From the inside out: fungal endophyte–grass associations and grassland communities

Fungal endophytes and other clandestine citizens that reside within plants are increasingly appreciated for the role they play in community ecology. In a well designed study examining the interaction between the fungal endophyte Neotyphodium lolii and perennial ryegrass, Lolium perenne, Rasmussen et al. (this issue; pp. 787–797) address the question of whether the fungal alkaloid content of the host plant is a function of enhanced alkaloid biosynthetic rates within the endophyte, or of increased endophyte populations in the plant. The authors’ investigations provide much-needed insight into how genetic and abiotic interactions affect the fungal endophyte–grass host relationship, and how these in turn could influence multidirectional biotic interactions within an agronomic grassland community. ‘This would suggest that the host plant's C and N metabolite pool status conditions its ability to effectively limit fungal colonization.’ Plants respond to a myriad of endogenous and environmental cues by modulating their utilization of available carbon and nitrogen resources, which serve to accommodate the resource demands of growth, development and reproduction, as well as interactions with other organisms. The metabolic networks that promote allocation of C and N resources to different plant parts, and partitioning of those resources into different biosynthetic pathways, not only are under exquisite genetic control, but also are incredibly responsive to cues such as C and N resource status. For example, genome-wide analyses of gene expression have demonstrated that extensive reconfiguration of the transcriptome occurs in response to differential N availability, which effects dramatic changes in primary and secondary metabolism as well as growth and developmental processes (e.g. Scheible et al., 2004). But how do changes to a plant's resource status affect the association of the plant with interacting organisms such as endophytic fungi? Rasmussen et al. investigated how the resource status of L. perenne affects communications between the host plant and its fungal endophyte, N. lolii. To this end, the authors manipulated the resource status of L. perenne through increasing N availability as well as by use of cultivars exhibiting contrasting levels of soluble carbohydrate. Two manifestations of resource-mediated interactions between the host plant and its fungal endophyte may be altered colonization of the host plant by N. lolii, and/or a change in the fungal endophyte's partitioning of resources to alkaloid biosynthesis. Without a reliable means of quantifying fungal biomass, these two scenarios remain confounded. Rasmussen et al. shed light on these questions by using quantitative PCR as a means to estimate N. lolii fungal biomass within the host, L. perenne. The sensitivity and specificity of quantitative PCR is having a transformative effect on the study of interacting genomes. The ability to quantify the extent of colonization of a plant host by microbial associates provides a powerful lens through which to view these interacting species (Schena et al., 2004; Mumford et al., 2006). The addition of spatial or temporal dimensions to these analyses allows a means of exploring the dynamics of plant–microbe interactions. The utility of quantitative PCR is nicely illustrated by the authors’ investigations of the N. lolii–L. perenne symbiotum. Rasmussen et al.'s quantitative PCR analyses revealed that fungal endophyte biomass was negatively correlated with increasing plant-soluble resource pools, measured either as soluble carbohydrates or N sources (amino acids and proteins). In the case of the soluble carbohydrate pools these differences are a function of host plant differences at the genetic level, whereas in the case of the soluble N pools these differences result from N fertilization. Thus fungal endophyte population levels can be influenced by both the host's genotype and the environment. The authors present a plausible argument that the difference in fungal DNA levels in these high soluble-N or high soluble-C plants is not a consequence of a ‘dilution effect’, that is, caused by increased N availability stimulating growth of the plant more than the endophyte (sensu Lane et al., 1997). This would suggest that the host plant's C and N metabolite pool status conditions its ability to limit fungal colonization effectively. How does this happen? Does the plant employ strategies akin to the defence mechanisms invoked in response to pathogen attack? Or do other mechanisms come into play? An understanding of the mechanisms that the plant uses to regulate fungal growth is surely one key to unlocking the mystery of fungal endophyte symbioses. As one step towards elucidating this relationship, two outstanding studies recently revealed that reactive oxygen species produced by the NADPH oxidase NoxA act to regulate hyphal branching in the fungal endophyte Epichloë festucae, and identified regulators of NoxA (Takemoto et al., 2006; Tanaka et al., 2006). The authors also used an elegant regression method to demonstrate that concentrations of fungal alkaloids are proportional to fungal DNA concentrations. In other words, plant resource availability does not influence the proportion of alkaloids synthesized per unit fungi. Interestingly, the fungal endophyte increased soluble pools of low molecular-weight carbohydrates in the host plant, but only for the plant cultivar with standard carbohydrate levels. Thus, at least in this study, the endophyte is capable of altering partitioning of C resources in the host plant, but the host plant does not appear to alter partitioning of C resources towards alkaloid biosynthesis in the endophyte associate. Bioactive alkaloids synthesized by fungal endophytes have been well characterized as antiherbivory agents (summarized by Schardl et al., 2004). Tanaka et al. (2005) recently provided genetic evidence that peramine, an alkaloid synthesized by N. lolii, provides the host plant with protection against insect herbivory. Considerable evidence exists that colonization of grass species with fungal endophytes has an impact on grass–herbivore interactions, and that these interactions, in turn, affect other species within the grassland community (e.g. de Sassi et al., 2006). Rasmussen et al. present compelling evidence that N availability affects fungal endophyte levels and, by extension, concentrations of three of the four fungal alkaloids that they examined. The finding that alkaloid concentrations decreased with increased N availability in endophyte-infected perennial ryegrass is consistent with the authors’ earlier report (Hunt et al., 2005), but contrasts with the reports of others (Belesky et al., 1988; Arachevaleta et al., 1992). On first glance, these results appear to be irreconcilable. However, Rasmussen et al. provide fresh insight that may help resolve this apparent conundrum. First, these various studies have often employed grass species and fungal endophytes of different genetic backgrounds (different species or genotypes). As the authors demonstrate, the genotype of both host plant and fungal endophyte may influence the degree of fungal colonization, and hence alkaloid accumulation in the plant. These observations lend credence to the notion that symbiota comprising different species may well respond differently to environmental variables such as N availability. Second, these published studies measure a range of alkaloids that, as Rasmussen et al. show, can exhibit distinct responses to N fertilization. Third, obtaining comparable levels of endophyte colonization between research groups – indeed, even from one experiment to the next – is challenging. The results of Rasmussen et al. suggest that this variability will affect alkaloid accumulation, which may have a strong impact on the outcome of the study. The issue is compounded by the practice of inducing alkaloid accumulation by clipping leaf blades of the host plant, which may introduce even more variability. Future efforts that quantify alkaloid concentrations as a function of both plant dry weight and fungal biomass, as estimated by quantitative PCR, should help to delineate relationships between N availability, fungal endophyte biomass, alkaloid concentrations and herbivory. As mentioned above, Rasmussen et al. showed that the genotype of the host plant influences fungal endophyte colonization levels. Because endophyte cell numbers were found to be correlated with fungal alkaloid concentrations, the plant genotype also potentially influences herbivores through its effect on endophyte levels. In a similar vein, Bailey et al. (2005) reported that the genotype of two Populus species and hybrids of these species conditions the tree's relationship with fungal endophytes. In this study, fungal endophyte infection of twigs from P. fremontii, P. angustifolia and their hybrids was negatively correlated with the concentration of condensed tannins in twig bark. Condensed tannin concentration, in turn, is a trait that is under strong genetic control in these poplars. We can thus begin to investigate fungal endophyte–host plant interactions within the context of community genetics; that is, genetic interactions between species in communities, and the influence of the abiotic environment on those interactions (Whitham et al., 2006). This may be especially true where the host is a foundational species within either a pasture or natural grassland community, as we might predict that a gene exerting large phenotypic effects in a foundational species will have a proportionately greater effect within its community. These fungal endophyte–host grass interactions, and the interactions of the symbiotum with other organisms in the grassland community, provide an exciting system with which to explore emerging concepts such as interspecific indirect genetic effects and community phenotypes. Innovations in the ways in which we explore the interactions between plants and their microbial colonizers such as fungal endophytes are instrumental in building a comprehensive model of plant host–microbial associate dynamics. The study by Rasmussen et al. provides a fine example where the insight that has been gained would not have been possible using conventional approaches to assess the fungal endophyte partner. Genomics resources are becoming available for an ever-expanding range of plant species and their microbial partners, and these resources are likely to form part of the toolbox for investigating grass–fungal endophyte associations. Assuredly, the power of molecular and genomics approaches will be indispensable in our quest to understand not only the mechanisms that condition fungal endophyte–host plant interactions, but also how these interactions affect community-level processes.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.