Managing friends and foes: Sanctioning mutualists in mixed‐infection nodules trades off with defense against antagonists

Successful plant growth requires plants to minimize harm from antagonists and maximize benefit from mutualists. However, these outcomes may be difficult to achieve simultaneously, since plant defenses activated in response to antagonists can compromise mutualism function, and plant resources allocated to defense may trade off with resources allocated to managing mutualists. Here, we investigate how antagonist attack affects plant ability to manage mutualists with sanctions, in which a plant rewards cooperative mutualists and/or punishes uncooperative mutualists. We studied interactions among wild and domesticated pea plants, pea aphids, an aphid‐vectored virus (Pea enation mosaic virus, PEMV), and mutualistic rhizobial bacteria that fix nitrogen in root nodules. Using isogenic rhizobial strains that differ in their ability to fix nitrogen and express contrasting fluorescent proteins, we found that peas demonstrated sanctions in both singly‐infected nodules and mixed‐infection nodules containing both strains. However, the plant's ability to manage mutualists in mixed‐infection nodules traded off with its ability to defend against antagonists: when plants were attacked by aphids, they stopped sanctioning within mixed‐infection nodules, and plants that exerted stricter sanctions within nodules during aphid attack accumulated higher levels of the aphid‐vectored virus, PEMV. Our findings suggest that plants engaged in defense against antagonists suffer a reduced ability to select for the most beneficial symbionts in mixed‐infection tissues. Mixed‐infection tissues may be relatively common in this mutualism, and reduced plant sanctions in these tissues could provide a refuge for uncooperative mutualists and compromise the benefit that plants obtain from mutualistic symbionts during antagonist attack. Understanding the conflicting selective pressures plants face in complex biotic environments will be crucial for breeding crop varieties that can maximize benefits from mutualists even when they encounter antagonists.

Methods
Overview of experiment. This dataset was collected from a greenhouse experiment performed at Washington State University, Vancouver in Fall 2020-Winter 2021. We grew different varieties of pea plants in individual pots, inoculated young plants with different strains of rhizobial bacteria (2 weeks post-sowing), exposed plants to different aphid treatments (6 weeks post-sowing), and then harvested plants (8 weeks post-sowing). The experiment followed a full-factorial design with 3 pea varieties x 5 types of rhizobial inoculum x 3 types of aphid exposure x 6 block replicates, for a total of 270 plants in the experiment.

Description of treatments. The 3 pea varieties were 1) a non-nodulating domesticated variety, 2) a nodulating domesticated variety, and 3) a wild variety. For rhizobial inocula we fluorescently labeled a nitrogen-fixing (Fix+) and non-nitrogen-fixing (Fix-) strain, reciprocally labeling each strain with either a red or green fluorescent marker. The 5 rhizobial inocula were as follows: 1) uninoculated control, 2) red Fix- and green Fix-, 3) red Fix+ and green Fix+, 4) red Fix+ and green Fix-, and 5) green Fix+ and red Fix-. Thus, inoculations #2 and #3 were "single inoculations" in which plants could only form one type of nodule (Fix- or Fix+, respectively), and inoculations #4 and #5 were "co-inoculations" in which plants could form Fix+ nodules, Fix- nodules, or "mixed-infection" nodules containing both strains. The 3 aphid treatments were 1) no aphid exposure, 2) exposure to virus-free aphids, and 3) exposure to aphids carrying the Pea enation mosaic virus (PEMV), which can be transmitted to pea plants.

Types of data collected. We measured shoot mass of harvested plants, assayed shoot tissue for accumulation of the aphid-transmitted plant virus (PEMV), photographed root systems with a fluorescence microscope to determine the number and color of individual nodules, and cultured rhizobia from a subset of nodules to check that the external color of nodules corresponded with the type of strain occupying the nodule. From the fluorescence photographs, we used ImageJ to calculate the size and color of each nodule on each plant's root system. We collated these data into three datasheets which are archived here.