Analysis of the Differences Among Camellia oleifera Grafting Combinations in Its Healing Process

Abstract

Grafting serves as a crucial propagation technique for superior Camellia oleifera varieties, where rootstock-scion compatibility significantly determines survival and growth performance. To systematically evaluate grafting compatibility in this economically important woody oil crop, we examined 15 rootstock-scion combinations using 'Xianglin 210' as the scion, assessing growth traits and conducting physiological assays (enzymatic activities of SOD and POD and levels of ROS and IAA) at multiple timepoints (0-32 days post-grafting). The results demonstrated that Comb. 4 (Xianglin 27 rootstock) exhibited superior compatibility, characterized by systemic antioxidant activation (peaking at 4-8 DPG), rapid auxin accumulation (4 DPG), and efficient sugar allocation. Transcriptome sequencing and WGCNA analysis identified 3781 differentially expressed genes, with notable enrichment in stress response pathways (Hsp70, DnaJ) and auxin biosynthesis (YUCCA), while also revealing key hub genes (FKBP19) associated with graft-healing efficiency. These findings establish that successful grafting in C. oleifera depends on coordinated rapid redox regulation, auxin-mediated cell proliferation, and metabolic reprogramming, with Comb. 4 emerging as the optimal rootstock choice. The identified molecular markers not only advance our understanding of grafting mechanisms in woody plants but also provide valuable targets for future breeding programs aimed at improving grafting success rates in this important oil crop.

Keywords: Camellia oleifera; antioxidant enzymes; auxin; grafting compatibility; transcriptomics.

Figure 1

Principal component analysis of growth-related traits in different grafting combinations. (A) The correlation analysis of growth-related indicators used for the comprehensive evaluation of graft compatibility among different combinations; (B) variance contribution rate of each principal component; (C) correlation between seedling growth-related indicators and principal components; and (D) comprehensive evaluation score of each combination. The size of circle in (A) represented correlation coefficient, while * represented p < 0.05, *** represented p < 0.001.

Figure 2

Antioxidant-related indicators of different grafting combinations in the healing process. (A) the enzyme activity of SOD; (B) the enzyme activity of POD; (C) the enzyme activity of APX; (D) the enzyme activity of GR; and (E) the reactive oxygen species (ROS) fluorescence intensity. According to Duncan’s multiple range test, each bar is labeled with both uppercase and lowercase letters to indicate statistical significance. Lowercase letters denote significant differences between the means of different combinations at the same time point (for example, ‘a’ differing from ‘b’ signifies (p < 0.05)). Vertical bars represent the standard deviation of the mean (n = 3).

Figure 3

Growth-related indicators of different grafting combinations in the healing process. (A) the content of soluble sugar; (B) the content of soluble protein; and (C) the content of IAA. According to Duncan’s multiple range test, each bar is labeled with both uppercase and lowercase letters to indicate statistical significance. Lowercase letters denote significant differences between the means of different combinations at the same time point (for example, ‘a’ differing from ‘b’ signifies (p < 0.05)). Vertical bars represent the standard deviation of the mean (n = 3).

Figure 4

Metabolism-related indicators of different grafting combinations in the healing process. (A) the content of gamma-aminobutyric acid (GABA); (B) the content of total phenol; and (C) the content of total flavonoid. According to Duncan’s multiple range test, each bar is labeled with both uppercase and lowercase letters to indicate statistical significance. Lowercase letters denote significant differences between the means of different combinations at the same time point (for example, ‘a’ differing from ‘b’ signifies (p < 0.05)). Vertical bars represent the standard deviation of the mean (n = 3).

Figure 5

Venn diagram of DEGs identified during the healing process of different grafting combinations (A) and DEGs identified in the comparison of different grafting combinations at different stages during the healing process (B).

Figure 6

WGCNA of DEGs between different combinations during its healing process. (A) Hierarchical cluster tree showing co-expression modules identified by WGCNAs; each branch in the tree represents an individual gene. (B) Correlation between module eigengenes and physiological/biochemical indicators during the healing process post-grafting.

Figure 7

Correlation analysis between coexpression gene modules and traits: (A) correlation analysis between green-yellow module and SOD activity; (B) correlation analysis between green-yellow module and TP content; (C) correlation analysis between green-yellow module and TF content; (D) correlation analysis between orange module and IAA content; (E) correlation analysis between orange module and TP content; (F) correlation analysis between orange module and TF content; (G) correlation analysis between turquoise module and ROS fluorescence intensity; (H) correlation analysis between grey60 module and GR activity; and (I) correlation analysis between sky-blue module and POD acitivity. The blue line in the each figure represents the threshold of the corresponding coordinate axis.

Figure 8

The PPI network and heatmap analysis of DEGs associated with physiological indicators: (A) gene co-expression networks of DEGs associated with physiological/biochemical indicators during the healing process post-grafting; (B) correlation network of top 10 nodes associated with physiological/biochemical indicators during the healing process post-grafting; and (C) heatmap showing the expression profiles of DEGs in modules significantly correlated to physiological/biochemical indicators during the healing process post-grafting. The color depth of each cell represents the gene expression level by log₂(fpkm + 1).