LRR-RLPs are modular proteins anchored in the cell membrane, with an ectodomain built from tandem leucine-rich repeats that grips pathogen-derived molecules, and a short membrane-proximal region that connects to intracellular signalling complexes. Unlike the better-known LRR receptor-like kinases, which carry an internal kinase domain, RLPs rely on partner kinases to transmit their signal. This architecture makes them exquisitely adaptable: the repeat region can evolve to recognize new pathogen ligands while the signalling module stays intact. Many famous disease-resistance traits in crops trace back to RLP-type receptors, including the Cf proteins that protect tomatoes against leaf mold fungus. Yet despite their agricultural importance, no one had systematically traced how RLP lineages, their domain architectures, their copy numbers and their chromosomal arrangements differ across the whole tomato clade, which includes cultivated tomato and a dozen or so wild species from the Andean region.

To fill that gap, the team took a pan-genome approach. Rather than studying a single reference genome, they compiled the primary proteomes of 76 accessions spanning the tomato clade and used the clustering tool OrthoFinder to group RLP candidates into orthogroups, sets of genes descended from a common ancestor. The result was a pan-gene family of 3,798 candidates distributed across 97 orthogroups. Strikingly, no single orthogroup was detected in every accession under the study’s annotation, detection and clustering criteria. Even when the analysis was restricted to the 72 accessions from tomato’s home section, Lycopersicon, no strict core orthogroup emerged. Instead, the family split into 14 soft-core groups present in most accessions, 42 shell groups present in some, six cloud groups present in few, and 35 private groups found only in individual lineages.

The dominance of the shell category is the headline statistic of the study. Shell orthogroups accounted for 43.3 percent of all orthogroups but a remarkable 61.8 percent of all candidate genes, meaning that most RLP sequences in any given tomato genome belong to lineages that come and go across the species panel. This is a hallmark of genes under pathogen-driven selection: immune receptors are gained, lost and reshuffled as plants and their enemies race to outmaneuver one another. The authors were careful about verification. Independent reciprocal-best-hit searches and synteny checks, which compare the chromosomal neighborhoods of genes to confirm true orthology, supported the gene pairs in 40 orthogroups, while flagging six nominally private orthogroups whose apparent uniqueness may reflect annotation or clustering artifacts rather than genuine lineage-specific invention.

Beyond counting genes, the researchers dissected the architecture of the receptors themselves. A distinctive feature of many plant LRR receptors is the island domain, a non-leucine-rich insertion embedded within the repeat array that often serves as the actual ligand-binding surface. Across the panel the team identified 6,024 island domains, and a recurring pattern emerged: an island domain followed by four leucine-rich repeats at the terminal end of the ectodomain, a configuration that extends earlier cross-species observations and appears to be a favored structural solution in this receptor family. When the researchers clustered island-domain events by the comparison groups in which they occurred, communities spanning four or five groups contained 93.6 percent of all events, suggesting that most architectural variation is shared across related lineages rather than invented de novo in each species.

The modular composition of the receptors also told an evolutionary story. The team built phylogenetic trees from two regions: the C3 module and the combined endomembrane-juxtamembrane, transmembrane and cytoplasmic juxtamembrane region, abbreviated eJM-TM-cJM. Comparing the two trees revealed partial topological discordance, meaning that different parts of the same receptor family appear to have followed somewhat different evolutionary histories, perhaps through ancient recombination or domain shuffling. The authors note, however, that two of the classification boundaries separating module types had weak bootstrap support, a caution that the precise number of module classes should be interpreted with care. Such methodological honesty is increasingly valued in pan-genome research, where clustering thresholds can materially change conclusions.

Copy-number variation proved to be another axis of diversity. Different tomato accessions carried comparable burdens of multi-copy orthogroups, but the specific receptor lineages involved differed from accession to accession, meaning that each genome has expanded its own subset of immune receptors. Tandem duplications, where a gene is copied next to its original, together with proximal duplications nearby, accounted for 86.5 percent of the within-orthogroup duplication pairs detected, pointing to local replication events as the dominant engine of RLP expansion. Relative to the reference cultivar Heinz, local structural variants affected 28 of the 3,539 callable candidates, a reminder that even reference-relative comparisons reveal meaningful genomic rearrangements in these receptor-rich regions.

One orthogroup stood out for its evolutionary signal. RLP-OG0013 showed the highest median pairwise ratio of nonsynonymous to synonymous substitution rates, Ka/Ks, at 1.55, a value above one that is conventionally read as a sign of positive selection, where amino-acid changes are favored over silent mutations. The authors are appropriately measured here, describing the figure as a descriptive divergence signal that requires lineage-aware selection tests before firm conclusions can be drawn, since averaging Ka/Ks across divergent lineages can obscure the true pattern of selection acting on individual branches. Still, the result marks RLP-OG0013 as a prime candidate for deeper evolutionary analysis, and potentially for functional studies of pathogen recognition.

The study did not stop at genomes. In two separate time-course experiments in which plants were inoculated with the necrotrophic fungus Botrytis cinerea, the gray mold pathogen that devastates tomatoes worldwide, 43 candidate receptor genes from 27 orthogroups were differentially expressed, linking a substantial slice of the pan-gene family to actual immune responses. One technical nuance deserves attention: reads from the wild accession LA2574 were quantified against reference genes from LA1777, another wild species, which permitted integration of expression data at the orthogroup level but did not allow any inference of LA2574’s own genomic copy number. This kind of cross-reference quantification is common in non-model wild species but constrains interpretation, and the authors flag it explicitly.

What emerges from the study is a portrait of the tomato immune receptor repertoire as a dynamic, patchwork system: recurrent receptor architectures shared across the clade, combined with accession-dependent lineage representation and copy expansion shaped by tandem duplication. For breeders, the orthogroup-level genomic and expression summaries provide a ranked shortlist of candidates for genotype-matched tests of pathogen recognition and resistance, particularly against threats like gray mold where cell-surface perception is a first line of defense. As wild relatives continue to supply resistance genes for cultivated tomato, knowing which receptor lineages are private, which are shared, which have expanded locally and which light up after infection turns a haystack of nearly 3,800 genes into a manageable set of leads. The pan-genome framework used here, with its careful attention to clustering artifacts and cross-reference limitations, may well become the template for cataloging immune receptor diversity in other crop clades.

Subject of Research: Pan-genomic diversity of LRR-RLP immune receptor genes in tomato and wild relatives

Article Title: Receptor-module diversity and lineage-specific expansion of the LRR-RLP pan-gene family in tomato and allied wild relatives

Article References: Jiang, T., Yang, W., Tang, B., Wang, Q., Ao, N., Mo, C., Deng, Y., & Zhang, B. (2026). Receptor-module diversity and lineage-specific expansion of the LRR-RLP pan-gene family in tomato and allied wild relatives. BMC Plant Biology. https://doi.org/10.1186/s12870-026-10132-5

Image Credits: AI Generated

DOI: 10.1186/s12870-026-10132-5

Keywords: tomato, wild relatives, LRR-RLP, pan-gene family, island domain, receptor modules, copy-number variation, coding-sequence divergence, pathogen response, Botrytis cinerea, orthogroups, tandem duplication