A study published in the advance online publication of Nature today (April 21) reports the first mouse created by parthenogenesis. But a leading researcher called the results very confusing, saying they raise more questions than answers.

Tomohiro Kono and colleagues from the Tokyo University of Agriculture, Japan, report that they knocked out one allele of H19—a maternally expressed gene thought to function as A noncoding mRNA that blocks IN CIS the expression of Igf2—in an oocyte derived solely from two maternal genomes. A normally developed and viable parthenote resulted, suggesting a pivotal role for the paternally imprinted H19 gene in allowing Igf2 expression from the paternal allele and controlling the requirement for a paternal genome, according to the authors.

“When you put these two sets of chromosomes together functionally, the individual would have a father-like genome with the original mother genome, and therefore it works and gives rise to a live offspring,” Patrick Tam, head of the embryology unit at the Children's Medical Research Institute, Westmead, Australia, who wrote an accompanying News and Views article, told The Scientist.

In prior experiments, the Tokyo group had used a maternally imprinted genome and a neutral genome, which still proved to be lethal, according to Wolf Reik, head of developmental genetics and imprinting at the Babraham Institute in Cambridge, UK. “What they are trying to do now is convert the neutral genome into a paternal one by giving it this H19 deletion allele, which allows the neutral genome to express Igf2—and this is clearly a key event that needs to happen here,” said Reik, who was not involved in the study.

But Rudolf Jaenisch, professor of biology at the Whitehead Institute and the Massachusetts Institute of Technology, Cambridge, US, found the paper flawed and confusing. “I really know this literature,” he told The Scientist. “Igf2 mutations—total knockout, no Igf2 whatsoever—gives you live mice and every mouse survives: they're all small.” This suggests that expression of Igf2 is not an absolute requirement for normal embryonic development, according to Jaenisch, in contrast to the paper's results.

In addition, Jaenisch said, it took almost 500 oocyte H19-deletion manipulations to produce two normal live parthenotes, yet only 12 to 15 manipulations with normal oocytes were attempted as controls. “I think is a stochastic event that sometimes makes mice survive, and it might help but Igf2 is not a key gene,” he said.

Jaenisch also said that in the offspring of the parthenote, the presence of the mutation that rendered the maternal allele paternal-like would be lethal in combination with a paternal allele as overexpression of Igf2 would occur—a condition that he said was previously described as lethal. But the genotype of these pups was not reported in the paper.

“This [parthenogenetic] mouse transmits this deletion—so which of the offspring survives? They should know that,” he said. “I should say that it doesn't invalidate the conclusion that this was a fertile mouse, but only half of its offspring should survive.”

Tomohiro Kono, principal author of the paper, and professor in the Department of Biosciences at the Tokyo University of Agriculture, said that he had not studied these pups except to observe that they were normal. “Some of the pups of course carry the mutation, I think,” he said. “Overexpression of Igf2 is not lethal, that is evident. [It would result in] maybe a little bigger pups or a little smaller pups.”

Reik said that the authors had limited success in trying to construct one of the genomes into a more paternal-like one, needed in order to survive. “They would probably need in addition to manipulate the Dlk locus—which is the other important paternally imprinted gene in oocytes—in order to give it better success,” he said.

“We are focusing on that gene,” said Kono. “But that is a state secret.”