Genetic Divergence between Camellia sinensis and Its Wild Relatives Revealed via Genome-Wide SNPs from RAD Sequencing

Abstract

Tea is one of the most popular beverages across the world and is made exclusively from cultivars of Camellia sinensis. Many wild relatives of the genus Camellia that are closely related to C. sinensis are native to Southwest China. In this study, we first identified the distinct genetic divergence between C. sinensis and its wild relatives and provided a glimpse into the artificial selection of tea plants at a genome-wide level by analyzing 15,444 genomic SNPs that were identified from 18 cultivated and wild tea accessions using a high-throughput genome-wide restriction site-associated DNA sequencing (RAD-Seq) approach. Six distinct clusters were detected by phylogeny inferrence and principal component and genetic structural analyses, and these clusters corresponded to six Camellia species/varieties. Genetic divergence apparently indicated that C. taliensis var. bangwei is a semi-wild or transient landrace occupying a phylogenetic position between those wild and cultivated tea plants. Cultivated accessions exhibited greater heterozygosity than wild accessions, with the exception of C. taliensis var. bangwei. Thirteen genes with non-synonymous SNPs exhibited strong selective signals that were suggestive of putative artificial selective footprints for tea plants during domestication. The genome-wide SNPs provide a fundamental data resource for assessing genetic relationships, characterizing complex traits, comparing heterozygosity and analyzing putatitve artificial selection in tea plants.

Fig 1. Neighbor-joining phylogenetic tree, plot of the principle component analysis (PCA) and genetic structures for the 18 tea accessions.

(a) Neighbor-joining phylogenetic tree based on 15,444 identified SNPs with bootstrap values calculated from 1,000 trees. (b) Principal component analysis of the 18 tea accessions. (C) Genetic structure of the 18 tea accessions. Different inferred populations are distinguished by different colors. Each accession is indicated by a vertical bar, and the length of each colored section in each vertical bar represents the proportion from ancestral populations.

Fig 2. Heterozygosity levels of the 18 tea accessions.

The heterozygous rates of 18 tested tea accessions were evaluated by calculating the ratio of the number of heterozygous SNPs to the length of the shared SNP-associated genome fragments from the RAD sequencing in each accession.

Fig 3. Gene Ontology classifications of the identified genic SNP-associated tea unigenes.

GO terms were assigned to C. sinensis unigenes based on the top BLASTX hits against the NCBI Arabidopsis protein database. The GO terms were classified into three main GO categories (i.e., biological process, cellular component, molecular function) that included 38 sub-categories. The left y-axis indicates the proportion of genes in the main category, and the right y-axis indicates the number of genes in the same category.

Fig 4. Validation of SNP identification and genotyping of the Tea_298263 SNP locus in the 18 tea accessions by PCR-based sequencing.

(a) Flanking sequences adjacent to SNP loci obtained from Sanger sequencing were aligned against tag sequences containing SNP loci from RAD-Seq data and unigene Singletons19310 based on the top BLAST hits of the consensus tag sequences from C. sinensis var. Longjing43 transcriptome [73] using DNAMAN software. N in the RAD tag sequence represents the SNP locus, which indicates the heterozygous genotypes in the SNP loci of the accessions Csa-2 and Ctb. (b) Confirmation of the heterozygous genotypes (A/T) of the Tea_298263 SNP locus in accession Csa-2 by Sanger sequencing. (C) Confirmation of the heterozygous genotype (A/T) of the Tea_298263 SNP locus in accession Ctb by Sanger sequencing.