Symbiotic fungi and clonal plant physiology

Fungal symbionts commonly occur in plants influencing host growth, physiology, and ecology (Carlile et al., 2001). However, while whole-plant growth responses to biotrophic fungi are readily demonstrated, it has been much more difficult to identify and detect the physiological mechanisms responsible. Previous work on the clonal grass Glyceria striata has revealed that the systemic fungal endophyte Epichloë glyceriae has a positive effect on clonal growth of its host (Pan & Clay, 2002; 2003). The latest study from these authors, in this issue (pp. 467–475), now suggests that increased carbon movement in hosts infected by E. glyceriae may function as one mechanism by which endophytic fungi could increase plant growth. Given the widespread distribution of both clonal plants and symbiotic fungi, this research will have implications for our understanding of the ecology and evolution of fungus–plant associations in natural communities. Well over 80% of vascular plant species form symbiotic associations with arbuscular mycorrhizal fungi (AMF) within their roots (Smith & Read, 1997). Also, AMF and leaf-inhabiting fungal endophytes are widely distributed in some ecologically important groups, such as the grasses (Newsham & Watkinson, 1998; Clay & Schardl, 2002). Given the preponderance of clonal plants in many habitats worldwide, the potential for fungal symbionts to impact plant population and community dynamics is enormous. In addition, by differentially affecting the sexual or asexual reproduction of host genotypes, infection by symbiotic fungi (whether pathogenic or mutualistic) is likely to influence microevolutionary processes within plant populations. For clonal plants, clonal growth is a major determinant of genet fitness through its effects on genet persistence and new ramet and seed production (Pan & Price, 2002). Only a few studies have described the potential role of symbiotic fungi in the growth and asexual reproduction of clonal plants. Streitwolf-Engel et al. (2001) showed that, for the stoloniferous herb Prunella vulgaris, AMF explained more of the variation in clonal traits such as ramet production than did host genotype. They concluded that the effects of AMF on plant growth and clonal reproduction were great enough to ‘affect population size and variation of clonal plants in communities’. For the stoloniferous grass Glyceria striata, Pan and Clay (2002) reported that plants infected by the systemic endophyte Epichloë glyceriae produced more stolons and clonal growth mass than uninfected plants. The results suggest that infected (+E) Glyceria genotypes should be spatially larger due to increased lateral spread than uninfected (–E) genotypes in natural communities. However, infection and host genotype were confounded because only naturally infected or uninfected individuals were utilized. In a follow-up study (Pan & Clay, 2003), effects of infection and host genotype were experimentally separated through the use of +E and –E replicates of the same host genotypes. This was achieved by manual separation of +E individuals into two ramet groups, one of which was treated with the systemic fungicide Benomyl®. Two host genotypes were replicated for each of three populations in southern Indiana, USA. Total dry mass did not differ between +E and –E plants; however, stolon numbers and lengths were significantly greater in +E plants (Pan & Clay, 2003). Furthermore, +E hosts allocated more of their total mass to clonal growth in all three populations (Fig. 1). Clearly in this system there was a shift in resource allocation associated with fungal infection. Putative physiological mechanisms that might account for this effect remained elusive. Mean (+ SE) percent of the total genet biomass allocated to clonal growth (stolons and daughter ramets) in three populations of the clonal grass Glyceria striata infected (shaded bars) or not infected (open bars) by the systemic fungal endophyte Epichloë glyceriae. Four infected and four uninfected replicates of two genotypes were used in each population. Data supplied by J.J. Pan, based on research reported in Pan & Clay (2003). Epichloë (Clavicipitaceae; Ascomycotina) and its asexual derivatives (Neotyphodium) are systemic fungal endophytes of cool-season grasses with effects on their hosts that span the continuum from antagonism to mutualism (Schardl, 1996; Schardl et al., 2004). Host benefits include protection from herbivores due to four classes of alkaloids produced by the endophyte; improved abiotic stress tolerance; and enhanced growth (Clay & Schardl, 2002). Antagonistic aspects to the symbiosis are due to the abortion of host inflorescences during stroma formation, which is part of the fungal sexual cycle (Schardl, 1996). The fungal hyphae are only located between the leaf cells of the host and do not penetrate host cell walls. Presumably, endophytic hyphae are able to utilize simple sugars, amino acids or other metabolites located within the intercellular spaces as nutrient sources (Clay & Schardl, 2002). The clonal grass Glyceria striata is widely distributed throughout most of the United States, typically occurring in moist woodlands or marshes (Gould & Shaw, 1983). An individual genet grows vegetatively by basal tiller production to form a clump of ramets, but can also exhibit lateral spread during clonal growth by ramet production along extending stolons. Tillers can develop large, open inflorescences (panicles) for sexual reproduction, but only when they are not infected by Epichloë glyceriae. In their paper in this issue, Pan & Clay note that, due to the replacement of inflorescences by fungal fruiting bodies (stromata), fertile inflorescences are never found on Epichloë-infected hosts. Hence, G. striata is sexually sterile when infected, but remains quite capable of vigorous clonal growth and asexual reproduction (Pan & Clay, 2003; Fig. 1). In efforts to document and understand physiological integration, that is, the sharing of resources between connected ramets within a genet, clonal plant biologists have often used 14C to trace the movement of carbon compounds (Jónsdóttir & Watson, 1997). Physiological integration may be ecologically advantageous to clonal plants in resource-poor environments (Jónsdóttir & Watson, 1997) or highly competitive communities (Gough et al., 2002). The new paper by Pan & Clay extends clonal plant research a bit further by asking whether or not changes in physiological integration can be mediated by a symbiotic fungus. Because they had shown previously that infection by Epichloë increased clonal growth of the host Glyceria striata (Pan & Clay, 2002, 2003), they hypothesized that increased resource movement in infected plants could be part of the mechanism responsible. Both severed and intact stolons of G. striata that were infected (+E) or not infected (–E) by Epichloë were labeled with 14C during the thirteenth week of the glasshouse experiment. Two weeks later, above-ground biomass was harvested; labeled stolons were divided into segments; and these were analyzed to determine the pattern of assimilate distribution. Stolon segments both proximal and distal to the original 14C-labeled leaf were assessed for radioactivity. Pan & Clay found that both proximal and distal movement of 14C away from the labeled leaf was greater for +E stolons; in contrast, more assimilate was retained by the labeled leaf of –E stolons. Although carbon movement was greater along +E stolons, there was no difference in stolon growth or new ramet production between +E and –E hosts, suggesting that the observed level of physiological integration did not play a major role in improving clonal growth in this study. The results of Pan & Clay's efforts are exciting because, by showing that symbiotic fungi change the dynamics of carbon within the host, they have opened up new avenues for research into the mechanistic basis for such an effect. The physiological reasons for the increase in carbon translocation in infected hosts is currently unknown, but Pan & Clay speculate that hormones or enzymes produced by Epichloë could be responsible. Although not explicitly mentioned by the authors, fungal-mediated changes in cytokinin levels may be a possibility, especially as cytokinins have been implicated in the regulation of source–sink relations in plants (Roitsch & Ehneß, 2000). Furthermore, both phytopathogenic and mycorrhizal fungi have been shown to produce cytokinins (Jameson, 2000). It is therefore possible that the endophyte affects assimilate partitioning by increasing the sink strength of meristems, which provide a nutrient-rich environment for hyphal growth (Schardl et al., 2004). Active uptake of carbon compounds by the fungus in regions of active host growth should increase sink strength. Alternatively, source strength may be increased if the endophyte somehow enhances leaf photosynthesis. Indeed, there is some evidence for greater photosynthetic rates in endophyte-infected tall fescue (e.g. Marks & Clay, 1996), but it is not certain that such effects would increase carbon translocation. However, assimilate movement is clearly driven by the pressure gradient between source and sink regions (Oparka & Santa Cruz, 2000), and the possibility that endophyte infection can alter source–sink relations cannot be discounted. Future investigations into the carbon physiology of the endophyte–grass symbiosis should consider fungal-mediated effects on carbohydrate production (Marks & Clay, 1996); translocation (Pan & Clay); and storage (Cheplick & Cho, 2003). Pan & Clay's new study provides a nice starting point for further investigations into the potentially important role that symbiotic fungi play in the physiology and ecology of clonal plants. There is no doubt that the ability to integrate resources impacts the population dynamics of clonal plants (Jónsdóttir & Watson, 1997). Provisioning of carbohydrate resources to new ramets by established genets should improve the probability of ramet- and whole-genet survival in a heterogeneous environment. Symbiotic endophytes might improve the odds for genet spread and persistence within the local community. It is also increasingly recognized that symbiotic associations can be significant determinants of plant community structure (Clay, 2001). Mycorrhizal associations, for example, can affect the ecology of clonal species (Streitwolf-Engel et al., 2001; Watson et al., 2002) and entire plant communities (Newsham et al., 1995). The community and ecosystem-level consequences of fungal endophyte infection are only just beginning to be documented and explored by ecologists (Clay & Schardl, 2002; Rudgers et al., 2004).