Restriction and recruitment - gene duplication and the origin and evolution of snake venom toxins

Abstract

The genetic and genomic mechanisms underlying evolutionary innovations are of fundamental importance to our understanding of animal evolution. Snake venom represents one such innovation and has been hypothesised to have originated and diversified via a process that involves duplication of genes encoding body proteins and subsequent recruitment of the copy to the venom gland where natural selection can act to develop or increase toxicity. However, gene duplication is known to be a rare event in vertebrate genomes and the recruitment of duplicated genes to a novel expression domain (neofunctionalisation) is an even rarer process that requires the evolution of novel combinations of transcription factor binding sites in upstream regulatory regions. This hypothesis concerning the evolution of snake venom is therefore very unlikely. Nonetheless, it is often assumed to be established fact and this has hampered research into the true origins of snake venom toxins. We have generated transcriptomic data for a diversity of body tissues and salivary and venom glands from venomous and non-venomous reptiles, which has allowed us to critically evaluate this hypothesis. Our comparative transcriptomic analysis of venom and salivary glands and body tissues in five species of reptile reveals that snake venom does not evolve via the hypothesised process of duplication and recruitment of body proteins. Indeed, our results show that many proposed venom toxins are in fact expressed in a wide variety of body tissues, including the salivary gland of non-venomous reptiles and have therefore been restricted to the venom gland following duplication, not recruited. Thus snake venom evolves via the duplication and subfunctionalisation of genes encoding existing salivary proteins. These results highlight the danger of the "just-so story" in evolutionary biology, where an elegant and intuitive idea is repeated so often that it assumes the mantle of established fact, to the detriment of the field as a whole.

The origin and diversification of a novel protein family in venomous snakes

“…Casewell (21) utilized phylogenetic analysis of viperid venom-expressed SVMPs to reveal the monophyletic origin of P-II SVMPs from a P-III SVMP ancestor, but P-I SVMPs appear to have arisen independently at least eight times from class P-II ancestors. The large number of SVMP proteins and their diverse activities have prompted many authors to suggest a role for gene duplication and sequence divergence during the expansion of advanced snakes (16,19,(22)(23)(24). In the king cobra (an elapid), Vonk et al (16) found two venom-expressed SVMPs adjacent to the ADAM28 locus and suggested that the two cobra SVMPs evolved by duplication from the ADAM28 ancestor.…”

Intragenome Diversity of Gene Families Encoding Toxin-like Proteins in Venomous Animals

“…Failing to account for this have multiple misleading effects, most notably regarding tests of the recruitment hypothesis (i.e. whether there has been one or multiple recruitment events, see [Casewell et al, 2012;Hargreaves et al, 2014b]) and the adequacy of selection tests performed on alignments merging both old and recent paralogs as these are likely to be influenced by the different evolutionary rates of both (see [Jordan et al, 2004;Aguileta et al, 2006;Pegueroles et al, 2013]) and affected by gene conversion leading to overestimation of positive selection [Casola and Hahn, 2009]. The quantitative differences between venom glands and pooled tissues expression of genes encoding toxin-like proteins in M. martensii and the restricted set of VZsame expressed genes in O.…”

WITHDRAWN: ToxVec: Deep Language Model-Based Representation Learning for Venom Peptide Classification

“…This gap motivates computational methods that can automatically and accurately identify venom peptides in the large protein datasets. The prediction of venoms versus non-venom sequences is not a trivial task protein classification task, where the use of BLAST-based approaches is challenging: venoms are often (i) evolved from non-toxic proteins (Hargreaves et al, 2014), (ii) and then have highly diverged (Linial et al, 2017). Several studies have proposed computational and machine learning-based methods for predicting or analyzing toxin/venom peptides (Cole and Brewer, 2019;Dao et al, 2017;Gacesa et al, 2016;Naamati et al, 2009;Ojeda et al, 2018;Pan et al, 2020;Wong et al, 2013).…”