Back

Genetic variation in host selectivity and adaptive strain enrichment within legume-rhizobia symbiosis: processes are host-dependent, far from perfect, and correlate with nodule morphology

Burghardt, L. T.; Sydow, P.; Sutherland, J.; Epstein, B.; Tiffin, P.

2025-06-20 evolutionary biology
10.1101/2025.06.16.659998 bioRxiv
Show abstract

Mutualism breakdown can be prevented if partner species preferentially select and reward partners that provide greater benefit. We examined these two components using the legume Medicago truncatula and its nitrogen-fixing symbiont Sinorhizobium meliloti. First, we reanalyzed data from 202 accessions to show significant genetic variation in the capacity of Medicago to restrict strain diversity, finding that hosts with shorter nodules were more selective. A genome-wide association study on host selectivity identified genes including the hormone leginsulin, pectin degradation, multidrug and toxic compound efflux, Zinc transport, and DNA methylation. Second, we used two well-studied Medicago genotypes with contrasting nodule morphologies to assess the effectiveness of adaptive enrichment mechanisms by sampling the relative frequencies of rhizobial strains in pools of small nodules (indicating a lack of host investment) compared to large nodules (indicating increased host investment) and pairing these results with previous single-strain assessments of strain benefits to hosts. While both hosts enriched beneficial strains in large nodules, the host that formed larger and more variably sized nodules and thus had greater potential to increase rhizobial populations was less effective. Our findings reveal that host genetic variation affects strain selectivity and suggest that nodule traits warrant attention when exploring mutualism evolution.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.