Human blastocyst-like structures generated entirely from pluripotent stem cells. Shifting the paradigm of developmental biology?

Published 02 April 2021

Four research groups have in the past few weeks independently reported that they have grown clusters of cells which mimic the function of human blastocysts. Susana Chuva de Sousa Lopes, co-ordinator of ESHRE’s SIG Stem Cells, and deputy Mina Popovic describe how these breakthroughs might extend the present limits on human embryo research.

Until now knowledge of early mammalian development has relied heavily on observing and manipulating human and animal embryos directly. Nevertheless, the relatively short timeframe available for analysis, coupled with the inaccessibility of research material, has inherently limited our understanding. Fired by their ambitious quest to elucidate the cellular and molecular complexities of human embryos, researchers are currently rethinking the way in which we study early human development in vitro.

Embryo-like structures are taking precedence. Unlike embryos resulting from the process of fertilisation, these structures are formed by stem cell coaxing, providing a novel, scalable platform for interrogating developmental pathways.

Human pluripotent stem cells (hPSCs) have been extremely valuable for understanding aspects of early human development. However, hPSCs have thus far only been successfully applied for capturing early human post-implantation development, recapitulating aspects of epiblast, trophoblast and amniotic cavity formation, and some features of axis development and gastrulation.(1) Traditional culture systems have lacked the complexity to model the spatio-temporal dynamics of a blastocyst.

Now, in a recent breakthrough, two research groups have harnessed the synergy between stem cell and developmental biology to generate the first human blastocyst-like structures, termed ‘blastoids’.(2.3) These papers were accompanied by two additional preliminary reports describing similar results.(4,5). Yu et al and Liu et al applied a 3D-microwell system and specific culture media to support the differentiation of hPSC towards structures that resembled human blastocysts in terms of their morphology, size and cell number.

In both studies, ~20% of the cell aggregates formed blastoids after 6-8 days. Detailed gene expression analysis revealed the presence of distinct embryonic lineages; however, the blastoids also contained many unidentified cell types. The authors thus cultured the blastocyst-like structures beyond the implantation stages in vitro. Interestingly, a small portion of outgrowths revealed phenotypes akin to the epiblast and amniotic cavity. Nevertheless, these findings warrant careful interpretation. Ultimately, thorough characterisation remains a challenge, as there are currently no optimal culture systems to mimic human peri-implantation in vitro.

With advances in high resolution genetic analysis and imaging technologies, research using blastoids certainly holds promise. For instance, blastoids may be generated in large numbers, allowing sufficient material for in-depth assays and high-throughput screens. Furthermore, they are more amenable to rapid genetic modifications than in experiments involving (natural) human embryos. However, as Heuser and Streeter elegantly wrote in 1941: ‘The embryo is a machine that needs to function while it is being built.’

Accordingly, it is important to appreciate that these models do not capture the full complexity of human blastocysts. Blastoids do not have a zona pellucida and, while some primitive endoderm (PE)-like cells were present, a defined PE cell layer could not be observed. Furthermore, immunofluorescence staining and transcriptomic analysis show inconsistencies for trophectoderm markers, while many of the blastoid cells cannot be correlated to in vivo counterparts. Several further limitations persist, such as poor efficiency and heterogeneity within and between different blastoids. Most importantly, the developmental potential of human blastoids remains to be determined. At present, blastoids generated from mouse PSCs do not have the capacity to develop beyond the early post-implantation stages.

Alongside scientific innovation, harnessing the full potential of human blastoids will also require urgent ethical reflection. While blastoids may overcome the destruction of human embryos, their genome is not individually unique, but rather represents a genetic clone of the stem cells or donor cells of origin. Hence, the legal and ethical implications associated with informed consent for the application of hPSCs will require revision. For instance, a donor may agree to his/her stem cells being used to generate tissues, but not for the creation of cloned embryos.

In addition, evaluating the extent to which the use of blastoids raises ethical concerns typical of human embryo research, such as the 14-day rule, will be crucial. If these structures were to acquire functionality, the definition of an embryo will require careful rethinking. Perhaps in the future, some of the ethical and legal restraints imposed on human embryo research may be overcome in blastoids by ensuring non-viability. For instance, gene editing may be used to introduce a necessary lethal mutation in the donor hPSCs.

In this context, the multidisciplinary approach offered by stem-cell based embryo models does provide a new edge. Depending on their functionality and moral status, blastoids may prove valuable in complementing human blastocysts for research. This integrated approach will be important not only for addressing fundamental biological questions, but perhaps also for improving ART, for studying implantation, modelling specific diseases related to early pregnancy and improving embryo selection. Armed with this potential, we are undoubtedly facing thrilling times ahead in human embryo research.

1. Rossant J, Tam PP. Opportunities and challenges with stem cell-based embryo models. Stem Cell Reports 2021;
2. Yu L, Wei Y, Duan J, et al. Blastocyst-like structures generated from human pluripotent stem cells. Nature 2021; 591: 620-626. doi:10.1038/s41586-021-03356-y
3. Liu X, Tan JP, Schröder J, et al. Modelling human blastocysts by reprogramming fibroblasts into iBlastoids. Nature 2021; 591: 627-632. doi:10.1038/s41586-021-03372-y
4. Fan Y, Min ZY, Alsolami S, et al. Generation of human blastocyst-like structures from pluripotent stem cells. bioRxiv 2021; preprint at doi:10.1101/2021.03.09.434313
5. Sozen B, Jorgensen V, Zhu M, et al. Reconstructing human early embryogenesis in vitro with pluripotent stem cells. bioRxiv 2021; preprint at

Get notified of new articles with our ESHRE newsletter.

Sign up and never miss an update