Open Access

Anjuli Calchera

• In the light of emerging resistances against common drugs, new drug leads are required. In the past natural sources have been more yielding in this respect than synthetic strategies. Fungi synthesize many natural products with biological activities and pharmacological relevance. However, only a fraction of the estimated fungal diversity has been evaluated for biological activity, and much of the Fungi’s natural chemical diversity awaits discovery. Especially promising in this context are lichenized fungi. Lichens are well known for their particularly rich and characteristic secondary chemistry which allows them to withstand intense UV radiation, protects them against herbivory, and prevents them from being overgrown. The slow growth rates of lichens and difficulties and infeasibility of large scale cultivations in the laboratory render lichens inaccessible for applied purposes. These experimental challenges have led to a poor understanding of the molecular mechanisms underlying the biosynthesis of characteristic lichen secondary metabolites. The recent development of improved sequencing techniques has enabled new strategies to address multi-species assemblages directly through metagenome sequencing and survey their biosynthetic potential through genome mining. However, whole genome sequencing of entire lichen thalli to metagenomically assess the lichen-forming fungus without the need of cultivation has not been evaluated for lichens before. This approach will enable the reconstruction of fungal genomes from mixed DNA from lichen thalli and allow the exploration of biosynthetic gene content. My thesis was conducted in two parts: a methodological evaluation of a metagenomic strategy to reconstruct genomes and gene sets of lichen-forming fungi, and the exploration of biosynthetic gene content with the help of comparative genomics and phylogenetics. For the first part, I evaluated the quality of metagenome-derived genome assemblies and gene sets by direct comparison to culture-derived reference assemblies and gene sets of the same species. I showed that metagenome-derived fungal assemblies are comparable to culture-derived references genomes and have a similar total genome size and fungal genome completeness. The quality of assemblies was affected strongly by the choice of assembler, but not by the method of taxonomic assignment or inference of non-mycobiont DNA sequences. The fungal gene space is well covered in metagenome-derived and culture-derived fungal gene sets and overlaps to 88-90 %. Finally, the metagenome-derived assemblies reliably recover gene families of secondary metabolism. This shows the suitability of metagenomically derived genomes for mining biosynthetic genes, and potentially also other gene families. Overall, the method validation showed a high similarity between metagenome- and culture-derived genome assemblies. For the second part of my thesis, I explored the biosynthetic gene content in two different systems: Between two sister-species with different ecological requirements but similar chemical profile, and between two species which are metabolite-rich and economically relevant in the perfume industry. I compared the diversity of biosynthetic gene clusters between the species and in the broader context of other lichenized and non-lichenized fungi. Overall, the whole genome mining revealed a large number of uncharacterised secondary metabolite gene clusters in fifteen genomes of lichen-forming fungi compared to other fungal classes. Their number highly outweighs the number of known synthesized metabolites and highlights the hidden biosynthetic potential in lichen-forming fungi. Many biosynthetic gene clusters in the ecological distinct sister-species showed a high homology in accordance with the high synteny in gene content and order in both genomes. These clusters represent ideal candidates for secondary metabolites synthesized by both species, while the remaining clusters may encode for metabolites relevant for the different ecological requirements of both species. The metabolite-rich species used in the perfume industry showed a particularly high number of biosynthetic gene clusters. An in-depth characterization of architecture and gene content of homologous gene clusters together with hints from phylogenetic relatedness to functional characterized metabolites provides promising insights into the biosynthetic gene content of these lichen-forming fungi. In conclusion, I showed that metagenome sequencing of natural lichen thalli is a feasible approach to reconstruct the fungal mycobiont genome of lichens and circumvent time-consuming and in some cases impossible cultivation of individuals. The genome mining for secondary metabolite gene clusters in lichen-forming fungi revealed a high biosynthetic potential for the discovery of new natural products. One of the focal species, Evernia prunastri, contained the highest ever reported number (80) of biosynthetic clusters in lichenized fungi. The comprehensive cluster characterizations through annotation, comparative mapping and phylogenetics provide first valuable hints for linking metabolites to genes in these lichen-forming fungi. My results pave the way for biotechnological strategies to unlock the vast richness of natural products from lichens for applied purposes.