First use of CRISPR-Cas9 genome editing in human embryos as a tool to understand basic principles of human development
Comparative analysis of embryos from different species shows notable similarities and differences, and determines which surrogate model organisms are most suitable for human preimplantation development
Pivotal contributions to global governance and policy surrounding research into human embryos
A deeper knowledge of the molecular mechanisms that govern human preimplantation development is of vital importance for understanding human infertility, for potential gene editing therapies, and for stem cell biology, with its enormous promise for disease modelling, drug screening and cell replacement therapies. Since the advent of single cell techniques, and the development of more efficient and precise tools to modulate gene expression such as CRISPR-Cas9, doing such research in the forensic detail necessary is now possible.
Many of the most significant advances in this field have come in the last quinquennium from the laboratory of Kathy Niakan. Niakan is dedicated to understanding the molecular mechanisms that regulate the very first cell fate decisions in the human preimplantation embryo. Given the sensitivity of the material, and the funding bans or significant limitations in many other countries, there are few groups conducting research directly on early human embryos, and so the Crick’s support of Niakan’s work has been pivotal.
Following a number of cleavage divisions, the first cell fate decisions in mammalian development occur at the morula stage (four days post-fertilisation in humans). The morula forms as cells undergo a process called compaction, whereby cells on the outside initiate a trophectoderm programme from which the placenta will eventually be made. Cells on the inside of the morula are still uncommitted. Around days five to seven post-fertilisation, and immediately prior to implantation, the blastocyst – a hollow ball of about 256 cells – has formed. The outer layer of the blastocyst comprises the trophectoderm, which surrounds the inner cell mass, where differentiation into two cell types occurs: the primitive endoderm makes yolk sac, and a small number (around 10-15) of pluripotent epiblast progenitor cells make the embryo proper.
As the gross morphology of preimplantation development appears very similar in humans and mice, it was assumed for many years that this similarity would extend to the molecular processes involved in early cell fate decisions. By single-cell RNA sequencing of mouse and human blastocysts, Niakan’s group demonstrated that this was not the case (Blakeley et al, 2015): while there were some conserved transcriptional programmes, there were many differences in expression of both lineage- and species-specific factors. The datasets and analysis from this work are now widely used internationally, but more importantly, the paper provided crucial evidence that studying human embryos, in addition to those from surrogate model organisms, is essential for a mechanistic understanding of preimplantation development. Its conclusions prompted the UK Human Fertilisation and Embryology Authority (HFEA) to allow the modification of DNA in human embryos to study gene function.
In 2017, Niakan and her colleagues published a landmark paper reporting the first use of CRISPR-Cas9 genome editing in human embryos with the sole purpose of understanding basic principles of human development (Fogarty et al, 2017). As proof-of-principle, Niakan chose to focus on blastocyst development, knocking out the pluripotency transcription factor OCT4, which in mice is required to specify the inner cell mass. To use as few human embryos as possible, the group identified an efficient OCT4-targeting guide RNA in an inducible human ES cell-based system, and optimised microinjection conditions using mouse zygotes. They were then able to efficiently and specifically target the gene encoding OCT4 in fertilised human eggs.
Even more content
Using live imaging, it was evident that OCT4-targeted human embryos initially underwent early cleavage divisions as normal. However, significantly fewer reached the blastocyst stage, suggesting that OCT4 is needed to correctly form a blastocyst. This is in marked contrast to mice, where the blastocyst is established, but maintenance is compromised.
Closer examination of human OCT4-null blastocysts showed that they went through iterative cycles of expanding and collapsing; they were unable to properly form the inner cell mass, and surprisingly, there was also a failure in development of the trophectoderm. At the molecular level, key trophoectoderm molecular markers were not expressed, and there was also dysregulated expression of tight junction proteins, which may explain the blastocyst defects - a taut trophectoderm is required to facilitate pumping of fluids into the cavity at the centre of the blastocyst in order to expand it.
Data from single cell transcriptomics analysis demonstrated further major defects. OCT4-targeted embryos were transcriptionally distinct from unedited embryos, with misexpression of genes in all three blastocyst lineages. Among the genes most altered were a number of putative developmental regulators, including NANOG, a regulator of the pluripotent epiblast, which was not expressed in targeted embryos. This is in striking contrast to OCT4-null mouse embryos, which continue to express NANOG, underlining the important mechanistic differences in how pluripotency is regulated between the species.
Niakan has also published important work concerning the first lineage decision in human development, which is to initiate the trophectoderm at the morula stage, days before the embryo cells become specialised. In a recent paper (Gerri et al, 2020), she and her colleagues showed using mouse, human and cow embryos that apical-basal polarity, established by cell-cell contact, is vital for trophectoderm formation. Furthermore, the cascade of molecular events at this stage is broadly conserved between the three species.
In the morulae of all three species, outer cells acquire an apical–basal cell polarity, with expression of atypical protein kinase C (aPKC) at their contact-free outer edges. Outer cells also show nuclear expression of Hippo signalling pathway effectors, and initiate expression of a trophectoderm programme, exemplified by the switching on of the transcription factor GATA3, which lies downstream of aPKC and the Hippo pathway.
The apical-basal polarisation has developmental significance. By inhibiting aPKC by small-molecule pharmacological modulation or Trim-Away protein depletion, the Hippo pathway effector YAP1 can no longer enter the nucleus. As a result, GATA3 expression is reduced, and trophectoderm initiation is impaired at the morula stage in all three species.
This comparative embryology analysis provides insights into early lineage specification and suggests that a similar mechanism initiates a trophectoderm programme in human, cow and mouse embryos. While the paper is important in its own right, it is also part of Niakan’s efforts to determine which species may provide better surrogates for human development, so that human embryos can be used only when absolutely necessary.
In parallel with her lab’s work on unravelling preimplantation development, Niakan has also been assessing the accuracy and efficiency of CRISPR-Cas9 gene editing in human embryos, something that is clearly of huge clinical significance. In a recent paper (Alanis-Lobato et al, 2020), her group has analysed on-target events in CRISPR-Cas9-mutated embryos to discover whether they encompass more complex mutations missed by conventional genotyping methods.
By computational analysis of single-cell genomics and transcriptomics datasets from OCT4-targeted human preimplantation embryos, the lab has observed loss-of-heterozygosity in edited cells that span regions beyond the targeted locus, suggesting that there were either large deletions or gene conversion events. Strikingly, chromosome 6, on which the OCT4 gene is located, had suffered segmental loss or gain in 20% of samples. These unintended genome editing events suggest a hitherto unrecognised complexity of mutation at on-target sites that must be addressed before widespread clinical usage of genome editing techniques can become a reality.
As the first holder of a national research licence to perform genome editing in human embryos, Niakan has created a framework for future functional investigations with the potential to transform understanding of human biology. The licence has attracted widespread governance and policy interest, both in the UK and abroad, and Niakan is recognised as a world authority on the legal, ethical and technical intricacies of working with early human embryos. Her lab’s international reputation has also resulted in multiple collaborations, both within and outside the Crick. Notably, the group generated extensive pre-clinical data that formed part of the evidence used to support a change in UK law regulating mitochondrial replacement therapy, a novel reproductive technology to prevent fatal inherited mitochondrial diseases (Hyslop et al, 2016).
Future breakthroughs in the understanding of human developmental programmes will come from a combination of the knowledge gained from working with in vitro stem cell models, model organisms and primary human embryos. In the last quinquennium, Niakan has made seminal contributions to human embryo research, giving the field a solid foundation on which to build.