features of small RNA species in kernel and endosperm small RNA libraries. (A) Proportions of distinct small RNA spceies in 12 sequencing libraries. (B) Proportions of abundance for 21-nt small RNAs with A, C, G, and U at the beginning of the sequence, during kernel and endosperm developmental stages. (C) Proportions of abundance for 24-nt small RNAs with A, C, G, and U at the beginning of the sequenceduring kernel and endosperm developmental stages. DAP, days after pollination.

Figure 2

Conserved miRNA expression pattern during maize kernel and endosperm development in reciprocal crosses. (A) Hierarchical tandem of 57 conserved miRNAs with relatively high expression levels in at least one developmental stage. Diverse and tissue-specific miRNA expression patterns were exhibited, and a large proportion of miRNAs were highly expressed in kernels whereas only a few miRNAs were abundantly expressed during endosperm stages. (B) Dynamic and differential expression patterns of maize miR156 family members. The Zma-miR156a group has the highest expression level compared with other family members during all the developmental stages. (C) Kernel-abundant expression patterns of maize miR166 family members. Despite different expression levels, Zma-miR166 family members showed similar expression trends: they were highly expressed in 0-, 3-, and 5-DAP kernels but not in endosperms. (D) Endosperm-abundant expression patterns of maize miR167 family members. All Zma-miR167 members exhibited higher expression levels in endosperm stages than in the kernel stages. DAP, days after pollination.

Figure 3

Developmentally dependent expression patterns of newly identified maize miRNAs negatively correlated with their targets expression. (A) 18 novel miRNAs were identified in 7-, 10-, and 15-DAP maize endosperms, which can be clustered into four groups according to their expression patterns, namely, one-step-down, one-step-up, two-step-down-up, and two-step-up-down. (B) Quantitative RT-PCR results of Zma-miR2006 and Zma-miR2011. Zma-miR2006 showed gradually decreased expression patterns in 7-, 10-, and 15-DAP endosperms in both reciprocal crosses, whereas Zma-miR2011 reached its lowest expression level in 10-DAP endosperm. (C) Boxplot of expression levels of novel miRNA and their putative targets in kernels and endosperms. MiRNA shows higher abundance in endosperm stages compared to kernel stages; in contrast, their targets exhibited lower expression levels in endosperm stages than in kernel stages, which is consistent with the negative correlation between miRNAs and their targets.

Figure 4

Phenotypic difference and temporal expression patterns of miRNAs in 0-, 3-, 5-DAP kernels and 7-, 10-, 15-DAP endosperms between B73 and Mo17 reciprocal crosses. (A) The kernel and endosperm of Mo17 × B73 grow faster than in B73 × Mo17 at all six stages. (B) Zma-miR528, Zma-miR397, and Zma-miR408 were more highly expressed in Mo17 × B73 compared to B73 × Mo17 in 5-DAP kernel and 7-DAP endosperm. (C) Newly identified Zma-miR2002, Zma-miR2017, Zma-miR2001, and Zma-miR2010 exhibited differential expression patterns between reciprocal crosses.

Figure 5

Biosynthesis and dynamic expression patterns of tandem miRNAs. (A) Stem-loop structures of tandem miRNAs. Zma-miR169&Zma-miR169.1, Zma-miR319 and Zma-miR319.1, Zma-miR2001 and Zma-miR2001.1, Zma-miR2013 and Zma-miR2013.1 were generated from a single precursor. (B) Varied expression patterns of Zma-miR319 and Zma-miR319.1 during kernel and endosperm developmental stages. (C) Dynamically changed expression patterns of Zma-miR169 and Zma-miR169.1 during kernel and endosperm developmental stages. (D) Temporal expression patterns of Zma-miR169 and Zma-miR169.1 during kernel and endosperm developmental stages. (E) Development-dependent expression patterns of Zma-miR2001 and Zma-miR2001.1 during kernel and endosperm developmental stages.

Figure 6

Biased expression pattern of Zma-miR408-5p in 7-, 10-, and 15-DAP endosperms in B73 and Mo17 reciprocal crosses. (A) SNP of Zma-miR408-5p between B73 and Mo17 confirmed by re-sequencing. (B) Zma-miR408-5p of B73 was more highly expressed compared to that of Mo17 during endosperm developmental stages in both reciprocal crosses based on the sequencing data. (C) Validation of biased expression pattern of Zma-miR408-5p using CAPS in 7, 10, and 15 DAP endosperms of reciprocal crosses.