George Church, a well known geneticist at Harvard, is renowned for his contributions to methods of sequencing DNA as well as of “bioengineering” DNA by changing it using the CRISPR technique, which he helped develop. CRISPR gives us the ability to precisely edit DNA, inserting individual nucleotides, bits of genes, or whole genes and groups of genes into precise locations in another genome. We can even use it to turn on inserted genes at will. This, of course, opens up a vast array of remarkable things we can do, has huge implications for things like human health and crop improvement, and surely the inventors of the technique will get a Nobel Prize. (Who owns the patents to this method has been the subject of bitter dispute. They were awarded this week to the Broad Institute in Cambridge, Massachusetts, but further battles remain.)
One of the more bizarre applications of CRISPR was suggested, and is apparently under development, by Church’s own lab. It is, as the Guardian just reported, an attempt to bring back the woolly mammoth (Mammuthus primigenius), a pachyderm that went extinct only about 4,000 years ago, probably due to a combination of climate warming and human hunting.
The mammoth, pictured below, is far more closely related to the Asian elephant (Elephas maximus, which diverged from the mammoth 6 million years ago) than to the African elephant (Loxodonta spp., which diverged from the mammoth 25 million years ago), and so Church is proposing to use Asian elephants to “resurrect” the mammoth.
The problem is that the Guardian headline,“Wooly mammoth on verge of resurrection, scientists reveal”, and its contents, produced by Guardian science writer Hannah Devlin, are completely erroneous. What she did originally was to just uncritically quote George Church and his research plan, and then leverage that into a clickbait piece. That piece in fact was the most quoted article in the Guardian yesterday, accruing over 1400 comments. But it was just wrong, or, to paraphrase Wolfgang Pauli, “not even wrong”, for reasons I describe below. After it was published, our own Matthew Cobb, ever vigilant for bad science, complained on Twitter to the journalist. Such is the power of Twitter that she then inserted a few critical comments from Matthew. But the piece remains misleading.
What Church intends to do is not even close to resurrecting the mammoth, which would mean producing a live animal containing an entire woolly mammoth genome. As I’ve said before, we don’t have the technology to do that, because if you try to do it by putting a mammoth genome into a “host” Asian-elephant egg whose own DNA has been removed, it wouldn’t work. That’s because you can’t just stick all the DNA of a mammoth willy-nilly (or should I say “woolly nilly”?) into a DNA-less elephant egg: the DNA has to be properly arrayed on chromosomes to function. Further, maternal-effect substances from a mother mammoth would have put in the egg, for the Asian elephant doesn’t have those.
The DNA of woolly mammoths can’t be synthesized on whole chromosomes, and the DNA from frozen mammoths themselves (several have been found that fell into crevasses in the ice, preserved for thousands of years), has degraded to the point that it’s not in the proper configuration on chromosomes. What we have is bits of preserved mammoth genome. From that we can get its genomic sequence, but we can’t get a usable genome ready to insert into an egg.
As I said, we can use those bits to sequence the entire mammoth genome, and thereby see the differences between it and its close relative, the Asian elephant. What can we do then? Well, we can’t resurrect the woolly mammoth—not by a long shot. What we can do right now is simply put a small number of mammoth DNA sequences (genes) into an Asian elephant’s DNA, and then rear an egg that would, in effect, develop—if it does develop—into an Asian elephant with some woolly mammoth traits, like smaller ears, more fat, and more hair.
But even to do that we must know exactly which woolly mammoth genes produce its difference in appearance and physiology from the Asian elephant, and we don’t even know that. What Church et al. have apparently done is picked some “candidate” genes whose DNA sequence differs between the two species, and then splice in about 45 of those candidates into an Asian elephant embryo using the CRISPR technique. What they’ve achieved so far, though, is limited to having put some mammoth genes into cultured skin cells from an Asian elephant, and gotten the mammoth genes to express themselves–to produce a proteins or messenger RNA. While this is nice, it’s not anywhere near creating a whole mammoth.
There are further problems. Astoundingly, Church plans to rear this mammothized Asian elephant embryo in an “artificial womb”, which is pure fantasy. Such wombs been used to rear mice for 10 days, but not to term (20 days). No mammal has ever been successfully reared to the “birth” stage from an artificial womb. To rear a mouse embryo for ten days is quite different from rearing a 100-kilogram mammoth embryo to term over a period of 22 months! To suggest that this is just around the corner is pure fantasy—and bad reporting. Need I add that Church has never published a paper giving details of his technique or of the artificial womb?
And why would they use an artificial womb rather than implanting the mammothized egg back into an Asian elephant, a form of in vitro fertilization? Well, that could be dangerous for both the embryo and the mother, and I doubt that any zoo would volunteer one of their Asian elephants to become the surrogate mom.
The whole technique would, if successful (and that’s a long, long shot) produce an Asian elephant might be hairy and have small ears and a few other mammoth-like traits. But it’s not a genuine mammoth by any means, and the Guardian was wrong to suggest so. What Devlin did in her first version was simply parrot what Church said, not asking any other geneticists to comment Church’s plan. (Bad form!) That’s when Matthew tweeted “foul” and Devlin added his comments to her credulous report.
Vestigial bits of that article are still there, and they’re misleading:
The woolly mammoth vanished from the Earth 4,000 years ago, but now scientists say they are on the brink of resurrecting the ancient beast in a revised form, through an ambitious feat of genetic engineering.
Speaking ahead of the American Association for the Advancement of Science (AAAS) annual meeting in Boston this week, the scientist leading the “de-extinction” effort said the Harvard team is just two years away from creating a hybrid embryo, in which mammoth traits would be programmed into an Asian elephant.
Maybe, but that embryo has to be reared to birth, and we’re not going to do that any time soon. Even developing an artificial womb that could succor a hybrid embryo for 22 months is an enormous undertaking.
Finally, Church suggested that this technique could be used, as Devlin wrote, “to help preserve the Asian elephant, which is endangered, in an altered form.” Presumably he means that they’d produce a lot of hairy, mammothy Asian elephants and then release them in northern Asia, where their hirsuteness and fat would preserve them in the cold. That’s ridiculous too. The way to save the Asian elephant is to preserve its habitat and stop people from killing them or, in the worst case, keep a bunch of them in zoos.
Here’s Matthew’s brief pushback in the Guardian article, added to the original credulous piece. He brings up a point I hadn’t considered:
Matthew Cobb, professor of zoology at the University of Manchester, said: “The proposed ‘de-extinction’ of mammoths raises a massive ethical issue – the mammoth was not simply a set of genes, it was a social animal, as is the modern Asian elephant. What will happen when the elephant-mammoth hybrid is born? How will it be greeted by elephants?”
. . . . “Church’s team is proposing to rear the embryo in an ‘artificial womb’ which seems ambitious to say the least – the resultant animal would have been deprived of all the pre-birth interactions with its mother,” said Cobb.
If you know British euphemistic language, you’ll know that by “seems ambitious to say the least”, Matthew means “this is going to fail big time, and that’s only one of the problems.” Matthew’s summary to me in an email, which I reproduce with permission, is this:
In fact [Church] is more circumspect than a rapid reading implies – he merely wants to use CRISPR to introduce some mammoth-like sequences into an asian elephant embryo. End of. The rest is fantasy – artificial uteruses and the rest. Sigh.
The big fault here is the journalist Devlin, who reports on an imminent “resurrection” of the woolly mammoth, a complete and utter fabrication. Church, of course, is also guilty—of wasting time and money on a hairbrained scheme that almost certainly won’t work, and even if it did work won’t resurrect anything meaningful. In fact, if you want a mammothy elephant, it’s far safer to select for Asian elephants to be hairier and have more fat and reduced ears. (That, of course, would take ages, but it wouldn’t have the dangers or expense of gene editing.)
After Matthew’s tweet and the Guardian’s insertion of his remarks, Matthew was besieged by radio and television stations to comment on this “exciting” story. (The public loves to contemplate the reappearance of extinct creatures.) As he said, “Yesterday turned into Mammoth Day.” He had to turn most of the requests down, but here he is on Newsnight last night commenting on the mammoth proposal.
But that had nothing to do with Mammoth Resurrection. Church is a good scientist and an accomplished one, but in this case he’s wasting a lot of time and money in a futile attempt to resurrect an animal which, even if resurrected, would have to be brought back in multiple copies of opposite sexes.
As the Germans would say, “Das ist ja Wahnsinn!”