Genome-wide characterization of cis-acting DNA targets reveals the transcriptional regulatory framework of opaque2 in maize

Abstract

Opaque2 (O2) is a transcription factor that plays important roles during maize endosperm development. Mutation of the O2 gene improves the nutritional value of maize seeds but also confers pleiotropic effects that result in reduced agronomic quality. To reveal the transcriptional regulatory framework of O2, we studied the transcriptome of o2 mutants using RNA sequencing (RNA-Seq) and determined O2 DNA binding targets using chromatin immunoprecipitation coupled to high-throughput sequencing (ChIP-Seq). The RNA-Seq analysis revealed 1605 differentially expressed genes (DEGs) and 383 differentially expressed long, noncoding RNAs. The DEGs cover a wide range of functions related to nutrient reservoir activity, nitrogen metabolism, stress resistance, etc. ChIP-Seq analysis detected 1686 O2 DNA binding sites distributed over 1143 genes. Overlay of the RNA-Seq and ChIP-Seq results revealed 35 O2-modulated target genes. We identified four O2 binding motifs; among them, TGACGTGG appears to be the most conserved and strongest. We confirmed that, except for the 16- and 18-kD zeins, O2 directly regulates expression of all other zeins. O2 directly regulates two transcription factors, genes linked to carbon and amino acid metabolism and abiotic stress resistance. We built a hierarchical regulatory model for O2 that provides an understanding of its pleiotropic biological effects.

Figure 1.

Genome-Wide Transcript Profile of 15 DAP Wild-Type and o2 Endosperm Determined by RNA-Seq Analysis. (A) Number of significantly expressed genes in 15 DAP wild-type and o2 endosperm. (B) Length distribution of RNA-Seq-identified potential lncRNAs. (C) Comparison of gene expression in 15 DAP wild-type and o2 endosperm. (D) Comparison of lncRNA expression in 15 DAP wild-type and o2 endosperm. (E) Validation of expression differences observed by RNA-Seq through qRT-PCR of 40 DEGs. Values are the means with se (n = three individuals). R² = 0.66.

Figure 2.

GO Classification for Genes with Altered Expression in o2. The most significantly related cellular components, biological processes, and molecular functions of the 607 functional annotated DEGs. The significance and number of genes classified within each GO term is shown.

Figure 3.

Summary of O2 ChIP-Seq Results. (A) Distribution of O2 binding regions in the maize genome. (B) O2 binding peaks for the positive control genes are shown in the Integrated Genome Browser. For each gene model, the T01 splicing variant is shown. Large boxes indicate exons. Aligned reads are indicated in black (O2) or gray (IgG). (C) Distribution of O2 binding peaks per 100-bp bin corresponding to the −1000- to +2000-bp region flanking the TSS. (D) Venn diagram representing a comparison between RNA-Seq and ChIP-Seq results and P value by χ² test is shown under the gene number of each intersection.

Figure 4.

O2 Binding Sites. (A) The specific sequence and origin of 13 previously reported O2 binding sites. The ACGT core is underlined. (B) EMSAs with 6XHis-O2 fusion protein using probes derived from O2 target gene promoters containing 13 previously reported O2 binding sites (Supplemental Table 2). Arrows on the left indicate the formation of O2 and O2 binding-element complexes. (C) O2 binding motif identified by MEME-ChIP in 1-kb flanking sequences around the genic peak summits and the density plot of this motif around the summits of peaks. (D) EMSAs with probe derived from the O2 target gene promoters containing Z3, B3, and four O2 binding sites identified in this study (Supplemental Table 2). Arrows on the left indicate the formation of O2 and O2 binding element complexes.

Figure 5.

ChIP-qPCR Assays Confirming in Vivo Binding Activity of O2 to the Promoter Region of Zeins. (A) Anti-O2 antibody was used to precipitate chromatin prepared from 15 DAP wild-type endosperm. IgG was used as an antibody control. Primers from b-32, cyPPDK1, or ubiquitin were used to detect the corresponding promoter fragments in ChIP products for positive and negative control analysis. The fold changes were calculated based on the relative enrichment in anti-O2 compared with anti-IgG. Values are the means with se (n = three individuals; *P < 0.05, Student’s t test). (B) Primers from the 19-kD α-zein z1A (GRMZM2G059620), the 19-kD α-zein z1B (AF546188.1_FG001), the 19-kD α-zein z1D (AF546187.1_FG001), the 22-kD α-zein gene (GRMZM2G346897), the 14-kD β-zein gene, the 27-kD γ-zein gene, the 50-kD γ-zein gene, and the 10-kD δ-zein gene were used to detect the corresponding promoter fragments in ChIP products by detecting direct binding of O2 to the promoter region of the genes of different zein classes. The fold changes were calculated based on the relative enrichment in the anti-O2 treatment compared with anti-IgG treatment. Values are the means with se (n = three individuals; *P < 0.05, **P < 0.01, and ***P < 0.001, Student’s t test).

Figure 6.

O2 Functions as a Transcriptional Activator of Different Zein Genes. (A) The 35S:REN-Pro zein:LUC reporter constructs were transiently expressed in onion epidermal cells together with control vector or 35S:O2 effector, respectively. (B) The expression level of Renilla (REN) was used as an internal control. The LUC/REN ratio represents the relative activity of zein promoters. Data are values of three independent experiments. Significant differences from the corresponding control values (using Student’s t test [n = 3]): *P < 0.05, **P < 0.01, and ***P < 0.001.

Figure 7.

O2 Functions as a Transcriptional Activator of Two Transcriptional Factors. (A) The 35S:REN-Pro Myb TF:LUC, 35S:REN-Pro GBF1:LUC, and 35S:REN-Pro NFYB/HAP3 TF:LUC reporter constructs were transiently expressed in onion epidermal cells, together with control vector or 35S:O2 effector, respectively. (B) The LUC/REN ratio represents the relative activity of the promoters. Data are values of three independent experiments. Significant differences from the corresponding control values (using Student’s t test [n = 3]): *P < 0.05 and ***P < 0.001.

Figure 8.

O2 Functions as a Transcriptional Activator of cyPPDK1, cyPPDK2, and Two Lactoylglutathione Lyases. (A) The 35S:REN-Pro cyPPDK1:LUC, 35S:REN-Pro cyPPDK2:LUC, 35S:REN-Pro GRMZM2G312877:LUC, and 35S:REN-Pro GRMZM2G028110:LUC reporter constructs were transiently expressed in onion epidermal cells together with control vector or 35S:O2 effecter. (B) The LUC/REN ratio represents the relative activity of the cyPPDK1 and cyPPDK2 promoters. Data are values of three independent experiments. Significant differences from the corresponding control values (using Student’s t test [n = 3]): **P < 0.01 and ***P < 0.001. (C) The LUC/REN ratio represents the relative activity of two lactoylglutathione lyase promoters. Data are values of three independent experiments. Significant differences from the corresponding control values (using Student’s t test [n = 3]): *P < 0.05 and **P < 0.01.

Figure 9.

A Hierarchical Genome-Wide Transcriptional Regulatory Framework for O2 Involving O2 Targeted Transcription Factors. O2 not only directly activates genes associated with nutrient storage and enzymes involved in carbon and amino acid metabolism but also controls transcription factors that regulate various other aspects of plant metabolism.