Dead or Alive? Challenges in Discriminating Dietary From Host‐Associated Community via RNA and DNA Metabarcoding in a Filter Feeder

ABSTRACT Environmental DNA metabarcoding has found application in a range of research areas by simplifying the collection of high‐quality field data in a cost‐efficient way. In biodiversity studies, environmental DNA metabarcoding of filter feeders like mussels and sponges has recently gained attention. Due to their continuous filtering activity, they accumulate a high amount of genetic signatures from their environment in their tissue. However, it can be difficult to separate genetic signatures of ingested dietary taxa from endobiotic ones originating from parasites or commensals living within the host. This issue parallels a broader problem in environmental DNA‐based biodiversity studies: the inability to differentiate between DNA derived from living and dead organisms. A recent attempt to address this problem is using environmental RNA metabarcoding, which is believed to specifically represent the live and active community of an ecosystem. Therefore, we tested whether endobionts, as metabolically active organisms, can be distinguished from dietary taxa, as the presumed dead community, in a parallel RNA and DNA metabarcoding approach. We targeted nuclear 18S rDNA and rRNA to amplify mussel‐associated communities (dietary and endobiotic taxa) in samples of Dreissena polymorpha and Mytilus edulis . Our results do not reveal any presence/absence or abundance pattern in the RNA and DNA library that could be used to distinguish dietary from endobiotic signatures. However, we found that over 40% of all genetic signatures were detected by RNA only and that those accounted for < 4% of the total reads. This study thus demonstrates the outstanding sensitivity of RNA metabarcoding in comparison to DNA metabarcoding and suggests that using (environmental) RNA may be a way of capturing a larger proportion of the biodiversity in a given ecosystem.