The Birth and Death of Toxins with Distinct Functions: A Case Study in the Sea Anemone Nematostella

Abstract

The cnidarian Nematostella vectensis has become an established lab model, providing unique opportunities for venom evolution research. The Nematostella venom system is multimodal: involving both nematocytes and ectodermal gland cells, which produce a toxin mixture whose composition changes throughout the life cycle. Additionally, their modes of interaction with predators and prey vary between eggs, larvae, and adults, which is likely shaped by the dynamics of the venom system. Nv1 is a major component of adult venom, with activity against arthropods (through specific inhibition of sodium channel inactivation) and fish. Nv1 is encoded by a cluster of at least 12 nearly identical genes that were proposed to be undergoing concerted evolution. Surprisingly, we found that Nematostella venom includes several Nv1 paralogs escaping a pattern of general concerted evolution, despite belonging to the Nv1-like family. Here, we show two of these new toxins, Nv4 and Nv5, are lethal for zebrafish larvae but harmless to arthropods, unlike Nv1. Furthermore, unlike Nv1, the newly identified toxins are expressed in early life stages. Using transgenesis and immunostaining, we demonstrate that Nv4 and Nv5 are localized to ectodermal gland cells in larvae. The evolution of Nv4 and Nv5 can be described either as neofunctionalization or as subfunctionalization. Additionally, the Nv1-like family includes several pseudogenes being an example of nonfunctionalization and venom evolution through birth-and-death mechanism. Our findings reveal the evolutionary history for a toxin radiation and point toward the ecological function of the novel toxins constituting a complex cnidarian venom.

The draft genome of Actinia tenebrosa reveals insights into toxin evolution

“…This is supported by the evidence of Nv1 copies being encoded by a cluster of at least 12 highly conserved sequences (Moran, Genikhovich, et al, 2012;Moran et al, 2008). Recent experimental evidence supports the adaptive evolution of these paralog escaping the process of concerted evolution (Sachkova et al, 2019). From our selection analyses, we could not infer that these highly homogenous Nv1 sequences are evolving under diversifying selection.…”

“…The GenBank accession numbers for the protein-coding gene used in this phylogenetic analysis are described in Sunagar and Moran (2015) and Sachkova et al (2019). This complete draft assembly is the first from any species of the superfamily Actinioidea.…”

“…Bootstrap values after 1,000 iterations are shown next to nodes, values under 70% not reported. The GenBank accession numbers for the protein-coding gene used in this phylogenetic analysis are described in Sunagar and Moran (2015) and Sachkova et al (2019). A corresponding bar plot is provided which shows the computed dN/dS value for orthologs and paralogs.…”

“…From our selection analyses, we could not infer that these highly homogenous Nv1 sequences are evolving under diversifying selection. While evidence supports that both Nv1 and Av2 are evolving through a process of gene conversion or unequal-crossing over consistent with concerted evolution (Moran et al, 2008), the escaped paralogs, however, may have resulted from lineage-specific duplications (Sachkova et al, 2019). Recent experimental evidence supports the adaptive evolution of these paralog escaping the process of concerted evolution (Sachkova et al, 2019).…”

“…This is further supported in the NaTx phylogeny we generated in this study, with Nv1 sequences clustering strongly together ( Figure 5). The Nv4 and Nv5 paralogs also exhibit divergent activity being highly toxic to fish, compared with Nv1 which has greater activity against arthropods (Sachkova et al, 2019). Divergently, the N. vectensis paralogs that escaped this homogenization are inferred to be evolving under diversifying selection, which consists of Nv3-8.…”