Gene Editing of Human Embryos with CRISPR/Cas9: Great Promise Coupled with Important Caveats

Main Text

Modern gene editing technologies with unprecedentedly high accuracy have now enabled precise editing of cellular genomes, and have tempted some to attempt to edit human embryos despite the controversy this engenders within the research community and beyond. Ma et al.¹ recently described the use of CRISPR/Cas9 technology to correct a pathogenic mutation in human embryos. The authors corrected a heterozygous MYBPC3 mutation in human preimplantation embryos through homology-directed repair. This autosomal dominant mutation leads to hypertrophic cardiomyopathy, which is a common cause of sudden cardiac deaths in young people, particularly athletes, who are carriers of the mutant allele.

Several research groups in China have previously reported genome editing with the CRISPR/Cas9 technique in human embryos.2, 3, 4 However, gene targeting efficiencies in these previous studies were quite low and accompanied by significant off-target effects as well as mosaicism. Ma et al.¹ reported that 42 (72.4%) out of 58 human embryos fertilized with sperm carrying the heterozygous MYBPC3 mutation carried two copies of the non-mutated gene copy after CRISPR/Cas9-treatment, as opposed to 47.4% in the control, untreated embryos. Moreover, only one mosaic embryo was detected, and the authors reported a 100% targeting efficiency based on the absence of detected heterozygous blastomeres.

One reason for the improved outcome is that the components of the CRISPRS/Cas9 system were injected directly into the MII-phase egg cells along with the sperm. Thus, gene editing took place much earlier than in the previous studies. Since recombinant Cas9 protein was used, it was probably rapidly degraded and had less time to cleave DNA at off target-loci, as compared to techniques where these components were delivered as DNA constructs.

Formation of mosaic embryos, which has hampered previous attempts with similar goals, was very low. It appears that a crucial factor underlying this improvement was the more optimal timing of the CRISPR/Cas9 injection into oocytes. Quite unexpectedly, the DNA repair machinery in the embryos preferentially used the maternal wild-type gene for mutation correction instead of the exogenous DNA template. Thus, early embryonal cells seem to protect and correct their DNA using special mechanisms that are not yet fully understood. According to the authors, a homology-directed repair mechanism is favored in embryos over a non-homologous end joining mechanism, although it should be pointed out that the latter mechanism also occurred at relatively high frequency, indicating that CRISPR/Cas9 can cause undesired off-target effects in the embryo. It also remains unknown whether epigenetic changes occur at this stage and after gene repair. The observed 22% increment in MYBPC3^wt/wt embryos was significant in comparison to controls, and the authors suggested the CRISPR/Cas9 technique could be applied to increase the number of available embryos for transfer in the context of in vitro fertilization and preimplantation genetic diagnosis. When only one parent carries a heterozygous mutation, 50% of the embryos are expected to be naturally mutation-free. It remains to be considered whether such increments are worthy of any interventions.

The report of Ma et al.¹ has engendered criticism. Egli et al.⁵ questioned whether efficient inter-homolog recombination can happen in early embryos where maternal and paternal genomes are still undergoing their own biological programs in distinct parts of the nucleus (i.e., pronucleae remain separated from each other). They also noted that detection of these recombination events involves significant technical challenges. If homology-directed repair was common in early embryos, it could also put cells at risk of unmasking deleterious recessive alleles via loss of heterozygosity. Egli et al.⁵ suggest that CRISPR/Cas9 gene editing might actually not have occurred in all cases via homology-directed repair and interaction between healthy maternal and mutant paternal alleles. Thus, replication and verification of the methods and findings of Ma et al.¹ are clearly warranted.

These experiments obviously raise broader ethical concerns. Fears of generating “designer babies” with genetically enhanced intellectual or other traits is a major concern with germline engineering. However, the current study may reduce such fears to some extent if the embryos preferentially used the healthy parent’s gene copy for the repair process instead of the exogenous DNA template. Nevertheless, fears of unjustified gene editing of human embryos are real and should be controlled and regulated by government and society. Interestingly, last year the National Academy of Sciences, Engineering, and Medicine endorsed the view that modifying human embryos should be allowed, but only to correct mutations that cause serious disease or condition and when no reasonable alternatives exist.⁶ This statement indicates a somewhat softened attitude toward genetic and germline engineering among researchers and ethicists. In this context, it is also interesting to note that, according to a recent study, antibodies against Cas9 from S. aureus and S. pyogenes are commonly found in human serum, and that several individuals also have T cells that can target SaCas9.⁷ Although these findings are more relevant for direct in vivo gene editing in children and adults, it should be kept in mind that the CRISPR/Cas9 technology is derived from bacteria and could also have some unknown effects on the embryos. Also, off-target effects are an obvious concern and clearly further studies are needed to determine their prevalence.

Clinical implications of gene editing in human embryos are clearly substantial. Even though much needs to be done before CRISPR/Cas9 techniques could become a reasonable and safe approach for clinical applications, the current study by Ma et al.¹ undoubtedly reports a step forward and opens theoretical possibilities to treat several human inborn errors and genetic mutations which otherwise cannot be cured with current medical practices.