Population DNA Methylation Analysis Reveals The Diversity Of Polyphenol Metabolism During Tomato Domestication

Tomato(Solanum lycopersicum)is the model plant,as well as the world’s most nutrientrich fruit and vegetable.According to the history of breeding,tomatoes can be mainly divided into two major processes: domestication and improvement.Among them,cherry tomatoes(S.lycopersicum var.cerasiforme,CER)are thought to be domesticated from Solanum pimpinellifolium(PIM),and cultivated tomatoes(BIG)are improved by human improvement of cherry tomatoes.As a crucial epigenetic marker,DNA methylation plays a key role in plant growth and development,as well as in responding to stress.However,the effects of DNA methylation in the processes of plant domestication and improvement have rarely been studied,especially the relationship between the role of DNA methylation on metabolic and breeding selection in the process of human breeding.Therefore,this study systematically analyzes the regulation and effect of DNA methylation on metabolism in the two main breeding stages of tomato domestication and improvement.In order to explore the changes of DNA methylation in human breeding behavior,the metabolic pathways and regulatory networks of tomatoes in coping with abiotic stress in different geographical environments were analyzed.In this study,96 representative cultivars were selected from the core planting resources of tomato for WGBS(Whole Genome Bisulfite Sequencing),and 9.663 billion pairs of raw PE reads were produced.Through DNA methylation site detection and signal level identification,the methylation levels of CG,CHG and CHH in tomatoes were 0.8,0.7 and less than 0.1,respectively.Comparing the domestication and improvement processes,a total of 8,375 DMRs(Differentially Methylated Region)were identified,and they were mainly distributed in the gene region.To dissect the role of population-level DNA methylation variants,transcriptome and metabolome were performed and analyzed in conjunction with variome data previously published in our lab.From the perspective of different omics,the expression matrix of more than 30,000 genes and the relative content of 339 metabolites were obtained.The m GWAS(metabolilte-based Genome-Wide Association Studies),m EWAS(metabolite-based Epigenome-wide Association Study)and me QTL(methylation-based Quantitative Trait Loci)methods were used to identify 971,711 and tens of thousands of significant large effect loci,respectively.Based on the results of three association analyses,a multi-omics association network of metabolite-SNP-DMR in tomato leaves was constructed,including265 metabolites,853 SNPs and 546 DMRs.This study provides valuable source data for the study of tomato population epigenetics,and the results of these association analyses have contributed to the process of resolving the metabolomics of tomato.Although human selection of tomatoes is primarily based on fruit weight,there are certain patterns in the geographical distribution of different subpopulations of tomatoes,for example,wild tomatoes are closer to the equator and suffer from greater UV-B stresses,which can cause severe damage to tomato seedlings,especially the leaves.Therefore,we selected different cultivars of tomato leaf tissue for in-depth analysis.We found that the number of DMRs was much greater in the domestication process than in the improvement process,and that the length of DMRs was significantly longer in the domestication process than in the improvement process.It was then found that more than 20% of DMRs in the tomato population overlapped with the selective sweeps,and that these DMRs had no preference in different human selection processes.These DMRs acted on multiple key transcription factors,such as Sl MYB12,which were affected by DMR and caused differential expression between populations and regulated phenylpropane metabolism.In addition,using me QTL analysis,it was found that more than 80% of the DMRs in tomato had no significant association with SNPs,while DMRs had a strong linkage with neighboring SNPs.Enrichment analysis showed that they mainly acted on phenylpropane metabolism and were responsible for the different tolerance of tomato plants to UV-B.To reveal the role of DNA methylation in tomato domestication and metabolic diversity,this study identified 13 novel candidate genes by m GWAS and m EWAS strategies to reconstruct the polyphenol metabolic network.The results of m EWAS,transcriptome data analysis and exogenous addition of 5’-azacytidine experiments revealed that high DNA methylation inhibited the transcriptional level of UGT71AV3,thereby inhibiting the glycosylation modification of tomato kaempferol.Whereas UGT73L8 could be captured by both SNPs and DMRs,haplotype as well as K-means analysis showed that both SNPs for the reference type and DMRs for high methylation levels could promote luteolin 7-Oglycoside biosynthesis.In short,our study revealed that DNA methylation variants were stronger than SNPs in affecting metabolite content and further demonstrated the function of candidate genes by in vitro enzyme activity assays.In this study,a landscape analysis of population DNA methylation in tomato breeding history was performed by multi-omics means.Using genome-wide analysis strategies,the metabolic diversity of the tomato leaf population was resolved to provide a valuable resource and guidance for breeding new superior cultivars of tomato,and the multi-omics data from this study will also advance research on the improvement of agronomic traits in tomato as well as other crops.