Metabolite signatures of diverse Camellia sinensis tea populations

Abstract

The tea plant (Camellia sinensis) presents an excellent system to study evolution and diversification of the numerous classes, types and variable contents of specialized metabolites. Here, we investigate the relationship among C. sinensis phylogenetic groups and specialized metabolites using transcriptomic and metabolomic data on the fresh leaves collected from 136 representative tea accessions in China. We obtain 925,854 high-quality single-nucleotide polymorphisms (SNPs) enabling the refined grouping of the sampled tea accessions into five major clades. Untargeted metabolomic analyses detect 129 and 199 annotated metabolites that are differentially accumulated in different tea groups in positive and negative ionization modes, respectively. Each phylogenetic group contains signature metabolites. In particular, CSA tea accessions are featured with high accumulation of diverse classes of flavonoid compounds, such as flavanols, flavonol mono-/di-glycosides, proanthocyanidin dimers, and phenolic acids. Our results provide insights into the genetic and metabolite diversity and are useful for accelerated tea plant breeding.

Fig. 1. Geographic origins and phylogenetic relationships of 136 representative tea plant accessions in China.

a Geographic origins of the tea plant accessions examined in this study. The map of China was generated using the R package “chinamap” (https://github.com/GuangchuangYu/chinamap). b An approximate Maximum Likelihood-based phylogenetic tree constructed using 45,162 fourfold-degenerate SNPs that were identified from mapped RNA-sequencing data. The tree was rooted using five tea relative species (in black) as outgroup. Numbers at the branch points represent support values (percentage) based on 1000 bootstrapping replicates. Five main clades were identified and indicated in different colors: group 1 (red), group 2 (green), group 3 (purple), group 4 (yellow), group 5 (blue). Source data underlying b are provided as a Source Data file.

Fig. 2. Population structure of 134 representative tea plant accessions in China.

a Model-based clustering analysis with different K-values (number of clusters). The optimal K-value is 5 as determined by the Harvester software. b PCA analysis using SNP allele data showing the genetic distances among tea plant accessions in five groups identified in a. Source data are provided as a Source Data file.

Fig. 3. Population differentiation and selective sweep regions across five groups of tea plant accessions.

a Boxplot of Fst values calculated from pairwise comparisons of five groups of tea plant accessions (identified in Fig. 1a). Fst was calculated on 100 kb sliding windows with a step size of 10 kb along all scaffolds in the tea reference genome. (n = 180,916, 181,969, 182,127, 182,256, 181,842, 181,136, 181,436, 181,989, 182,235, 181,300 sliding windows for plots ordered from left to right). Boxes = interquartile ranges, middles = medians, whiskers = 1.5 × the interquartile range. b–e Plot of selective sweep (b, c) and nucleotide diversity values (d, e) along scaffold859 and scaffold1492 that contain selective sweep regions. The dotted line in b, c indicates a threshold value of 18.40 and the dotted line in d, e indicates a threshold value of 1.054. Source data are provided as a Source Data file.

Fig. 4. Metabolites that showed significant changes in concentration in pairwise comparisons of five groups of tea accessions.

a, b Number of metabolic features that were detected under NEG (a) and POS (b) modes, respectively, and were identified as differentially accumulated in pairwise comparisons of five groups of tea accessions. Red and blue bars indicate the numbers of metabolic features showing increase and decrease in concentrations, respectively. c Heatmap showing the abundance patterns of annotated metabolites with an average relative abundance greater than 500 in at least one tea group and with significant changes in abundance in at least one comparison. d–f Box plots of abundance of epicatechin gallate, theanine and caffeine in different tea groups. (n = 29 for Group 1, 11 for Group 2, 32 for Group 3, 26 for Group 4, and 36 for Group 5). Boxes = interquartile ranges, middles = medians, whiskers = 1.5 × the interquartile range, single points = outliers. Source data underlying c–e, and f are provided as a Source Data file.

Fig. 5. Differentially expressed genes identified in pairwise comparisons of five groups of tea accessions.

a Number of differentially expressed genes in pairwise comparisons of five groups of tea accessions. Numbers of up- and downregulated genes are indicated in red and blue, respectively. b Heatmap showing the expression patterns of genes that are known to be involved in the biosynthesis of catechins (in red), caffeine (in blue), and theanine (in black), and show significant changes in expression in at least one pairwise comparison. c, d Box plots of gene expression values (log2-transformed counts per million) of TCS and F3′5′H in different tea groups. (n = 25 for Group 1, 9 for Group 2, 28 for Group 3, 26 for Group 4, and 34 for Group 5). Boxes = interquartile ranges, middles = medians, whiskers = 1.5 × the interquartile range, single points = outliers. Source data underlying b, c, and d are provided as a Source Data file.