Exploring Morphological, Transcriptomic, and Metabolomic Differences Between Two Sister Lines with Contrasting Resistance to Orange Rust Disease in Sugarcane

Abstract

Sugarcane (Saccharum spp.) hybrid, one of the most important crops in Florida, has been affected by orange rust (OR) disease caused by Puccinia kuehnii since 2007, resulting in significant yield loss. Developing resistant cultivars to this disease has become an important goal in sugarcane breeding programs. However, the specific genes and molecular mechanisms underlying the resistance to OR disease in sugarcane are still not clear. In this study, we selected two sugarcane sister lines with different genotypes-showing contrasting resistance responses to the disease-from a major quantitative trait loci (QTL) region controlling OR disease resistance. Morphological and anatomical observations revealed that the resistant line (540) had significantly smaller stomatal size and lower stomatal density than the susceptible line (664). Transcriptomic analyses showed that resistant line 540 had increased cell surface modification activity, suggesting possible increased surface receptors. Differentially expressed gene and coexpression analyses also revealed key genes involved in the biosynthesis of anti-fungal molecules, such as hordatines, arabidopyrones, and alkaloids. They also showed a strong increase in long non-coding RNA expression, playing a role in transcriptional regulation. Transcriptomic-metabolomic joint analysis suggested that the biosynthesis of phenylpropanoid derivatives with purported antioxidant and anti-fungal capabilities increased in line 540, especially those deriving from ferulate. Genes, pathways, and some single-nucleotide polymorphisms identified in this study will provide fundamental information and resources to advance the knowledge of sugarcane molecular genetic mechanisms in relation to OR disease, supporting breeding programs in developing cultivars with improved resistance to OR.

Keywords: Pucinnia kuehnii; metabolomics; orange rust; sugarcane; transcriptome.

Figure 1

(A) Adaxial and abaxial surfaces of the two lines (resistant line 540 and susceptible line 664). (B) Leaf width of the two lines. (C) Stomata density of the two lines. (D) Stomata size of the two lines. (E) Microscopic observation of stomata of the two lines. Each bar in the figure includes the mean value along with the standard deviation. (*** p < 0.001).

Figure 2

(A) Principal component analysis of the mRNA datasets. (B) Numbers of differentially expressed genes (DEGs) at different time points in two lines. (C) Numbers of differentially expressed genes between the two lines (DEG2Ls) at different time points. (D) Heatmap of log₂ (fold change) of DEGs at different time points in two lines. (E) Heatmap of log₂ (fold change) of DEG2Ls at different time points.

Figure 3

GO enrichment analysis of DEG2Ls. Top significantly enriched GO terms in the biological process category. Cellular components and molecular functions are shown for DEG2Ls. Bar plots represent the number of genes of different GO terms.

Figure 4

Top 20 metabolic pathways with the smallest p-values of the DEG2Ls identified by KEGG enrichment.

Figure 5

(A) Numbers of differentially expressed lncRNAs (DE-lncRNAs) at different time points in the two lines (resistant line 540 and susceptible line 664). (B) Numbers of DE-lncRNAs differentially expressed between the two lines (DE-lncRNAs-2L) at different time points.

Figure 6

(A) Selection of soft-thresholding powers (β). (B) Gene co-expression network gene clustering numbers and modular cutting. (C) The proportion of hub genes from different modules (grey module was excluded). (D) Association analysis of gene co-expression network modules with genotype/resistance ability and time-dynamic traits. There are two numbers in each unit: the upper one is the correlation value and the lower number is the p-value. (* p < 0.05, ** p < 0.01, *** p < 0.001).

Figure 7

GO enrichment of turquoise module. Top significantly enriched GO terms in the biological process category. Cellular components and molecular functions are shown for genes in the turquoise module. Bar plots represent the number of genes of different GO terms in the turquoise module.

Figure 8

GO enrichment of blue module. Top significantly enriched GO terms in the biological process category. Cellular components and molecular functions are shown for genes in the blue module. Bar plots represent the number of genes of different GO terms in the blue module.

Figure 9

Top 20 metabolic pathways with the smallest p-values of the blue module identified by KEGG enrichment.

Figure 10

Blue module network. The selection was made by first identifying the top 10 genes with the strongest interactions, followed by selecting up to 10 of the strongest interactions for each of these genes, resulting in a maximum of 100 interactions in total.

Figure 11

(A) Venn diagram of the induced significant and repressed significant metabolites in two sister lines (resistant line 540 and susceptible line 664) after inoculation. (B) Venn diagram of the induced significant metabolites in two sister lines (line 540 and line 664).

Figure 12

Joint analysis of transcriptome and metabolome illustrating the overview of pathway analysis based on DEGs and induced metabolites in line 540 according to the p-values from the pathway enrichment analysis and pathway impact values from the pathway topology analysis. (A stronger red color indicates higher statistical significance).

Figure 13

Joint analysis of transcriptome and metabolome illustrating the overview of pathway analysis based on DEG2Ls and induced metabolites in line 540 according to the p-values from the pathway enrichment analysis and pathway impact values from the pathway topology analysis. (A stronger red color indicates higher statistical significance).