The International Society for Stem Cell Research, which promotes excellence in stem cell science and its applications to human health, hosted an online event on 'The Science of Integrated Human Embryo Models' to examine the science, methods and future of this fast-moving area of stem cell research.
All invited speakers gave overviews of how and why embryo models were developed. In brief, much of human early development occurs during weeks that follow implantation. This is sometimes considered the 'black box' of embryonic development, as it is difficult to study the significant and highly dynamic changes that occur during the time when the embryo is obscured by the uterine wall. Following implantation, embryos undergo a process called gastrulation, in which the embryo reorganises itself from a single layer of cells into a multidimensional structure.
It is estimated that 70 percent of conceptions are miscarried around this time. Thus, uncovering the processes at play during implantation and gastrulation advances our understanding of the principles of early embryonic development, formation of the placenta, and the reasons behind early loss of pregnancy. In addition to technical difficulties, there are also legal limitations in accessing embryonic material at this stage of development, in the volume required for scientific research.
The technology of stem-cell based embryo models (SCBEMs) has been developed to help address and overcome some of these issues. SCBEM technology is based on the recent discovery that under the right conditions, stem cells can self-organise into three-dimensional structures that mimic aspects of early embryos. Since this was discovered, many SCBEMs have been developed, each focusing on different aspects of early development. Importantly the creation of these SCBEMs has involved stem cells from different species including mice, nonhuman primates and humans.
SCBEMs can be divided into the categories 'integrated' and 'non-integrated'. 'Integrated' SCBEMs contain, or can develop, extra-embryonic cell types that contribute to extra-embryonic tissues that are critical for an embryo to develop during pregnancy, such as the placenta and the yolk sac. 'Non-integrated' SCBEMs only model the embryo proper, and are unable to form extra-embryonic tissues.
Speaking first, Professor Kathy Niakan from the University of Cambridge shared her lab's findings on the earliest stages of human embryonic development (see papers by Fogarty, Gerri and colleagues). Using donated human embryos, this work highlighted species differences in how the first lineage decision is made, and also the importance of benchmarking research findings to human embryos.
Unlike embryos formed from fertilised eggs, SCBEMs self-assemble from groups of stem cells. Professor Niakan reminded us that this fundamental difference – and how this difference affects critical processes, such as embryonic genome activation – places limitations on the use of SCBEMs to study how tissues of the developing embryo interact during very early stages. Professor Niakan concluded that while SCBEMs have great potential use for investigating certain research questions, they cannot and should not replace the use of human embryos in establishing the ground truth of early human development.
Dr Yasuhiro Takashima from the Centre for iPS Cell Research and Application, Kyoto University, Japan, presented findings from a human SCBEM modelling the bilaminar epiblast, which – along with the bilaminar hypoblast – eventually forms the fetus. They called this particular SCBEM a 'bilaminoid'. They discovered that when cultured with trophoblast cells, this bilaminoid was able to form cavities akin to those typically seen in early embryonic development. A role for the IL6 gene was also revealed. Interestingly, by day six of culture, these bilaminoids express genes typically seen in 14-day-old human embryos. This research showcases notable differences in developmental timing between SCBEMs and actual embryos (formed from eggs fertilised by sperm).
Professor Magdalena Zernicka-Goetz from the California Institute of Technology and the University of Cambridge shared the development of her team's most recent 'integrated' SCBEM. They detailed principles of self-assembly, such as the importance of an adhesion code and the role of cortical tension, which they discovered by building the ETX model – a 'non-integrated' SCBEM. They went on to generate an integrated SCBEM using mouse stem cells, and found that these models were able to develop brains, hearts and even germ-cell-like cells. Consequently, they pursued the possibility of similarly integrated SCBEMs made from human stem cells. Further technical details were presented for their recent proof-of principle integrated SCBEM, and through its generation, they discovered an antagonistic role for SOX17 in anterior hypoblast formation.
Initially, Professor Jacob Hanna from the Weizmann Institute of Science, Israel commented on the debate surrounding the stem-cell state required to generate SCBEMs – namely, naïve or primed (see BioNews 1204). He was of the opinion that the main issue is the difficulty of directed differentiation of the stem cells, rather than their initial state.
Professor Hanna then explained how his team explored the question 'Can mammalian gastrulation be fully captured ex utero?'. Importantly, in advancing conditions for culturing mouse embryos ex utero, they learned that embryos can develop substantially without maternal interaction and are therefore self-organising. Interestingly, the efficiency of self-assembling SCBEMs is low, with 0.5-1 percent efficiency at day eight and only 40 percent of mouse embryonic stem cell lines being able to be used for embryo model generation.
In developing his team's most recently reported integrated SCBEM, Professor Hanna found evidence for a human alternative naïve-like pluripotency state, discovering that the model generated was the equivalent of a human embryo at 13-14 days post fertilisation. Importantly while the morphology of the SCBEM they generated is very similar to that of actual human embryos, the SCBEM seems to bypass the blastocyst-like stage in achieving this feat.
In the concluding remarks and questions, Professor Zernicka-Goetz reiterated that SCBEMs are 'not natural embryos, they are models', and that research using actual human embryos is still needed. The diversity of the different SCBEMs is interesting in itself, and Professor Zernicka-Goetz believes they will shed light on different processes.
Professor Niakan commended researchers for self-regulating, and refraining from going beyond the primitive streak stage (typically found in actual human embryos at around day 14). This is important, as current UK law – and corresponding regulation by the Human Fertilisation and Embryology Authority – stipulates that human embryos may not be cultured beyond 14 days of development. While public engagement goes on, self-regulation within the community is critical to maintaining public trust in scientists.
The science, ethics and regulation of SCBEMs will be discussed at the 2023 PET Annual Conference – How Much Change Do We Want? Updating Fertility, Embryo and Surrogacy Law – taking place on Wednesday 6 December 2023.
Note that due to recently announced UK rail strikes, this conference will now be taking place entirely online.
Find out more and register here.
Leave a Reply
You must be logged in to post a comment.