Functional Redundancy in Ocean Microbiomes Controls Trait Stability

Abstract

Theory predicts that functional redundancy in microbial communities increases trait stability, meaning that traits or functions are less likely to be lost from the community when species go extinct. However, few experiments have empirically tested this prediction, especially in the context of microbial communities and at the landscape scale. In part, the lack of metrics for functional redundancy in microbial ecosystems has prevented addressing this question. In a companion manuscript we proposed a quantitative metric for functional redundancy called Contribution Evenness (CE) that is optimized to reflect trait stability. Here, we use CE to predict the stability of marine microbial functions to species and transcription loss. Using transcriptomes deposited in the Ocean Microbial Reference Gene Catalog (OM-RGC.v2), a catalog of genes and transcripts sequenced by the TARA Ocean expedition, we quantified the functional redundancy for 4,314 KEGG Orthologs (KOs) across 124 marine sites. Functional redundancy was highly correlated with a latent variable consisting of four ocean physiochemical parameters: oxygen and chlorophyll a concentrations, depth, and salinity. Functional redundancy was higher at the poles than in non-polar regions. Simultaneously, regional {beta}-diversity for individual functions was higher for functions with higher functional redundancy. These observations provide evidence that higher functional redundancy indicates increased stability of microbial ecosystem functions on spatiotemporal scales consistent with surface ocean mixing. We suggest that future changes in ocean physiochemistry could likely influence this stability for functions with lower functional redundancy. ImportanceFunctional redundancy describes the state of multiple species performing the same function. Theory suggests functional redundancy stabilizes microbial community functions from disturbances leading to species loss or other changes to the microbiome. Previous work suggests that functional redundancy is common in ocean microbiomes which implies traits should be more stable among metacommunities. Some laboratory experiments demonstrate this idea, but it is difficult to test in the natural world. In a companion manuscript, we proposed a functional redundancy metric sensitive to trait stability. Here, we used this metric to show that functional redundancy varied substantially among ocean microbiomes and that regions with higher functional redundancy had higher regional trait stability. Last, we noted that variations in functional redundancy strongly correlated to ocean physiochemistry. Thus, changes in ocean physiochemistry via climate change may alter community traits to become more or less resistant to disturbance relative to contemporary conditions.