OOS 12-3 - Similarity in gene expression predicts species interactions, coexistence and ecosystem functioning

Tuesday, August 8, 2017: 8:40 AM
E145, Oregon Convention Center
Anita Narwani1, Bastian Bentlage2, Markos Alexandrou3, Keith J. Fritschie4, Charles Delwiche2, Todd H. Oakley5 and Bradley J. Cardinale6, (1)Eawag, Swiss Federal Institute of Aquatic Science and Technology, Dubendorf, Switzerland, (2)Cell Biology and Molecular Genetics, University of Maryland, College Park, MD, (3)Wildlands Conservation Science, Lompoc, CA, (4)Department of Biological Sciences, Dartmouth College, Hanover, NH, (5)Ecology, Evolution and Marine Biology, University of California, Santa Barbara, CA, (6)School of Natural Resources & Environment, University of Michigan, Ann Arbor, MI

Phenotypic variation controls the nature of species interactions, which in turn determine whether or not species coexist, and how they influence ecosystem functioning. We tested the hypothesis that as phenotypic similarity in patterns of gene expression across an organism’s entire transcriptome declines across species, they will be more likely to experience weaker competition, and will therefore be more likely to coexist and increase ecosystem functioning. To test this, we grew eight species of freshwater green algae in monocultures and bicultures for 46 days in a lab microcosm experiment. We quantified the strength of species interactions and coexistence by: 1) fitting Lotka-Volterra models to time-series densities and estimating interaction coefficients, and 2) calculating relative densities that compare species’ steady-state densities in biculture to those in monoculture. We used Illumina sequencing to quantify the expression of 1,253 families of homologous genes, including a set of 17 candidate genes that we hypothesized a priori to be involved in competition or facilitation. To assess levels of ecosystem functioning, we measured the community total biovolume, primary production, respiration and nutrient uptake.


We found that closely related species had greater similarity in gene expression than did distantly related species, but as gene expression became more similar across the transcriptome, species experienced weaker competition or greater facilitation, and were more likely to coexist. Species with greater similarity in gene expression also tended to increase levels of all ecosystem functions measured. We further identified gene functional categories that were contrastingly regulated in species experiencing different types of interactions. Transporter genes for a variety of nutrients including nitrate, sugars, and other micronutrients tended to be upregulated in species experiencing competition. By contrast, species experiencing facilitation upregulated genes associated with core cellular metabolism (e.g. the Citric Acid Cycle) and carbon fixation through the Calvin Cycle. In summary, contrary to common thinking in ecology and evolution, similarity in gene expression across the whole transcriptome, and not differentiation, was associated with weaker competition, stronger facilitation, coexistence, and higher ecosystem functioning. However, differential expression of gene families with particular gene functions could be identified that were uniquely upregulated in species experiencing different interaction types.