Larger plants promote a greater diversity of symbiotic nitrogen-fixing soil bacteria associated with an Australian endemic legume

A major goal in microbial ecology is to understand the factors that structure bacterial communities across space and time. For microbes that are plant symbionts, community assembly processes can lead to either a positive or negative relationship between plant size or age and soil microbe diversity. Here, we evaluated the extent to which such relationships exist within a single legume species (Acacia acuminata) and their naturally occurring symbiotic nitrogen-fixing bacteria (rhizobia). We quantified the diversity of rhizobia that associate with A. acuminata trees of variable size spanning a large environmental gradient in southwest Australia (72 trees in 24 sites spread across similar to 300,000 km(2)), using metabarcoding. We modelled rhizobia diversity using 16S exact genetic variants, in a binomial multivariate statistical framework that controlled for climate and local soil characteristics. We identified two major phylogenetic clades of rhizobia that associate with A. acuminata. Soil sampled at the base of larger Acacia trees contained a higher richness of rhizobia genetic variants. Each major clade responds differently to environmental factors (climate and soil characteristics), but the positive association between tree size and rhizobia genetic diversity was mainly driven by responses from one of the two clades. Overall tree size explained more variation than any other factor, resulting in a similar to 3-fold increase in total richness and clade diversity from the smallest to the largest trees. Synthesis. Previous studies have shown that plant host species is important in structuring microbial soil communities in the rhizosphere. Our results show that host size or age within a single plant species can also structure diversity of at least one group of soil microbes. A positive relationship between plant host size and rhizobia diversity suggests that hosts may modify the niche space of their surrounding soil (niche construction hypothesis) enabling a higher richness of microbial taxa. That different rhizobial groups responded differently to host size and other ecological factors suggests that rhizobia is not an ecologically uniform group, and that entirely neutral explanations for our results are unlikely. Host plants may be analogous to "islands," where larger plant hosts may accumulate diversity over time, through migration opportunities.