Do native and invasive plants differ in their interactions with arbuscular mycorrhizal fungi? A meta‐analysis

Abstract

Summary Divergent hypotheses have been proposed that suggest plant invasions either enhance or degrade the mutualism between plants and arbuscular mycorrhizal ( AM ) fungi, but their relative support remains unknown. We conducted a meta‐analysis using 67 publications, involving 70 native and 55 invasive plant species to assess support for the enhanced mutualism hypothesis, the degraded mutualism hypothesis and an alternative hypothesis that factors other than invasive status (such as plant functional group) better predict AM function following invasion. We used multiple measurements to test these hypotheses: AM fungal colonization, growth responses to AM fungi and AM fungal‐mediated shifts in competitive interactions among native and invasive plants. Additionally, we assessed whether invasive plants alter AM associations in native plants and whether native and invasive plants host different AM fungal abundances and communities. Arbuscular mycorrhizal fungal colonization (%) and average growth responses did not differ between native and invasive plants. However, growth responses (±) were dampened among invasive plants, and the positive correlation between AM fungal colonization and growth response in native plants was absent in invasive plants. Rather than plant invasive status, plant functional group was a significant explanatory factor; forbs were generally more colonized and exhibited positive growth responses (when grown alone and in competition), whereas grass responses were neutral to negative. Arbuscular mycorrhizal fungal abundance (measured by percentage colonization, extraradical hyphal and spore densities, as well as neutral lipid fatty acid and glomalin concentrations) did not differ between native and invasive plants, but invasive plants hosted different AM fungal communities in 78% of studies. AM fungal colonization of native plants was lower when grown with, or after, invasive plants, likely due to the prevalence of non‐mycorrhizal plants in studies of neighbour and legacy effects. Synthesis . Neither the degraded nor the enhanced mutualism hypothesis was supported, suggesting that invasions do not select for directional shifts in AM associations. Instead, our results indicate that AM fungi are most likely to influence invasion trajectories when native and invasive plants belong to different functional groups.