The Perils of Cloning

By Alice Park

It was 10 years ago this week, on a warm July night, that a newborn lamb with an unique pedigree took her first breath in a small shed tucked in the Scottish hills a few miles south of Edinburgh. From the outside, she looked no different from thousands of other sheep born each summer on surrounding farms. But Dolly, as the world soon came to realize, was no ordinary lamb. She was cloned from a single mammary cell of an adult ewe, overturning long-held scientific dogma that had declared such a thing biologically impossible. Her birth set off a race in laboratories around the world to duplicate the breakthrough. It also raised the specter--however distant--of human cloning.

A decade later, scientists are starting to come to grips with just how different Dolly was. Dozens of animals have been cloned since that first little lamb--mice, cats, cows, pigs, horses and, most recently, a dog--and it's becoming increasingly clear that they are all, in one way or another, defective.

It's tempting to think of clones as perfect carbon copies of the original--down to every hair and quirk of temperament. It turns out, though, that there are various degrees of genetic replication. That may come as a rude shock to people who have paid thousands of dollars to clone a pet cat only to discover that their new kitten looks and behaves nothing like their beloved pet--with a different-color coat of fur, perhaps, or a completely different attitude toward its human hosts.

And these are just the obvious differences. Not only are clones separated from the original template by time--in Dolly's case, six years--but they are also the product of an unnatural molecular mechanism that turns out not to be very good at making identical copies. In fact, the process can embed small flaws in the genomes of clones that scientists are only now discovering. The more scientists have learned about the inner workings of the procedure that created Dolly, the more they are amazed that she survived at all.

"We are still surprised that cloning works," says Ian Wilmut, the embryologist who led the team that created Dolly. Ten years and 15 mammalian species later, the efficiency of the process is no better than it was at Dolly's birth: only 2% to 5% of the eggs that start out as clones end up as live animals. For each clone born, hundreds of others never make it past their first days and weeks, the victims of defects in development too severe to allow them to survive.

Clones are vulnerable throughout the cloning process, from their first days in a culture dish to their final moments in the womb to their first weeks after birth. (By contrast, embryos created by in vitro fertilization, which also start out in a petri dish, are pretty much home free if they make it past the first month in the womb.) Dolly, in fact, was the sole survivor of 277 cloning attempts. Clones, as the scientists who make them are fond of saying, are the exception rather than the rule.

It's not hard to appreciate why. Mammalian cloning is an intricate process involving at least three animals, hundreds of eggs, hundreds of more mature cells and not a single sperm. The key challenge is to undo the development of an adult cell--which, like all cells, contains in its DNA the genetic blueprint of the entire organism--that has been programmed or "differentiated" to be one kind of cell (skin or bone or nerve) and no other kind. Somehow, scientists must trick this mature, fully developed cell into resetting its genetic clock so that it can begin life anew as an embryo.

The process by which that is achieved is called nuclear transfer. The first step is to remove the nucleus from an egg and replace it with the nucleus of an adult cell (in Dolly's case, a cell from a ewe's udder). The two components are electrically fused and chemically activated to trick the hybrid cell into dividing like an embryo. Not surprisingly, the process doesn't always go right. "I call it a lottery," says Wilmut. "Even if you use the same method as consistently as you can, you may get some clones with severe abnormalities and some that have only minor ones."

The most common defect--seen across most of the species that have been cloned so far--is a condition known as large-offspring syndrome. Those clones are born larger than normal and have trouble breathing in their first few weeks. Most of the surrogates that gave birth to them experience prolonged pregnancies and sluggish, difficult labors, which may have something to do with their distended and enlarged placentas. Some of Wilmut's cloned sheep were born with incomplete body walls, with muscles and skin around their abdomen that failed to properly join. Other scientists have reported abnormalities in kidney and brain function. In still other clones, the heart does not develop normally, and the walls that are supposed to separate fresh blood from deoxygenated blood do not form.

The good news, as far as cloning's future is concerned, is that those problems seem to be limited to the clones and are not passed on to the next generation. When clones mate with ordinary animals, their offspring are created by the natural merging of egg and sperm--not by the reprogramming of a mature cell--which may erase any reprogramming errors in the clone. The proof is that Dolly gave birth to five healthy lambs. Cloned cows, pigs and mice are also bearing normal offspring. But when clones mate with other clones, all bets are off. Mice created this way appear to accumulate more abnormalities with each generation.

Most of the errors in reprogramming, scientists say, can be traced to a process known as DNA methylation. During normal development, molecules called methyl groups attach themselves to DNA in precisely timed patterns that regulate which genes are expressed at which times. During cloning, however, those patterns are not always reconstructed in exactly the same way. It's a bit like taking all the words in a novel, jumbling them up and then trying to re-create the original book, putting sentences, pages and chapters back in the right order. The chances of that happening with 100% accuracy are minuscule, which helps explain why cloning is so inefficient. Rudolf Jaenisch, a geneticist at the Massachusetts Institute of Technology, estimates that 4% to 5% of the genes in a cloned animal's genome are expressed incorrectly--probably because of faulty methylation. "If you reprogram, it affects the whole genome," he says. "From what we know, I would argue that cloned animals cannot be normal. They can be closer to normal, but not normal."

The mammalian body is surprisingly forgiving and can often compensate for minor programming errors. That's why some genetic changes in clones may not have any measurable functional effects on the animals.

Dolly seems to have been one of those lucky ones. She showed just two signs of her unusual provenance. One was the arthritis she developed at an early age. The other was shortened telomeres in her cells. Telomeres are bits of DNA that sit at the ends of chromosomes and serve as a biological clock chronicling a cell's age. In general, the shorter the telomeres, the older the cell. Dolly, a clone of a 6-year-old ewe, had cells whose telomeres were closer in length to those of her biological mother than to those of a baby lamb. We will never know, though, whether her shortened telomeres would have shortened her life. In 2003 Wilmut and his team decided to put Dolly to sleep after she developed lung cancer caused by a viral infection common among sheep. An autopsy revealed that she was otherwise normal.

But the fact that clones have defects--however minor--only bolsters the arguments that scientists have made against human cloning. Based on his studies of the faults introduced by reprogramming, Jaenisch, for one, thinks human cloning is now out of the question. "I think we cannot make human reproductive cloning safe," he says. "And it's not a technological issue. It's a biological barrier. The pattern of methylation of a normal embryo cannot be re-created consistently in cloning."

But Jaenisch and Wilmut both see a role for cloning in treating human diseases--and perhaps someday conquering some of man's most intractable conditions. Wilmut and others have already created cow, sheep and pig cells genetically engineered to produce a particularly beneficial human protein, then cloned those cells to generate live animals able to make copious amounts of the target protein in their milk. It may be another 10 years or more before that approach yields anything safe and reliable enough to be used in real patients, and there is no guarantee that it will ever be successful. But as Wilmut points out, nobody thought Dolly was possible until she made history that warm July night 10 years ago.