DNA-Barcoding – Pflanzen im wald

To the non-expert, forests may seem monotonous aggregations of ever the same botanical species; while in fact, they harbor much diversity in the ground flora composition. Plant communities of forests can be grouped by their tree species composition, but a refined phytosociological classification, reflecting soils, site type and lastly also productivity, is based on characteristic and differential species mostly of the herb and shrub layer (e.g., Willner & Grabherr 2007, Staudinger & Willner 2014). Nevertheless, the identification of the ground vegetation species in forests often is difficult, because under a shady canopy only vegetative parts are found and flowers or fruits are missing. The current estimate of spermatophyta naturally present in Austria’s forest is ca. 1150 (41 % of all spermatophyta), and similarly for ferns and bryophytes (42 %). Five of 51 tree species are considered “endangered”, six more are “vulnerable”. For the rest of the spermatophyta (in forests), three are “extinct in the wild”, 22 “critically endangered”, 86 are “endangered”, and 159 “vulnerable” (Niklfeld 1986, 1999). The Austrian Federal Research Centre for Forests (BFW) is currently trying to initiate a DNA barcoding project for “forest plants”. If funded, the available “research infrastructure” of BFW can be utilized; the Austrian forest inventory is permanently sending small groups to a network of forest sampling points, for measurements and observations; sampling plants for barcoding would be rather easy for them. A herbarium and DNA laboratories are also available. We are offering DNA-based identification of plant species as a paid-for service to the public. Chloroplast DNA sequencing is one of the methods for this service. A pilot trial for barcoding of forest plants was initiated in Northern Sardinia (Italy), in pubescent oak (Quercus pubescencs s.l.) forests, which cover sites from the mountain slopes (approx. 1500 m a.s.l) down to coastal valleys, corresponding to different vegetation types (Bacchetta et al. 2004). The idea is to characterize these forests according to any variation found in chloroplast DNA of the ground vegetation. Up to 20 recurring species (multiple samples per species if possible) were collected from each of 20 sites. Frequencies of the species are being analyzed for use as a classification tool. The assessment is clearly hampered by human-induced changes (agriculture, pasture, wood extraction, settlements). In a first next generation sequencing experiment, 384 specimens of eight species were included. Up to eight successfully amplified DNA fragments were mixed, tagged and sequenced in bulk on an Illumina MiSeq v3 machine. The resulting approx. 17 million single sequences are currently being analyzed. Average read length is 81 basepairs (after quality filtering), and 25,000 to 50,000 reads were obtained for each DNA sample. A first check seems to confirm the presence of the desired sequences, though the known issues of this technique, e.g. with mononucleotide repeat tracts, are there as well. Based on these circumstances and experiences, the perspectives for success will be discussed in this presentation.