Human Germline Editing – What’s the big deal?

Author: Dr Jonathan Appleby, 14th May 2019

In November 2018, He Jiankui, gave a presentation at the second international summit on Human Genome Editing (1).  During the presentation, social media led claims that he had successfully used CRISPR Cas9 gene editing to modify the germ line of human embryos which were then implanted back into the donor and followed through to birth, were confirmed.  Being the first known example of deliberate human germline editing, the research has attracted considerable attention.  In this blog I will try to summarise and explain why the research has been widely condemned as inappropriate and explore some of the issues that this research highlights.

So what exactly happened?  Whilst the details have not been extensively published in peer reviewed literature, He’s presentation (1) is a good source of information.  During his presentation, and the follow up Q&A, it was explained that preclinical work was conducted in order to develop a CCR5 gene editing procedure.  The CCR5 gene produces a cell surface protein which some HIV strains need to infect T cells.  The CCR5 gene was made famous in 1996 by the “Berlin Patient”, who after a bone marrow transplant from a homozygote donor who carried two naturally occurring mutations in the CCR5 gene known as Delta 32, became HIV free (2).  The objective of He’s gene editing was to change both CCR5 loci in the embryo so that they were the same as the delta 32 variant, which is present (as a single copy) in about 10-13% of Northern Europeans.  If this germline editing could be achieved, then it would provide some degree of protection against HIV infection for the lifetime of the recipient.

Having generated data in preclinical models and on test embryos He Jiankui started a clinical trial that enrolled seven couples, with the objective of generating homozygous Delta 32 CCR5 gene edited embryos which would then be implanted back into the mother and followed to term.  This is highly irregular because, although most countries do allow embryo gene editing research, it is not permitted to grow the embryo beyond a few weeks or allow such gene edited embryos to be implanted into humans. 

During the conference presentation He explained that one mother had been implanted with two embryos that had been followed to term and been born fit and healthy. One child had two gene edited alleles for the CCR5 gene (a compound heterozygote) and one child was gene edited at one locus, all three modifications were slightly different.  During questioning after his presentation, it emerged that a second mother was also pregnant, the details of the genetic modification in the second woman were not discussed.

Much of the published discussion that followed has focussed on the choice of target and the clinical situation.  Typically, potentially high-risk medical techniques are first used in clinical situations where the patient has a very poor prognosis, where the risk to the patient from the procedure or treatment seems outweighed by the potential benefit.  In this CCR5 gene editing situation, the benefit side of the decision comes from a desire to protect a child from future risk of infection. There was no benefit to the child prior to birth because the IVF procedure itself has steps within it to prevent HIV transmission to the embryo.  Without trivialising the seriousness of HIV, in particular in circumstances were access to treatment is poor, the response of commentators clearly suggests that protection of a person from future risk of HIV is not widely supported as a justifiable rationale for the application of germline editing. 

The choice of target has also been questioned on grounds that we do not understand enough about the function of CCR5 and how changing the gene may put the recipient at risk.  In this respect the community seems to be somewhat conflicted.  On the one hand there is evidence that the delta 32 variant creates a susceptibility to West Nile Virus (WNV) infection (3,4).  WNV can cause CNS damage, to give some idea of incidence, upward of a thousand serious infections occur every year in the USA.  Any increase in susceptibility would indeed be grounds for questioning the ethics of the research.  On the other hand, clinical trials are ongoing in the USA and China attempting to gene edit T-cells and bone marrow stem cells (i.e. CD34cells) so that they carry the delta 32 allele (5,67).  Note also that the gene editing approach was not well controlled and therefore probably not ready for introduction into the clinic.  The investigator knew that the changes introduced into the embryo were not identical to the normally occurring delta 32 variant.  Indeed, one embryo was unaltered at one allele (i.e. was heterozygote) and would therefore expect to gain no benefit from the modification, only risk from the procedure.

After learning more about this example of human germline editing I expected there to be broader public engagement and debate.  I thought perhaps, that we would hear about significant calls from the public to allow germline editing as a treatment option for parents whose children are at risk from life threatening rare diseases.  I have experience of developing non germline gene therapy treatments for children with rare disease, so understand all too well how distressing paediatric rare disease are.  I wanted to know if there was sufficient justification for germline gene editing to prevent inheritance of debilitating and frequently fatal paediatric disease.  As part of my research to address this question I started to learn about a service offered by the Human Fertilisation and Embryology Authority called PGD, or Pre-implantation Genetic Diagnosis.

As a novice in embryology I found PGD to be hugely impressive.  Essentially, embryos are generated with standard IVF, then at approximately 5 days when the embryo has reached 100 to 150 cells (the blastocyst stage) a small area of the trophectoderm (i.e. the area which will develop into the placenta) is cut away.  You can see a video of this being done in ref 8, the cells are then tested for the presence of the disease allele.  When I watched the procedure, I thought there might be some detrimental impact on the developing foetus but published success rates for embryos that have been selected through PGD vs embryos that don’t undergo PGD are very similar.   

Here in the UK, several hundred diseases are eligible for screening with PGD (9), and the process for adding new diseases is relatively quick and simple. However, PGD is not a panacea for the prevention of genetic disease.  Many cases are spontaneous, i.e. the parents are not carriers in the first instance.  Most cases derive from recessive inheritance patterns were parents are unaware that they are carriers.  PGD can only be applied as a screen for disease when the parents are aware of some risk, such as following the birth of a first affected child or through an awareness due to family history. 

When discussing germline editing it is necessary to think not just about the “safety” of the technique.  It is an over simplification to say germline editing has not been proven to be safe for human use. This recent study with CCR5 may become a marker of safety if the children stay healthy. Rather the application of new technology should be done after consideration of the benefit risk balance.  In geographies in which PGD is available there seems little to gain in medical terms from germline editing.  In February 2019 Chinese authorities announced an explicit ban on the implantation of germline edited embryos and additional regulation of “high risk” procedures (10).  Whilst germline editing for research will continue this case involving CCR5 has also prompted calls for an international moratorium blocking human implantation of germline edited embryos.

By Dr. Jonathan Appleby, PhD; Chief Scientific Officer, Cell and Gene Therapy Catapult.


  2. Hutter, G., et al. (2009). “Long-term control of HIV by CCR5 Delta32/Delta32 stem-cell transplantation.” N Engl J Med 360(7): 692-698.
  3. Glass, G. et al. (2006). “CCR5 deficiency increases risk of symptomatic West Nile virus infection.” J Exp Med 203(1):35-40.
  4. Durrant, D.M. (2015). “CCR5 limits cortical viral loads during West Nile virus infection of the central nervous system.” J Neuroinflammation 12:233.