The Inefficient Clone 

LAURA SPINNEY / ELSO 2003 22sep03

Investigators: Ian Wilmut, Philippe Collas and Mellissa Mann
Day 2 - Monday 22 September 2003

If you want to change a cell's identity, you can tinker with its external environment or you can manipulate its insides. Researchers are homing in on genes that could pull a cell back to the embryonic stage, thereby artificially expanding its plasticity and range of potential identities, but for now, one British biologist believes the most promising approaches are external.

Ian Wilmut of the Roslin Institute in Edinburgh, Scotland - the brains behind Dolly the sheep - points to one approach in particular, exemplified, he says, by the work of Philippe Collas at the University of Oslo, Norway, and his American collaborators.

Collas and colleagues have shown that fibroblasts taken from human skin can be partially reprogramed as immune cells simply by bathing them in an extract of T cells. The reprogramed skin cells express T-cell-specific receptors, for instance. And when the same fibroblasts are exposed to extracts of a neuronal precursor cell, they extend neurite-like projections.

"The treated cells acquired some of the characteristics of the cells whose extracts they were bathed in," explained Wilmut. "Not all the characteristics - nobody's claiming that they 'made the change'. But the research showed that it is at least possible to bring about this kind of permanent change. And that was simply by immersing the cells in the extract for less than two hours. I think that is the way things will go in the future."

But Wilmut also believes that within five to ten years, researchers will have candidates for genes with the power to turn a cell back to its embryonic stage: two nuclear transcription factors called Oct-4 and Nanog are already under investigation, he says.

For now, though, "We have very, very little understanding of the processes which are involved, either in reprograming itself, or in the errors occurring in the methylation of the DNA," he said.

And he points to a study carried out by Mellissa Mann of the University of Pennsylvania School of Medicine in Philadelphia and colleagues, and published this month in Biology of Reproduction, which suggests one reason why the cloning procedure might be so inefficient.

Mann's group studied five genes in cloned mouse embryos that are known to be imprinted by the parent of origin - in other words, they show different expression patterns depending on whether they were inherited paternally or maternally, and those differing characteristics are mediated by alterations in the methylation patterns of the DNA.

They found that the patterns of DNA methylation in the clones were severely disrupted compared with those of normal controls. And only in 4% of the clones that reached the blastocyst stage did the pattern of gene expression match that of blastocysts produced by natural fertilization.

Wilmut's lab is now concentrating on adding to the catalogue of abnormalities that crop up during the development of clones, such as increased weight, immune deficiency and subtle developmental defects. These abnormalities appear to be epigenetic, rather than genetic, says Wilmut, because no cases have been documented in which they are transmitted to offspring.

"This is a stochastic process - it's a lottery," he said. "It's as if there are five or six events which are going on after nuclear transfer, and rarely do they all work out correctly together. Some embryos have a lot of abnormalities, sometimes some of the abnormalities are more harmful than others, and [the embryos] don't begin to develop at all. Others have fewer abnormalities and develop to term, perhaps even to adulthood."

Wilmut hopes that his research may eventually lead to the identification of internal candidate genes that would assist in reprograming cells, and also that it will throw light on why attempts to clone rats, dogs and rhesus monkeys have so far failed.