by Ben Mezrich
The largest flying animal ever was Quetzalcoatlus northropi at 200 kg. It is estimated to have been able to fly for ten days at an altitude of 4.5 km and speed of 129 km/hr. While a conventional pig can get up to 300 kg, the smallest adult mini-pig is 25 kg. So, plenty of room between the 25 kg minimum for pigs and 200 kg maximum for flight. Bat morphogenetic pathways could be moved over to the pig (or elephant) genome to elongate their arms. Another option would be making a bat that looks like a pig—indeed there is a “hog-nosed bat” that would be a good starting point.
Animal swim bladders can generate gas mixtures up to 75 percent oxygen, differing greatly from our normal atmosphere of 21 percent oxygen. Furthermore, many microbes make hydrogen gas as their major metabolic output. So, if an animal grew large, thin bladders filled with hydrogen (rather than air or oxygen), then it could fly like a dirigible. But it might live in mortal fear of thorns and winter static.
So what about flying Mammoths? The Cyprus dwarf elephant (Elephas cypriotes) went extinct thirteen thousand years ago. The adults had a weight of only 200 kg. This is at the edge of the Quetzalcoatlus limit, and wings could add extra mass. Nevertheless, this is tiny compared to the largest elephant family member (Palaeoloxodon namadicus), which weighed in at a whopping twenty-two tons.
Dwarf elephants are an example of a more widespread phenomenon in which many large animals evolve toward miniature versions on islands—while small animals tend to become much larger on islands. Being small is not just about being cute (or flyable), it also tends to track with the rate of development and aging. Mice weigh about one gram at birth and take only twenty days to gestate from fertilized egg to birth. Elephants take twenty-two months to gestate and emerge at 100 kg. Wouldn’t it be great for research if the elephant egg-to-birth process took only twenty days?
Many genetics experts (especially those studying natural human DNA variations) will say that body-size traits do not have simple DNA causes or easy manipulation. But the complexity of thousands of natural genetic and environmental factors does not exclude the option of a single manipulated gene having huge impact. For example, increasing one gene for human growth hormone (or the receptor for human growth hormone) results in the largest and smallest human beings and is used clinically to treat a variety of diseases (for example, Turner syndrome, chronic renal failure, Prader–Willi syndrome, intrauterine growth retardation, idiopathic short stature, and AIDS muscle wasting).
An interesting phenomenon kicks in at this point, called Peto’s paradox, which is that even though elephants and cetaceans (blue whales at 180 tons) are 100 million times larger than mice and similar creatures (1.8-gram adult Etruscan shrews), the larger animals seem to be much more cancer resistant and resistant to aging. Each time a cell divides (replicates), there is a chance that it will mutate in a way that will make it start dividing without limit, which often is the start of cancer. So 100 million times the number of cells could/should mean that much higher chance of cancer. To some this can be rationalized by the needs of a particular ecosystem niche, such that small, easy prey (like mice) have large litters (twelve) and short gestations (twenty days) so Darwinian selection for long life is scant.
There are many ways to get short life, but what are the mechanisms behind the superlong lives? And can humans benefit from such knowledge? One intriguing new observation from Ting Wu’s lab that might intersect with elephant biology is that certain regions of mammalian genomes, appropriately called ultra-conserved elements (aka UCE), are nearly unchanging in normal development and evolution, but highly changeable leading up to cancer. Ting’s lab is testing if these UCEs can be harnessed to reduce mutations leading to cancer and aging or radiation health issues in space.
While in the early stages of planning radically new forms of life, we should keep in mind humane treatment of humans and animals. Even though humans and our works are part of nature, “natural” could be defined as prehuman. Natural is not necessarily kind or benignant, as seen herein in chapters 24 and 26 with the elephant endotheliotropic herpes virus (EEHV) or human smallpox, naturally occurring diseases. If we make these viruses extinct, the animals and humans can, arguably, lead happier, longer lives. The same may go for UCEs, pain pathways, and so on. Ideally, all of the intermediate steps between here and there will also be humane. An interesting question is whether such genetically modified organisms will be regulated in the same manner as “transgenics,” which are banned in some countries and from “organic foods.” The key difference is that transgenics introduce whole genes from distant species, while “cis-genics” are smaller changes that can happen in normal mutations and interbreeding species. More than thirty such cis-genic GMOs have been approved by the USDA as exempt from the usual transgenic definitions. This may be quite relevant to preventing extinction of elephants and other species and making their lives more pleasant.
Another frequently asked question is: Can we go farther back than the seven-hundred-thousand-year record, so far, for ancient DNA, and in particular, harvest the DNA of dinosaurs? Some information in ancient proteins seems to last much longer than that of ancient DNA. Proteins are studied by electron microscopy, immunological assays, mass spectrometry, and Fourier transform infrared spectra. The oldest such evidence is from 195-million-year-old ribs of a sauropodomorph dinosaur, Lufengosaurus, reported in 2017. (For perspective, Tyrannosaurus rex lived relatively recently—a mere 67 million years ago.) Another option for getting very ancient information is based on comparing the genomes of living species. The oldest of all is from scientists Betül Kaçar, Lily Tran, Xueliang Ge, Suparna Sanyal, and Eric Gaucher, who in 2016 resurrected a 700-million-year-old version of a protein called EF-Tu, inferring its protein sequence from comparisons of DNA from living species.
Could something that looks and acts like a dinosaur (ideally a herbivore) be a few years away? In comparing the genomes and developmental biology of living species of birds and reptiles, with dinosaurs in mind, a good starting point would be the ostrich. The ability to grow teeth and front clawed hands is not far off in terms of developmental potential, either by small mutations in the bird genes or by replacing genes lost by bird ancestors with the equivalents found in living species, such as alligators. Similarly, we are learning how to make featherless bodies with long tails. The scales on the feet of an ostrich are of similar developmental origins to the feathers. Many mutations interconvert related tissue types—for example, antennae convert into legs in fruit fly mutants (in a gene aptly called antennapedia). Old adult mouse cells morph into young embryolike cells by inducing only four genes. These examples argue that mutations could be found and optimized to turn all ostrich-feather-producing cells into scale-producing cells.
In contrast to dinosaurs, we have excellent information on DNA from Woolly Mammoths (which lived 5 million to five thousand years ago). Like much of modern science, this is a team effort, often including team members who have never communicated directly. We are currently using four Mammuthus primigenius genomes, one from a 44,800-year-old in northeastern Siberia, one that died 4,300 years ago at Wrangel Island (both from Eleftheria Palkopoulou, Swapan Mallick, Pontus Skoglund, Jacob Enk, Nadin Rohland, Heng Li, Ayca Omrak, Sergey Vartanyan, Hendrik Poinar, Anders Götherström, David Reich, and Love Dalén), as well as two from twenty thousand and sixty thousand years ago (from Vincent Lynch, Oscar Bedoya-Reina, Aakrosh Ratan, Michael Sulak, Daniela Drautz-Moses, George Perry, Webb Miller, and Stephan Schuster). We compare these with three Asian elephant genomes and put them in the context of the more distant relative (and more carefully annotated) African elephant reference genome. We are looking for genetic variations (unique to Mammoths) present in four out of four Mammoth genomes and zero out of three Asian elephant genomes.
Kevin Campbell, Alan Cooper, and their coworkers characterized the Mammoth blood protein called hemoglobin. Starting with the Elephas maximus HBB/D gene (from Opazo, Sloan, Campbell, Storz, below) more than 2,100 bp long, only three changes are seen and only these were needed t
o resurrect the properties of cold tolerance in oxygen exchange. These three changes are underlined and bold, A to G, G to T, and G to C. Despite the fact that 80 percent of the region (in lowercase) does not code for the hemoglobin protein, all of the changes are the coding regions (uppercase and in brackets). You’ve made it all the way to this point in the book and deserve to see some real Mammoth DNA (in contrast to fake dinosaur DNA mentioned herein in chapter 9).
ttctgggcctcagtttcctcatttgtataataacagaattggagagtaaattcttaagaggcttaccaggctgtaattctaaaaaaaatgcataaataaacttgccaaggcagatgtttttagcagcaattcctgaaagaaacgggaccaggagataagtagagaaagagtgaaggtctgaaatcaaactaataagacagtcccagactgtcaaggagaggtatggctgtcatcattcaggcctcaccctgcagaaccacaccctggccttggccaatctgctcacaagagcaaaaagggcaggaccagggttgggcatataaggaagagtagtgccagctgctgtttacactcact-tctgacacaactgtgttgactagcaactacccaatcagacacc[ATGGTGAATCTGACTGCTGCTGAGAAGACACAAGTCACCAACCTGTGGGGCAAGGTGAATGTGAAAGAGCTTGGTGGTGAGGCCCTGAGCAG]gtttgtatctaggttgcaaggtagacttaaggagggttgagtggggctgggcatgtggagacagaacagtctcccagtttctgacaggcactgacttcctctgcaccstgtggtgctttcaccttcag[GCTGCTGGTGGTCTACCCATGGACCCGGAGGTTCTTTGAACACTTTGGGGACCTGTCCACTGCTGACGCTGTCCTGCACAACGCTAAAGTGCTGGCCCATGGCGAGAAAGTGTTGACCTCCTTTGGTGAGGGCCTGAAGCACCTGGACAACCTCAAGGGCACCTTTGCCGATCTGAGCGAGCTGCACTGTGACAAGCTGCACGTGGATCCTGAGAATTTCAGG]gtgagtctaggagacactctattttttcttttcactttgtagtctttcactgtgattattttgcttatttgaatttcctctgtatctctttttactcgactatgtttcatcatttagtgttttttcaacttataccattttgtattacttttctttcaatattcttccttttttcctgactcacattcttgctttatatcatgctctttatttaatttcctacgtttttgctcttgctctccctttctcctagtttccttccctctgaacagtacccaaattgtgcataccacctctcgtccactatttctgcactggggcaaatccccacccctcctccatatgagggttggaaaggactgaatcaaagaggagaggatcatggtgctgttctagagtatgtgattcatttcagacttgaaggataacttgaataatataaaatcaggagtaaatggagaggaaagtcagtatctgagaatgaaagatcagaaggtcatagacgagatggggagcagaagttactaagaaactgaccattgtggctataattaatcacttaattagttaattaatatgtttgttatttattcacgtttttcattttggtgggagtaaatttgggctagtgtgtgggcaacataaatgggtttcaccccattgtctcagaggccaagctggattgctttgttaaccatgtctgtgtatgtatctacctcttcccca-tag[CTCCTGGGCAATGTGCTGGTGATTGTCCTGGCCCGCCACTTTGGCAAGGAATTCACCCCAGATGTTCAGGCTGCCTATGAGAAGGTTGTGGCAGGTGTGGCGAATGCCCTGGCTCACAAATACCACTGA]gatcctggcctgttcctggtatccatcggaagccccatttcccgagatgctatctctgaatttgggaaaataatgccaactctcaagggcatctcttctgcctaataaagtactttcagctcaactttctgattcatttattttttttctcagtcactcttgtggtgggggaagttcccaaggctctatggacagagagctcttgtgccttataggaaaagttcaagggaaattggaaaataaagggaaccatacacagatattaatgggaacaattctacttcaaaggcataaagattgggaaggtttggcaaataggatactggtactacagggattccatgggcctcaggcctaagacatagccccagggctaactttcagattcaattccagaaattactcacaaaataatgga
We could edit the elephant genome one DNA base pair change at a time (for example with CRISPR), or we could make all three changes at once by swapping in the whole 2,100+ bp gene—or even a million base pairs at once. At least three groups (led by teams at Harvard, the J. Craig Venter Institute, and New York University) are already engineering genome pieces in this “mega-base” range. Because of the larger extent and easier access to radically novel sequences, this is often called genome “writing” rather than “editing.” As few as three thousand of these mega-base segments could cover the whole elephant genome, allowing nearly perfect conversion to a Mammoth genome. If all we want is a cold-resistant elephant, then we might want to make only a few dozen small changes, but as the technology gets cheaper we will probably try many versions, including some that are with obsessive-compulsive levels of molecular realism. If we get that good at genome engineering, what about the unsequenced part of genomes, plus the epi-genomes and microbiomes?
Experts in ancient DNA reading have been dismissive of the prospects for sequencing a whole genome from ancient DNA. This is a fairly reasonable position today, since no mammalian genome has been completely sequenced even for living species (even for humans), and ancient DNA is shredded into millions of pieces per cell by radiation, which make it even more daunting. Beth Shapiro said in her 2015 book, How to Clone a Mammoth, “Because we cannot know the complete genome sequence of an extinct species, synthesizing a complete genome from scratch would not be an option.” Svante Pääbo said in his 2014 New York Times op-ed piece, “Neanderthals Are People Too”: “Since the DNA preserved in ancient bones has degraded into short pieces, we cannot tell from which copies of these repeated sequences they come and so we cannot reconstruct exactly how they were arranged.”
But here is another great opportunity for thinking out of the box. Visualize the machines that cleanly punch a two-dimensional cardboard image into a million jigsaw puzzle pieces, all still in place and readable. The machine then shakes up the pieces into a box for sale, and the puzzle becomes very hard to assemble (and then read). So if you can read the puzzle after cutting but before shaking, then maybe reading is easy. We hope to try this idea on a variety of ancient genomes soon, using an amazing new method developed in Ting Wu’s laboratory called “Oligopaints” (and related in situ sequencing methods). We can lock the cut pieces in their original places by using chemical “fixatives” and restraining polymers and then scan the clear three-dimensional cells using fluorescent confocal microscopy.
Okay. But haven’t we lost the Mammoth microbiomes, and isn’t our ability to understand the biomes of viruses, bacteria, and fungi in our bodies too primitive? It should be noted that we have been engineering these invisible communities since the 1500s, with the dawn of smallpox vaccines in China. Today, the engineering of body ecosystems has become so advanced that companies have been founded around the process, such as Seres, SynLogics, AOBiome, Fitbiomics, and Holobiome. We study and use the microbiomes of modern elephants that eat and play in snow as well as other herbivores.
Finally, isn’t the epigenome even more perplexing than the missing genomic DNA and the microbiome? Well, we can read important parts of the epigenome via the methylated cytosine bases that persist in ancient Mammoth DNA in a variety of body tissues. We also can leverage the epigenome of Asian elephants, which were probably genetically interfertile with Mammoths, just as breeds of dogs with wildly different traits (for example, nine-hundred-fold range in body size) have compatible epigenomes. Whether using genome writing or editing, we change only a very small fraction of the elephant genome to make it identical to Mammoth DNA (for example, the three out of 2,100 bases in the example above)—and we give plenty of time for the epigenome to spread onto the new DNA in the cells. Also, much of the epigenome is reset during the moving of the engineered genome into the pluripotent embryo or eggs en route to making a fetal GMO elephant.
AFTERWORD: MAMMOTH PLUS BY STEWART BRAND
Reviving and restoring Woolly Mammoths—and their climate-stabilizing Mammoth steppe—is the most spectacular wildlife project that Ryan Phelan and I have taken on for our California nonprofit called Revive & Restore, and thanks to George Church’s marvelous team, it is the furthest along in terms of actually editing genes from an extinct species into the genome of a living relative. What they are doing is brilliant, breakthrough science, and therefore exactly the kind of high-visibility, proof-of-concept example that will show conservation biologists and the general public what a potent new toolkit biotech is bringing to wildlife conservation.
Still, restoring Woolly Mammoths all the way back to life and to the wild will take many decades, probably most of this century. Identifying and editing all the right genes for the first round of Mammothlike embryos will take time. Developing a successful artificial uterus will take time. Rearing by Asian elephant parents will take time. An elephant generation takes fifteen years from a newborn female to its sexual maturity and then another twenty-two months to the first daughter. Acclimatizing to the far north will take time (though Asian elephants already love snow, as can be seen at a zoo in Ontario). The multiple stages of release to the northern wild will take time. (Russia? Canada? Both?) Each step of the way will be thrilling news. Each ponderous step.
Easier de-extinctions are also under way at Revive & Restore. If ambitiously funded, the first proxy passenger pigeons could be alive as early as 2022. Since they are sexually mature in just seven months, there could be enough birds to start flocking into the wild by 2032. Along the way, an extinct grouselike bird of the American East Coast called the heath hen will probably be brought back to life as a pioneer species to develop the
capability for other birds. (Heath hens are genetically close enough to domestic chickens to build on the sophisticated primordial-germ-cell technology invented for chickens.)
Other extinct species are leading candidates for revival. Tasmania might be able to welcome back its apex predator, a marsupial wolf known as the Tasmanian tiger—hunted to extinction in the 1930s. New Zealand’s famous ostrichlike moas might make a comeback. In Europe, the impressive mother of all cattle, the aurochs, which has been extinct since 1627, could return. The entire northern Atlantic Ocean was once fished by a flightless, penguinlike bird, the great auk, until the last ones were killed by 1852. They might be revived via their close relative, the razorbill. Other candidates in North America are the colorful Carolina parakeet (extinct by 1918) and the presumed-extinct “Lord God” bird, the ivory-billed woodpecker. The DNA for all these species is well preserved in museum specimens.
De-extinction is dramatic, but it represents only a small part of the benefits that genomic technology can bring to wildlife conservation. Many remnant populations of animals in the wild and in captive breeding programs are facing what is called an “extinction vortex,” as inbreeding forces them into an accelerating loss-of-fitness spiral. America’s most endangered mammal, the black-footed ferret, might be approaching that situation. That’s why Revive & Restore is exploring with U.S. Fish & Wildlife, the San Diego Frozen Zoo, and the biotech company Intrexon the possibility of cloning back to life two ferrets whose tissue was cryopreserved thirty-five years ago. When revived and breeding, they will enrich the ferret gene pool by increasing the number of founders of the current population from seven to nine. Yet further genetic diversity might be mined from museum specimens that harbor healthy gene variants (alleles) now missing from living ferrets. If this approach works, it could be applied to a variety of endangered species that need their gene pools restored to healthy variability.