A Wild Soybean Cross Has Produced Plants That Seem to Rewrite Their Own Genetic Rules

A soybean breeding experiment has uncovered an unexpected source of genetic chaos: plants carrying two extra chromosomes from a wild perennial relative began producing descendants with strikingly different appearances, altered gene activity and genome-wide genetic changes. The finding could reshape how scientists think about transferring useful traits from wild plants into one of the world’s most important crops—and raises a deeper mystery about how pollen moved through the research fields at rates far above those normally seen in soybean.

The study examined disomic alien addition lines, or DAALs, created by crossing the cultivated soybean Glycine max with Glycine tomentella, a perennial wild relative. Ordinary soybean plants have 40 chromosomes, arranged as 20 pairs. The experimental DAALs contained those 40 soybean chromosomes plus two homologous chromosomes from G. tomentella, bringing their total to 42. The extra chromosomes were intended to act as genetic cargo vessels, carrying wild genes that might eventually contribute disease, pest or environmental stress resistance to cultivated soybean.

Glycine tomentella is an unusually complicated breeding partner. Unlike annual soybean, it belongs to a different subgenus and exists in several chromosome forms, including diploid and polyploid types. It also carries traits that plant breeders would like to access, including resistance to diseases, pests and difficult environmental conditions. Yet the new research shows that adding an entire pair of foreign chromosomes can have consequences far beyond introducing one or two desirable genes. In the field, the DAAL plants were short, late-maturing and developmentally unusual, with shortened internodes, rugose leaves, reduced pod set, delayed leaf drop and green stems at maturity.

The researchers followed the experimental lines for years at the University of Illinois Crop Science Research and Education Center. They found that the extra wild chromosomes did not always remain in the descendants. In some offspring, both G. tomentella chromosomes disappeared, leaving plants with the normal soybean chromosome number of 40. This kind of chromosome loss is known from other wide hybridization experiments, where alien chromosomes may fail to pass reliably through the pollen or egg cells during meiosis. But the soybean descendants in this study did more than simply return to a familiar genetic state: many displayed traits absent from both parents and from the original DAAL plants.

One especially revealing comparison involved two related lines. LG13-7552 retained 42 chromosomes, including the two G. tomentella chromosomes, while LG12-7063 had 40 chromosomes and no detectable wild chromosomes. Fluorescent in situ hybridization, which uses labeled DNA probes to visualize chromosomes under a microscope, marked the centromeres of soybean chromosomes in one color and the entire G. tomentella chromosomes in another. The images showed the two foreign chromosomes in LG13-7552 and their complete absence in LG12-7063. The 40-chromosome descendant was not simply a smaller version of its progenitor: it was taller, later maturing and more prone to lodging, and it had gray pubescence, white flowers and a buff hilum—features not found in the parents or the DAAL line.

To investigate what the chromosome difference was doing inside the plants, the team sequenced RNA from young seedlings. RNA sequencing provides a snapshot of which genes are active and how strongly they are being transcribed. Using both direct alignment to a soybean reference genome and a reference-free assembly method, the researchers identified thousands of genes whose activity differed between the 42-chromosome DAAL and the 40-chromosome descendant. The reference-based analysis detected 2,499 differentially expressed transcripts, while the de novo assembly identified 4,206 differing contigs. Hundreds of genes showed at least a fourfold difference in expression.

The altered genes pointed toward a plant under internal strain. Gene Ontology analysis found enrichment for biological processes involving defense and immune responses, oxidation-reduction reactions and programmed cell death. Several differentially expressed genes encoded leucine-rich repeat receptor-like kinases, proteins that help plants recognize pathogens and activate defensive signaling. Other highly altered genes were involved in meristem development, flowering time, branching and leaf formation. In the DAAL, six developmental genes were expressed at least 16 times more strongly than in the 40-chromosome descendant. The pattern suggests that the extra chromosomes disrupted the normal regulatory balance of the soybean genome, producing a form of genetic stress even though the seedlings were grown under controlled conditions without an external challenge.

The researchers also found a molecular explanation for one of the most visible differences. Gray pubescence in soybean is associated with reduced activity of the T locus, which encodes flavonoid 3′-hydroxylase, or F3′H, an enzyme in the pathway that produces plant pigments. In LG12-7063, DNA sequencing revealed a previously undocumented deletion of a single adenine base in the third exon of the F3′H gene. Because genetic information is read in three-base units, the missing base shifted the reading frame, changing the encoded amino acids and introducing a premature stop signal. The resulting protein lacks 124 amino acids at its carboxyl end, including a conserved heme-binding region required by cytochrome P450 enzymes such as F3′H. The defective transcript was also present at only about one-quarter the level seen in the DAAL, consistent with nonsense-mediated mRNA decay, a cellular quality-control process that destroys RNA containing an early stop codon.

The most surprising result emerged from field observations of the wider descendant population. Between 2013 and 2016, the frequency of progeny that lost the two alien chromosomes ranged from 3 to 9 percent, averaging about 4 percent. Among more than 100 phenotypically unusual offspring examined, all had 40 chromosomes. Some showed rare traits such as appressed or deciduous pubescence, altered stem termination, sterility or extensive pod shattering. Others switched between black and yellow seed coats, or produced tawny hairs instead of gray ones. The changes occurred at rates that were too high and too widespread to be easily explained by isolated spontaneous mutations.

Genome-wide genotyping strengthened the case for unexpected cross-pollination. Using thousands of SNP markers, the researchers found that many reverted plants differed from their DAAL sisters by heterozygous genetic variants distributed across all 20 soybean chromosome pairs. In one striking comparison, two sister plants from the same row had only modest visible differences, yet polymorphisms separated them throughout the genome. A newly observed tawny-pubescence plant and its gray-pubescence sister were nearly identical overall, but their genetic distance still indicated a cross-pollination event. In other lines, the genome-wide differences were much larger, consistent with pollen from a more distant soybean source.

Soybean is normally a strongly self-pollinating crop. Published estimates commonly place natural cross-pollination below 1 percent in neighboring plants, whereas the DAAL rows showed an average reversion rate of 4 percent and, in some years, as high as 9 percent. The researchers therefore argue that natural cross-pollination is the most likely explanation for the unexpected variation. Yet the proposed explanation creates a second puzzle. Some rare alleles in the descendants are not present in either the cultivated parent or G. tomentella. The likely source may have been soybean germplasm plots located roughly 100 to 500 meters away, but soybean pollen movement over such distances has not been documented under ordinary conditions.

Other mechanisms were considered. Transposable elements—mobile DNA sequences that can insert into genes or change their activity—were more abundant and more often uniquely expressed in the DAAL than in the 40-chromosome descendants. Some of the sequences that matched G. tomentella were retrotransposons, suggesting that the additional chromosomes brought active repetitive DNA into the hybrid genome. Tissue culture used during development of the DAALs can sometimes activate transposable elements, although the researchers note that the culture conditions did not include 2,4-D, a compound frequently associated with tissue-culture-induced variation. Somaclonal variation and spontaneous mutation remain possible contributors, but neither readily explains the widespread heterozygosity found across the genomes of many revertants.

The study does not show that the wild chromosomes directly supplied the new traits. In fact, the researchers found no evidence of expressed G. tomentella sequences in the 40-chromosome descendants, although the limited number of available wild reference sequences makes complete exclusion impossible. The gray-pubescence allele and several altered stem-termination alleles were not found in the perennial parent. Instead, the data suggest that the presence of the alien chromosomes may have destabilized reproduction or genome behavior, allowing unusual pollination and subsequent segregation to generate a burst of variation. Some seed-coat changes also failed to follow the expected three-to-one Mendelian ratios, hinting that additional genetic or developmental factors were involved.

For plant breeders, the results are both promising and cautionary. Alien addition lines can expose a crop to genetic diversity that is otherwise inaccessible, and G. tomentella remains a potential source of resistance traits for soybean improvement. But adding whole chromosomes can disturb gene regulation, alter development and compromise genetic stability. The researchers emphasize that the favorable effects of transferring individual wild genes have yet to be demonstrated in these lines. Before such material can be used reliably, scientists will need to determine why the DAALs cross-pollinate so readily, how pollen traveled across the field, whether the elevated outcrossing persists after the alien chromosomes disappear and which genomic changes are responsible. For now, the soybean experiment has delivered something more dramatic than a new breeding line: a living demonstration of how a plant genome can become unexpectedly fluid when species boundaries are crossed.