Grasses hosting Epichloë endophytes are protected against herbivores due to the production of various fungal alkaloids. Previous research has found that high foliar phosphorus concentrations reduce the level of the alkaloid ergovaline, thereby reducing the endophyte-mediated herbivore resistance. Yet, the impact of phosphorus on ergovaline biosynthesis versus its influence on endophyte growth and synthesis of other fungal alkaloids remains unresolved. Our objective was to elucidate these relationships. We grew endophyte-symbiotic and non-symbiotic Festuca arundinacea plants and fertilized them with different doses of phosphorus. Later, half of the plants from each treatment were challenged with larvae of the generalist chewing insect Spodoptera frugiperda. We assessed the relationships between foliar phosphorus levels, fungal mycelium, and alkaloid concentrations, as well as their impacts on larvae performance, herbivore-caused damage, and plant biomass. Endophyte mycelial biomass in plant tissue was found to be independent of foliar phosphorus concentration. The alkaloids lolines and peramine showed a linear relationship with mycelial biomass but no correlation with foliar phosphorus. Surprisingly, high ergovaline concentrations were positively associated with an interaction between endophyte mycelial biomass and foliar phosphorus concentration. Although herbivory increased loline concentration, only high concentrations of ergovaline and peramine were related to reduced S. frugiperda larvae weight gain. However, endophyte presence did not reduce herbivory damage to plants. Contrary to expectation, we did not find a negative but a positive association between concentrations of foliar phosphorus and ergovaline alkaloid, through its interaction with endophyte mycelial biomass. Alternatively, our findings suggest that phosphorus plays a crucial role in modulating the Epichloë-mediated defensive mutualism, primarily through its effects on ergovaline rather than on endophyte concentration or production of other alkaloids.

Herbivory index

We used an herbivory index to estimate the damage caused by herbivores to plants (Boege & Dirzo, 2004; Salgado-Luarte et al., 2023). Plants not exposed to herbivores were not included in this calculation as they did not show signs of damage (personal observation). For each estimation, five leaves were randomly selected and photographed. Then, the damaged area was estimated using the software ImageJ (Abràmoff et al., 2004). As in Boege & Dirzo (2004), each leaf was scored according to the level of damage: 0 = no damage, 1 = 1-6% damage, 2 = 6-12% damage, 3 = 12-25% damage, 4 = 25-50% damage, and 5 = more than 75% of damage. These categories were selected because low levels of damage were most frequent and high levels of damage were less common. Thus, categories with narrower ranges were needed in the lower part of the range to capture differences in plant damage (Boege & Dirzo, 2004). Adapted from Boege and Dirzo (2004), the herbivory index for each plant was calculated as: Herbivory index = d/N ⨯ ∑ ni ⨯ Ci, where d is the number of leaves with damage, N the total number of evaluated leaves, i is the ith damage score, ni the number of leaves with the ith damage score, Ci is the midpoint of the ith damage score (i.e., C0 = 0, C1 = 3.5, C2 = 9.0, C3 = 18.5, C4 = 37.5, and C5 = 87.5).

References

• Abràmoff, M. D., Magalhães, P. J., & Ram, S. J. (2004). Image processing with ImageJ. Biophotonics International, 11(7), 36-42.

• Boege, K., & Dirzo, R. (2004). Intraspecific variation in growth, defense and herbivory in Dialium guianense (Caesalpiniaceae) mediated by edaphic heterogeneity. Plant Ecology, 175(1), 59-69. https://doi.org/10.1023/B:VEGE.0000048092.82296.9a

• Salgado‐Luarte, C., González‐Teuber, M., Madriaza, K., & Gianoli, E. (2023). Trade‐off between plant resistance and tolerance to herbivory: Mechanical defenses outweigh chemical defenses. Ecology, 104(1): e3860. https://doi.org/10.1002/ecy.3860