Advanced genes expression pattern greatly contributes to divergence in Verticillium wilt resistance between Gossypium barbadense and Gossupium hirsutum

Abstract

Verticillium, representing one of the world's major pathogens, causes Verticillium wilt in important woody species, ornamentals, agricultural, etc., consequently resulting in a serious decline in production and quality, especially in cotton. Gossupium hirutum and Gossypium barbadense are two kinds of widely cultivated cotton species that suffer from Verticillium wilt, while G. barbadense has much higher resistance toward it than G. hirsutum. However, the molecular mechanism regarding their divergence in Verticillium wilt resistance remains largely unknown. In the current study, G. barbadense cv. Hai7124 and G. hirsutum acc. TM-1 were compared at 0, 12, 24, 48, 72, 96, 120, and 144 h post-inoculation (hpi) utilizing high throughput RNA-Sequencing. As a result, a total of 3,549 and 4,725 differentially expressed genes (DEGs) were identified, respectively. In particular, the resistant type Hai7124 displayed an earlier and faster detection and signaling response to the Verticillium dahliae infection and demonstrated higher expression levels of defense-related genes over TM-1 with respect to transcription factors, plant hormone signal transduction, plant-pathogen interaction, and nucleotide-binding leucine-rich repeat (NLR) genes. This study provides new insights into the molecular mechanisms of divergence in Verticillium wilt resistance between G. barbadense and G. hirsutum and important candidate genes for breeding V. dahliae resistant cotton cultivars.

Keywords: Verticillium dahlia; cotton; defense response; disease resistance; transcriptome analysis.

FIGURE 1

Gossypium barbadense (G. barbadense) cultivar Hai124 is more resistant to Verticillium wilt than Gossupium hirsutum (G. hirsutum) TM-1. (A) G. barbadense cv. Hai124 and G. hirsutum acc. TM-1 were used in this experiment. Disease symptoms of cotton plants inoculated with V. dahliae for 50 days. (B) Seedlings were incubated in growth chambers at 22°C under a 16–8 h day-night period for about 2 weeks (until they grew two true leaves) and then treated with V. dahliae and water, and harvested at 0 h, 12 hpi, 24 hpi, 72 hpi, 96 hpi and 144 hpi. (C) PCA depicting pairwise Pearson correlation of gene expression values of all samples.

FIGURE 2

Homoeologous expression bias difference between Hai7124 and TM-1. (A) Line chart of DEGs, upregulated and downregulated genes of Hai7124 and TM-1 in each hour post infection. (B) Venn diagrams of total DEGs in Hai7124 and TM-1. (C) The significant KEGG pathways of the total DEGs identified in Hai7124, the specific pathways were marked in red. (D) The significant KEGG pathways of the total DEGs identified in TM-1, the specific pathways were marked in red. (E) The top 30 significant GO terms enriched in Hai7124. (F) The top 30 significant GO terms enriched in TM-1.

FIGURE 3

Dynamic gene expression changes during Verticillium wilt infection process. (A) Venn diagrams of DEGs with interspecies expression bias at each time point. (B) The top 15 significant GO terms enriched in DEGs uniquely expressed in Hai7124 at 12 hpi; (C) The top 15 significant GO terms enriched in DEGs uniquely expressed in TM-1 at 12 hpi. (D) The top 15 significant GO terms enriched in DEGs commonly expressed in Hai7124 and TM-1 at 12 hpi. (E) The top 15 significant GO terms enriched in DEGs uniquely expressed in Hai7124 at 24 hpi; (F) The top 15 significant GO terms enriched in DEGs uniquely expressed in TM-1 at 24 hpi. (G) The top 15 significant GO terms enriched in DEGs commonly expressed in Hai7124 and TM-1 at 24 hpi. (H) The top 15 significant GO terms enriched in DEGs uniquely expressed in Hai7124 at 96 hpi; (I) The top 15 significant GO terms enriched in DEGs uniquely expressed in TM-1 at 96 hpi. (J) The top 15 significant GO terms enriched in DEGs commonly expressed in Hai7124 and TM-1 at 96 hpi.

FIGURE 4

The quantitative and expression level relationship of the four categories’ genes with time changes. (A) Number of DEGs in transcription factors with infection time. (B) Heatmap of gene expression patterns of TFs uniquely expressed in Hai7124. (C) Heatmap of gene expression patterns of TFs uniquely expressed in TM-1. (D) Heatmap of gene expression patterns of TFs commonly expressed in Hai7124 and TM-1. (E) Variation of the number of DEGs in plant hormone signal transduction with infection time. (F) Heatmap of gene expression patterns of DEGs in plant hormone signal transduction. (G) Variation of the number of DEGs in plant-pathogen interaction with infection time. (H) Heatmap of gene expression patterns of DEGs in plant-pathogen interaction. (I) Variation of the number of DEGs in NLRs with infection time. (J) Heatmap of gene expression patterns of DEGs in NLRs.

FIGURE 5

Weighted gene co-expression network analysis (WGCNA). (A) WGCNA of DEGs from Hai7124 and TM-1 at each hour post-inoculation. The modules showing a high degree of correlation (P < 0.05) with samples are in red. (B) Hierarchical cluster tree of WGCNA analysis, the co-expression modules labeled with seven different colors based on the calculation of eigengenes. (C) The correlation network of major genes from the green modules. The genes related to phytohormones were in purple, genes relative to plant-pathogen interactions were in blue, genes belonged to NLRs were in green, and hub genes from other categories were in pink. (D) Box-plots of hub genes in plant hormone showing expression pattern of the network. (E) Box-plots of hub genes in NLRs relative subnetworks showing expression pattern of the network. The x-axis indicates time points (0 h, 12 hpi, 24 hpi, 72 hpi, 96 hpi, and 144 hpi), and gene expression data were normalized to Log2(FPKM + 1).

FIGURE 6

Defense-related molecules involved in response of cotton to V. dahliae during the first 144 hpi.