A global perspective on the fungal community structure in green coffee beans

Background/Question/Methods

Discerning whether or not microorganisms have geographic dispersal limitation remains at the fore-front of microbial ecology. Studying the geographic distributions of fungi associated with highly-mobile agricultural crops provide the opportunity to understand how microbial communities assemble, and whether “everything is everywhere.” We are developing coffee as a model system to understand how global food trade facilitates gene flow in fungi, and how different farming practices and commodity processing methods structures fungal communities. Fungal communities within green coffee beans from six different countries were assessed using culture-based and culture-independent methods. For each country, comparisons were also made between conventional and USDA-certified organic coffee. Culture-based sampling methods involved plating 200 surface-sterilized beans per lot on two media types, MEA and DG18. Fungi were isolated into pure culture within the first two weeks of growth. Genomic DNA was extracted from each culture and the ITS region was amplified using fungal-specific primers. Sequence data was used to identify recovered species by BLAST querying GenBank.  Culture-independent sampling methods involved drying and grinding 20 surface-sterilized coffee beans per lot, extracting total genomic DNA, and amplifying with fungal-specific primers for ITS. PCR products were cloned into a plasmid vector and 24 inserts were sequenced. The software package MOTHUR was used to analyze community similarity and diversity.

Results/Conclusions

Incidence rate (percentage of total bean infection per lot) varied significantly between coffee growing regions and between individual farms within African and Central American countries. Dry-processed coffee had greater incidence than wet-processed coffee, and dry-processed samples were dominated by Aspergilus niger and Aspergillus tubingensis. Incidence rates did not vary between conventional and organic samples, although conventional coffee was host to a greater diversity of fungi than organic coffee. Culture-based and culture-independent methods of sampling gave distinct views of community profiles with limited overlap of species. A. niger, A. tubingensis, Eurotium rubrum, Wickerhamomyces anomalus, and A. fumigatus were the most abundant species in cultured samples, while W.anomalus, Eurotium amstelodami, Pichia pijperi, Cladosporium sp., and Dipodascaceae sp. were most abundant in culture-independent samples. From these results we were able to detect differentiation of fungal communities associated with coffee plants from different geographic locations. Fungal communities in coffee appeared to be structured based on processing methods more so than production methods. Using both culture-based and culture-independent methods to sample fungi in coffee beans allows for a more thorough investigation of the diversity and abundance of fungi associated with global coffee production.