Integrated transcriptome, small RNA, and degradome analysis reveals the complex network regulating starch biosynthesis in maize

Abstract

Background: Starch biosynthesis in endosperm is a key process influencing grain yield and quality in maize. Although a number of starch biosynthetic genes have been well characterized, the mechanisms by which the expression of these genes is regulated, especially in regard to microRNAs (miRNAs), remain largely unclear.

Results: Sequence data for small RNAs, degradome, and transcriptome of maize endosperm at 15 and 25 d after pollination (DAP) from inbred lines Mo17 and Ji419, which exhibit distinct starch content and starch granule structure, revealed the mediation of starch biosynthetic pathways by miRNAs. Transcriptome analysis of these two lines indicated that 33 of 40 starch biosynthetic genes were differentially expressed, of which 12 were up-regulated in Ji419 at 15 DAP, one was up-regulated in Ji419 at 25 DAP, 14 were up-regulated in Ji419 at both 15 and 25 DAP, one was down-regulated in Ji419 at 15 DAP, two were down-regulated in Ji419 at 25 DAP, and three were up-regulated in Ji419 at 15 DAP and down-regulated in Ji419 at 25 DAP, compared with Mo17. Through combined analyses of small RNA and degradome sequences, 22 differentially expressed miRNAs were identified, including 14 known and eight previously unknown miRNAs that could target 35 genes. Furthermore, a complex co-expression regulatory network was constructed, in which 19 miRNAs could modulate starch biosynthesis in endosperm by tuning the expression of 19 target genes. Moreover, the potential operation of four miRNA-mediated pathways involving transcription factors, miR169a-NF-YA1-GBSSI/SSIIIa and miR169o-GATA9-SSIIIa/SBEIIb, was validated via analyses of expression pattern, transient transformation assays, and transactivation assays.

Conclusion: Our results suggest that miRNAs play a critical role in starch biosynthesis in endosperm, and that miRNA-mediated networks could modulate starch biosynthesis in this tissue. These results have provided important insights into the molecular mechanism of starch biosynthesis in developing maize endosperm.

Keywords: Endosperm; Maize; Regulatory network; Starch biosynthesis; miRNA.

Fig. 1

Expression profiles of starch biosynthetic genes. a The expression of 40 known starch biosynthetic genes and the pathways in which they could be involved in the endosperm of maize inbred lines Mo17 and Ji419 at 15 and 25 DAP. The heat map represents the expression of each gene in a specific line. Red boxes represent up-regulated genes in Ji419 compared with Mo17 at both 15 and 25 DAP; orange boxes represent genes up-regulated in Ji419 at only one stage; yellow boxes represent genes up-regulated in Ji419 at one stage, but down-regulated at the other stage; green boxes represent genes down-regulated in Ji419 at only one stage; blue boxes represent genes with no differential expression between Mo17 and Ji419 at both 15 and 25 DAP. b Venn diagram of starch biosynthetic genes that are differentially expressed in the endosperm of Mo17 and Ji419 at 15 and 25 DAP. Red represents genes down-regulated in Ji419 compared with Mo17 at 15 DAP; blue represents genes up-regulated in Ji419 at 15 DAP; yellow represents genes down-regulated in Ji419 at 25 DAP; green represents genes up-regulated in Ji419 at 25 DAP. c-d Expression levels of eight key starch biosynthetic genes in the endosperm of Mo17 and Ji419 at 15 DAP (c) and 25 DAP (d). Black bars represent the expression of genes in Ji419; grey bars represent the expression of genes in Mo17. The data in c and d are means ± SD (n = 3). ** significant at p ≤ 0.01 by the Student’s t test

Fig. 2

Analysis of the expression patterns of two miRNAs and their target genes, and eight key starch biosynthetic genes in the endosperm of maize inbred lines Mo17 and Ji419 at 15, 20, 25, 30, 35, and 40 DAP. Red boxes represent miRNAs, yellow boxes represent target genes, and green boxes represent starch biosynthetic genes

Fig. 3

Differential expression of miRNAs and their target genes in the endosperm of maize inbred lines Mo17 and Ji419 at 15 DAP. FC, fold-change of the expression of the miRNAs/target genes in Ji419 relative to that in Mo17 at 15 DAP based on the sequencing data. Black bars represent expression of miRNAs; grey bars represent expression of target genes

Fig. 4

Differential expression of miRNAs and their target genes in the endosperm of maize inbred lines Mo17 and Ji419 at 25 DAP. FC, fold-change of the expression of the miRNAs/target genes in Ji419 relative to that in Mo17 at 25 DAP based on the sequencing data. Black bars represent expression of miRNAs; grey bars represent expression of target genes

Fig. 5

Co-expression network of miRNA-target-starch biosynthesis in developing maize endosperm. Blue boxes represent miRNAs, yellow boxes represent target genes, and green boxes represent starch biosynthetic genes. PCC, Pearson’s correlation coefficient

Fig. 6

Transient expression analysis in maize endosperm. a The expression of miR169a in miR169a-overexpressing endosperm (pUbi:miR169a) and the control (pUbi:Gus); b The expression of miR169o in miR169o-overexpressing endosperm (pUbi:miR169o) and the control (pUbi:Gus); c The expression of NF-YA1 in miR169a/NF-YA1-overexpressing endosperm (pUbi:miR169a and pUbi:NF-YA1) and the control (pUbi:Gus); d The expression of GATA9 in miR169o/GATA9-overexpressing endosperm (pUbi:miR169o and pUbi:GATA9) and the control (pUbi:Gus); e The expression of GBSSI in NF-YA1-overexpressing endosperm (pUbi:NF-YA1) and the control (pUbi:Gus); f The expression of SSIIIa in NF-YA1/GATA9-overexpressing endosperm (pUbi:NF-YA1 and pUbi:GATA9) and the control (pUbi:Gus); g The expression of SBEIIb in GATA9-overexpressing endosperm (pUbi:GATA9) and the control (pUbi:Gus). All data are means ± SD (n = 3). *, ** significant at p ≤ 0.05 and p ≤ 0.01 by the Student’s t test

Fig. 7

Transactivation assays of the promoters of GBSSI/SSIIIa and SSIIIa/SBEIIb by NF-YA1 and GATA9, respectively. a Diagram of the effector, reporter, and internal constructs; b Regulatory interaction between NF-YA1 and the promoter of GBSSI; c Regulatory interactions between NF-YA1/GATA9 and the promoter of SSIIIa; d Regulatory interaction between GATA9 and the promoter of SBEIIb. All data are means ± SD (n = 3). ** significant at p ≤ 0.01 by the Student’s t test