Wild tomato genomes reveal how structural variants and retrotransposon explosions block crossbreeding
Wild tomato species carry valuable genetic traits, drought tolerance, disease resistance, salt tolerance, that breeders would love to introduce into cultivated tomatoes. There is one persistent obstacle: when you cross a wild tomato with a domestic one, the chromosomes refuse to swap genetic material at key locations, blocking the transfer of desirable genes.
A team led by the Max Planck Institute for Plant Breeding Research in Cologne has now produced the highest-quality genome assemblies ever for two wild tomato species, and in doing so, has identified the structural features responsible for those recombination barriers. The findings, published June 28 in Nature Communications, provide both a genomic toolkit and a mechanistic understanding of why some wild traits are so difficult to breed into modern tomatoes.
Two wild genomes, 146 Mbp of new sequence
The researchers sequenced Solanum pennellii (accession LA0716, a stress-resistant species from the Andean region) and Solanum cheesmaniae (LA1039, a salt-tolerant species from the Galapagos Islands) using a combination of PacBio HiFi (high-accuracy, ~20-kb reads) and Oxford Nanopore ultralong reads (up to ~200 kb), scaffolded with Omni-C chromatin conformation data.
The resulting assemblies set a new benchmark for tomato wild-relative genomes: both achieved BUSCO completeness above 98%, Merqury quality values above 69, and LAI scores above 14. The S. pennellii assembly, at 1,109 Mbp across 12 chromosomes, added 145.9 Mbp (~146 Mbp) of sequence compared with the previous reference, filling more than 41,000 gaps and newly assembling inversions on chromosomes 3, 6, 7, and 8 that were previously missed. The S. cheesmaniae assembly came in at 862 Mbp.
Corresponding author Charles J. Underwood, who holds affiliations at the Max Planck Institute and Radboud University, said the improvement is dramatic: “What we had before for S. pennellii was essentially a draft. Now we have a reference-grade assembly that lets us see exactly where and how wild and cultivated genomes differ.”
The Tekay retrotransposon explosion
One of the most striking findings is the difference in transposable element content between wild and cultivated tomatoes. Total TE content reaches ~65% in S. pennellii compared with approximately 55% in cultivated tomatoes. The most active lineage across the entire tomato clade is the Tekay family of Ty3/Gypsy LTR retrotransposons.
S. pennellii carries approximately 3,200 Tekay elements, more than double the count in cultivated tomatoes (1,261–1,380). Phylogenetic clustering of integrase sequences confirmed that independent Tekay explosions occurred in S. pennellii and in S. peruvianum, another wild species. These repeat expansions are also recent: LTR retrotransposons in wild tomatoes are younger and more recently integrated than their cultivated counterparts.
Notably, the researchers found that IPT (cytokinin biosynthesis) genes are significantly enriched near Tekay elements in S. pennellii (p = 1.96 × 10⁻¹⁵), suggesting these wild retrotransposon insertions may have regulatory effects on stress-related hormone pathways, a potential breeding target for stress resilience.
Inversions and recombination coldspots
To understand how these structural features affect breeding, the team generated whole-genome sequence data from 709 recombinant plants derived from three hybrid backcross populations: S. pennellii × cultivated tomato, S. cheesmaniae × cultivated tomato, and an intraspecific cultivated × cultivated cross.
The result: ~64% of the genome (529 Mbp) is composed of recombination coldspots, regions where crossovers rarely or never occur. These coldspots are conserved across all three hybrids and are driven by three factors:
1. Megabase-scale inversions, regions where the chromosome segment is flipped relative to the other parent, making crossover physically impossible. The team identified 202 inversions totaling >200 Mbp between S. lycopersicum and S. pennellii, including newly discovered inversions of 4.7 Mbp, 2.2 Mbp, 7.1 Mbp, and approximately 20 Mbp.
2. Insertion-deletion polymorphisms, large stretches of sequence present in one species but absent in the other disrupt homologous pairing during meiosis.
3. Pericentromeric repeats, dense clusters of Gypsy/Copia retrotransposons around the centromeres that suppress crossover initiation.
Heterochiasmy: a practical finding for breeders
A finding with immediate practical relevance: female backcross populations showed 28–36% more crossovers than male populations across all three hybrids. This means that using the female side as the recurrent parent in interspecific crosses could substantially increase the chance of breaking linkage drag, the undesirable genetic associations that come along with a desired wild trait.
“This isn’t something breeders have been able to factor in without detailed genomic information,” said first author Willem van Rens. “Now we know that crossing direction matters, and it matters a lot.”
What this means for tomato breeding
The three-way combination of structural variants, repeat expansions, and conserved coldspots explains a long-standing puzzle in tomato improvement: why certain stress-tolerance and disease-resistance traits from wild relatives have been nearly impossible to introgress into elite varieties through traditional backcrossing.
The high-quality assemblies now make it possible to track wild introgressions with single-base precision, to design markers that target, or avoid, specific inversions, and to exploit the female crossover advantage. For the approximately 146 Mbp of newly accessible wild genetic sequence, the breeding bottleneck shifts from “can we see it?” to “can we unlock it?”.
Sources:
1. van Rens, W.M.J., Fuentes, R.R., Zangishei, Z., Primetis, E. et al. “Wild tomato genome assemblies reveal structural variants and repeat content act as recombination barriers.” Nature Communications 17, 5590 (2026). DOI: 10.1038/s41467-026-74784-5
2. Max Planck Institute for Plant Breeding Research. Press release, June 28, 2026.