New Paper on CRISPR of Human Embryos Highlights Serious Tech Problems

Human embryo CRISPR

A new paper was recently published on CRISPR of human embryos for genetic modification. I’ve got to read this publication more carefully, but here’s a quick initial take on this human CRISPR 2.0 study.

The paper is Kang, et al. and is entitled “Introducing precise genetic modifications into human 3PN embryos by CRISPR/Cas-mediated genome editing”. This is only the second paper reporting genetic modification of human embryos. The first paper, also from China, created quite a stir last year and reported a lot of technical problems with the genetic modification of the human embryos.

Human embryo CRISPR
Part of Figure 1 from Kang, et al.

This new paper comes from a team led by Yong Fan from Third Affiliated Hospital of Guangzhou Medical University, Guangzhou, China. The researchers focused on using CRISPR to try to create a specific form of the CCR5 gene allele called CCR5Δ32 that reduces human susceptibility to HIV. As in the first human embryo CRISPR paper, these scientists also used so-called 3PN human embryos that are generally believed to be non-viable. The point of the use of 3PN embryos is to reduce ethical concerns.

How did the CRISPR’ing of human embryos turn out this time?

In rare cases the team were able to create the CCR5Δ32 mutation within the human embryos (5-15%) and they did not observe in a limited screen any so-called off-target effects (i.e. CRISPR making edits in other genomic spots such as other genes where it might go). These are positive outcomes from the study. At the same time, however, they ran into trouble with another serious issue and that is that their approach more often than not led to the creation of other CCR5 mutations besides CCR5Δ32.

These other mutations called Indels were observed 36-63% of the time and could lead to complete inactivation of CCR5 or production of abnormal forms of the protein. CRISPR can lead to Indels instead of precise point mutations due to cells utilizing a specific type of DNA repair in response to cutting by Cas9 called non-homologous end joining.

In order for an attempt to make human embryos that bear CCR5Δ32 to be successful even in an in vitro experiment with no intent to produce a genetically modified person, you could not have the embryos have CCR5 Indels. It is also important to point out that the authors did not do whole genome sequencing to look for off-target effects, but only looked at a couple dozen or so predicted off-target sites. Furthermore, a successful production of CCR5Δ32 human embryos would need to have both CCR5 alleles get changed to have the CCR5Δ32 mutation. The authors did not find such homozygous CCR5Δ32 embryos had been created in their studies. Another problematic issue was that they observed evidence of mosaicism in these embryos as well.

Thinking about this in the big picture and longer term, the idea to change the CCR5 allele to make humans resist HIV infection would most likely invoke creation of a genetically modified person, but I suppose could also involve modifying blood cells ex vivo for subsequent infusion in more of a gene therapy approach.

Overall my quick first take on this paper is that more than anything it shows the many problems with hoping for success with this kind of research. The paper does not in my opinion strengthen the case that CRISPR’ing of human embryos with reproductive intent is ever something that could work well enough to be done clinically.

The authors seem to realize that and they write, “We believe that any attempt to generate genetically modified humans through the modification of early embryos needs to be strictly prohibited until we can resolve both ethical and scientific issues.”

The key word there seems to be “until”, which seems to leave the door open at least a crack.

I expect we will see many more papers like this one and frankly I’m not quite sure how I feel about that.

For my take on human genetic modification using CRISPR more broadly including the targeting of CCR5, check out my new book, GMO Sapiens.


  1. What really strikes me in this amazing paper is not that somatic mtDNA mutations are carried to iPSC cells but how oocytes escape mtDNA degradation? A 25 year old woman already has a lot of mutations in her somatic mtDNA, but the

Leave a Reply