Traits along the leaf economics spectrum are associated with communities of foliar endophytic symbionts

Abstract

Leaf traits of plants worldwide are classified according to the Leaf Economics Spectrum (LES), which links leaf functional traits to evolutionary life history strategies. As a continuum ranging from thicker, tough leaves that are low in nitrogen (N) to thinner, softer, leaves that are high in N, the LES brings together physical, chemical, and ecological traits. Fungal endophytes are common foliar symbionts that occur in healthy, living leaves, especially in tropical forests. Their community composition often differs among co-occurring host species in ways that cannot be explained by environmental conditions or host phylogenetic relationships. Here, we tested the over-arching hypothesis that LES traits act as habitat filters that shape communities of endophytes both in terms of composition, and in terms of selecting for endophytes with particular suites of functional traits. We used culture-based and culture-free surveys to characterize foliar endophytes in mature leaves of 30 phylogenetically diverse plant species with divergent LES traits in lowland Panama, and then measured functional traits of dominant endophyte taxa in vitro. Endophytes were less abundant and less diverse in thick, tough, leaves compared to thin, softer, leaves in the same forest, even in closely related plants. Endophyte communities differed according to leaf traits, including leaf punch strength and carbon and nitrogen content. The most common endophyte taxa in leaves at different ends of the LES differ in their cellulase, protease, chitinase, and antipathogen activity. Our results extend the LES framework for the first time to diverse and ecologically important endophytes, opening new hypotheses regarding the degree to which foliar symbionts respond to, and extend, the functional traits of leaves they inhabit.

Keywords: Ascomycota; endophytic fungi; leaf functional traits; symbiosis; tropical forest.

Figure 1

Endophyte abundance, diversity, and richness varied with leaf traits along the leaf economics spectrum. First two latent variables (LV) from partial least squares regression (PLSR) showing correlations between leaf functional traits (blue) and (A) endophyte abundance (culture-based survey; 24 host species), (B) endophyte diversity (culture-based survey; 24 host species), and (C) endophyte richness (culture-free survey; 11 host species). Axis labels indicate the variance in endophyte response explained by each LV. Model description and output can be found in the section “Materials and Methods,” Supplementary Tables S5, S6. Labels: Leaf mass per area (LMA); leaf dry-matter content (LDMC); leaf punch strength (P. Strength); leaf thickness (Thick); percent leaf carbon content (%C); and percent leaf nitrogen content (%N).

Figure 2

Endophyte community composition was associated with leaf traits from the leaf economics spectrum (culture-free data set). Endophyte community variation within and between 11 host species in six genera (n = 29), from distance-based redundancy analyses (dbRDA) models constrained by leaf traits. Solid arrows represent significant associations (p < 0.05). Each point represents an endophyte community sampled from one host tree; colors represent host species; and symbols represent host genus (for a list of host species abbreviations, see Table 1).

Figure 3

The two most common fungal orders showed contrasting relationships with leaf functional traits. Associations between the relative abundance of fungal orders (A) Xylariales (Sordariomycetes) and (B) Botryosphaeriales (Dothideomycetes) to leaf mass per area (LMA, g m⁻²), and leaf punch strength (“Tough,” log N mm⁻¹), leaf thickness (Thick, log μm), leaf dry matter content (LDMC, g g⁻¹), leaf carbon (%C, log %), and leaf nitrogen (%N, log %). Grey bands represent the SEs of the regression lines. Analyses are from the culture-based survey (n = 51).