Building a baby: are stem cell-based, embryo-like models the key to unlocking the secrets of human development?

It is estimated that one in four pregnancies will end in miscarriage [1]. Most cases of miscarriage occur in the first 12 weeks of pregnancy and are due to errors in embryonic formation. The consequences of developmental errors are not limited to miscarriage; one in 33 babies will be born with a birth defect due to the incorrect formation of the embryo [2]. Worldwide, 240,000 babies die within 28 days of birth due to birth defects, and a further 170,000 children will die between the ages of 1 month and 5 years old [3].

Both miscarriage and the development of defects most commonly occur in the gastrulation period of development. During this phase, which occurs 15 days post fertilization, the blastula – a single-layered, hollow sphere of cells – folds inwards on itself and is reorganized into a multilayered structure called a gastrula. The three distinct layers of the gastrula will each give rise to a group of the body's major organ systems; the ectoderm becomes the nervous system, the endoderm the gut and the mesoderm the muscles [4].

Miscarriage or the death of a child is a devastating experiences for mother and father; however, these areas remain vastly under researched. Limitations on research that utilizes human embryos means that our understanding of the developmental process of humans remains unclear – known as the ‘black box’ period of development, gastrulation is a particular mystery as restrictions have prevented the culture of human embryos in a lab for longer than 14 days [4]. Existing knowledge in the field has been gleaned from histological specimens, medical images or cultured non-human mammalian embryos. Interspecies differences and limitations in imaging technology have thus impeded our understanding of human development [5] (Figure 1).

Relaxing the 14-day rule

What became known as the 14-day rule is internationally upheld and in some countries, the UK included, is now enshrined in law. However, in 2021, the International Society for Stem Cell Research (ISSCR; IL, USA) released guidelines proposing a relaxation in the rule, replacing the restriction with a suggestion that studies growing human embryos beyond the 14-day period be assessed on a case-by-case basis, with reviewers determining the point of embryo termination [6].

The ISSCR research committee had good reason for proposing a relaxation of the 14-day rule: stem cell models. In the past decade, scientists have developed increasingly complex models of the human embryo from human stem cells – enabling further study of human development while avoiding the controversial use of fertilized human embryos (Figure 2). Such structures lack the ability to grow into a human, though provide a more accurate model for human development than can be offered by a mouse or other non-human mammalian model [6].

Modeling the first weeks of development

Stem cell-based embryo models are organized 3D structures that mimic the developmental processes of the early-stage human embryo, offering insight into the critical stages required for a successful pregnancy: implantation, gastrulation and neurulation [5].

In 2020, a collaborative group of researchers from the University of Cambridge (UK) and the Hubrecht Institute (The Netherlands) developed a model of human gastrulation, mimicking the key elements of a human embryo at around days 18–21 post conception [4,7].

Publishing their work in Nature, the researchers used human embryonic stem cells to generate a 3D-assembly of cells, known as gastruloids, which differentiate into the three distinct layers seen in a human gastrula (Figure 3). To generate the gastruloids, defined numbers of cells were placed into small wells, in which they formed tight cellular aggregates. Following chemical treatment, the gastruloids lengthened along the anteroposterior axis and genes were up-regulated in patterns that reflected elements of a mammalian body plan. When studying the genes expressed, the researchers were able to identify a clear signature of the developmental event that gives rise to key bodily structures, such as thoracic muscles, bone and cartilage [7].

In creating this ‘black box’ stage of development in vitro, it is possible to observe processes never before seen, potentially shedding light on the causes of birth defects that develop during this period. “This is a hugely exciting new model system, which will allow us to reveal and probe the processes of early human embryonic development in the lab for the first time,” commented first author Dr Naomi Moris (University of Cambridge); “Our system is a first step towards modeling the emergence of the human body plan, and could prove useful for studying what happens when things go wrong, such as in birth defects.” [4].

These lab-grown gastruloids are unable to develop into fully-formed embryos as they lack both brain cells and the ability to implant into the womb, therefore do not run afoul of the 14-day rule that remains law in the UK. When judging the age of their models, the researchers compared them to those found in the Carnegie Collection of Embryology (DC, USA), an official continuum of human embryos that includes day-by-day growth over the first 8 weeks of development [7].

Prior to the relaxation of the 14-day rule, this was the only method for aging such models; however, with the extension allowed by the ISSR, researchers are now able to compare their stem-cell based models with real embryos, and test them as feasible stand-ins for research, noted Robin Lovell-Badge, Chair of the ISSCR steering committee and researcher at the Francis Crick Institute (London, UK) [6]. Whether the model developers take advantage of this development remains to be seen.

Inducing gastrulation

A limitation of the Cambridge group's model is the requirement for human embryonic stem cells. While research on human embryos is vital for understanding human development, the use of such cells in research is controversial and can be challenging to access. To overcome these barriers, in 2021 a group of researchers from the California Institute of Technology (Caltech; CA, USA) developed similar models from induced pluripotent stem cells (iPSCs) [8,9].

Originally derived from human embryonic tissue, the iPSCs retain the ability to develop embryonic structure, if exposed to the right environmental conditions. “The ability to assemble the basic structure of the embryo seems to be a built-in property of these earliest embryonic cells that they are simply unable to ‘forget’”, explained lab leader Magdalena Zernicka-Goetz (Caltech). “Nevertheless, either their memory is not absolutely precise or we don't yet have the best method of helping the cells recover their memories. We still have further work to do before we can get human stem cells to achieve the developmental accuracy that is possible with their equivalent mouse stem cell counterparts.” [8].

As these second-generation models do not require the use of donated embryonic stem cells for each future iteration, they can be developed more easily and in larger quantities than previous versions. Thus, the model system may open up more doors in the understanding of embryonic development, mitigating the need for human embryos as a starting point [8].

All of the models developed to date are rudimentary; however, they hold vast untapped potential. Modeling the human embryo could allow researchers to answer questions that until now have been legally, and ethically, unanswerable: how does the embryo implant into the uterus? How do drugs affect the developing embryo? And how does the embryo ensure that cells are in the right position? [2].

All models are wrong, yet some are useful

As with the development of any new technique or model, success brings with it more questions than answers. With new doors now open, the fields of birth defects, infertility, in vitro fertilization protocols and drug screening all look set to advance from the development of this research technique, potentially impacting wider society in a myriad of ways [10]. The core principles of early-stage embryonic development also provide signposts for tissue engineering and regenerative medicine [5,11].

That being said, such models come with concerns. The extent to which such models accurately mimic in vivo human development remains to be seen [10] and while the relaxation of the 14-day rule will allow for comparison with human embryos, it will still not shed light on the in vivo process. Further, current models require improvements of their controllability, scalability, reproducibility and standardization protocols before they may be rolled out for widespread reproducible research [11]. The inclusion of bioengineering tools to biodynamically control the cellular environment will be an essential development in the next few years, and will allow for the improvement of embryoid models and help researchers to glean new insights in their developments [11].

However, further advances raise a paradoxical dilemma; stem cell-based embryo-like models should aim to be as human-like as possible in order to be useful research models and yet, should remain sufficiently different as to preserve the distinction between model and human and thus evade an ethical quandary. As models become closer and closer to the real thing, the line between human and research tool begins to blur, raising the question: what does it mean to be human?

What it means to be human

“We will have to confront ourselves with the question of what is a human embryo, and whether these models really have the potential to develop into one,” noted Zernicka-Goetz in a 2019 Nature article that discussed this emerging technology [12].

Is an embryo more valuable if it is formed from the union of sperm and egg, or does its lab-grown counterpart, which can perform the same biological processes, hold the same legal rights? Who is qualified to answer such a question may also be up for discussion. In the battle of science against ethical beliefs, the matter of what it means to be human is a hotly contested and polarizing issue, with each side firmly set in their opinions and beliefs.

For some the question is moot; “It's like putting four wheels on a frame and saying it's a car, even though there is no engine”, commented Eric Siggia (Rockefeller University, NY, USA) [12]. The synthetic embryos lack a placenta and other core components required for human development, therefore lack the ability to grow into people. However, in this ever-advancing field the development of a more human-like model may not be too far away. As models approach a more human-like state, we then veer into cloning territory – and the ethical minefield continues to grow.

Conclusion

The development of our complex human selves from the unity of a singular sperm and egg pairing remains one of the wonders of nature, the intricate process having been fine tuned over millenia via the process of evolution. Yet despite nature's best efforts, the process still goes wrong and often in the most devastating of ways. With these stem-cell models, we have the opportunity to explain and potentially correct these developmental abnormalities. However, this field is still in its infancy and there are still hurdles to overcome.