A review of dietary DNA metabarcoding in marine vertebrates: a new frontier in sea turtle foraging ecology?

22 pages, 2 tables, 5 figures

Diet characterisation is important for understanding trophic roles of animals across space and time, including in response to climate change. This has led to the development of a large range of dietary analysis techniques, from centuries-old morphological stomach analysis to recent molecular techniques. Given the difficulties and limitations of direct analysis in marine animals, here we review DNA-based methods of marine vertebrate diet analysis, examining the proliferation of studies over the last two decades. We identify a keystone taxon, sea turtles, where DNA-based approaches have had limited use, but offer great potential for characterising diet across species, life stages and regions. We show that contemporary molecular techniques can overcome some limitations of traditional methods based on morphological identification, such as the ability to identify rapidly digested food items. We report on the development of DNA metabarcoding protocols that enable simultaneous identification of many diet item sequences from heterogeneous samples. DNA metabarcoding can increase taxonomic resolution, improve the identification of certain items (e.g., gelatinous organisms), and increase the comprehensiveness of diet characterisation, particularly in combination with other techniques. However, careful methodological development and finer optimisation of metabarcoding protocols (e.g., appropriate primer selection, blocking of host DNA amplification) are necessary to improve results. Combination approaches to sea turtle dietary analysis and further experimentation with metabarcoding methodology will help to characterise variations and effectively monitor shifts in diet composition in response to environmental changes such as rising sea temperatures and displacement to alternative foraging grounds

This work was funded by the Bertarelli Foundation as part of the Bertarelli Programme in Marine Science (project 820633). The work was partially supported by a Swansea University Faculty of Science and Engineering scholarship

Peer reviewed