Detection of terrestrial organic matter using ancient DNA from marine sediments by trnL metabarcoding

Sedimentary ancient DNA (sedaDNA) extracted from sediment cores is more frequently used to reconstruct past ecosystems and their biodiversity. When reconstructing past vegetation components, cores are most often taken within terrestrial and lake systems. Research on marine environments looking at the terrestrial input has so far been frequently based on biogeochemical analyses. While studies looking into marine eukaryotes, especially diatoms, have been frequently conducted, it is unknown whether or not terrestrial plant sedaDNA from marine sediments can be used to reconstruct past ecosystems and the prevailing conditions characterizing the surrounding areas as successfully. Here, sedaDNA was extracted from 18 samples taken from the sediment core SO202-18-6 from the Bering Slope within the Bering Sea, representing the past drainage basin of the Yukon River. The selected samples were dated based on previously published age models and cover the Late Deglaciation to the early Holocene. Via metabarcoding, the short P6 loop of the chloroplast trnL (UAA) intron, a barcode unique in terrestrial plant DNA, was amplified and sequenced using HTS. Throughout the sediment record, the Yukon River vegetation was dominated by Salicaceae. Rosaceae contributed a further significant part to the identified vegetation. While their trends throughout time are consistent, an exception is found within the colder Younger Dryas, in which Rosaceae increases at the expense of Salicaceae. While further taxa are recorded at low abundances, a stable record is found to have characterized the Yukon River vegetation within the Late Deglaciation based on the two dominating taxa, indicating ecological stability, even though major climatic shifts took place since the LGM. Thus, this study provides a first look into using sedaDNA as a tool for reconstructing past terrestrial ecosystems derived from a marine sediment core and based on a riverine input source.