Viper's venom is tailored to its prey

18 May 2009 by Tom Marshall

A snake has venom adapted to kill its favoured quarry, new research suggests.

Saw-scaled viper

This might seem obvious, but it contradicts a common idea among biologists. Many experts on the evolution of snakes have thought they are so toxic, and inject so much more venom than is strictly needed to kill their prey, that the composition of their venom isn't subject to evolutionary selection.

But Axel Barlow, now a PhD student in the molecular ecology and evolution group in the School of Biological Sciences at Bangor University, found evidence to the contrary while researching his undergraduate dissertation, which was part of a wider project on the evolution of snake venom funded by the Leverhulme Trust.

His work suggests that a snake's venom is specifically adapted to the kind of prey it eats - although this specialisation seems to take the form of needing to use less venom rather than killing that prey more quickly.

"Some people have argued that these venoms are so toxic that minor changes in venom composition are not significant." says Barlow. "But there is an increasing body of evidence suggesting that venom is under strong selective pressure. This research is the final nail in the coffin for the hypothesis that venom composition is neutral in evolutionary terms."

"This study provides one of the most convincing pieces of evidence to date for the role of natural selection for diet in shaping snake venom composition, a key question in our understanding of venom evolution in snakes," says Dr Wolfgang Wüster, a lecture at Bangor's School of Biological Sciences and another of the paper's authors.

This turns on its head the classic scenario of a venomous snake killing its prey, in which you imagine a snake striking and the prey dropping dead almost immediately.

- Axel Barlow, Bangor University

The research could have important implications for how doctors treat snake bites. Differences in venom between species of snake can complicate the treatment of bites; understanding how these differences arose in the course of evolution could help create better antivenoms.

The research appears in Proceedings of the Royal Society B, and combines several approaches to the problem. Barlow chose to work with the genus of saw-scaled vipers; these snakes, which live in North Africa, Arabia and the Indian subcontinent, are closely-related but eat very different things.

Some species have settled on the diet of scorpions and other arthropods - an unusual choice of prey among snakes. These snakes are also of interest because they are so dangerous to humans, probably accounting for the majority of deaths from snake bites in Africa. But hospitals here often rely on imported antivenom from Asia, where saw-scaled vipers have very different venom, and this unsuitable antivenom has led to many unnecessary deaths in Africa.

The researchers studied the genetic differences between four different species within the genus to produce a phylogenetic tree showing the evolutionary relationships between different kinds of saw-scaled viper and indicating the order in which the major lineages diverged from each other. "This is the first study that shows that shifts in diet over the evolutionary history of the saw-scaled vipers go alongside changes in venom toxicity," Barlow says.

The scientists then performed chemical analysis on the venom of the four species, and dissected preserved museum specimens of these snakes to look at their stomach contents and hence learn about their favoured diets, looking out for species that eat a lot of arthropods - these are creatures with hard exoskeletons, like insects or scorpions. Lastly they tested how lethal each kind of snake's venom was to scorpions.

Initially this involved determining the median lethal dose for each kind of venom - that is, the amount of venom that, when injected into a group of scorpions, kills half of them within 24 hours.

This is a standard way of measuring how toxic a substance is to a particular organism. But the scientists realised that snakes need a greater than 50% chance of killing their prey, and to do so in much less time than 24 hours, so they also measured how long the amount of venom delivered during a normal bite took to incapacitate and kill scorpions.

They found that species that had already been shown to eat a lot of arthropods did indeed have venom that's better at killing them, when measured in terms of median lethal dose. This is due to specialisation in killing arthropods; these venoms are not simply more toxic to everything.

Efficiency, but not speed

But strangely, this adaptation showed itself only in the amount of venom needed for a kill, rather than in incapacitating and killing scorpions more quickly than the venoms of other species.

This means that saw-scaled vipers that have evolved to eat arthropods need to inject less venom into a scorpion to kill it, but this venom doesn't kill them any faster; these snakes can still bite a scorpion and have it run away for some time before dying.

Barlow says more work is needed to understand why this is; one might imagine that killing prey quickly would be a major advantage to snakes, reducing the risk of losing their quarry and the time they are exposed to predators while following it.

It could be that not having to produce as much venom is a more significant advantage to snakes than killing their prey more quickly would be; many snakes have excellent tracking abilities and they may have no problem following their prey until it is incapacitated.

"We know that venom is incredibly costly to make in metabolic terms, so it's an advantage for a snake not to have to make so much of it" explains Barlow. "But it appears that speed of prey death is not under the same selective pressure. This turns on its head the classic scenario of a venomous snake killing its prey, in which you imagine a snake striking and the prey dropping dead almost immediately."

Other Bangor students are now repeating the work with locusts rather than scorpions for their own dissertations; the results should confirm whether this is a general pattern among arthropods or whether scorpions are an unusual case. Barlow is now working on genetic variation in the snakes of southern Africa.