Monday, May 25, 2015


Do you want to learn about another interesting application of metabarcoding? If so, keep reading:

Paleoenvironmental studies are essential to understand biodiversity changes over long timescales and to assess the relative importance of anthropogenic and environmental factors. Sedimentary ancient DNA (sedaDNA) is an emerging tool in the field of paleoecology and has proven to be a complementary approach to the use of pollen and macroremains for investigating past community changes.

The idea is that organismal DNA can be directly obtained from recent and past sediments even in the absence of visible fossils. The development of metabarcoding methods now allows massive sequencing of ancient DNA contained in such sedimentary archives. Reconstructions of ancient environments often rely on indicator taxa which requires expert knowledge, but quantitative ecological analyses through direct sequencing of environmental DNA might provide more objective information. 

A new study by a group of French colleagues reports on the use of sedaDNA to investigate plant community changes in the catchment of Lake Anterne in the northern French Alps (2063m above sea level) over the last 6400 years. 

Multivariate analysis of ecological communities is a powerful tool that allows an objective understanding of the processes determining environmental variation (Legendre & Legendre 2012), but they have seldom been applied to sedaDNA data. In this study, we combined these two approaches, using metabarcoding data from a subalpine lake, along with data on present plant species assemblages. Our purpose was to infer past vegetation dynamics in the catchment during the Holocene, and to assess the relative importance of human impacts, natural evolution of communities and variation in climatic conditions in the observed trajectories.

The study provided some in-depth insight into the recent history of plant communities and the extend of human impact. The authors conclude that subalpine pine forests and tall-herb communities were the dominant vegetation in the lake catchment before human arrival. An indicator for a this is a change to taxa from open environments, typically subalpine meadows which is the result of grazing by cattle or sheep. Indeed DNA analyses have proven the presence of cattle and/or sheep herds in the Lake Anterne catchment during the late Iron Age and the Roman period, as well as during the Middle Ages and the Modern Times.

The sedaDNA-based approach currently has some limitations but offers, in addition to a relatively fast and easy protocol, some interesting perspectives. In this study, only sequences with a perfect assignment score (100%) were considered; consequently, data were highly dependent on the completeness of the reference database. In addition, the taxonomic resolution of the trnL P6 loop does not always allow identification to the species level, with barcodes often shared by several species. However, this fragment represents only a very limited part of the chloroplast DNA, less than one-one-thousandth of the whole molecule. The fact that such a small part of the chloroplast DNA can be reliably amplified demonstrates that a very large number of chloroplast DNA fragments are present in the DNA extract, and thus in the core slice. The next challenge will be to more efficiently extract the taxonomic information from those DNA fragments contained in the core. One possibility might be to directly sequence the DNA extracts using a shotgun approach (Taberlet et al. 2012c; Zhou et al. 2013). An ongoing project exists to sequence the whole chloroplast genome for approximately 4500 species of alpine flora. Shotgun sequencing combined with the availability of an exhaustive reference database containing the entire chloroplast DNA molecules creates the potential for a much more precise reconstruction of past plant communities, with identifications to the species level, and opens unprecedented avenues in paleoecology.

1 comment:

  1. I'm reading a 2009 paper by Willerslev et al on sedaDNA
    from mammoths and horses in Alaska. It occurs to me that
    perhaps human sedaDNA might also exist in the same layers
    if humans were abundant. Is this possible?