▪ Abstract Endophytic fungi living asymptomatically within plant tissues have been found in virtually all plant species. Endophytes are considered plant mutualists: They receive nutrition and protection from the host plant while the host plant may benefit from enhanced competitive abilities and increased resistance to herbivores, pathogens, and various abiotic stresses. Limited evidence also indicates that endophytes may influence population dynamics, plant community diversity, and ecosystem function. However, most of the empirical evidence for this mutualism and its ecological consequences has been based on a few agronomic grass endophytes. More recent studies suggest that endophyte-host plant interactions are variable and range from antagonistic to mutualistic. A more comprehensive view of the ecology and evolution of endophytes and host plants is needed. This article discusses how life history traits—such as fungal reproduction and pattern of infections and genotypic variation and ecological factors—influence the direction and strength of the endophyte-host plant interaction.

Reactive oxygen species are in all types of organisms from microbes to higher plants and animals. They are by-products of normal metabolism, such as photosynthesis and respiration, and are responsive to abiotic and biotic stress. Accumulating evidence suggests reactive oxygen species play a vital role in programmed cell death, stress responses, plant defense against pathogens and systemic stress signaling in conjunction with antioxidant production. Here, we propose that reactive oxygen species and antioxidants, as both universal and evolutionarily conserved, are likely to play important role(s) in symbiotic interactions. To support this hypothesis we review the root and foliar fungal endophyte literature specific to fungal-plant symbiotum production of reactive oxygen species and antioxidants in response to stress. These asymptomatic fungi can produce antioxidants in response to both biotic and abiotic stress when grown in culture as well as in planta. In addition, there is a growing but nascent literature reporting a significant impact of endophyte colonization on the antioxidant activity of colonized (E+) hosts when compared to uncolonized (E-) hosts, especially when exposed to stress. Here we summarize general patterns emerging from the growing literature specific to antioxidant activity of endophytes in colonized hosts and bring up possible future research questions and approaches. The consequences of changes in reactive oxygen species production and increased antioxidant activity in the symbiotum appear to be beneficial in many instances; but costs are also indicated. Unexplored questions are: 1) to what extent do antioxidants originating from the fungal endophyte mediate host metabolism, and thereby control host responses to endophyte colonization; (2) what role do fungal, plant, or symbiotum produced reactive oxygen species and antioxidants have in determining symbiotic outcome between extremes of pathogenicity and mutualism; and (3) what role if any, do the production of reactive oxygen species and their antioxidant counterparts play in the symbiotum’s ability to respond to changing selection pressures? If as the literature suggests, such endophyte imposed mediation can be utilized to foster increases in plant production in resource limited habitats then the utilization of fungal endophytes may prove useful in agronomic and conservation settings.

Endophyte-grass symbiosis is generally considered to be a classic example of microbe-plant symbiosis in which the fitness of the microbial symbiont and its host plant is closely linked, and thus, presumed to align the interests of partners toward mutually beneficial cooperation. Accumulating evidence seems to suggest that defensive mutualism provides the best framework for understanding plant-endophyte interactions in general. We conducted a meta-analysis of 99 published studies on 36 plant (inc. both grass and tree species), 62 herbivore and 17 predator or parasitoid taxons to test the importance of defensive mutualism in multitrophic interactions. In general, statistical perusal revealed that we still know little about these seemingly well-studied biological interactions. The conceptual framework for endophyte-grass interactions has largely been based on endophyte-plant-herbivore studies of two, economically important, artificially selected and introduced agricultural grass species, tall fescue and perennial ryegrass, and two generalist invertebrate pests. Only 10 original publications provided data of higher trophic levels. Consistent with the defensive mutualism hypothesis, the meta-analysis indicates that endophytes slightly increase grass resistance to herbivores, and the defensive mutualism appears to be most commonly detected in systemic and vertically transmitted grass endophytes compared to horizontally transmitted tree endophytes. However, variation appears to increase when higher trophic levels are considered. In addition to taxonomical bias, the literature is strongly biased toward short-term laboratory and greenhouse experiments rather than field conditions. Thus, current literature is insufficient to capture the breadth of variability inherent in the wild grass-endophyte populations and communities, and the general importance of defensive mutualism remains to be solved in future studies.