Current research has shifted the long-established paradigm of plant phenotype as the sole product of interactions between a plant’s genetics and the abiotic environment to that of a “holobiont” interpretation in which microbes from roots and soils serve as a reservoir of additional microbial genes and functions for the host plant. Findings from both plant-soil feedback research and single inoculation studies have shown that plant species can condition unique soil microbiomes and that particular microbes confer targeted benefits for plants, suggesting functional relationships between plants and soil microbes. However, identifying functional phenotypic linkages between soil microbes and plants requires examining whole-soil community dynamics rather than simply pair-wise comparisons. To examine how whole-soil community dynamics holistically affects plant phenotype, we examined the microbial taxonomic composition of soil from natural populations of Solidago species that vary in phenotypes (two understory species, S. caesia and S. flexicaulis, and one meadow species, S. gigantea). We then conducted a reciprocal greenhouse experiment where each Solidago species was grown in soil conditioned from each species to assess if plant phenotype is modified when grown by distinct soil microbial taxa.
Results/Conclusions
Abundance of particular fungal phyla varied between Solidago species such that Basidiomycota fungi were more prevalent in S. caesia and S. flexicaulis soil than S. gigantea soil, and Ascomycota fungi were more prevalent in S. gigantea soil than in S. caesia and S. flexicaulis soil. Under experimental conditions, varying the origin of the microbiome altered root biomass and leaf area traits but not other growth or reproductive traits. For example, S. caesia produced 30% larger root biomass when grown in conspecific soil compared to soil of S. gigantea (p = 0.04), and S. flexicaulis produced 35% more root biomass when grown in S. caesia soil compared to S. gigantea soil (p = 0.04). S. gigantea exhibited slightly greater specific leaf area when grown in conspecific soil compared to either S. caesia or S. flexicaulis soil (p = 0.07). These data suggest that whole soil microbial communities may shift phenotypes in some functional traits but whole communities from the field may not play as large a role in determining plant phenotypes as controlled experimental work suggests. Future work comparing microbial taxonomic and functional-plant phenotype interactions is needed to clarify these relationships under more natural conditions. This may provide more accurate understanding of the drivers of plant phenotypes.