WELCOME to the farm where the mating season never comes. Here the ewes, cows and sows form a celibate sisterhood, yet somehow they manage to carry on the business of procreation anyway. Without testosterone-pumped suitors. And sans sperm. Each year, they spawn another generation, all female, all with impeccable genetic credentials, because long ago scientists helped to purge their lineage of destructive mutations. This farm is the futuristic vision that virgin births were supposed to build.
That was the view from 1957, when geneticist Richard Beatty of the University of Edinburgh published a book on the scientific study of parthenogenesis – the development of an embryo without the help of sperm. He argued that creating generation after generation of virgin-born mammals would not only help to improve and accelerate the breeding of livestock (producing big males for the sake of a little sperm slows down breeding), but would yield new insights into disease, genetics and the function of sex itself.
Today, the pursuit of fatherless propagation by farm animals seems bizarre. But before words like “genome project” and “DNA diagnostics” were ever uttered, biologists were betting that parthenogenesis would help them to change the world.
Advertisement
The search for a mammalian virgin mother has gone through many incarnations since those days: from a great hope of agriculture to the obsession of top reproductive scientists, from an impossible dream to the reality of sex-free replication with the help of cloning technology. And although the world is unlikely to change as dramatically as Beatty once envisaged, it may soon be possible to genetically engineer mammals that reproduce purely through parthenogenesis.
Of course, asexuality was hardly a new idea. The ability of microbes, insects and plants to reproduce without a hint of carnality was well known. But pioneering experiments in the 1920s and 30s had raised a new question: just how far up the evolutionary tree had parthenogenesis managed to climb?
Good tidings
Some evidence suggested it might have reached the very top. żěè¶ĚĘÓƵs discovered that in the ovaries of vertebrates, even in some mammals, eggs would occasionally begin to develop without a whiff of sperm about. There were even isolated reports that these parthenogenetic embryos could develop into full-blown animals. “This work turned out to be dead wrong,” says Christopher Graham of Oxford University, whose own studies of parthenogenesis in mice began in the 1970s. “But they made for very exciting, optimistic times.”
No episode reflected this optimism better than when geneticist Helen Spurway from University College, London, unintentionally sparked off a serious hunt for a human virgin birth. In 1955, she argued that virgin births might be discovered most easily in humans. She said that women who suspected they had parthenogenetic children might be encouraged to step forward if they knew that their claims could be vindicated beyond a reasonable doubt.
She pointed out that science made some clear predictions for a human virgin birth. Biblical precedent apart, the infant born of mammalian parthenogenesis should be female. Mammalian sex is determined by the X and Y chromosomes. Males are XY, females XX. Since a mother’s egg can only contribute X chromosomes, the child must be XX. And this child could only have genes that her mother also possessed.
The Sunday Pictorial newspaper repeated Spurway’s musing and called for any woman who believed she had given birth to such a child to step forward. Accordingly, 19 women were marched to a blood-typing specialist, one S. Balfour-Lynn, to substantiate their claim. Most were easily discounted after a few tests showed a blood protein mismatch between mother and daughter. However, an impressed Balfour-Lynn reported in the medical journal The Lancet that one daughter matched her mother on 14 different genetic criteria. This mother’s claim must not only be considered seriously, Balfour-Lynn concluded, “but it must also be admitted that we have been unable to disprove it”.
If he’d had the knowledge of modern immunologists at his finger tips, Balfour-Lynn would have been far less impressed. For instance, his potential virgin mother had rejected a skin graft from her daughter after several weeks. This was slow for tissue rejection and an “obscure” finding to the esteemed blood specialist. But it is now clear that rejection, however long it took, showed that the daughter had genes the mother did not, and meant sudden death to a claim for parthenogenesis.
Evidence for parthenogenesis in other mammals was also unravelling. For example, many laboratories were unable to reproduce perhaps the most impressive claim of the 1940s, the production of parthenogenetic rabbits.
But there was enough good news to keep some top scientists interested. In the late 1950s, American and Russian researchers tracked down populations of all-female lizards that reproduced exclusively by parthenogenesis. Meanwhile, researchers in the US made a discovery that brought virgin births closer to home, and closer to the dinner table – parthenogenetic turkeys.
Beginning in 1952, Marlow W. Olsen and his colleagues at the US Department of Agriculture in Beltsville, Maryland hatched 1100 parthenogenetic turkeys. The group found that a talent for fatherless development was strangely common in some breeds of turkey, where up to 40 per cent of eggs developed without sperm. About 1 in 5 of the birds that hatched was fertile. Perpetuating the lineage by parthenogenesis was a little difficult, however, because all the offspring were male. That’s because sex in birds is determined by a different set of chromosomes to mammals, called Z and W. Males are ZZ and females ZW. Because of the way the eggs develop, parthenogenetic embryos have only one type of sex chromosome, and of the two possibilities only ZZ embryos can survive.
The turkey’s odd reproductive ability is often dismissed as a quirk of extensive inbreeding during its domestication. But Tom Savage, a turkey researcher at Oregon State University in Corvallis, points out that it might be useful in the wild. If female turkeys are inseminated with semen low in sperm, for example, the incidence of parthenogenetic egg production goes up. The incidence also increases if the birds are infected with viruses. This suggests that virgin births are a response to any threat to the she-turkey’s chance of successful sexual reproduction.
In fact, another instance of parthenogenesis that produces only male offspring was reported just over a year ago by David Chiszar at the University of Colorado, Boulder, and Gordon Schuett of Arizona State University West in Phoenix (This Week, 27 September 1997, p 23) – in a virgin timber rattlesnake. Since then, the researchers say they have found evidence that other snakes may occasionally adopt the same type of reproduction, as might the green iguana and – appropriately enough – certain species of the Basilisk lizard, which is sometimes called the “Jesus Christ lizard” for its ability to walk on water. Chiszar and Schuett argue that parthenogenetic reproduction has played an important role in the evolution of these reptiles. “If a female gets isolated and there is little chance of her mating, this is plan B,” says Schuett.
Even with the few examples of vertebrate parthenogenesis available in the 1970s, some researchers were undaunted. If our feathered and scaly friends could pass on their genes without bothering with sex, surely mammals could pull off the same trick. Hopes rose when Graham and others showed that mouse eggs could be stimulated to begin spermless development in test tubes by treating them with chemicals, enzymes, temperature shifts or electric shocks, which meant the process could be studied in the laboratory. “This began a really enormous effort to produce parthenogenetic mice,” remembers Davor Solter of the Max-Planck Institute for Immunology in Freiburg, Germany.
But progress certainly wasn’t easy. “No matter what we did, the embryos stopped developing during gestation,” says Azim Surani of the Wellcome and Cancer Research Campaign Institute of Developmental Biology and Cancer Research in Cambridge. “Understanding why became an obsession.” One possibility was that most parthenogenetic embryos inherit a deadly combination of genes.
This poor genetic legacy would not be surprising because of the effect of lethal recessive genes. These genes are only deadly if the embryo inherits two copies, one from each parent – usually a rare event. But parthenogenesis can be thought of as a highly inbred mating between the egg and itself. So in parthenogenesis, the probability of inheriting any recessive lethal trait is much higher than for the offspring of two parents. This was both good and bad news for parthenogenetic enthusiasts. It meant that if generations of parthenogenetic mice could be produced, their lineage would quickly be cleansed of any deadly traits. But equally, if a strain contained several recessive lethal mutations, a parthenogenetic lineage wouldn’t have much chance of succeeding.
So the researchers devised a clever new way to “mate” genetic material from different animals, which gave them more genetic freedom than parthenogenesis. A recently fertilised mouse egg contains mum and dad’s chromosomes in separate compartments called pronuclei. Both research groups learnt how to remove pronuclei and switch them with pronuclei from other embryos. This meant they could “mate” pronuclei from two unrelated female mice, or two unrelated male mice, which should have eliminated any problems with inbreeding.
Lost dream
What they discovered seemed to sound the final death knell for mammalian parthenogenesis. Only when pronuclei from both sexes were present could the embryos complete development. This led to a remarkable idea: that male and female mammalian chromosomes are somehow marked or “imprinted” differently, so that either genome by itself is incomplete (“Hidden inheritance”,żěè¶ĚĘÓƵ, 28 November, p 27). They reasoned that organisms with a gift for parthenogenesis must lack imprinted genes. Sure enough, among vertebrates, imprinting appears to be exclusive to mammals.
Solter also thought imprinting might explain why the cloning of adult mammals hadn’t been accomplished. In cloning, a single cell from an animal is persuaded to begin development again to produce the original organism’s genetic twin. But Solter reasoned that if essential parental imprints were lost during development, cloning would become “biologically impossible”.
Now, after the successful cloning of Dolly the sheep in 1997, followed by the cloning of adult cows and several dozen mice reported in the past year, it is clear that Solter was mistaken – but only in part. Researchers now believe that some adult cells probably retain their imprints throughout development. In that sense, Dolly and company didn’t overcome the sexual imperatives introduced by imprinting. They simply borrowed imprints from the previous generation.
With mammalian parthenogenesis a lost dream, biologists like Surani and Solter began to study its nemesis – imprinting. Beginning in 1991, researchers uncovered the molecular basis of imprinting: certain genes are switched on in sperm, but not in eggs, and vice versa. There are now more than two dozen such sex-biased genes known in mammals. This genetic division of labour thwarts parthenogenesis, but that’s probably not why it evolved. Some think that the evolution of the placenta somehow required the activity levels of certain genes to be drastically reduced and that imprinting was a way to achieve this. Because this tissue is a purely mammalian invention, this would explain why other vertebrates didn’t evolve imprinting.
Another theory, developed by David Haig of Harvard University and his colleagues, is that imprinting is the result of an evolutionary battle between the sexes. When it comes to the growth of a mammalian fetus, each parent has a different agenda. It is in the father’s best interest for his offspring to grow large and use up the mother’s resources, so they won’t be wasted on another male’s offspring. Such favouritism doesn’t benefit a mammalian mother, however. To her, the unlimited growth of the fetus could be deadly. She is better off treating each embryo equally.
The result, Haig says, is a genetics arms race, with paternal genes beefing up the offspring and maternal genes attempting to keep growth in check. According to this theory, the development of parthenogenetic embryos fails because they contain a double copy of maternal genes without paternal rivals to act as a counterbalance. “This is a tug of war,” says Rudolf Jaenisch of the Massachusetts Institute of Technology. “So if one side stops pulling, things are thrown into a sudden crisis.”
Some genetic evidence bolsters this theory. For example, the gene for insulin-like growth factor 2, Igf2, is expressed in sperm and promotes embryonic growth, while the egg expresses Igf2r, a gene that codes for a protein that inhibits the growth factor. So when researchers genetically engineer mice without Igf2r from their mum, the embryo grows excessively large and dies before birth, while mice lacking the growth factor from their dad are runts. Intriguingly, embryos with neither gene are of normal size and fertility. Jaenisch explains this using the tug-of-war analogy; the battle can end peacefully, so long as both sides drop the rope at the same time. Which leads Jaenisch to a remarkable conclusion. If biologists reverse millions of years of evolution by eliminating imprints from mammalian genes, development should proceed without a hitch.
Brains and brawn
Jaenisch has already tested his theory using mouse embryonic stem (ES) cells, which have the capacity to become any tissue when injected into a normal, developing mouse embryo. The researchers stripped these ES cells of all imprinted markers. With data yet to be published, the researchers show that when these cells are added to a developing embryo they transform into many tissues, including brain (the formation of which requires maternally expressed genes) and muscle (which requires paternal genes).
But in that experiment, the ES cells were in the company of normal, imprinted cells in the embryo which may have assisted their development. Jaenisch now wants to go one step further by using the unimprinted ES cell in a cloning experiment like the one that produced Dolly. Because the resulting animal would have come from a single cell, it would have to develop in the complete absence of imprinted genes. If this creature is viable, this would be the strongest proof possible that Haig’s theory is correct.
And if that experiment works, Jaenisch will be only one step away from fulfilling Beatty’s vision of producing generation after generation of parthenogenetic mammals. To make the lack of imprinting heritable, the researchers would only have to cripple the as-yet undiscovered proteins that imprint the genes in the developing sperm and egg. “Then you would have animals that could replicate through parthenogenesis forever,” says Jaenisch.
- Further reading: The production of parthenogenetic mammalian embryos and their use in biological research by C. F. Graham, Biological Reviews, vol 49, p 399 (1974)
- DNA Methylation and Imprinting: Why bother? by R. Jaenisch, Trends in Genetics, vol 13, p 323 (1997)
