Mitochondrial genomes of Pleistocene megafauna retrieved from recent sediment layers of two Siberian lakes

Abstract

Ancient environmental DNA (aeDNA) from lake sediments has yielded remarkable insights for the reconstruction of past ecosystems, including suggestions of late survival of extinct species. However, translocation and lateral inflow of DNA in sediments can potentially distort the stratigraphic signal of the DNA. Using three different approaches on two short lake sediment cores of the Yamal peninsula, West Siberia, with ages spanning only the past hundreds of years, we detect DNA and identified mitochondrial genomes of multiple mammoth and woolly rhinoceros individuals-both species that have been extinct for thousands of years on the mainland. The occurrence of clearly identifiable aeDNA of extinct Pleistocene megafauna (e.g. >400 K reads in one core) throughout these two short subsurface cores, along with specificities of sedimentology and dating, confirm that processes acting on regional scales, such as extensive permafrost thawing, can influence the aeDNA record and should be accounted for in aeDNA paleoecology.

Keywords: genetics; genomics; mammal; mammoth; woolly rhinoceros.

Figure 1.. aDNA of Mammuthus in recent lake sediments.

(A) Read counts assigned to Mammuthus (square-root-transformed proportion of the respective number of raw reads per library) after hybridization capture enrichment of aeDNA of core LK-001 (shown are results of 22 libraries; one library was excluded as it did not produce any reads assigned to mammals); square-root transformation of percentage. Indicated are sample depths (in cm; 1.5–80 cm) and approximate ages as per ²¹⁰Pb chronology (Appendix 1—table 7; to a maximum depth of 39.5 cm). The solid line indicates the general trend. Across the 22 libraries: (B) Fragment length distribution and (C) damage patterns (red indicates C-to-T transitions, blue G-to-A transitions. the Y-axis indicates the percentage of positions with a nucleotide change, the X-axis indicates the position along the fragment).

Figure 2.. Locations of the sediment cores of the present study (Yamal peninsula, Siberia) and previously retrieved mammoth remains and their haplo(sub)groups (Appendix 1—table 6).

The bar chart indicates a maximum-likelihood estimate of the haplogroup proportions derived from the reads from the three sediment core libraries with the most mammoth reads.

Appendix 1—figure 1.. Comparison of transformed metabarcoding read counts (A) and ddPCR concentration estimations (B) of Mammuthus primigenius DNA in sediment cores LK-001 and LK-007 from the Yamal peninsula.