Despite the ecological importance and pervasiveness of fungal endophytes, little is known concerning the assembly of their communities. We hypothesized that fungal assembly is driven by dispersal because, in deciduous plants, endophytes must colonize new leaves every year. We also hypothesized that deterministic processes affect assembly, as endophytes must have finite tolerance ranges for climate variables, for example.
We tested the influence of both neutral and deterministic processes with respect to leaf fungal endophytes by characterizing fungal communities within Gambel oak (Quercus gambelii) on north- and south-facing aspects within four canyons along a 15 km gradient on the Wasatch front, Utah. North- and south-facing aspects within each canyon were approximately 200 meters apart, though with dramatically different climates, allowing us to assess the effect of climate independent of dispersal limitation. We used aspect as a proxy for climate. On the small geographic scale of this survey, climate variation between canyons was not correlated with geographic distance. Thus, by comparing fungal communities between canyons, we additionally assessed the influence of dispersal on these communities independent of climate.
Results/Conclusions
Analysis of fungal communities within a single canyon showed aspect was a significant factor in determining community structure, explaining between 8% and 18% of variation between communities. As aspect is a proxy for climatic differences, this suggests a significant but limited impact of deterministic processes in controlling endophyte communities.
A multifactorial analysis of fungal communities from all canyons, including canyon, aspect, and their interaction as factors showed canyon and the interaction term explained 10% and 8% of variation between communities respectively, while aspect explained only 2% of community variation. These results suggest that, while climate was an important factor within a canyon, its impact was dependent on canyon, suggesting a potential dispersal limitation. This is further supported by Mantel tests that indicated geographic distance between communities was significantly and positively correlated with community dissimilarity, explaining between 3% and 8% of community variation. Thus, our results suggest both neutral and deterministic processes determine leaf fungal endophyte community assembly.