Normal and Abortive Buds Transcriptomic Profiling of Broccoli ogu Cytoplasmic Male Sterile Line and Its Maintainer

Abstract

Bud abortion is the main factor affecting hybrid seeds' yield during broccoli cross breeding when using ogura cytoplasmic male sterile (ogu CMS) lines. However, the genes associated with bud abortion are poorly understood. We applied RNA sequencing to analyze the transcriptomes of normal and abortive buds of broccoli maintainer and ogu CMS lines. Functional analysis showed that among the 54,753 annotated unigenes obtained, 74 and 21 differentially expressed genes in common were upregulated and downregulated in ogu CMS abortive buds compared with ogu CMS normal buds, maintainer normal, and abortive buds, respectively. Nineteen of the common differentially expressed genes were enriched by GO terms associated with glycosyl hydrolases, reactive oxygen species scavenging, inhibitor, and protein degradation. Ethylene-responsive transcription factor 115 and transcriptional factor basic helix-loop-helix 137 were significantly upregulated; transcription factors DUO1 and PosF21/RF2a/BZIP34 were downregulated in ogu CMS abortive buds compared with the other groups. Genes related to polygalacturonase metabolism, glycosyl hydrolases, oxidation reduction process, phenylalanine metabolism, and phenylpropanoid biosynthesis were significantly changed in ogu CMS abortive buds. Our results increase our understanding of bud abortion, provide a valuable resource for further functional characterization of ogu CMS during bud abortion, and will aid in future cross breeding of Brassica crops.

Keywords: RNA-Seq; broccoli; bud abortion; cytoplasmic male sterile; gene expression; transcriptome.

Figure 1

Morphological characterization of two broccoli lines with different degrees of bud abortion. (A,B) ML93219, (C,D) CMS93219. B-I and B-II represent the normal and abortive bud of ML93219, respectively. D-I and D-II represent the normal and abortive bud of CMS93219, respectively. The bars in (E,F) represent the standard deviation (n = 15). Asterisks indicate that the average number of abortive buds and the abortive buds rate per branch are very significantly different between ML93219 and CMS93219 (unpaired t test, p < 0.01).

Figure 2

Clusters of eukaryotic orthologous groups (KOG) functional classification of the broccoli bud transcriptome.

Figure 3

Gene Ontology (GO) classifications of the broccoli bud transcriptome.

Figure 4

Boxplot of Fragments per kb per million fragments (FPKM) distribution for the four samples. Five statistics are represented by different regions of the Boxplot; from the top down they are the maximum, upper quartile, median, lower quartile, and minimum, respectively.

Figure 5

Cluster analysis of differentially expressed genes among the four samples. Heatmap of differentially expressed genes among the four samples. Red indicates high expression, and blue indicates low expression. Color from red to blue represents descending log10 (FPKM + 1).

Figure 6

Differentially expressed genes (DEGs) between normal and abortive buds from ML93219 and CMS93219. (a) DEGs between CMS AB and CMS NB. (b) DEGs between CMS AB and ML AB. (c) DEGs between CMS AB and ML NB. The x-axis indicates the log₂ (fold change) between the two samples. The y-axis indicates −log₁₀ (padj) (p-adjusted). The scatter points in the figure represent individual genes, the green dots indicate genes with no significant differences, the red dots indicate up-regulated genes with significant differences, and the blue dots indicate down-regulated genes with significant differences. The screening condition for DEGs is padj < 0.05.

Figure 7

DEGs between normal and abortive buds from broccoli maintainer and ogu cytoplasmic male sterile (CMS) lines.

Figure 8

Verification of the DEGs by qRT-PCR. Eleven DEGs with higher expression and ten DEGs with lower expression in CMS_AB were selected for qRT-PCR validation. The relative expression level of each gene was expressed as the FPKM among four samples in the RNA-Seq data (black line) and qRT-PCR data (blue bar). To normalize the expression data, the broccoli β-actin gene was used as the internal control. The bars represent the standard deviation.

Figure 9

Heatmap of the 19 enriched common differentially expressed genes in normal and abortive broccoli buds. The bar represents the expression levels for each gene (log₁₀ (FPKM + 1)) in the ML_NB, ML_AB, CMS_NB, and CMS_AB groups, as indicated by red or green rectangles. Red means upregulation of genes and green means downregulation.