Assessment of environmental DNA as a biomonitoring tool for vulnerable deep-sea habitats

High-throughput sequencing of targeted “barcode” loci in DNA extracted from marine environmental samples such as water and sediments has recently exploded in popularity due to the ability to generate a taxonomic community profile that usually surpasses what is obtainable from using traditional monitoring methods in terms of biodiversity and detection of rare taxa. Indeed, given the existing poor biodiversity inventory of many marine habitats coupled with their high cost and difficulty of access, environmental DNA (eDNA) biomonitoring of these remote ecosystems is attractive if reliable taxonomic inventories are obtained for a modest price and with little required sample material. As part of the DEEP SEARCH (DEEP Sea Exploration to Advance Research on Coral/Canyon/Cold seep Habitats) project, we collected water samples from eight JASON2 remotely operated vehicle (ROV) dives at sites within canyon, cold-seep, and cold-water coral reef habitats along the US Atlantic coast in April 2019. The use of an ROV allowed controlled water sample collection near features of interest. Duplicate one-liter seawater samples filtered through a 0.2 µm Sterivex and ten liters of seawater filtered through a 0.8 µm cellulose nitrate filter were collected from each Niskin. Metabarcoding of microbial 16S rRNA, metazoan 18S rRNA, and mitochondrial 12S and cox1 sequences from eDNA extracted from water samples were performed on an Illumina MiSeq. The 1L and 10L samples recovered largely the same communities. Differentiation among habitats in both microbial and metazoan community structure evidenced through multiple ordination and comparative analyses was clear. The strongest differentiation occurred between the shallow ( about 200 meters) and deep (about 2,000 meters) cold-seep habitats. The implications of our findings for biomonitoring in the deep sea, as well as plans for expanding our analyses to include other barcode loci, water column eDNA samples collected via CTD, and reference database augmentation through genome skimming, will be discussed during this seminar.