An efficient technique to create healthy mice without sperm promises insights on differentiation and the role of paternal genes in development.

The mice above have two mothers, no father. Credit: Kawahara et al

The epic battle of the sexes known as genomic imprinting begins long before a mammalian embryo is ever fertilised. The tug of war starts during sperm and egg production, when specific genes are switched off within each gamete.

According to one theory, the selective gene inactivation within the resulting embryo balances dad's demands of solid growth with mom's cautionary resource conservation, but the equilibrium can be easily upset if either parent's contribution disappears. This extra layer of genetic control may help explain why producing live mice from “virgin birth” parthenogenesis has been so difficult and why imprinting has caused so many headaches for scientists trying to improve the efficiency of somatic cell nuclear transfers in which a nucleus from a differentiated cell is placed in an enucleated egg.

A Japanese team led by Tomohiro Kono at Tokyo University of Agriculture now seems to have engineered an end-run around the problem and delineated two of the most important paternally controlled imprinting regions in the process1. Imprint-free oocytes collected from day-old mice served as stand-ins for sperm, which were then fused to fully mature and maternally imprinted oocytes to yield fatherless bi-maternal embryos.

Scientists have so far identified three imprinting-control regions created during sperm production. These regions, on chromosomes 7, 9 and 12, co-ordinate the silencing of specific genes before the DNA is passed on by dad to his offspring.

In the absence of these paternal silencing mechanisms, such as during parthenogenesis, the corresponding gene alleles are active for both sets of chromosomes, leading to gene overexpression and dooming the developing embryo. A past sperm-free effort in which the Japanese researchers paired a mature oocyte with an immature oocyte yielded two normal mice out of 371 tries, but only if the immature oocyte donor harboured a knockout of chromosome 7's imprinting-control region2.

The scientists' new effort suggests that a double deletion of the imprinting-control regions on chromosomes 7 and 12 is enough to do the trick. Whereas the maternal imprinting process occurs normally within the mature oocyte, Kono and his colleagues found that deleting both regions within the immature oocyte's genome has roughly the same effect as paternal silencing and can restore some semblance of balance to gene expression.

Embryos containing both genetic deletions, in fact, developed into healthy adult mice almost as often as mice created by in vitro fertilization of normal embryos. The survivors were fertile and healthy, and they made normal progress in exercise and learning, although they grew somewhat more slowly than wild-type mice. Genes that had been abnormally expressed within imprinting-deficient mice, with a few exceptions, were back to near-normal levels.

The results bode well for understanding how genomic imprinting prevents fatherless mice from developing normally, perhaps ensuring that dad's genetic input is included in the final product. Furthermore, the technique suggests how genetic engineering may effectively remove that barrier from nuclear transfer manipulations in the future.