April 12, 2009

experts comment on research into female germline stem cells in mice, as published in Nature Cell Biology

Scientists in China have found a way to produce viable eggs from stem cells contained in the ovaries of adult and newborn mice, and that it’s possible to fertilise such eggs to produce healthy offspring. The findings suggest the possibility of similar techniques in humans, which could extend human female fertility beyond its normal limits.

Prof Azim Surani, Mary Marshall and Arthur Walton Professor of Physiology and Reproduction, Gurdon Institute, University of Cambridge, said:

“Sperm are produced continuously in men but the number of eggs in women is fixed at birth. This new study in mice now suggests that there are also stem cells present in ovaries that can be cultured in a dish, which upon transfer to ovaries can develop into viable eggs and give rise to offspring. This finding, if confirmed independently, could advance understanding of these ovarian stem cells and advance research on female infertility.”

Prof Robin Lovell-Badge, Head of Division, MRC National Institute For Medical Research, said:

“This paper will stimulate lots of activity in the scientific community, as happens when any dogma is challenged. This is a good thing. But what would be unfortunate is if this paper is hyped as a cure for female infertility. A lot more work is needed to understand what these new cells really are, and to verify the findings and the claims.

“This is another chapter in what has been a very controversial story, where some scientists have gone against the dogma which states that in mammals, such as mice and humans, all the progenitor germ cells that could give rise to eggs (oocytes) have already done so by birth. This dogma explains menopause, where women run out of eggs, and fits with a lot of experimental studies, for example data showing that germ cells have to enter meiosis early during embryo development if they are to give rise to eggs, and that no new eggs are added to the pool of eggs already existing at birth. [N.B. Meiosis is the special cell division that occurs only in germ cells, which reduces the two sets of chromosomes found in most cells down to the one set found in eggs and sperm.] Several recent studies have suggested that this dogma is wrong, or incomplete, and claim to provide evidence that there is a population of dividing cells in the postnatal and adult ovary that can still rise to eggs. These studies have then been countered by arguments based on theory or further experiments.

“The present paper describes a new set of data, this time suggesting that there is a small population of cells (referred to as FGSCs) persisting in the ovaries of mice from birth to adulthood that can be isolated, grown in culture, frozen and thawed if required, returned to ovaries of other mice that have been depleted of their own eggs, and then give rise to offspring after mating to normal male mice. If true, and especially if applicable also to humans, then this is very important. For example, it could provide a means to restore fertility to women who have few eggs or who have had to undergo cancer treatments, by isolating these cells, expanding their number in culture and keeping them frozen until needed for IVF, etc.

“But extraordinary claims require extraordinary evidence, whereas to me this is a very incomplete piece of work that will only add to the confusion. According to the authors, these FGSCs are meant to be persisting as a “stem cell” population in the ovary. This would imply that they normally divide infrequently, but the one assay they carry out on this suggests that they must be dividing very rapidly. The FGSCs express an incomplete set of the marker genes or proteins that are normally characteristic of germ cells. The authors use a method to isolate the cells that relies on magnetic beads coated with an antibody to one of the marker proteins (MVH) that is expressed. However, MVH is found within cells and not on their surface, so it is a mystery to me how the method could work. The authors then show that they can culture the cells, but not how they determined which culture medium is best for them – it usually takes a lot of effort to define appropriate culture conditions.

“The authors then look at “genomic imprints”. In somatic cells, the DNA we inherit from our fathers has a paternal imprint, which makes certain genes inactive, while the DNA from our mothers has a maternal imprint, making a different set of genes inactive. This is why both a maternal and paternal set of genes is required for normal development (parthenogenetic embryos die early). These imprints are erased in germ cells early in development so that they can be reset as paternal imprints when sperm are formed or maternal imprints in eggs. All the available evidence shows that these maternal imprints are only added when oocytes start to grow, and they are certainly not present in germ cells prior to entry into meiosis. However, the present authors find that their FGSCs, which have neither entered meiosis nor have started to grow, have a maternal imprint. No explanation is given.

“For the grafting experiments, the authors use a chemical treatment to kill the eggs present in the ovaries of recipient female mice. We are not provided with the data to say how efficient this was – indeed the evidence presented suggests that many endogenous eggs survived. The finding that about a quarter of offspring carried the GFP gene introduced into the FGSCs in culture is the most compelling piece of data to support the authors claims, but this is only one inherited trait – they could have set up the experiment in a way that would have allowed thousands of genetic markers to have been followed. And the GFP gene was introduced in a retrovirus – which could have jumped from their FGSCs to host oocytes. This could have been tested directly by looking at the genomic DNA sequences flanking the retrovirus, but this was not done.

“And then there are other important issues the authors did not look at at all. For example, X chromosome inactivation is the mechanism used in mammals to equalise X-linked gene dosage between males (XY) and females (XX). This occurs in all cells in early XX embryos, including the early germ cells. However, by the time germ cells are found in the early fetal ovary, they have reactivated the inactive X. They are happy like this, unlike other cell types, possibly because they have essentially stopped dividing. So what is the X chromosome status of these apparently permanently dividing FGSCs? This is an important question, but one the authors do not address.”