The Scientist

Researchers have shown that cloned embryos undergo complete nuclear reprogramming and resemble fertilized embryos. The report, appearing in the December 6 issue of Proceedings of the National Academy of Science, appears to refute the long-held belief that the high failure rate of cloning is due to faulty nuclear reprogramming of the donor cell nucleus to an embryonic state. Researchers suggest that the findings may only deepen the mystery of why cloning often fails, and demonstrate how far scientists have to go before improving the process.

"It's a surprising result that reprogramming obviously works very nicely," Heiner Niemann at the Institute for Animal Breeding in Neustadt-Mariensee, Germany, told The Scientist.

Thirteen mammal species have been successfully cloned to date, but only 1-5% of cloned embryos ever produce live young, regardless of species. In the current study, Xiangzhong "Jerry" Yang, at the University of Connecticut, Storrs, Ct., along with colleagues at the University of Illinois in Urbana, Il. and the National Institute for Agricultural Research (INRA) in France, used cDNA microarray analysis to determine if abnormal gene expression among cloned embryos could explain why cloning often fails. "Many papers have looked at a few genes, but this is the first time anyone has looked at a large number of genes," said Eckhard Wolf, at Ludwig-Maximilian University in Munich, Germany.

The researchers compared the gene expression profiles of cow embryos obtained by artificial insemination (AI), in vitro fertilization (IVF), and somatic (donor) cell nuclear transfer (NT), in which the nucleus of a donor cell is transferred into an egg cell whose nucleus has been removed.

One week after cloning (blastocyst stage), NT embryo expression profiles differed completely from the donor cells used to create them, indicating that nuclear reprogramming had been successful. "What was shocking to us was that the cloned embryos were more similar to naturally fertilized embryos – and among cloned embryos, there was very little difference" between them, Yang told The Scientist. Furthermore, the NT embryos resembled AI embryos much more closely than IVF embryos.

Using bovine microarrays developed by colleagues at the University of Illinois, the researchers found that less than 1% of 5000 genes analyzed differed more than two-fold between NT and AI embryos – comparable to differences observed between genetically unrelated AI embryos. However, 25 genes were uniquely expressed in the NT embryos. "We want to extensively study these genes and see if they are responsible for the later reprogramming problems," said Yang. He added that cloning problems may happen at a later stage during differentiation.

Niemann cautioned that these are early findings. "I would anticipate 8000-10000 genes involved in" fetal development, said Niemann. "We will learn more about critical genes involved when we have complete bovine genome arrays."

Wolf agreed that further studies are needed. "We cannot conclude that cloned blastocysts are more normal than IVF blastocysts because there was a lot of genetic variability in the IVF group," he said.

These findings don't bode well for those hoping to see stem cell therapy a reality anytime soon either, according to Wolf Reik, at the Babraham Institute in Cambridge, U.K. "Even if cloned embryos are born, many are not normal and die prematurely – these late effects are not caused by early deficits in gene expression, and that's what this study shows," said Reik. Consequently, stem cells used therapeutically may appear perfectly normal initially, with problems arising only later, he cautioned.

"This study clearly indicates that technical improvements cannot overcome the post-implantation problems of clones," Atsuo Ogura, at the Riken Bioresource Center in Tsukuba-shi, Ibaraki, Japan, told The Scientist. "It reminds us of the fact that no eventual progress has been made in cloning efficiencies in the 10 years since Dolly the sheep was born."