Explore the vast range of titles available.

PREMISE OF THE STUDY: Multiple mutualist effects (MMEs) are common in nature, yet we lack a predictive understanding of how two mutualists on the same host will influence each other and whether these effects will be positive or negative. Leguminous plants maintain root symbioses with two nutritional mutualists: rhizobia that fix atmospheric nitrogen and arbuscular mycorrhizal fungi (AMF) that increase phosphorus uptake. Both symbionts receive plant carbon, and host genetic networks that regulate colonization are partially shared by both symbioses; whether these factors generate trade-offs or facilitation between rhizobial and AMF symbionts of legumes is not well known. METHODS: We evaluated host allocation to each symbiont in three settings. First, in situ in a remnant prairie, then in a greenhouse experiment with multiple plant populations, and finally under manipulated rhizobium densities in the greenhouse. KEY RESULTS: In the remnant prairie, rhizobium nodule number and colonization of AMF were positively correlated, and plants with increased nodule number had higher fitness in the field, generating indirect selection on the colonization of AMF. In the greenhouse experiment, allocation to each symbiont was genetically variable among populations, with some suggestion that rhizobium and AMF colonization are positively genetically correlated. Finally, increasing the number of rhizobia in the soil decreased AMF colonization. CONCLUSIONS: Our results suggest that trade-offs between plant colonization by rhizobia and AMF are context dependent and might not be common under field conditions, but that physiological and/or genetic drivers couple these two symbioses in nature.

Mutualisms between hosts and multiple symbionts can generate diffuse coevolution if genetic covariance exists between host traits governing multiple interactions. Rhizobia and arbuscular mycorrhizal fungi (AMF) both interact with legume hosts, providing complementary nutrients (nitrogen and phosphorous). Molecular approaches have revealed extensive pleiotropy in the plant genetic pathways required for colonization of both symbionts; however, a quantitative genetic approach is required to understand whether this pleiotropy shapes evolution in natural populations. In a greenhouse experiment with 75 families of Chamaecrista fasciculata grown in two phosphorous (P) environments (fertilized and unfertilized), positive covariance between nodule number and plant aboveground biomass within and across environments indicates selection for increased colonization by rhizobia. Genetic variation for host restriction of AMF colonization in response to P suggests that this aspect of context dependency can evolve in host populations and that selection in this mutualism varies with P. Despite the existence of gene-level pleiotropy during rhizobium and AMF infection, we find no evidence for genetic covariance in symbiont colonization or its response to P—suggesting that genetic variation at other, nonpleiotropic loci governs variation in colonization and thus that these traits likely evolve independently in plant populations.

Expertise in the broader impacts of scientific research is an increasingly important aspect of professional development, particularly because federal grant proposals are commonly reviewed using both the Intellectual Merit and the Broader Impacts Criteria. Unfortunately, training in broader impacts, such as science communication and outreach, is not typically part of undergraduate or graduate curricula. We initiated one of the first graduate-level biology courses on broader impacts, focusing on giving graduate students firsthand, authentic experiences with grant writing, science communication, and educational outreach. Students in this interdisciplinary course learned from experts, wrote for a broad audience about their own research, and proposed and implemented outreach in collaboration with local organizations. We outline our approach, discuss outcomes from each activity, assess our impact, and finally consider how future programs might expand on this model.