Contribution of transposable elements (TEs) to tomato domestication and improvement – tomaTE

Tomato is a prime model for fleshy fruit species and the highest-value fruit and vegetable crop worldwide. Despite the recurrent genetic bottlenecks that have occurred since its domestication, tomato exhibits extensive phenotypic variation and a large part of the diversity seen among the ~25000 cultivars of today likely results from selection of alleles with large effects. The genetic source of this striking phenotypic diversity is unclear, given the narrow genetic pool existing in cultivated tomato. In comparison to SNPs and other small DNA sequence changes, transposable elements (TEs) have a much greater potential to generate large effect alleles, notably because they can disrupt genes and modulate their expression epigenetically. However, the contribution of TE mobilization as well as of its epigenetic control to tomato domestication and improvement is not known.

Here, we propose to use novel genomic approaches to characterize comprehensively the “mobilome” of cultivated tomato as well as of its earliest forms, S. lycopersicum var. cerasiforme, and its closest wild relative S. pimpinellifolium and to determine the role of TEs in gene expression variation among accessions. Our specific aims are to:
(i) Determine the mobilome composition of domesticated tomato as well as of its closest wild relative
(ii) Characterize the contribution of TE mobilization to gene expression variation during tomato domestication and improvement
(iii) Assess the relative amount of gene expression variation among cultivated tomatoes that results from TE presence/absence or TE-dependent epigenetic variants rather than small DNA sequence changes.

Aim 1 relies on an in depth bioinformatics analysis of whole genome sequencing data sets that are publically available for hundreds of tomato accessions and will lead to the production of a TE-sequence capture custom design to interrogate the activity of the tomato mobilome. Aim 2 involves the production of new genetic material (F1 hybrids). Aims 2 and 3 will generate molecular data using TE-sequence capture, RNA-seq (including allele-specific expression (ASE) analysis) and locus-specific DNA methylation assays. Genome-wide association studies (GWAS) will be performed to identify TE-associated changes in gene expression. Crossing the data generated in this project with the hundreds of quantitative trait loci (QTLs) of agronomic importance will enable us to identify candidate causal TE-associated alleles or epialleles. In Aim3 we will further undertake state of the art genome and epigenome editing methodologies whenever possible in order to confirm causality.

Thanks to the integrative and highly synergistic nature of this ambitious project, we expect to obtain major insights into the role played by TE mobilization crop domestication and improvement. In turn, the knowledge, methods and paradigms we hope to generate should have far reaching implications for breeding programs. More generally, our findings should provide answers to one of the fundamental, unresolved questions in the genomics era, namely that of the contribution of TEs and of their epigenetic control to the generation of phenotypic variation within any given species.