Immune snakes outrun toxic newts in evolutionary arms races

Blogging on Peer-Reviewed Research

The story of evolution is filled with antagonists, be they predators and prey, hosts and parasites, or males and females. These conflicts of interest provide the fuel for 'evolutionary arms races' - cycles of adaptation and counter-adaptation where any advantage gained by one side is rapidly neutralised by a counter-measure from the other. As the Red Queen of Lewis Carroll's Through the Looking Glass said to Alice, "It takes all the running you can do, to keep in the same place."


The Red Queen analogy paints a picture of natural foes, wielding perfectly balanced armaments and caught in a perpetual stalemate. But this is an oversimplified view. It is entirely possible for one combatant to develop such a significant advantage that it completely outruns the other and temporarily wins the race.

Charles Hanifin from Utah State University has found one such example among garter snakes and newts living along North America's west coast. Even though some of the newts pack one of the most powerful poisons used by any animal, they still fall prey to garter snakes that have evolved extreme levels of resistance to them.  In some locations, the snakes' immunity is so complete that the not a single newt is poisonous enough to overwhelm them.

Snake v. newt

The three species of newts from the genus Taricha defend themselves with a lethal poison called tetrodotoxin. It kills by plugging up molecular pores on the surface of nerve and muscle cells that act as channels for sodium ions. If these ions are denied passage, nerve cells can't fire and muscles can't contract. The heart stops, breathing becomes impossible and death soon follows. There is no antidote.

The skin of a single newt is laced with enough tetrodotoxin to kill 10-20 humans, or thousands of mice. But not the common garter snake (Thamnophis sirtalis); some individuals have become immune to tetrodotoxin, by changing the structure of their sodium channels so that the poison no longer blocks them.


To study the arms race between snake and newt, Hanifin surveyed different populations across their entire shared range, a 2,000 km stretch of land between British Columbia and the southern tip of California. While many arms-race studies look at a single pair of populations, that's a bit like spotlighting on two actors on a crowded stage; instead, Hanifin wanted to look at a large geographical stage to watch populations at different stages of escalation.

Together with two Edmund Brodies (Jr and III), he measured the levels of tetrodotoxin in newts from 28 locations across the west coast. They also measured how resistant local snakes were by injecting them with the poison and measuring its effect on their slithering speed.

As expected, they found massive differences in both toxicity and resistance. Some populations haven't entered the arms race at all; in British Columbia, for example, non-resistant snakes live alongside poisonless newts. As the newts become more toxic, the snakes become more resistant and the conflict escalates until both poison and resistance are magnified by a thousand times.

In general, the most resistant snakes lived alongside the most toxic newts. But Hanifin also found that the animals' abilities were often mismatched and in every single case, it was the snakes that came out ahead. In a third of the locations they sampled, even the least resistant snakes were more than capable of eating the most toxic newts. Taking mouthfuls of one of the most lethal of animal poisons barely slowed them down.

In these locations, the snakes have escaped from the cyclic nature of the evolutionary arms race. Their advantage is so great that there isn't a newt toxin they can't handle, and as such, they are under no impetus to become even more resistant.

A genetic upper hand

It seems surprising that the newts never developed an overwhelming advantage themselves. After all, you might assume that they would be under even greater pressure to develop better defences for they stand to lose their lives while the snakes merely risk losing their dinner.

But the snakes have a genetic advantage. Their ability to shrug off the effects of tetrodotoxin depends on the structure of their sodium channels and these in turn are governed by a small number of genes. The upshot is that it takes a very small number of simple genetic changes to turn a susceptible snake into a resistant one and these chances can spread rapidly throughout a population.


In at least one group of extremely resistant snakes, the altered sodium channel differs from the basic model by a single amino acid in its entire length. The effect is like making a fort invulnerable by changing the position of a single brick.

It's altogether more complicated for newts to evolve more powerful poisons. Some scientists have suggested that the various animals that wield tetrodotoxin may accumulate it from an environmental source rather than making it themselves, and that would limit the amount that an individual could build up.

Tetrodotoxin is also so powerful that the newts themselves aren't immune to it. They safely store the chemical in their skin but it would be physically impossible for them to house enough poison to overwhelm the defences of the most resistant snakes.

What happens next is unclear but while the race has been temporarily suspended, it isn't over. While the snakes can take a breather from all the relentless innovation, the newts are still very much in the game and under strong pressure to develop even more lethal defences, if they can.

Alternatively, the snakes may even find it beneficial to become less resistant. Their altered sodium channels open up an exclusive menu of newts unavailable to other predators, but they carry a cost too. The changes to the snakes' nerves and muscles make them move more slowly and Hanifin speculates that if this drawback is significant enough, the snakes could begin to lose resistance. This de-escalation of arms could bring them back to a level where they could once again be poisoned by the newts, and the race is rejoined.

Images by Edmund Brodie III, Ivan Tortuga, and Eugene van der Pijll in order.

Hanifin, C.T., Brodie, E.D., Brodie, E.D. (2008). Phenotypic Mismatches Reveal Escape from Arms-Race Coevolution. PLoS Biology, 6(3), e60. DOI: 10.1371/journal.pbio.0060060

If you're interested, check out two more posts from the classic site which show evolutionary arms races in action - between ants and Alcon blue butterflies, and between water fleas and parasites.

More like this

According the Edmund Brodie junior, his interest in the rough-skinned newts began after hearing stories of drunken men eating newts that resulted in the death of the men. Additionally he tells a story about a run in between three hunters and a boiled newt.

"There was an old wise tale from coastal Oregon that three hunters had been found dead at their campsite and in their coffee pot was a dead newt, boiled along with the coffee," recalled Utah State's Brodie.

By Jerry Hodge (not verified) on 10 Mar 2008 #permalink

Interesting. Tetrodotoxin is highly selective, only interacting with cholesterol in the plasma membranes of cells. From this, it's believed that sodium ion channels at least partly correspond to the hydrophilic and hydrophobic routes in the axolemma of the cell membrane. That is, during a nervous impulse a widening of hydrophobic pathways takes place. I'd be interested in knowing from a molecular perspective how garter snakes circumvent the poison. For instance, do they use potassium channels, instead? Or maybe these snakes only have fatty acid hydrocarbons and no cholesterol in their plasmid membranes. Or maybe it has something to do with pH levels in their tissue.

From my understanding of the paper, the resistance is down to a change in the structure of the sodium channel that stops tetrodotoxin from binding to them.

It will be interesting to see if, in a few years/decades, these garters will be able to interbred with non-resistant garters.

Very interesting. I wonder if there's an alternative cyle: 1. newts develop toxicity. 2. snakes develop overwhelming immunity. 3. newts completely lose their now-useless toxicity. 4. snakes completely lose their now-useless immunity, and the cycle starts over.

Instead of the British Columbia newts evolving behind the rest, they may just be at a different point in the cycle.

One thing to keep in mind is that there are other potential newt predators besides Garter Snakes so the toxin is likely to continue to be advantageous.

There has been a whole bunch of interesting evolutionary biology done on Garter Snakes (of several species). I vaguely recall hearing about other work, also on western Garter Snakes that involved selection for avoidance of grey slimy things as food items. Slugs are delicious and nutritious (escargots without the troublesome shell). However, some leeches, on the other hand, respond to being swallowed by slicing there way out through the body wall of the snake. In places where GS and certain leeches are sympatric GS avoid grey slimy things but then they miss out on the slugs. In places where there are no leeches the GS gorge with impunity. The food preferences are based on olfactory cues and are partially genetic. I think there was also something about introgression of the genes through hybridization so that some unfortunate snakes found themselves with slug eating genes but in places where the potentially fatal leeches were abundant. Likewise other snakes missed out on abundant slugs in places where there were no leeches. Selective forces acted against introgression and confined it to a narrow contact zone. Unfortunately a quick google search didn't turn up the source and I'm too tired to be more thorough tonight.

Good point Matt, although it harms my beautiful theory as to why there's no race going on in British Columbia, so I'm forced to oppose it. :)

Could be that GS are predominant newt predators in some areas, and drive the cycle, but not others. I live in the San Francisco Bay Area - it's easy on a rainy day hike in a forest here to see a half-dozen newts, but I've seen less than a dozen GS in as many years in the area, and see them mostly in sunnier, lower-elevation areas where I don't see newts.