Break no eggs

Behind the scenes of an astonishing leap in embryonic stem cell science

This mouse grew from an embryo containing adult cells that had been reprogrammed to an embryonic state. Photos: Sam Ogden	Marius Wernig (left) and Alexander Meissner were among five lead authors on the Nature paper.

As an undergraduate at the Institute for Molecular Pathology in Vienna, Konrad Hochedlinger was thinking big.

The young Austrian scientist was determined to devote his entire career to solving one of biology’s most fundamental and difficult questions.

Unfortunately, he wasn’t sure what that question was.

“All I knew for certain was that I did not want to spend my life making small contributions to areas we already knew lots about,” says Hochedlinger, a faculty member at Harvard Medical School whose lab is at Massachusetts General Hospital (MGH). “I wanted to pursue an aspect of biology that was absolutely new—and vital.”

But finding the next big thing before it gets big requires not only ambition but a generous helping of luck. Fortunately for Hochedlinger, luck struck.

At Whitehead, Konrad Hochedlinger studied how, in cloning, an egg reprograms the nucleus of an adult cell.

One day in 1999, he attended a packed university lecture by Whitehead Member Rudolf Jaenisch. The message couldn’t be clearer: In the world of developmental biology, everything had changed.

Three years earlier, Dolly the sheep had been cloned, a feat that contradicted the scientific mainstream thinking of the time. Dolly was created through a process called nuclear transfer, in which the nucleus from a single skin cell was inserted into an egg that had been stripped of its own nucleus, and then coaxed into developing into an embryo. Dolly was the genetic twin of the sheep that had donated the skin cell.

But how did this happen? How did the egg take a fully mature cell and send it back down its ancestral lineage to that developmental moment when it was once again a blank slate, devoid of all memories of ever having been skin?

What, exactly, was going on inside that skin-cell nucleus as the egg turned back the clock and reprogrammed it into an embryo?

That, Jaenisch told the audience, was one of the great biological questions of our time.

“I knew immediately that I was going to devote my career to answering that question,” says Hochedlinger.

Hochedlinger moved to the United States and joined Jaenisch’s lab at Whitehead Institute, first as a graduate student and then as a postdoctoral researcher. He was a lead author on a number of key papers that explored the dynamics of nuclear reprogramming, including the first proof-in-principle in mice of using somatic cell nuclear transfer for therapeutic purposes. In 2005 he left to start his own lab at MGH.

Then came the shocker.

A tale of four factors

For embryonic stem cell research, 2006 was the best of years and the worst of years.

The best came at a meeting of the International Society for Stem Cell Research in June 2006, when Shinya Yamanaka of Kyoto University reported that it’s possible to take a mature mouse skin cell and do with it exactly what the scientists had done with the donor cell that created Dolly a decade earlier.

The genome of a skin cell can be manipulated to reverse its developmental clock, resulting in a cell identical to an embryonic stem cell. [view graphic] Image: Christina Ullman

Without the egg.

Through long and arduous genomic screenings, Yamanaka had found four genes belonging to a category called “transcription factors” that were excellent candidates for reprogramming a cell. These factors—Oct4, Sox2, c-Myc and Klf4—are gene master regulators, meaning that they preside over large groups of genes. Manipulating any factor results in a cascading effect through any number of genetic networks.

Using a gene therapy technique, Yamanaka introduced additional copies of each of the four factors into mature skin cells using viral vectors—tiny synthetic viruses that shuttled the genes right into the chromosomes. (See illustration at right.)

Yamanaka found that the combined activation of the four factors caused the skin cells to de-evolve back into cells that bore a striking similarity to embryonic stem cells.

Rethinking reprogramming

For years, creating embryonic stem cells without embryos like this had been a major goal for Jaenisch. He had concluded that using human embryos to generate embryonic stem cells for actual medical use would never be practical, both for technical reasons and because of societal concerns about such cells.

This is your brain Well, no. The above cells are motor neurons, the nuts and bolts of your central nervous system. But these particular human cells will never have the pleasure of helping someone walk, swallow or breathe. Instead, these motor neurons started out as human embryonic stem cells and were subsequently cultivated into neurons by Maya Mitalipova, a member of the Jaenisch lab and director of Whitehead’s human embryonic stem cell facility. Growing neurons out of human embryonic stem cells is an extraordinary feat in itself, but a team consisting of Whitehead Members Rudolf Jaenisch and Richard Young and Affiliate Member David Gifford plans to do far more than that. In collaboration with Columbia University neurobiology expert Thomas Jessell, the team received a five-year, $6.8 million grant last year to unpackage the molecular strategies that cells use to graduate from the “be all you can be” embryonic stem cell to one of the most specialized human cells: the neuron. Team members believe that prying open each step of this process will yield insights into certain motor-neuron diseases, such as amyotrophic lateral sclerosis (Lou Gehrig’s disease) or spinal muscular atrophy. “So far these diseases have been hard to understand because of our sketchy information on the normal molecular profile of motor neurons,” says Jessell. “We plan to change that.” – David Cameron

He was astonished by the prospect of accomplishing this with a mere four genes. So were Hochedlinger and their peers in the inner circle of embryonic stem cell research.

But Yamanaka’s accomplishment also drew considerable skepticism and relatively little interest from the outside world. One big reason was what had made 2006 the worst of years—the final collapse of Korean scientist Hwang Woo-Suk’s claim to have created human embryonic stem cells through nuclear transfer.

Yamanaka’s work was solid science, but “to be perfectly honest, many of us didn’t even believe it,” admits Hochedlinger. “Four transcription factors reprogramming a cell? It just seemed too simple.”

Additionally, the new cells were limited when compared with naturally derived embryonic stem cells, mainly in that they couldn’t produce live chimeric mice. And Yamanaka was unable to generate live mice—the definitive experiment for demonstrating that a stem cell is embryonic.

Even after Yamanaka published a paper in Cell in August 2006 describing exactly how he had done the reprogramming, most reseachers sat on their hands.

But as Yamanaka’s lab charged forward, two other groups eagerly plunged ahead to reproduce—and improve—the technique.

One group was led by Hoched-linger and Kathrin Plath from the Institute for Stem Cell Biology and Medicine at the University of California/Los Angeles, and another by Jaenisch’s Whitehead lab. That group included postdoctoral researchers Marius Wernig, Alexander Meissner and Tobias Brambrink; graduate student Ruth Foreman; and Manching Ku, a research fellow from Bradley Bernstein’s lab at MGH.

And this June, all three labs published papers confirming and extending Yamanaka’s work.

Although all the experiments occurred in mice and have yet to be demonstrated in human cells, the field of embryonic stem cell research got a major jolt of energy—and headlines around the world.

Irving Weissman, one of the world’s leading stem cell scientists, declared to the New York Times that “this is about as big a deal as you could imagine.”

Success on the big screening

The scientists drew on well-established genetic and biochemical procedures, as well as almost endless patience, to create and test the cells.

Rudolf Jaenisch was surprised by the power four genes brought to reprogramming an adult cell into an embryonic state.

Jaenisch’s group successfully used the same technique as the Yamanaka lab to activate Oct4, Sox2, c-Myc and Klf4 in mouse skin cells. The key difference in its approach was the screening technique used to sort through these cells.

“We were working with tens of thousands of cells, and we needed to devise a precise method for picking out those rare cells in which the reprogramming actually worked,” says Wernig. The odds were not in their favor—only about one in 1,000 cells made the cut.

The group focused on Oct4 and another transcription factor called Nanog, two identifying hallmarks that are active in embryonic stem cells that are fully pluripotent (able to spin off cells that differentiate into almost every cell in the adult body). The trick was to figure out a way to harvest Oct4- and Nanog-active cells from the rest of the population.

The answer came in a common laboratory technique called “homologous recombination.”

Here, the scientists took genetic material known to be resistant to the toxic drug neomycin and spliced it into the genomes of each cell right beside Oct4 and Nanog. If Oct4 and Nanog were switched on, the drug-resistant DNA would also spring into action, conferring immunity. When they added the drug to the batch, only the Oct4- and Nanog-active cells could resist it. Thus, the researchers ensured that only fully reprogrammed, pluripotent embryonic stem cells survived.

The team ran these cells through a battery of tests, searching for any substantial differences from normal embryonic stem cells. Their genetic markers were identical. So were the markers for epigenetic effects (which differentiate cells without changing the underlying genes).

Prove it

But definitive evidence would come only by proving that these cells could develop into any kind of cell type or body tissue.

The Jaenisch group approached this question in three ways.

First, they fluorescently labeled the reprogrammed cells and injected them into early-stage embryos, which eventually gave rise to live mice. While these mice consisted of both the reprogrammed cells and the natural cells from the original embryo, the fluorescent tags indicated that the reprogrammed cells contributed to all tissue types—everything from blood to internal organs to hair color.

Next, they bred these mice and found lineages of the reprogrammed cells in the subsequent generation, adding an additional level of proof for these cells’ pluripotency.

Finally, the team exploited another lab technique that involves creating a genetically abnormal embryo whose cells all consist of four chromosomes, rather than the usual two. Such an embryo can only form a placenta, and cannot develop into a full-term fetus.

The researchers injected the reprogrammed cells into this embryo, and then implanted it in a uterus. Eventually, viable late-gestation fetuses could be recovered—created exclusively from the reprogrammed cells. “This is the most stringent criteria anyone can use to determine if a cell is pluripotent,” says Jaenisch.

Both Hochedlinger and Yamanaka also created chimeric mice from such reprogrammed cells and then bred these mice and created new generations. In addition, Hochedlinger described reactivation of the silent X chromosome, which is shut down in differentiated female cells, adding another level of pluripotency.

Any one of these papers would have caused quite a stir. But all three papers were published simultaneously—Jaenisch and Yamanaka in Nature, and Hochedlinger in Cell Stem Cell—and that ensured that the results could not be ignored.

“All three papers validated each other,” says Jaenisch. “This drove home the point that reprogramming without eggs is a biological fact.”

It’s all about the science

Researchers in the Jaenisch lab were among those showing that naturally derived embryonic stem cells from mice (top) are morphologically identical to reprogrammed skin cells (bottom). [view larger images] Images: Marius Wernig

Jaenisch cautions that the experiments are only a starting point, with no guarantee that the techniques will be effective with humans. “The questions that we’ve asked in this study are fundamental questions in developmental biology,” he says. “From the scientific perspective, they stand on their own.”

He emphasizes that research on conventionally derived human embryonic stem cells needs to continue. “We don’t know about the future, but for now at least, it’s ‘both/and,’ not ‘either/or,’” agrees Hochedlinger.

Such caveats, however, immediately were buried in an onslaught of political and media reaction.

Most visibly, two weeks after the papers appeared, when President Bush announced he would veto legislation that promised to lift many barriers on stem cell research, he stated in a White House press release that “researchers are now developing promising new techniques that offer the potential to produce pluripotent stem cells, without having to destroy human life—for example, by reprogramming adult cells to make them function like stem cells.”

Jaenisch is no stranger to seeing others misinterpret his work for political gain. In the fall of 2005, he co-published a paper with Meissner demonstrating how embryonic stem cells could be culled from a nonviable embryo-like entity.

Honing for humans

Nevertheless, the research has continued moving forward in a way that brings it one step closer to possible medical applications.

In a paper published in Nature Biotechnology this summer, Wernig and Meissner report a way to simplify their earlier experiment. Previously, they had been forced to genetically manipulate the original donor cells to later select for those that had been thoroughly reprogrammed.

This was problematic for two reasons. First, genetically manipulated cells would never be approved for therapies, and second, even if they were, the techniques used to manipulate them have never been successfully applied to human cells.

In the new experiment, however, they isolated reprogrammed cells from non-reprogrammed cells solely by examining the cells’ physical attributes. They noticed, for example, that while the non-reprogrammed cells are large and flat, the embryonic cells are small and round and tend to form tight colonies.

“We still have some challenges,” cautions Wernig. “The mouse cells were originally reprogrammed with retroviruses. That’s something we’d never do in humans. Still, it’s nice to know that we can now, theoretically at least, overcome one hurdle.”