In a recent paper, Olivera et al analyze the venom of Cone snails, which represent a fairly large and diverse genus of venomous marine snails.
What they found was very interesting. The venom genes had the following properties:
- The gene was hypermutable – in other words, within the genus, the variability of the venom was immensely more variable than other gene regions
- The hypermutability of the gene was localized – in other words, there was a specific target of hypermutability in the gene. There were some parts which were highly conserved.
- The localization of the hypermutability was not due to selection – in other words, it isn’t that the whole thing was hypermutable and selection only kept the ones which match the pattern. The static region didn’t even have synonymous mutations.
- Structurally-significant amino acids were kept – cysteine is an important amino acid for proteins that are excreted – they make disulfide bonds which help stabilize the protein. The cysteine amino acids were untouched, despite being scattered throughout the protein. This was also not due to selection, because, again, there were not even synonymous substitutions.
This evidence suggests that the evolution of the Conus genus is tightly regulated. In other words, the evolutionary process itself is directionalized.
Interestingly, the paper also notes that a similar pattern also emerges in other members of the cone snail’s suborder, such as Turridae. These molluscs have a different superfamily of toxins, but the mutational pattern is very similar – a highly conserved region within a hypermutated gene, untouched Cysteine amino acids, etc. So, the open question is, are these two related by ancestry or by common design patterns?
I will offer a completely unsubstantiated suggestion on this point – I propose that the toxin superfamily is based on genes from a symbiont, and the suborder has sufficient biochemical information to take the genes from a symbiont and process them into proper venom. This is based on the fact that these organisms have highly similar feeding morphology, similar to the uniqueness of the feeding morphology of ruminantiae (not that the morphology is similar between these two groups, but that the patterns of similarity are somewhat analogous – both are suborder ranks in which a unique feeding mechanism is shared by all members of the suborder). It also matches Margulis’ observation that a wide variety of speciation is based on shuffling of symbiotic relationships.