Artsprosjekt Endofyttisk sopp i trær 23-19

Trees harbor an enormous diversity of fungi spending at least part of their life-cycles as symptomless and secretive inhabitants. These, the so-called endophytic fungi, often form very species rich communities, but most species remain unidentified or undescribed. The primary objective of the project was to map fungal endophyte diversity associated with living sapwood and twigs of beech, linden, maple, and oak trees in Norway. A total of 193 trees were drilled and sapwood samples were taken for metabarcoding. Since identifying environmental sequences of Ascomycota involves many challenges and pitfalls (e.g., most species globally remain to be named and described, many described species do not have any available reference sequences, a relatively high percentage of reference sequences are known to be mis-labelled as well as numerous taxonomic issues), great care is needed when interpreting lists of species names generated by environmental sequencing and run against databases. By using 99.5% similarity with a known reliable sequence in UNITE and/or Genbank, by only considering sequences of over >250 base pairs in length, and checking original publications, we minimized the risk of misidentification. Sequencing of wood samples resulted in 105000 sequences and 1837 fungal species hypotheses. By applying our strict criteria for sequence data, we identified 60 Ascomycota species new to Norway. While relaxing the criteria for “safe identification” somewhat, more species could have been recorded, but we suggest that our criteria are reasonable. For the rest of the species hypotheses, the majority may represent so called “dark taxa”. These are equally scientifically important but how to handle them remains a great challenge. Knowledge of morphology and taxonomy are indispensable for identification, and it was our hope that complementing environmental sequencing with sampling of ascomata on trunks and twigs, and pure culturing from the drill samples, would increase the chances of reliable identification of fungal species. However, as the latter methods often failed to detect as many and the same species as the sequencing, the value of these complimentary methods was limited. Nevertheless, in a minority of cases we were able to confirm the identity by studies of anamorph and/or teleomorph morphology. Sampling of ascomata from trunks and twigs of the drilled trees yielded two additional species new to Norway including one undescribed species but showed little overlap with sequence data. Sampling from additional trees yielded two undescribed genera and several undescribed species (descriptions in progress). To generate data valuable for management strategies and the conservation of biodiversity, we compared species composition and species richness of our entire sequence dataset among the four studied tree species, trees of different sizes and among trees in three different regions of S Norway. Controlling for sample size, oak had significantly higher species density than Norway maple, beech, and linden. The effect of tree size on species density was small and not statistically significant and there was no effect of region. Regarding species composition, oak was significantly separate from the other three tree species, which all had largely over-lapping species communities. Tree size had a significant overall effect on species composition and there were significant interactions among tree species and the size of the trees. There was no significant difference in composition among the three regions. Our results emphasize the unique value of oak and large temperate deciduous trees for biodiversity and show that metabarcoding techniques can be very relevant for conservation-oriented studies of for instance the endophytic fungal communities in old trees. As such they can also generate a lot of new specific information about species new to Norway, distribution and ecology. However, only for a minority of species the identity can be decided with certainty, and complimentary morphology-based methods proved to be of limited value in this respect. For the future, Ascomycota biodiversity mapping projects based on ascomata collection, and were relevant, single spore isolation, pure culturing, morphological studies of the fungal holomorph, and multi-marker sequencing are needed. In addition, metabarcoding datasets can be useful for biodiversity mapping purposes if studied by taxonomists.