97th ESA Annual Meeting (August 5 -- 10, 2012)

COS 41-5 - Contrasting fungal communities in asymptomatic vs. symptomatic forest seedlings within the context of Janzen - Connell hypothesis

Tuesday, August 7, 2012: 9:20 AM
E146, Oregon Convention Center
Maria Soledad Benitez1, Michelle H. Hersh2, Brantlee Spakes Richter3, Rytas Vilgalys4 and James Clark1, (1)Nicholas School of the Environment, Duke University, Durham, NC, (2)Department of Biology, Eastern Michigan University, Ypsilanti, MI, (3)Plant Pathology, University of Florida, Gainesville, FL, (4)Department of Biology, Duke University, Durham, NC
Background/Question/Methods

Fungal pathogens can impact plant community composition through negative density dependence. The Janzen-Connell hypothesis (JC) assumes that host-specific natural enemies accumulate when and where their tree hosts are abundant, thus tending to promote diversity. In Duke Forest (NC), for instance, germination of Liquidambar styraciflua increases with distance from adult trees under high light conditions. Quantifying the effects of pathogens requires identification of the pathogens that can infect each host species and their impacts singly and in combination.  The objective of this work is to identify potential seedling pathogens through the comparison of fungal communities associated with symptomatic and asymptomatic seedlings.  Symptomatic and asymptomatic seedlings from seven plant species were sampled from hardwood stands at the Duke Forest (NC) on three years. Surface sterilized plant tissue was used for fungal pure culture isolation. Taxonomic identification of pure cultures was performed based on sequence information of the internal transcribed spacer region (ITS). We hypothesized that the structure of fungal communities from symptomatic versus asymptomatic seedlings differ and that the most abundant fungal groups present in symptomatic seedlings represent potential pathogen groups

Results/Conclusions

Over 1300 pure cultures with sequence information from the ITS region were obtained from approximately 320 seedlings studied. Based on 97% similarity at the ITS region, the isolates were classified into 316 fungal taxa. The most abundant isolates from symptomatic seedlings often represent known plant pathogenic groups and were isolated from at least five of the seven host species studied with variation in the frequency of isolation per host. Potential pathogens were also isolated from asymptomatic seedlings, but at lower abundances. The most abundant isolates from asymptomatic seedlings most closely matched environmental samples from GenBank with no clear taxonomic identification. In addition, for most individual seedlings studied, the fungal community appears to be dominated by one or two taxa. At the isolate level, no host-specificity was observed; however, preliminary results from culture-independent analyses reveal community structure separation according to host species. The implications of host specificity of fungal pathogens in forest seedlings are discussed within the context of plant community dynamics.