A comparative study of ancient environmental DNA to pollen and macrofossils from lake sediments reveals taxonomic overlap and additional plant taxa Mikkel Winther Pedersen a, * , Aurélien Ginolhac a , Ludovic Orlando a , Jesper Olsen b , Kenneth Andersen a , Jakob Holm a , Svend Funder a , Eske Willerslev a , Kurt H. Kjær a a Centre for GeoGenetics, The Natural History Museum of Denmark, Øster Voldgade 5-7, 1350 Copenhagen K, Denmark b Department of Physics and Astronomy, Aarhus University, Ny Munkegade 120, 8000 Aarhus, Denmark article info Article history: Received 16 October 2012 Received in revised form 30 May 2013 Accepted 12 June 2013 Available online 13 July 2013 Keywords: aDNA Lake sediments sedaDNA Metabarcode TrnL abstract We use 2nd generation sequencing technology on sedimentary ancient DNA (sedaDNA) from a lake in South Greenland to reconstruct the local floristic history around a low-arctic lake and compare the re- sults with those previously obtained from pollen and macrofossils in the same lake. Thirty-eight of thirty-nine samples from the core yielded putative DNA sequences. Using a multiple assignment strategy on the trnLgeh DNA barcode, consisting of two different phylogenetic and one sequence similarity assignment approaches, thirteen families of plants were identified, of which two (Scrophulariaceae and Asparagaceae) are absent from the pollen and macrofossil records. An age model for the sediment based on twelve radiocarbon dates establishes a chronology and shows that the lake record dates back to 10,650 cal yr BP. Our results suggest that sedaDNA analysis from lake sediments, although taxonomically less detailed than pollen and macrofossil analyses can be a complementary tool for establishing the composition of both terrestrial and aquatic local plant communities and a method for identifying additional taxa. Ó 2013 Elsevier Ltd. All rights reserved. 1. Introduction During the last few decades reliable methods for analysing sedimentary ancient DNA (sedaDNA) have widened our under- standing of past environments and the potential (Willerslev, 2003; Hofreiter et al., 2003; Willerslev et al., 2004a,b; Lydolph et al., 2005; Haile et al., 2009; Hebsgaard et al., 2009; Thomsen et al., 2009; Sønstebø et al., 2010; Andersen et al., 2011; Jørgensen et al., 2011, 2012; Taberlet et al., 2012; Parducci et al., 2012; Epp et al., 2012; Yoccoz et al., 2012). However, only a few studies have dealt with plant DNA in lacustrine sediments and none have made a direct temporal comparison to pollen and macrofossils (Matisoo- Smith et al., 2008; Anderson-Carpenter et al., 2011; Jørgensen et al., 2011; Parducci et al., 2012). Lacustrine sediments contain both autochthonous and allochthonous plant and minerogenic material washed in from within the drainage basin hereby creating a vertical stacked temporal archive. However, DNA is subjected to various types of degradation such as fragmentation, oxidation and hydrolyzation both in and outside the cell (Willerslev and Cooper, 2005). The rate of these decomposing factors is dependent on temperature, oxygen availability and microbial activity (Willerslev et al., 2004b; Hansen, 2006). Studies of DNA behaviour in both soil and waterlogged sediments suggest that this fragmented extracellular DNA is quickly adsorbed onto charged surfaces or absorbed into the crystal lattice of fine particles, amorphous crys- tals and particulate organic compounds (Poté et al., 2007, 2009). However once incorporated the DNA becomes partly protected reducing the rates of degradation (Poté et al., 2007). This study is founded on extraction of sedaDNA and PCR ampli- fication of short fragments of chloroplast DNA (cpDNA), followed by 2nd generation sequencing and identification against the reference database GenBank (http://www.ncbi.nlm.nih.gov/, visited 2nd February 2012). Ancient DNA analysis of environmental samples has been facilitated by the use of tagged primers (Binladen et al., 2007) followed by next generation high-throughput sequencing (pyrose- quencing) of amplicons e.g. (Haile et al., 2009; Parducci et al., 2012). Recent sedaDNA studies have shown inconsistent relationships be- tween pollen records and DNA results, suggesting that sedaDNA derives from mainly somatic tissues and not from pollen when using conventional extraction protocols (Jørgensen et al., 2011). * Corresponding author. Tel.: þ45 2927 5342. E-mail address: mwpedersen@snm.ku.dk (M.W. Pedersen). Contents lists available at SciVerse ScienceDirect Quaternary Science Reviews journal homepage: www.elsevier.com/locate/quascirev 0277-3791/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.quascirev.2013.06.006 Quaternary Science Reviews 75 (2013) 161e168