Initial Virome Characterization of the Common Cnidarian Lab Model Nematostella vectensis

Abstract

The role of viruses in forming a stable holobiont has been the subject of extensive research in recent years. However, many emerging model organisms still lack any data on the composition of the associated viral communities. Here, we re-analyzed seven publicly available transcriptome datasets of the starlet sea anemone Nematostella vectensis, the most commonly used anthozoan lab model, and searched for viral sequences. We applied a straightforward, yet powerful approach of de novo assembly followed by homology-based virus identification and a multi-step, thorough taxonomic validation. The comparison of different lab populations of N. vectensis revealed the existence of the core virome composed of 21 viral sequences, present in all adult datasets. Unexpectedly, we observed an almost complete lack of viruses in the samples from the early developmental stages, which together with the identification of the viruses shared with the major source of the food in the lab, the brine shrimp Artemia salina, shed new light on the course of viral species acquisition in N. vectensis. Our study provides an initial, yet comprehensive insight into N. vectensis virome and sets the first foundation for the functional studies of viruses and antiviral systems in this lab model cnidarian.

Keywords: Cnidaria; Nematostella vectensis; RNA-seq; viruses.

Figure 1

Taxonomic classification of N. vectensis viral sequences. Distribution of viral families (a) and groups (b) in the sequence assembly from all merged datasets and (c) relative abundance of viral families within each studied dataset. Sequences representing fragments of the same viral species were collapsed into one entry.

Figure 2

Relative abundance of viral sequences in each analyzed dataset presented as log₂ of transcript per thousand (TPM/1000). Sequences representing fragments of the same viral species were collapsed into one entry.

Figure 3

Venn diagram of core viral sequences for all non-embryonic datasets [31,33,34,35,36] revealed by remapping filtered reads to the viral contigs assembled from the merged dataset. For simplicity, only numbers of contigs common to all datasets and specific to each dataset are shown. Values in brackets indicate the total number of different viral contigs to which reads from each dataset were successfully remapped.

Figure 4

The number of reads mapped to Nematostella viral sequences assembled from all merged datasets presented as Reads Per Million (RPM) (a) and the fraction of total reads which mapped to A. salina viruses (b).