Seedborne fungal communities in a fragmented landscape depend on host identity, fragment size, and temperature variation

Background/Question/Methods

Habitat fragmentation has the potential to alter soil fungal communities, and consequently, how soilborne fungi affect plant community structure. Smaller, more isolated fragments typically experience lower plant species richness, altered plant community composition, and more extreme environmental conditions. These changes may mediate subsequent shifts in fungal community structure and perhaps, function. This is particularly relevant in light of how fungal pathogens help maintain plant diversity through negative feedbacks. In this research, we explore how fungal communities on dormant seeds are altered by fragmentation. In 2013, we buried seeds of three herbaceous plant species – Andropogon gerardii, Chamaecrista fasciculata, and Rudbeckia hirta -- in small and large fragments in the Kansas Fragmentation Study (KFS), a large-scale experimental fragmentation study in Lawrence, Kansas. During seed burial, we measured temperature variation in burial sites using iButton dataloggers. Seeds were unearthed the following year, and fungi were cultured from seeds and identified using DNA barcoding. Aboveground vegetation at burial sites was censused in 2014 by estimating percent cover for each plant species. We assessed changes in seedborne fungal community structure in response to seed host, fragment size, aboveground vegetation, and environmental variables using non-metric multidimensional scaling (NMDS) and permutational multivariate analysis of variance (PERMANOVA).

Results/Conclusions

We successfully isolated 579 fungal cultures which we identified using DNA barcoding. Comparing seedborne fungi isolated from small patches, the interiors of large patches, and the edges of large patches, we found differences in fungal community composition by landscape location (large patch interiors, large patch edges, and small patches; F_2,45=2.43, p<0.001; PERMANOVA). This indicates that habitat fragmentation has measurable effects on fungal community structure. Further, we found that fungal community composition varied according to host species (PERMANOVA; F_2,44=3.52, p<0.0001). This demonstrates that seedborne fungi show host preferences—a necessary precondition for negative feedbacks that maintain plant diversity. We explored the influence of several environmental variables typically associated with patch size to evaluate the mechanisms through which fragmentation impacts fungal community structure. Two related environmental variables – maximum temperature and temperature range – were correlated with fungal community structure when tested individually (Maximum temperature: R²= 0.1368, p=0.046; Temperature range: R²= 0.1341, p=0.05). However, neither aboveground plant species richness (R²= 0.0279, p=0.524) nor aboveground plant diversity (R²= 0.0151, p=0.702) were correlated with fungal community structure. These analyses highlight the combined role of environmental variability, host identity, and fragment size on influencing the structure of fungal communities inhabiting buried seeds.