Tiny Signaling Peptides: Rice DEVIL Genes Reveal Hidden Breeding Potential

Rice feeds more than half of humanity, and every improvement in its yield or resilience ripples through global food security. Yet some of the most important players in the crop’s genome are also among the least studied. A new analysis published in BMC Plant Biology by Xinting Liu, Yingyao Shi and colleagues at Anhui Agricultural University turns the spotlight on one such overlooked group: the DEVIL, or DVL, family genes. These genes encode small signaling peptides that help coordinate plant development and stress responses, and the study provides the first comprehensive portrait of how they are organized, expressed and diversified across the rice genome and across thousands of rice varieties.

The DVL family belongs to a broader class of small peptide-encoding genes that have attracted growing attention in plant biology. Unlike classic protein-coding genes that produce enzymes or structural components, small peptide genes typically produce short molecules that act as signals, relaying information between cells and fine-tuning developmental programs. In Arabidopsis, the founding member of the family was named ROTUNDIFOLIA3, later grouped with DEVIL, because mutations in these genes visibly reshape leaves and organs. Since then, DVL-like genes have been identified in many plant species, where they influence everything from organ shape to responses to environmental challenges. What remained unclear was how many of these genes rice carries, where they sit in the genome, and whether their natural variation has any bearing on the traits farmers care about.

To answer those questions, the team carried out a systematic bioinformatic survey of the rice reference genome. The search identified five DVL genes, which the authors designate as members of the OsDVL family. Five may sound like a small number, but small peptide families are often compact, and each member can carry substantial functional weight. The researchers characterized the physical properties of the encoded peptides and examined the structure of each gene, establishing a baseline catalog that other rice geneticists can now build on. Identifying the complete membership of a gene family is a deceptively simple but essential step: without it, expression data and genetic associations can be misassigned or missed entirely.

With the family cataloged, the team turned to the regulatory regions upstream of each gene. Promoters, the DNA sequences that sit before a gene and control when and where it is switched on, are rich in short motifs called cis-regulatory elements that serve as binding sites for transcription factors. Scanning the OsDVL promoters revealed a wealth of stress-responsive elements, including motifs typically associated with responses to salinity, cold, drought and hormonal signals. This regulatory architecture suggests that rice DVL genes are not passive developmental housekeepers but are wired to respond when the plant encounters adverse conditions. In an era when saline soils and erratic temperatures increasingly threaten rice paddies across Asia, that wiring is of more than academic interest.

Expression profiling reinforced that suspicion. By examining transcript levels across different tissues, the researchers found that OsDVL genes show tissue-specific patterns, meaning individual family members are active in particular organs or developmental stages rather than uniformly throughout the plant. More strikingly, when the team looked at how selected OsDVL genes behaved under salt and cold stress, transcript levels changed significantly. Genes whose expression rises or falls under stress are prime candidates for mediators of tolerance, because a plant’s ability to survive salinity or chilling often depends on how quickly and appropriately it reprograms its gene activity. The data position OsDVL genes as plausible contributors to abiotic stress responses, though the authors are careful to frame this as a role suggested by the evidence rather than one proven by direct functional experiments.

The most ambitious part of the study, and the part with the clearest implications for breeding, is the haplotype analysis. A haplotype is a specific combination of genetic variants at a locus, inherited together, and haplotype diversity within a species reflects its evolutionary history and the selective pressures breeders have applied. The researchers mined sequence data from 3,010 rice accessions, a panel that spans the remarkable genetic breadth of cultivated rice, including the indica and japonica subspecies and landraces collected across decades of germplasm exploration. At the OsDVL loci, they uncovered substantial allelic diversity, with multiple distinct haplotypes segregating across the panel.

That diversity tells a story. The distribution of haplotypes across the accession panel appears to have contributed to population differentiation, meaning different rice groups carry different favored versions of these genes. Even more intriguingly, the patterns suggest that genetic improvement during rice breeding has left its mark on OsDVL loci. When breeders select plants for higher yield, better grain quality or stress tolerance, they inadvertently select the haplotypes carried by those plants. Detecting the signature of such selection at DVL loci implies that these small peptide genes have been riding along with, or possibly contributing to, the agronomic progress of the past century of rice improvement.

The connection to agronomy became explicit when the team linked haplotypes to measurable traits. At most OsDVL loci, the predominant haplotypes were significantly associated with key yield-related traits, the suite of characteristics such as grain number, panicle architecture and plant height that together determine how much harvestable rice a plant produces. In practical terms, this means a breeder could, in principle, look at which haplotype a rice line carries at an OsDVL locus and gain information relevant to its yield potential. That is the essence of marker-assisted breeding, a technique in which DNA markers are used to guide the selection of plants carrying favorable alleles long before those traits can be seen in the field. The authors highlight DVL-associated haplotypes as potential targets for exactly this kind of marker-assisted selection.

It is worth pausing on why small peptides keep surfacing in studies of yield and stress. Development and environmental response are not separate programs in a plant; they are deeply intertwined. A rice plant facing salinity must decide whether to keep growing, slow down, or redirect resources, and those decisions are mediated by signaling networks in which small peptides act as messengers. A peptide that influences cell proliferation or organ growth could easily have downstream consequences for how many grains a panicle bears, and the same signaling pathway could be co-opted under stress to adjust growth to available resources. The OsDVL findings fit this logic: a compact gene family, stress-responsive promoters, tissue-specific expression, and haplotypes correlated with yield traits all point toward genes that sit at the junction of development and adaptation.

The study also illustrates a methodological trend reshaping plant genetics. Rather than focusing on a single gene in a single variety, the work integrates genome-wide identification, promoter analysis, expression profiling and population-scale haplotype mining into one pipeline. This systems-level approach allows researchers to move from gene discovery to breeding relevance within a single study, and it depends on the availability of massive public datasets such as the 3,010-accession rice panel, which has become a workhorse for resequencing-based association analyses. As similar panels accumulate for other crops, expect more gene families to receive the same treatment, revealing hidden layers of diversity that single-reference-genome studies cannot see.

There are, of course, open questions. The evidence linking OsDVL genes to stress tolerance and yield is associative and expression-based; direct functional validation, through gene knockout, overexpression or CRISPR editing followed by phenotyping under stress, will be needed to establish causation and to determine which of the five genes matters most for which trait. The molecular receptors and downstream partners of the OsDVL peptides in rice also remain to be identified. Still, the study delivers a valuable and actionable starting point. It catalogs the family, maps its regulatory logic, documents its natural diversity across three thousand rice varieties, and flags haplotypes that breeders can begin testing. For a gene family that has lived in the shadow of rice’s better-known developmental regulators, the DEVIL genes have now stepped into the light, and the path from small peptide to bigger harvests looks considerably shorter than it did before.

Subject of Research: Identification, expression and haplotype diversity of DEVIL (DVL) small peptide-encoding genes in rice

Article Title: Identification and haplotype diversity analysis of DVL family genes in rice

Article References: Liu, X., Zhou, Y., Zhu, M., Lv, W., Wang, C., Zhang, M., Tang, C., Yang, S., & Shi, Y. (2026). Identification and haplotype diversity analysis of DVL family genes in rice. BMC Plant Biology. https://doi.org/10.1186/s12870-026-10051-5

Image Credits: AI Generated

DOI: 10.1186/s12870-026-10051-5

Keywords: rice, DVL genes, DEVIL family, small signaling peptides, haplotype diversity, abiotic stress, salt stress, cold stress, yield traits, marker-assisted breeding, genetic diversity, plant genetics