“Toxodon is perhaps one of the strangest animals ever discovered,” wrote Charles Darwin, a man who was no stranger to strangeness. He first encountered the creature in Uruguay on November 26th, 1834. “Having heard of some giant’s bones at a neighbouring farm-house…, I rode there accompanied by my host, and purchased for the value of eighteen pence the head of the Toxodon,” he later wrote.
The beast’s skeleton, once fully assembled, was a baffling mish-mash of traits. It was huge like a rhino, but it had the chiselling incisors of a rodent—its name means “arched tooth”—and the high-placed eyes and nostrils of a manatee or some other aquatic mammal. “How wonderfully are the different orders, at present time so well separated, blended together in different points of the structure of the toxodon!” Darwin wrote.
Those conflicting traits have continued to confuse scientists. Hundreds of large hoofed mammals have since been found in South America, and they fall into some 280 genera. Scientists still argue about when these mysterious beasts first evolved, whether they belong to one single group or several that evolved separately, and, mainly, which other mammals they were related too. “That’s been difficult to address because they have features that they share with a lot of different groups from across the mammalian tree,” says Ian Barnes from the Natural History Museum in London. “To some degree, people have circled around the same set of evidence for 180 years.”
Now, Barnes’ team, including student Frido Welker from the Max Planck Institute for Evolutionary Anthropology and Ross MacPhee form the American Museum of Natural History, have found a way to break out of the circle. They recovered a hardy protein called collagen from the fossil bones of Toxodon and Macrauchenia, another South American oddity that resembled a humpless camel. By comparing these molecules to those of modern mammals, the team concluded
that the two ancient beasts are most closely related to perissodactyls—odd-toed hoofed mammals like rhinos, tapirs, and horses.
“Toxodon looks a bit like a hippo and we now know that the features they share with hippos are probably due to convergence,” says Barnes. “Macrauchenia looks a bit like a camel, but we can now see that it’s not particularly well related to camels.. This has been a longstanding mystery and we have an answer, and that’s satisfying.”
The discovery has bigger implications, though. Many scientists, Barnes included, have recovered DNA from very old fossils. They have sequenced the full genomes of mammoths and Neanderthals, worked out the evolutionary relationships of giant birds, and even discovered entirely new groups of early humans. But ancient DNA has its limits.
To fish it out of fossils, you need molecular bait, and to design that bait, it really helps to know what kind of animal you’re looking for and what they’re related to. If you don’t, and your only clue is “er, some kind of mammal”, then recovering ancient DNA is hard. It becomes harder if the fossils are also very old, since DNA has a half-life of around 521 years. And it becomes absurdly hard if the bones come from warm climates, like most of South America, where DNA degrades even faster than usual.
Collagen, however, is exceptionally durable. This rope-like protein gives strength and elasticity to our skin, ligaments, tendons, and other tissues. “In principle, it should survive ten times longer than DNA in bone,” says Barnes. And while bones contain just a tiny amount of DNA, they carry a huge amount of collagen—this one molecule makes up 25 to 30 percent of the total proteins in our body. (This bounty also means that contamination of ancient samples by collagen from scientists or other creatures isn’t really a problem.)
To analyse the ancient collagen, Barnes worked with Matthew Collins from the University of York. They assembled near-complete sequences of COL1—the most common form of collagen—from both Toxodon and Macrauchenia. They then compared these sequences to COL1 from 76 mammal species (and one chicken) to build a family tree, with Toxodon and Macrauchenia perched at the base of the perissodactyl branch.
But Maureen O’Leary from the Stony Book School of Medicine, who studies mammal evolution, says that this approach is “not methodologically sound”. Before creating their tree, she says, the team should have combined their data with published information on other molecules and physical features.
Alan Cooper, an ancient DNA specialist from the University of Adelaide, adds that ancient proteins have their problems. Collagen does such an important job that it doesn’t change very easily; when it does, it is severely constrained in how it can change. As such, similarities between two collagen sequences might imply that their owners are genuinely related, or that they were independently forced down the same convergent routes. “The phylogenetic resolution of protein sequences”—that is, their ability to tell you what’s related to what—“is pretty ropey,” says Cooper.
That said, he notes that the team’s family tree makes sense. Toxodon and Macrauchenia are so weird that they could sit anywhere, but the rest of the branches and twigs are (mostly) where they should be. “It’s good to see something come out on Macrauchenia at last,” he adds. “A lot of us having been trying to get ancient DNA out of remains for a while. I suspect that’ll finally happen, at which point it’ll be really interesting to see how the [family trees] compare.”
Toxodon and Macrauchenia belong to two separate orders of South America’s weird hoofed mammals. There are three more, and, “unfortunately, the likelihood is that we won’t get similarly nice results from looking at any of them,” says Barnes. They went extinct too early; their fossils are too old. The only option is to identify the features in Toxodon and Macrauchenia’s bones that hint at their affinities with the horses and rhinos. They might then be able to focus at the same features in their other enigmatic neighbours.
Collins hopes that sequencing ancient proteins will provide researchers with another way of understanding prehistoric life, especially in situations where DNA preserves badly. His team, for example, have looked at proteins in ancient dental plaque to study the history of human milk consumption and our longstanding battles against oral infections. “A lot of people are aware that DNA sequencing is changing, but protein sequencing is undergoing a similar revolution in the sensitivity of the instruments,” adds Barnes. “Who’s to say what we can do?”
Reference: Welker, Collins, Thomas, Wadsley, Brace, Cappellini, Turvey, Reguero, Gelfo, Kramarz, Burger, Thomas-Oates, Ashford, Ashton, Rowsell, Porter, Kessler, Fischer, Baessmann, Kaspar, Olsen, Kiley, Eillott, Kelstrup, Mullin, Hofreiter, Willerslev, Hublin, Orlando, Barnes & MacPhee. 2015. Ancient proteins resolve the evolutionary history of Darwin’s South American ungulates. http://dx.doi.org/10.1038/nature14249
PS: Another team of scientists led by Mary Schweitzer claims to have recovered and sequenced collagen from dinosaurs like Tyrannosaurus and Brachylophosaurus. Those results have always been controversial and still are. Toxodon lived a couple of million years ago; these dinosaurs lived at least 65 million years ago. Even collagen has its limits, although Schweitzer’s team argues that the protein’s rope-like structure protected some parts of it. “We’re working within the accepted limits of collagen survival,” says Barnes. “Schweitzer’s work requires a different mechanism, that I don’t understand, and they’d be the first to admit that the results they get are very, very fragmentary.”