On the islands of the Dampier Archipelago, just off the coast of north-west Western Australia, giant piles of rusty, iron-rich boulders tumble into the brilliant turquoise waters of the Indian Ocean. Six thousand years ago, these islands were hilltops emerging from a wide coastal plain teeming with life. Aboriginal people recorded these animals by carving petroglyphs into the deep-red rocks.
Among the images are more than 20 thylacines, also known as Tasmanian tigers. These wolf-like, carnivorous marsupials carried their young in a pouch like kangaroos, sported tiger-like stripes on their backs and had jaws capable of an impressive 120-degree gape. They were once common across much of Australia and New Guinea.
The thylacine vanished from the Australian mainland about 3,000 years ago, probably as a result of a drying climate and the loss of dense vegetation. It maintained a toehold in forested Tasmania, only to be hunted to extinction by Europeans from the 1800s. The last known tiger died in Hobart Zoo in 1936.
Australia’s roll call of extinct species includes car-sized relatives of the wombat, lion-like predators and giant flightless birds. But the thylacine holds a special place in the public consciousness. Frequent ‘sightings’ and quests to find evidence of a living thylacine manifest hopes it might not really be lost.
In recent times, that hope has translated into possible ‘de-extinction’ through cloning.
Specimens from 450 thylacines are in museums around the world. Most are skin and bones, but 13 pouch young (joeys) were preserved in alcohol or formaldehyde. The Melbourne Museum has one so well-preserved that a team led by Andrew Pask at the University of Melbourne announced, in 2017, the successful sequencing of its entire genome. It is the most intact genome obtained for an extinct species.
The Melbourne joey’s own life might have been cut short, but its DNA may be a blueprint to resurrect the entire species. No one thinks it will happen soon but, as University of New South Wales palaeontologist and incurable ‘de-extinction’ champion Michael Archer puts it: “It’s a brave geneticist these days who’ll say what’s impossible in the next decade or two.”
Archer was perhaps the first person to dare to dream of cloning the thylacine. In 1996, when Dolly the sheep made history as the first mammal to be cloned, he declared doing the same with a thylacine was “a matter of not if but when”.
Dolly’s DNA originated from the mammary cell of an adult ewe. The cell’s nucleus, containing the DNA, was sucked out and transferred into a sheep egg whose own nucleus had been removed. The transferred nucleus ‘rebooted’ the egg’s development, creating a clone of the original ewe.
There is no chance of doing the same with a thylacine. Museum specimens can deliver thylacine DNA but not a viable nucleus or egg. So how do you clone something without these seemingly essential ingredients? Geneticist George Church, at Harvard University, has pioneered a way.
It is somewhat like the cloning strategy imagined in Jurassic Park. The fictional genetic engineers source dinosaur DNA from amber-preserved mosquitoes that dined on dinosaur blood. Gaps in the dinosaur DNA are filled by reptilian, bird or amphibian DNA.
In a similar manner, Church is heading an effort to clone the mammoth by using the DNA of its closest living relative, the Asian elephant, to fill in the missing bits of mammoth DNA.
What takes the scenario from fiction to reality is CRISPR. This latest tool in the genetic engineer’s kit is a set of enzymes used by bacteria to target and destroy foreign DNA. In 2015 genetic engineers co-opted CRISPR to target and alter DNA within living cells. Church’s goal is to ‘edit’ key tracts of elephant code to convert them into mammoth code, rather like turning a modern novel into medieval-era prose.
Church’s team have identified 1642 genes that differ between the species. In February 2017 Church announced the successful conversion of 45 of those genes. “We already know about the ones to do with small ears, subcutaneous fat, hair and blood,” he said, predicting a hybrid elephant-mammoth embryo “could happen in a couple of years”.
Once an edited facsimile of a mammoth nucleus has been created, it could be placed into an Asian elephant egg and then into a womb. Church is also looking into technologies for artificial wombs.
By the time Dolly the sheep was cloned, acquiring a thylacine’s DNA blueprint from a museum specimen was a tantalising possibility. Short sequences of DNA were already being extracted from mammoths and other long-dead specimens. Archer, then at the Australian Museum in Sydney, attempted to extract DNA from a thylacine in the museum’s collection – a six-month-old pup preserved in alcohol in 1886 – but the DNA was too fragmented to be useful.
Given those difficulties, Pask in Melbourne thought sequencing the thylacine genome would be impossible. His team focused instead on sequencing the genomes of living species – the platypus, tammar wallaby and dunnart. The goal was to compare their blueprints to placental mammals like us and trace how genes had evolved since these mammalian relatives had diverged.
Success at reading marsupial genomes emboldened the scientists to take another shot at the thylacine. In 2008 they reported a milestone: isolating a fragment of thylacine DNA so intact its code was still readable. A computer program recognised the DNA as the code for a gene – Col 2A1 – that directs the development of cartilage and bone. The researchers inserted the gene fragment into a mouse embryo, together with a chemical tag that made the gene glow blue wherever it was active. Blue patterns appeared in the embryo’s developing skeleton, meaning the code was good enough to work in a living creature.
The finding was encouraging. Even if scientists could never read a complete thylacine genome, they might glean important information from studying its genes – such as clues about how this cousin of the kangaroo evolved the body shape of a wolf.
Pask’s team spent 10 years taking samples from 40 thylacine specimens worldwide. “Most of the museum samples had really, really badly damaged DNA,” he says. He had almost given up hope when, in 2010, he came across a specimen on his doorstep. In a dusty cabinet in the bowels of the Melbourne Museum, preserved in a jar of ethanol, was a four-week-old joey taken from its dead mother’s pouch in 1909.
Pask’s team sampled its DNA. Unlike all the other specimens, the joey retained strings of DNA 1,000 letters in length – long enough to mean the entire three-billion-letter genome might be puzzled back together. Pask believes the DNA’s good condition might be due to the specimen missing the standard formalin fixation, instead going straight into ethanol.
The sample not only yielded long strings of DNA but plenty of them. Crucially that allowed Pask’s team to read every bit of the DNA sequence 60 times over using different strands. This enabled them to correct inevitable errors in the century-old material.
Imagine finding an old car manual with many pages missing. You would struggle to make use of it. But with 60 tattered incomplete copies you could probably compile a whole manual. Pask is similarly confident the blueprint is accurate enough to instruct the building of a thylacine. So too is Archer, who has lost none of his enthusiasm for bringing back extinct species. “It’s the roadmap for getting a thylacine back,” he says.
Cloning a thylacine will be more challenging than Church’s project to resurrect the mammoth using the Asian elephant. Their ancestors diverged just six million years ago, and they share about 99% of their genes. There is no equivalent species for the thylacine.
Pask suggests Western Australia’s numbat, whose genome he plans to sequence, might provide the best starting DNA blueprint. It is one of the thylacine’s closest living relatives, last sharing a common ancestor 30 million years ago. The diminutive termite-eating creature has stripes, but that’s where the similarity ends. Adult numbats are slightly bigger than a squirrel, whereas adult thylacines weighed about 30 kg. Despite this, Pask says as much as 95% of their DNA may be identical.
That still leaves an awful lot of numbat DNA to edit, making it an expensive proposition. But, as with all other genetic technologies, the costs are likely to fall fast. Pask will wait and watch while other de-extinction projects, particularly that of the mammoth and a similarly advanced effort to resurrect the passenger pigeon of North America, perfect the technologies.
The next series of steps are the most unpredictable: cloning an embryo, implanting it into a surrogate and gestating the pouch young.
Getting cloning to work is a major challenge. The techniques used to create Dolly are notoriously difficult to apply to different species. It was only in 2017 – more than 21 years after Dolly – that it was successfully replicated in a primate, with Chinese scientists producing two genetically identical long-tailed macaques.
Once researchers get a thylacine-recoded numbat egg to start developing into an embryo, gestating it is also far from straightforward. For humans and sheep, both placental mammals, the science of implanting embryos into a womb is well-established. Not so for marsupials, where implantation takes place much later. In placentals we know how to prime a mother with hormones to accept an embryo, but this knowledge is completely lacking in marsupials.
To master assisted reproduction in marsupials, Pask has turned to a different thylacine relative, the tiny mouse-like dunnart. They breed well in captivity and produce a litter of up to 20 young twice a year. Nevertheless, he says, “it will be a decade before we get a really good handle on a lot of this stuff in marsupials”.
Pregnancy is also a very different proposition to placental mammals. A marsupial still looks something like a foetus when it is born, typically two weeks after conception. About the size and shape of a pink jellybean, it must crawl up its mother’s abdomen and into her pouch, where it latches onto a teat to suckle. Its mother’s milk, like a placenta, changes its composition to guide most of the joey’s development.
This two-stage gestation does offer intriguing possibilities. A thylacine embryo might be gestated in the uterus of a smaller marsupial, and then transferred to the pouch of a larger one – perhaps a kangaroo. Cross-fostering is a well-established technique to help bolster the populations of endangered rock wallabies. In 2014 a rock wallaby successfully fostered a baby tree kangaroo in its pouch.
Another option is hand rearing, already widely employed for rescued kangaroos and also for Tasmanian devils captive bred to save the species from the devil facial tumour disease (DFTD) that has decimated wild populations.
Once a thylacine joey has weaned, at about nine months, there would be a new set of hurdles. Would it behave like a thylacine?
Little is known about natural behaviours, such as hunting or mating, as the thylacine was scarcely observed in the wild. “Many behaviours are innate,” Pask says, “but there would be a large subset that they probably learnt from individuals around them. Learned behaviour is more common in species that use complex decision-making to hunt prey, and preserved thylacine brains reveal a well-developed frontal cortex, indicating good memory and capacity to learn.”
We do know thylacines did not fare well in captivity. The Royal Zoological Society of NSW noted in 1939: “The thylacine does not take kindly to captivity, and rarely lives under such conditions for any length of time.” From 1850 to 1931, 224 were kept at zoos in cities including Washington DC, New York, Berlin and Paris. London Zoo had 20 over the years. Some died during journeys, others stopped eating and fell ill. None bred. While our skill at keeping animals has increased enormously, there is no guarantee resurrected thylacines would do better.
Understanding how a species might fare is important, says Beth Shapiro, an evolutionary biologist at the University of California, Santa Cruz, and author of How to Clone a Mammoth: The Science of De-Extinction (2015). “Populations living in captivity, possibly for decades, need not only to survive but must also learn how to live,” she says. “They need to learn how to feed and protect themselves, how to interact with others, how to avoid predation, how to choose a mate, and how to provide parental care.”
You also need a population with genetic variety, Shapiro says. Pask suggests it might be possible to edit such variation into the genome. “If you can get over the hurdle of making all those millions of edits to the genome to make it look like a thylacine in the first place,” he says, then introducing variability into immune system genes “is nothing”.
If all these hurdles can be overcome, the end goal of any de-extinction effort surely must be to reintroduce animals to the wild. One potential issue for some de-extinction candidates – appropriate habitat – is not a problem. Reserves cover about half of Tasmania today. “The habitat is the same, the animals they ate are still there,” says Archer. “There’s no question it could be put back into the bush of Tasmania.” There is also good reason to do so: “The thylacine was Tasmania’s key carnivore. Getting it back is about restabilising ecosystems currently under threat.”
That still may not be enough to convince everyone we should bring back thylacines. Many argue de-extinction projects take the focus away from the vital work to save other species from extinction.
“If you have the millions of dollars it would take to resurrect a species and choose to do that, you are making an ethical decision to bring one species back and let several others go extinct,” Canadian conservation biologist Joseph Bennett has said. “It would be one step forward, and three to eight steps back.”
Yet what is true today may not be true tomorrow. Pask agrees that, right now, resources should go to saving endangered marsupials. “If, however, in 10 to 15 years’ time it becomes relatively inexpensive, then I think it is definitely worth pursuing.” Having hunted the thylacine to extinction, he says, “we owe it to the species to bring it back”.
It may not be entirely thylacine, but one day, a century or so from now, a creature that looks and behaves like one might be found quietly slipping between piles of rusty rocks that bear its likeness, etched millennia ago.