October 17, 2016

expert reaction to study reporting production of functional mouse eggs in culture

Scientists publishing in the journal Nature reported the production of functional mouse eggs from embryonic and induced stem cells in mice.

Prof. Azim Surani, Director of Germline and Epigenomics Research, The Gurdon Institute, University of Cambridge, said:

“This study represents a major accomplishment; it is very impressive indeed. Hayashi had previously taken major steps towards achieving full development of mouse eggs in culture, and he has accomplished that objective in this new paper.

“The work is robust and in my view will be reproduced by other labs. Now we need additional data and observations, particularly on term embryos to see how many develop normally and what happened to those that did not. Do we anticipate or observe any developmental abnormalities?

“Ethically, this issue has yet to be discussed fully by the scientists and society. These discussions have occurred in the past, and are continuing within the regulatory bodies, certainly in the UK. This indeed is the right time to start a debate and involve the wider public in these discussions, long before and in case the procedure becomes feasible in humans.

“To replicate this work in humans poses further challenges, and it is futile to speculate when this will become possible. There are some key differences that have become evident between mouse and humans germ lines, which we described in our publications last year, which have to be taken into account. Second, it is possible that the procedure may require a considerably long culture period, which while not insurmountable, adds to the difficulties. For example, environmental factors during culture could cause aberrant epigenetic modifications (modifications that do not affect the DNA sequence but are potentially heritable and can affect gene expression in the progeny), which could affect development of synthetic eggs. Third, the mouse procedure requires co-culture of germ cells with fetal gonadal cells for their maturation into oocytes. This step might be difficult if it requires human fetal gonadal cells. We may need to find alternatives, for example replacement with additional factors or finding alternative cells for co-culture.

“We in my lab are working on the initial stages of developing human primordial germ cells currently.  We have much to learn about the human germ cell biology in order to ask questions, such as the mechanism of the unique epigenetic program of the germline that is essential for producing normal gametes. Other related research areas include mitochondrial biology, infertility and germ cell tutors. Advances in these areas could be widely applicable towards understanding age related diseases and in regenerative medicine.”

Prof. Martin Johnson, Professor of Reproductive Sciences, University of Cambridge, said:

“This paper reports the generation of fertile mice from eggs derived entirely in-vitro from stem cells – both embryonic and induced stem cells – and is a remarkable achievement, involving good quality research, conservatively interpreted, in which the conclusions are supported by reliable data.

“The data are primarily of interest to scientists, although potentially of clinical interest to those patients who lack eggs of their own.

“However, there are major limitations for the application of this technology to the clinic, acknowledged in the paper, and including: 1. a requirement to use granulosa cells during the in-vitro maturation phase; 2. the fact that the gene expression patterns in many or most of the fully matured in-vitro derived eggs differs significantly from that in control eggs; 3. the fact that chromosomal abnormalities are more common in the in-vitro derived eggs; and 4. that, probably as a result of 2 and 3, only a small number of embryos formed from in-vitro derived eggs developed into ‘normal’ young.

“For these reasons alone, the application of this technology to the human is years away.”

Prof. Richard Anderson, Honorary Consultant in Obstetrics and Gynaecology & Professor of Clinical Reproductive Science, MRC Centre for Reproductive Health at the University of Edinburgh, said:

“This new report from Professor Hayashi and colleagues provides a substantial advance in our understanding of how artificial eggs might be made. They start from embryonic stem cells, and make an artificial ovary in which the stem cells turn into immature oocytes, and then mature within follicles just as they would normally. They were able to fertilise these eggs using IVF, and generate healthy fertile pups. This is the first report of anyone being able to develop fully mature and fertilisable eggs in a laboratory setting right through from the earliest stages of oocyte development. Importantly, they carried out a number of safety checks along the way, which as might be expected showed that these eggs do not develop quite as well as normal eggs, but these findings will lead to new avenues to explore the genes and pathways that control normal egg development.

“Although we are a long way from making artificial eggs for women at the moment, this study also provides us with a basis for experimental models to explore how eggs develop from other species, including in women. This is extremely challenging at the moment due to the difficulties of getting eggs to study, so being able to develop them in the laboratory would greatly improve this.

“One day this approach might be useful for women who have lost their fertility at an early age, as well as for improvements in more conventional infertility treatments. But the very careful analyses in this paper show the complexity of the process and how it is a long way from being optimised.”

Dr Dusko Ilic, Reader in Stem Cell Science, King’s College London, said:

“Scientists from Japan were able to reconstitute the full process of making an egg in a dish. Cells from a mouse tail were reprogrammed into embryonic-like induced pluripotent stem cells (iPSC) lines, which were then used to generate in the lab fully potent mature eggs capable of producing offspring.

“To do that, at one stage the scientists co-cultured iPSC-derived primordial germ cell-like cells with female gonadal somatic cells from 12.5-day old mouse embryos. The next step would be to circumvent the use of embryonic tissue and reconstitute the entire process in a dish from iPSC. However, it may take years before we reach this point.

“If (when) this happens, developing similar culture systems in other species should be only a matter of technicality and with such technology we might be able to rewind the process of mammalian extinction.”

Prof. Robin Lovell-Badge, Group Leader, The Francis Crick Institute, said:

“This is an interesting piece of work, but like the topic – how to make eggs in vitro – it should be considered as a beginning, albeit a very promising one, and not an end. This is the first convincing evidence that it is possible to go all the way from pluripotent stem cells, via a primordial germ cell state, to functional oocytes entirely in vitro, which can then be fertilised and give rise to apparently healthy mice. The authors have carefully monitored progress through all the steps, ensuring that critical milestones are achieved by at least a proportion of the developing oocytes. And the methods they have used all make biological sense.

“It does seem to be a rather inefficient process, with many losses at each step of oocyte development, and notably with very few fertilized eggs giving rise to pups, but then so is oocyte development in vivo. Loss of germ cells occurs naturally in human ovary development with only about 30% surviving between 20 weeks and birth, and then only about 1 in 1000 of the primordial follicles at birth will mature to ovulation. This is not well understood, although it may be a quality control process to ensure that only the best survive. But having a way to observe germ cells in vitro as they become oocytes and then grow and mature will be important to allow research into the reasons why and how this happens.

“There is also a long way to go before these methods could be adapted and used in humans. Although the first step, notably deriving primordial germ cells (PGCs) from pluripotent stem cells in vitro has been achieved, there is no guarantee that the rest will work. Indeed, it will be of interest and perhaps of benefit in its own right to explore the similarities and differences between human and mouse oocyte development, of which we already know there are some. Moreover, there is the question of time. It usually takes longer than a decade to have fully grown oocytes in humans. Will it take this long to recapitulate the process in vitro, which would pose immense practical challenges, or will parts of the process speed up in vitro where the constraints that normally operate in vivo will be absent; such as the need for pituitary hormones to synchronise what goes on in the ovary with the rest of the body? Oocytes don’t usually begin to mature until puberty, but they may do so much sooner if the right conditions can be found in vitro.

“Clearly, if applied to humans, being able to get functional eggs via reprogramming skin cells to iPS cells, would have importance in overcoming female infertility, e.g. due to cancer treatment as a child, but it also opens up many other uses in research, in regenerative medicine, and potentially in avoiding genetic disease. But there are still many practical and ethical challenges to be resolved. One of these would be a source of fetal ovarian cell types, but as the authors suggest, it may also be possible to derive these from the same starting population of pluripotent stem cells.

“For some potential uses and ethical challenges, see: http://www.hinxtongroup.org/Consensus_HG08_FINAL.pdf  Although this was published in 2008, it is still very pertinent and it recognised many of issues, which have been debated at length subsequently.”

Further technical notes from Robin Lovell-Badge on the paper:

To derive functional oocytes (eggs) from pluripotent stem cells (embryonic stem (ES) or induced pluripotent stem (iPS) cells) entirely in vitro required several steps. The first, which is to derive primordial germ cell-like cells (PGCLCs) from chromosomally female (XX) cell lines, has already been published by several groups, with some of the present authors describing the most robust methods. To get these PGCLCs to go to the next stage, where they begin on the path to make oocytes requires the influence of somatic cells from the embryonic ovary. This can be carried out in simple co-cultures over a period of a few days. The same group had also done this previously; when they showed that these early oocytes (oogonia) could go on to make functional mature eggs after they grafted these co-cultures into the ovaries of newborn female mice. What is different in this paper, is that rather than grafting them back into ovaries, they have developed a much more complex 3-D co-culture system, and this, together with appropriate changes of factors present within the culture medium over time, allows the oocytes to associate with granulosa cells in the cultures to form normal-looking primordial follicles. Some of these then eventually grow to form mature follicles, from which normal looking oocytes can be obtained. Importantly, it is found that at least some of the oocytes undergo relatively normal meiosis, growth, and finally maturation. (N.B. aspects of this were also carried out previously in vitro by other groups (notably by John Eppig), but beginning with primordial follicles isolated directly from ovaries.) The authors then show that a proportion of these oocytes can be fertilised in vitro and when the resulting early embryos are implanted in females, a small proportion of these will go on to give offspring. It is clear that it is not an efficient process, with only around 1% of embryos giving pups. Some differences in gene expression were seen in the in vitro-derived oocytes compared to in vivo-derived control oocytes, moreover the offspring had some subtle differences from controls. However, they were fertile and able to produce entirely normal offspring of their own.

The authors also showed that early embryos obtained by fertilisation of the in vitro-derived oocytes can be used to derive embryonic stem cell lines. This means that it would be possible to carry out multigenerational genetics experiments entirely in culture (assuming a source of sperm), before then allowing the embryo to generate a mouse. Of course, the same lab and others are also working to have spermatogenesis occur entirely in vitro (there have been papers published claiming to have done this, but inefficiently, and with methods that seem difficult to replicate).

‘Reconstitution in vitro of the entire cycle of the mouse female germ line’ by Hikabe et al. published in Nature on Monday 17th October. 

Declared interests

Prof. Azim Surani: “I have shared interest based on our work on the specification and development of human germ cell lineage but I don’t have any conflicts of interest.”

Prof. Martin Johnson: “No relevant interests to declare.  Honorary fellow of the RCOG; on the advisory board of the Progress Educational Trust; Chief Executive of the Edwards and Steptoe Research Trust; secretary of the Anne McLaren trust fund managers.”

Prof. Richard Anderson: “I have no conflict of interest, although we undertake related work related to understanding early oocyte development, funded by MRC.”

Dr Dusko Ilic: “I declare no interests.”

Prof. Robin Lovell-Badge: “I am employed by the Francis Crick Institute and part of my research programme is on related topics of sex determination and gonadal development; but otherwise I have no financial interest in the topic of research. I am a member of the HFEA’s Scientific and Clinical Advances Advisory Group, which will find the topic of this paper of interest, along with other methods of obtaining in vitro-derived gametes, but this is an advisory role only. I am also a member of the steering group of the Hinxton Group, which considered the topic, as highlighted above.”

No others received