Haplotype-Resolution Transcriptome Analysis Reveals Important Responsive Gene Modules and Allele-Specific Expression Contributions under Continuous Salt and Drought in Camellia sinensis

Abstract

The tea plant, Camellia sinensis (L.) O. Kuntze, is one of the most important beverage crops with significant economic and cultural value. Global climate change and population growth have led to increased salt and drought stress, negatively affecting tea yield and quality. The response mechanism of tea plants to these stresses remains poorly understood due to the lack of reference genome-based transcriptional descriptions. This study presents a high-quality genome-based transcriptome dynamic analysis of C. sinensis' response to salt and drought stress. A total of 2244 upregulated and 2164 downregulated genes were identified under salt and drought stress compared to the control sample. Most of the differentially expression genes (DEGs) were found to involve divergent regulation processes at different time points under stress. Some shared up- and downregulated DEGs related to secondary metabolic and photosynthetic processes, respectively. Weighted gene co-expression network analysis (WGCNA) revealed six co-expression modules significantly positively correlated with C. sinensis' response to salt or drought stress. The MEpurple module indicated crosstalk between the two stresses related to ubiquitination and the phenylpropanoid metabolic regulation process. We identified 1969 salt-responsive and 1887 drought-responsive allele-specific expression (ASE) genes in C. sinensis. Further comparison between these ASE genes and tea plant heterosis-related genes suggests that heterosis likely contributes to the adversity and stress resistance of C. sinensis. This work offers new insight into the underlying mechanisms of C. sinensis' response to salt and drought stress and supports the improved breeding of tea plants with enhanced salt and drought tolerance.

Keywords: Camellia sinensis; allele-specific expression; co-expression network; salt and drought stress; transcriptome.

Figure 1

The transcriptome feature of C. sinensis′ response to continuous salt and drought stress. (a) All samples were clustered based on the filtered gene profile with FPKM > 1 using the Spearman correlation coefficient. (b) The proportion of gene expression was counted with four scales (FPKM ≤ 5, 5 < FPKM ≤ 20, 20 < FPKM ≤ 100, and FPKM > 100) in all samples.

Figure 2

The transcriptomic divergent and dynamic changes in C. sinensis’ response to salt and drought stress. (a) The histogram illustrates the total number of the DEGs of each treated sample compared with the corresponding control. (b,c) Boxplots of upregulated (b) and downregulated (c) DEGs and the overlap in or between salt and drought stress. The top histogram shows the number of DEGs for each overlapping combination as indicated by the connected circles below, and the bottom left horizontal histogram displays the total number of DEGs identified in each treated sample. (d,e) The heatmap shows the GO enrichment results of upregulated and downregulated DEGs in C. sinensis under each time point treatment of salt and drought stress. The top 20 non-redundant and significant GO terms of C. sinensis under each time point treatment of salt and stress are shown. The color intensity in both heatmaps indicates −log10 transformed ‘adjusted p-values,’ where darker red/blue indicates a more significant enrichment, and gray indicates missing enrichment.

Figure 3

WGCNA co-expression network analysis for the dynamic transcriptome of C. sinensis’ response to salt and drought stress. (a). Hierarchical cluster tree exhibiting co-expression modules identified by WGCNA. The major tree branches constitute 10 modules labeled by different colors, and module ‘Grey’ represents unassigned genes. (b). The barplots indicate the expression of module eigengenes. (c). Heatmap representing the Pearson correlation between different modules. The eigengenes and all samples are indicated by rows and columns, respectively.

Figure 4

The Cytoscape visualization of four co-expression network modules. The correlation network of the (a) MEturquois, (b) MEdarkmagenta, (c) MEpurple, and (d) MEgreen modules were represented by the co-expression genes with an edge weight ≥ 0.3. The size of the circle indicates the number of edges of the gene, and all hub genes were labeled with the gene ID and gene name around the circle.

Figure 5

The ASE genes categorization in C. sinensis’ response to salt and drought stress. (a) The boxplot illustrates the distribution of salt- or drought-responsive ASE genes across C. sinensis genome. The x-axis indicates the 15 chromosomes, and the y-axis indicates number of ASE genes. (b) The count of ASE genes that responded to salt or drought stress at 24, 48, and 72 h, respectively. (c) The boxplot shows the ASE of salt- and drought-specific genes, as well as the intersection of these two stresses that represents in network modules with high connection (with edges ≥5 and ≥50). (d) The Venn diagram displays the intersection between salt-, drought-responsive, and heterosis-related ASE genes [49].