Black Bone Illustration

Building a 'doomsday vault' to save the kangaroo and koala from extinction

Incredible feats of genetic engineering and frozen biobanks could prevent Australia's iconic marsupials from disappearing for good.

The road into Batlow is littered with the dead.

In the smoky, gray haze of the morning, it's hard to make out exactly what Matt Roberts' camera is capturing. Roberts, a photojournalist with the Australian Broadcasting Corporation, keeps his lens focused on the road as he rolls into the fire-ravaged town 55 miles west of Canberra, Australia's capital. At the asphalt's edge, blackened livestock carcasses lie motionless.

The grim scene, widely shared on social media, is emblematic of the impact the 2019-20 bushfire season has had on Australia's animal life. Some estimates suggest "many, many billions" of animals have been killed, populations of endemic insects could be crippled and, as ash washes into riverways, marine life will be severely impacted. The scale of the bushfires is so massive, scientists are unlikely to know the impact on wildlife for many years.

But even before bushfires roared across the country, Australia's unique native animals were in a dire fight for survival. Habitat destruction, invasive species, hunting and climate change have conspired against them. Populations of native fauna are plummeting or disappearing altogether, leaving Australia with an unenviable record: It has the highest rate of mammal extinctions in the world.

A large share of Australia's extinctions have involved marsupials -- the class of mammals that includes the nation's iconic kangaroos, wallabies, koalas and wombats. A century ago, the Tasmanian tiger still padded quietly through Australia's forests. The desert rat-kangaroo hopped across the clay pans of the outback, sheltering from the sun in dug-out nests.

Now they're gone.

Man carrying a kangaroo

Australia's 2019-20 bushfire season has been devastating for wildlife.

John Moore/Getty

In a search for answers to the extinction crisis, researchers are turning to one lesser-known species, small enough to fit in the palm of your hand: the fat-tailed dunnart. The carnivorous mouse-like marsupial, no bigger than a golf ball and about as heavy as a toothbrush, has a tiny snout, dark, bulbous eyes and, unsurprisingly, a fat tail. It's Baby Yoda levels of adorable -- and it may be just as influential.

Mapping the dunnart's genome could help this little animal become the marsupial equivalent of the lab mouse -- a model organism scientists use to better understand biological processes, manipulate genes and test new approaches to treating disease. The ambitious project, driven by marsupial geneticist Andrew Pask and his team at the University of Melbourne over the last two years, will see scientists take advantage of incredible feats of genetic engineering, reprogramming cells at will.

It could even aid the creation of a frozen Noah's Ark of samples: a doomsday vault of marsupial cells, suspended in time, to preserve genetic diversity and help prevent further decline, bringing species back from the brink of extinction.

If that sounds far-fetched, it isn't. In fact, it's already happening.

Of mice and marsupials

Creating a reliable marsupial model organism is a long-held dream for Australian geneticists, stretching back to research pioneered by famed statistician Ronald Fisher in the mid-20th century. To understand why the model is so important, we need to look at the lab mouse, a staple of science laboratories for centuries.

"A lot of what we know about how genes work, and how genes work with each other, comes from the mouse," says Jenny Graves, a geneticist at La Trobe University in Victoria, Australia, who has worked with marsupials for five decades.

The mouse is an indispensable model organism that shares many genetic similarities with humans. It has been key in understanding basic human biology, testing new medicines and unraveling the mysteries of how our brains work. Mice form such a critical part of the scientific endeavor because they breed quickly, have large litters, and are cheap to house, feed and maintain. 

Lab mouse

The lab mouse has been indispensable in understanding physiology, biology and genetics.

Getty/Picture Alliance

In the 1970s, scientists developed a method to insert new genes into mice. After a decade of refinement, these genetically modified mice (known as "transgenic mice") provided novel ways to study how genes function. You could add a gene, turning its expression up to 11, or delete a gene entirely, shutting it off. Scientists had a powerful tool to discover which genes performed the critical work in reproduction, development and maturation.

The same capability does not exist for marsupials. "At the moment, we don't have any way of manipulating genes in a devil or a kangaroo or a possum," says Graves. Without this capability, it's difficult to answer more pointed questions about marsupial genes and how they compare with mammal genes, like those of mice and humans.

So far, two marsupial species -- the Tammar wallaby and the American opossum -- have been front and center of research efforts to create a reliable model organism, but they both pose problems. The wallaby breeds slowly, with only one baby every 18 months, and it requires vast swaths of land to maintain.

The short-tailed opossum might prove an even more complicated case. Pask, the marsupial geneticist, says the small South American marsupial is prone to eating its young, and breeding requires researchers to sift through hours of video footage, looking for who impregnated whom. Pask also makes a patriotic jab ("they're American so we don't like them") and says their differences from Australian marsupials make them less useful for the problems Australian species face.

But the dunnart boasts all the features that make the mouse such an attractive organism for study: It is small and easy to house, breeds well in captivity and has large litters.

"Our little guys are just like having a mouse basically, except they have a pouch," Pask says.

Researchers in a lab facility with containers of dunnarts.

Pask (front) and Frankenberg inspect some of their dunnarts at the University of Melbourne.

Jackson Ryan/CNET

Workhorse, work mouse

A stern warning precedes my first meeting with Pask's colony of fat-tailed dunnarts.

"It smells like shit," he says. "They shit everywhere."

I quickly discover he's right. Upon entering the colony's dwellings on the third floor of the University of Melbourne's utilitarian BioSciences building, you're punched in the face by a musty, fecal smell.

Pask, a laid-back researcher whose face is almost permanently fixed with a smile, and one of his colleagues, researcher Stephen Frankenberg, appear unfazed by the odor. They've adapted to it. Inside the small room that houses the colony, storage-box-cages are stacked three shelves high. They're filled with upturned egg cartons and empty buckets, which work as makeshift nests for the critters to hide in.

Frankenberg reaches in without hesitation and plucks one from a cage -- nameless but numbered "29" -- and it hides in his enclosed fist before peeking out of the gap between his thumb and forefinger, snout pulsing. As I watch Frankenberg cradle it, the dunnart seems curious, and Pask warns me it's more than agile enough to manufacture a great escape.

In the wild, fat-tailed dunnarts are just as inquisitive and fleet-footed. Their range extends across most of southern and central Australia, and the most recent assessment of their population numbers shows they aren't suffering population declines in the same way many of Australia's bigger marsupial species are.

Fat-tailed dunnart in a researcher's gloved hand

Move over, Baby Yoda.

Jackson Ryan/CNET

As I watch 29 scamper up Frankenberg's arm, the physical similarities between it and a mouse are obvious. Pask explains that the dunnart's DNA is much more closely related to the Tasmanian devil, an endangered cat-sized carnivore native to Australia, than the mouse. But from a research perspective, Pask notes the similarities between mouse and dunnart run deep -- and that's why it's such an important critter.

"The dunnart is going to be our marsupial workhorse like the mouse is for placental mammals," Pask says.

Reprogramming the cell

For that to happen, Pask's team has to perfect an incredible feat of genetic engineering: They have to learn how to reprogram its cells.

To do so, they collect skin cells from the dunnart's ear or footpad and drop them in a flask where scientists can introduce new genes into the skin cell. The introduced genes are able to trick the adult cell, convincing it to become a "younger," specialized cell with almost unlimited potential.

The reprogrammed cells are known as "induced pluripotent stem cells," or iPS cells, and since Japanese scientists unraveled how to perform this incredible feat in 2006, they have proven to be indispensable for researchers because they can become any cell in the body.

"You can grow them in culture and put different sorts of differentiation factors on them and see if they can turn into nerve cells, muscle cells, brain cells, blood vessels," Pask explains. That means these special cells could even be programmed to become a sperm or an egg, in turn allowing embryos to be made.

Implanting the embryo in a surrogate mother could create a whole animal.


It took about 15 minutes to get this dunnart to sit still.

Jackson Ryan/CNET

Although such a technological leap has been made in mice, it's still a long way from fruition for marsupials. At present, only the Tasmanian devil has had iPS cells created from skin, and no sperm or egg cells were produced.

Pask's team has been able to dupe the dunnart's cells into reverting to stem cells -- and they've even made some slight genetic tweaks in the lab. But that's just the first step.

He believes there are likely to be small differences between species, but if the methodology remains consistent and reproducible in other marsupials, scientists could begin to create iPS cells from Australia's array of unique fauna. They could even sample skin cells from wild marsupials and reprogram those.

Doing so would be indispensable in the creation of a biobank, where the cells would be frozen down to -196 degrees Celsius (-273°F) and stored until they're needed. It would act as a safeguard -- a backup copy of genetic material that could, in some distant future, be used to bring species back from the edge of oblivion, helping repopulate them and restoring their genetic diversity.

Into the unknown

Underneath San Diego Zoo's Beckman Center for Conservation Research lies the Frozen Zoo, a repository of test tubes containing the genetic material of over 10,000 species. Stacked in towers and chilled inside giant metal vats, the tubes contain the DNA of threatened species from around the world, suspended in time.

It's the largest wildlife biobank in the world.

"Our goal is to opportunistically collect cells ... on multiple individuals of as many species as we can, to provide a vast genetic resource for research and conservation efforts," explains Marlys Houck, curator at the Frozen Zoo.

The Zoo's efforts to save the northern white rhino from extinction have been well publicized. Other research groups have been able to create a northern white rhino embryo in the lab, combining eggs of the last two remaining females with frozen sperm from departed males. Scientists propose implanting those embryos in a surrogate mother of a closely related species, the southern white rhino, to help drag the species back from the edge of oblivion.

For the better part of a decade, conservationists have been focused on this goal, and now their work is paying off: In the "coming months," the lab-created northern white rhino embryo will be implanted in a surrogate.


Sudan, the last male northern white rhinoceros, was euthanized in 2018.

Tony Karumba/Getty

Marisa Korody, a conservation geneticist at the Frozen Zoo, stresses that this type of intervention was really the last hope for the rhino, a species whose population had already diminished to just eight individuals a decade ago.

"We only turn to these methods when more traditional conservation methods have failed," she says.

In Australia, researchers are telling whoever will listen that traditional conservation methods are failing.

"We've been saying for decades and decades, many of our species are on a slippery slope," says John Rodger, a marsupial conservationist at the University of Newcastle, Australia, and CEO of the Fauna Research Alliance, which has long advocated for the banking of genetic material of species in Australia and New Zealand.

In October, 240 of Australia's top scientists delivered a letter to the government detailing the country's woeful record on protecting species, citing the 1,800 plants and animals in danger of extinction, and the "weak" environmental laws which have been ineffective at keeping Australian fauna alive.

Institutions around Australia, such as Taronga Zoo and Monash University, have been biobanking samples since the '90s, reliant on philanthropic donations to stay online, but researchers say this is not enough. For at least a decade, they've been calling for the establishment of a national biobank to support Australia's threatened species.

"Our real problem in Australia ... is underinvestment," Rodger says. "You've got to accept this is not a short-term investment."

The current government installed a threatened-species commissioner in 2017 and committed $255 million ($171 million in US dollars) in funding to improve the prospects of 20 mammal species by 2020. In the most recent progress report, released in 2019, only eight of those 20 were identified as having an "improved trajectory," meaning populations were either increasing faster or declining slower compared to 2015.

A spokesperson for the commissioner outlined the $50 million investment to support immediate work to protect wildlife following the bushfires, speaking to monitoring programs, establishment of "insurance populations" and feral cat traps. No future strategies regarding biobanking were referenced.



Researchers believe we need to act now to preserve iconic Australian species like the koala.

Fairfax Media/Getty

In the wake of the catastrophic bushfire season and the challenges posed by climate change, Australia's extinction crisis is again in the spotlight. Koalas are plastered over social media with charred noses and bandaged skin. On the front page of newspapers, kangaroos bound in front of towering walls of flame.

Houck notes that San Diego's Frozen Zoo currently stores cell lines "from nearly 30 marsupial species, including koala, Tasmanian devil and kangaroo," but that's only one-tenth of the known marsupial species living in Australia today.

"Nobody in the world is seriously working on marsupials but us," Rodger says. "We've got a huge interest in maintaining these guys for tourism, national icons... you name it."

There's a creeping sense of dread in the researchers I talk to that perhaps we've passed a tipping point, not just in Australia, but across the world. "We are losing species at an alarming rate," says Korody from the Frozen Zoo. "Some species are going extinct before we even know they are there."

With such high stakes, Pask and his dunnarts are in a race against time. Perfecting the techniques to genetically engineer the tiny marsupial's cells will help enable the preservation of all marsupial species for generations to come, future-proofing them against natural disasters, disease, land-clearing and threats we may not even be able to predict right now.

Pask reasons "we owe it" to marsupials to develop these tools and, at the very least, biobank their cells if we can't prevent extinction. "We really should be investing in this stuff now," he says. He's optimistic.

In some distant future, years from now, a bundle of frozen stem cells might just bring the koala or the kangaroo back from the brink of extinction.

And for that, we'll have the dunnart to thank.

Originally published Feb. 18, 5 a.m. PT.