Long-term fungus-plant co-variation from multi-site sedimentary ancient DNA metabarcoding in Siberia

Abstract

Climate change has a major impact on arctic and boreal terrestrial ecosystems as warming leads to northward treeline shifts, inducing consequences for heterotrophic organisms associated with the plant taxa. To unravel ecological dependencies, we address how long-term climatic changes have shaped the palaeo-ecosystems at selected sites in Siberia. We investigated sedimentary ancient DNA from five lakes spanning the last 47,000 years, using the ITS1 marker for fungi and the chloroplast P6 loop marker for vegetation metabarcoding. After bioinformatic processing with the OBItools pipeline, we obtained 706 unique fungal operational taxonomic units (OTUs) and 243 amplicon sequence variants (ASVs) for the plants. We show higher OTU numbers in dry forest tundra as well as boreal forests compared to wet southern tundra. The most abundant fungal taxa in our dataset are Pseudeurotiaceae, Mortierella, Sordariomyceta, Exophiala, Oidiodendron, Protoventuria, Candida vartiovaarae, Pseudeurotium, Gryganskiella fimbricystis, and Trichosporiella cerebriformis. The overall fungal composition is explained by the plant composition as revealed by redundancy analysis. The fungal functional groups show antagonistic relationships in their climate susceptibility. The advance of woody taxa in response to past warming led to an increase in the abundance of mycorrhizae, lichens, and parasites, while yeast and saprotroph distribution declined. We also show co-occurrences between Salicaceae, Larix, and Alnus and their associated pathogens and detect higher mycorrhizal fungus diversity with the presence of Pinaceae. Under future warming, we can expect feedback between fungus compositional and plant diversity changes which will affect forest advance and stability in arctic regions.

Regional opportunities for tundra conservation in the next 1000 years

“…Permafrost degradation/abrupt thaw and warming ( Biskaborn et al, 2019 ; Stuenzi et al, 2021 ) could lead to thermokarst processes that would locally cause swamping/waterlogging, leading to suppressed establishment/growth ( Rees et al, 2020 ), which is not explicitly simulated and unlikely to limit tree invasion on a broader scale. Further limitations could arise from the lack of a suitable mycorrhiza symbiosis partner, but as shown by Hippel et al, 2021 even in northernmost areas on the Taimyr Peninsula mycorrhiza are present even before the invasion of trees in the Holocene. Additionally, dispersal rates of mycorrhiza with small spores are a magnitude faster than for taxa with larger seeds so that geographic distance is unlikely to be a limiting factor, similar to diatoms as shown by Stoof-Leichsenring et al, 2015 .…”

Evaluation of lake sedimentary ancient DNA metabarcoding to assess fungal biodiversity in Arctic paleoecosystems

“…OTUs assigned to the genus Mortierella were among the 15 most abundant genera in each core, and these fungi typically occur as saprotrophs in soil, on decaying leaves, and other organic material (Domsch et al, 2007). Among mycorrhizal fungi, the genus Inocybe (Deacon et al, 1983) dominated some of the samples of lake CH12 and also occurred sporadically in samples of other lakes (von Hippel et al, 2021).…”

“…Here, we developed metabarcoding primers to investigate past fungal biodiversity, and we assessed the specificities of fungal sedimentary ancient DNA (sedaDNA) extracted from lake sediment cores in Siberia (for details on the cores see von Hippel et al, 2021).…”

“…Furthermore, the increase in available sequences potentially also offers the possibility to optimize metabarcoding assays in comparison to the primers that are currently in common use. However, fungal metabarcoding primers have not been re-evaluated in close to a decade (Bellemain et al, 2010(Bellemain et al, , 2013Epp et al, 2012), even though fungal metabarcoding may help reveal community turnovers and trace ecosystem changes on very long timescales (von Hippel et al, 2021). This is particularly critical regarding the high diversity of fungi and the resulting lack of truly universal but exclusive fungi metabarcoding primers (see UNITE primer notes, https://unite.ut.ee/prime rs.php).…”

“…DNA was isolated from approximately 2-5 g sediment using the PowerMax Soil DNA Isolation kit (Qiagen), and DNA extracts were purified and normalized to 3 ng/μl using a GeneJET Genomic DNA Purification Kit (Thermo Fisher Scientific); for details, see von Hippel et al (2021). Seventy samples were used (CH12: 28 samples; Lake Lama: 15 samples; Lake Levinson-Lessing: 9 samples; Lake Kyutuynda: 10 samples; Lake Bolshoye Toko: 8 samples; Table S2).…”