select search filters
briefings
roundups & rapid reactions
before the headlines
Fiona fox's blog
Research, published in Nature, looked at a new human embryo-like model created from human stem cells.
Prof Magdalena Zernicka-Goetz FMedSci, Professor of Mammalian Development and Stem Cell Biology, University of Cambridge, said:
“This is a highly important advance. It shows the incredible ability of human embryonic stem cells to be cajoled into undertaking their inherent ability to embark upon events that remarkably resemble natural development. Much of human development remains a mystery and so this breakthrough gives us direct access to the study of processes that go awry in a huge range of human diseases when the embryo is just three weeks old.”
Prof Jeremy Green, Professor of Developmental Biology, King’s College London, said:
How significant is this study?
“This is a nice piece of work, but no surprise. It is more of a technical extension, significant because it brings a relatively well developed field into the domain of human embryology. It does, though, highlight the amazing power of self-organisation of cells and tissues given the right conditions and reinforces the notion that we are beginning to have a pretty good handle on how to set the self-organisation process in motion. It also begins to extend studies on the formation of cell types to the next step, which is the formation of real structures and tissues in the body.
How does this study fit with previous discoveries in this field?
“This is the human version of mouse gastruloid work that has been going on (led by the same groups) for a few years, but that started with work in frog cells going back to the late 1980s using frog cells. Nature published something not too different in 2004 (Ninomiya et al. from the Winklbauer lab). The add-ons are looking in more detail at gene expression (previously done the same way for mouse gastruloids, although not for frog cells), but Moris et al. doesn’t add a lot conceptually.
Is this study good science?
“I did spot that a lot of the gene expression is based on just two replicates. This is suspicious on two levels. First of all, this makes any of the conclusions statistically quite dubious. The authors got away with it because their results are similar to those in mouse gastruloids and therefore plausible, but plausibility is an extremely bad reason in science for conclusion making. Secondly, this raises the question of why they only did two replicates when one of their claims is that this is a new ‘system’ implying that it is reproducible, reliable and robust. If I had been a referee or even an editor, I would not have permitted this to pass peer review.
Are there any caveats or limitations that journalists and the public should be aware of?
“This is a tool for understanding the basic biology and is quite a long way away from being useful for things like drug testing (which requires scalability and reproducibility). Also, gastrulation is such a comprehensive and whole-body event in development that it is unlikely that any embryo defective in this process is ever born – that means it doesn’t have much directly to do with birth defects. The next stage, which is known as neurulation and forms the spinal cord and brain is another story – neural tube defects that arise when that goes wrong constitute a major class of clinically relevant birth defects. Having said that, there are some rare classes of neural tube defects that may arise from subtly defective gastrulation processes. To have a handle on those, one would need this system to go as far as neurulation stages.
What will this study help us to understand that we previously could not understand?
“This is a fantastic window into the gastrulation and early body extension processes in human embryos at a stage that is currently inaccessible in whole embryos for regulatory/ethical reasons.”
Prof Robin Lovell-Badge FRS FMedSci, Group Leader, The Francis Crick Institute, said:
“This is a very interesting and well done, but nonetheless rather tantalising study. The authors use human embryonic stem cells in culture to mimic aspects of early human embryo development. This is important, because there is much we do not know about our beginnings, in particular the period from implantation at about 7 days, when there are just three very early cell types, to about 28 days, by which time the heart, the central nervous system (CNS), and many other tissues are becoming recognisable, albeit in still primitive form. We know more after this point, because of the availability of terminated embryos for study, even if these cannot be maintained intact. Having a robust model of this rather mysterious period of human embryo development would be important to allow researchers to understand why it all too frequently goes wrong, leading to miscarriage or congenital defects, and as a much better way to test new procedures used in IVF than the current practice of looking only at preimplantation stages – or asking retrospectively if something worked or not.
“The current study follows previous work published earlier this year where mouse embryonic stem cells were cultured in conditions that led them to largely self-organise into structures, termed gastruloids, that resemble the posterior parts of mouse embryos undergoing the critical process of gastrulation (at embryonic day 8.5). This is the process whereby various founding cell types are specified and much of the body plan is laid down. Critically, it was possible to carry out detailed comparisons between the gastruloids obtained in culture and mouse embryos at an equivalent stage, which allowed the authors to conclude that they were indeed rather similar. They were not identical, partly because the gastruloids fail to develop the anterior CNS (brain), but they looked to be a quite robust model of some aspects, notably the formation of somites, which are paired structures laid down sequentially adjacent to the midline that give rise to much of the axial skeleton, including bone, muscle, etc.
“The human gastruloids resemble those derived using mouse cells, reaching an equivalent fairly early stage during somite development, and similarly without formation of anterior structures, notably the brain. They were not identical, however, which might reflect real differences between mouse and human embryos. Even though the processes involved may look superficially similar, we now know that there are molecular and mechanistic differences already between preimplantation stages of mouse and human development when the embryos are much simpler.
“The authors make the argument in a note for the media, that the human gastruloids are not real embryos, which is clearly true, suggesting that they are more similar to a flight simulator than a plane. However, I imagine a flight simulator works because it relies on well understood physics and data obtained from real planes in flight. The model here is rather lacking in this respect. They can compare them with embryos and gastruloids from mice, as they have done, but these reveal differences. Do these reflect substantive differences between human and mouse embryos at this stage, which would be very interesting and of potential importance, or are they trivial in terms of the starting cells, medium or physical support being not quite right for the human cultures ?
“The only way to know for certain is to compare the molecular signatures of human gastruloids and human embryos obtained from terminations, which are hard to obtain at this stage, or by beginning with preimplantation embryos that are surplus to IVF treatment and donated for research, and culturing these to gastrulation stages, which may now be possible – indeed some of the methods developed in this paper might help. This cannot be done in the UK and many other countries for legal reasons, as this would cross the 14-day limit. If the law were changed, however, even a few such cultured embryos might suffice to validate the model. Until then, there will always be doubt about this or other models of human development during this critical, but hidden period. I would not trust someone to fly a jumbo jet if all they had learned on was a flight simulator for a Cessna.”
Prof Magdalena Zernicka-Goetz FMedSci, Professor of Mammalian Development and Stem Cell Biology, University of Cambridge, said:
“This is a highly important advance. It shows the incredible ability of human embryonic stem cells to be cajoled into undertaking their inherent ability to embark upon events that remarkably resemble natural development. Much of human development remains a mystery and so this breakthrough gives us direct access to the study of processes that go awry in a huge range of human diseases when the embryo is just three weeks old.”
Dr Teresa Rayon, Postdoctoral Training Fellow, The Francis Crick Institute, said:
“The work of Moris et al. establishes an in vitro system in 3D from human stem cells that will be essential to increase our understanding of human gastrulation. Previous models developed to recapitulate human embryo development from Eric Siggia and others were done in 2D, thus not allowing the study of key processes such as axial elongation.
“This work adds into the ‘in vitro toolkit’ that scientists can now use to study the most unknown stages of human pregnancy – between weeks 2 and 4, where women wouldn’t normally know if they are pregnant. Over this time period, many birth defects originate. Therefore, models that mimic this process and are reproducible will help to increase our understanding and allow testing how and when things can go wrong. In addition, this in vitro model subtracts some ethical issues that may arise from working with donated samples. It is remarkable how similar the process is to that in mice, and it will be really exciting to understand why the process takes 3 days in vitro and ~ 6 days in vivo (from day 14 to day 18-21).”
Prof Joyce Harper, Head of the Reproductive Science and Society Group, Institute of Women’s Health at University College London (UCL), said:
“Professor Lewis Wolpert refers to gastrulation as the most important stage of our life, but up until now we have never been able to study it in humans. This exciting work will allow many key studies to be done so we can learn about early human development and when it goes wrong. This will help us learn more about genetic diseases and infertility.”
‘An in vitro model for anteroposterior organisation during human development’ by Naomi Moris et al. was published in Nature at 10:00 UK time on Wednesday 11th June 2020.
DOI: 10.1038/s41586-020-2383-9
Declared interests
Prof Magdalena Zernicka-Goetz: “None”
Prof Robin Lovell-Badge: “I am a co-opted member of the Scientific and Clinical Advances Advisory Committee of the HFEA, for which I receive a small per diem.”
Dr Teresa Rayon: Nothing to declare
Prof Joyce Harper: Nothing to declare
None others received.