select search filters
roundups & rapid reactions
factsheets & briefing notes
before the headlines
Fiona fox's blog

expert reaction to study attempting to correct beta-thalassemia mutation using base editing in human embryos

Scientists publishing in Protein & Cell report the first study using base editing to correct a mutation in human embryos, for the disease beta-thalassemia.


Dr Dusko Ilic, Reader in Stem Cell Science, King’s College London, said:

“Chinese scientists have demonstrated a moderate success in correcting a mutation in hemoglobin subunit beta gene that is linked to beta-thalassemia. Although the technology will be getting better and more precise with each new attempt, it is difficult to imagine that would replace clinical practice currently available for the genetic profiling of embryos, which can distinguish healthy embryos from the embryos carrying single gene mutation(s) linked to disease. Such screening, called Preimplantation Genetic Diagnosis (PGD) is available for several hundreds of monogenetic diseases (diseases linked to single gene mutation) and has been in place for years resulting in the births of thousands of healthy babies, without exposing embryos to risks of gene editing.”


Prof. Matthew Cobb, Professor of Zoology, University of Manchester, said:

“The most striking thing about this study is that it was done on human embryos. The real potential of this technique is not to change the germ line of a future generation, but instead to cure people who are currently living with thalassaemia or sickle cell disease. Using CRISPR to change a patient’s blood stem cells could lead to life-changing possibilities, without any of the ethical worries associated with manipulating embryos. If this technique can be made more effective – at the moment only 1 in 5 attempts succeeded – and it was focused on blood stem cells, real progress could be made.”


Dr Helen Claire O’Neill, Programme Director, Reproductive Science and Women’s Health, University College London, said:

“Many hereditary diseases, such as β-thalassemia, have life-threatening implications and yet are the result of a mutation which causes a single letter change in an individual’s genome. The work presented here uses a technique called “base editing” which attempts to repair a mutation at the single letter (or base), resulting in the hopeful correction of the mutation. It does this without the need for breaking the double-stranded DNA which attempts to initiate innate (but inefficient) repair pathways. This powerful study sheds new light on precise gene correction for single gene disorders. It remains to be seen whether the efficiency, as reported here, for cloned human embryos (between 7.0% and 25.9%), can be improved upon.

“More work is needed to assess the precision of this base editing technology using genome-wide assays (here, only certain parts of the genome were analysed for unwanted genetic changes), to fully investigate both efficiency and specificity of the technique.”


Prof. Darren Griffin, Professor of Genetics, University of Kent, said:

“For many years, we have been saying that direct gene editing in embryos is some way in to the future.  Now the future is here and there is much to consider.  The paper itself represents a significant technical advance because, rather than using the classic CRISPR-Cas9 technology previously reported, the current “base editor” technology is an adaptation that chemically alters the DNA bases themselves.  The authors freely admit that they have not successfully achieved 100% success in human embryos (although that have in mouse).  Equally the presence or absence of “off target” effects needs a more thorough examination.

“While this is undoubtedly a highly significant advance, it is important not to get carried away about its widespread utility if put into clinical practice.  An embryo would still need to be diagnosed as abnormal (if there were other embryos in the cohort that were normal they would presumably be used instead), then the base editor applied, then re-diagnosed to make sure that it had worked.  This would be an involved procedure that would be very expensing.

“In the meantime, the ethical implications of gene manipulation in embryos need a thorough examination where safety is of paramount concern.”


Prof. Shirley Hodgson, Professor of Cancer Genetics, St George’s, University of London, said:

“This is another breakthrough in molecular editing from a Chinese group, in which they have used CRISPR with a base editor complex to cause single base changes in the DNA sequence at specific sites in cells cultured from skin cells and human embryos. This was not very efficient, correcting about 20% of the cells. The cell should then correct the gene in the complimentary DNA strand to the appropriate base to “correct” the mutation. However, sometimes it corrected the induced base change, thus reverting to the original DNA sequence. In addition, there is still the possibility of off target mutations as a result of this technique. Notably this method can only correct genetic disease mutations which are caused by a single base change in the DNA sequence, which is the case in only a very small minority of genetic disorders. However this is an important demonstration of a novel gene editing technique which could be an important tool in correcting genetic disorders in the future.

“This method has advantages over the technique previously used, CRISPR/cas9, which uses homologous gene repair, but has the disadvantages of potential mosaicism, off-target mutations and recombination with different DNA sequences, thus possibly being inaccurate and inefficient. The CRISPR/cas9 system could not repair diseases caused by mutations in both copies of the genes, as in autosomal recessively inherited diseases, because of the absence of a normal copy of the gene in question.

“This new technique could thus provide an important new tool for the treatment of a minority of genetic disorders caused by single base changes including autosomal recessive disorders such as sickle cell disease and cystic fibrosis, but clearly much more research needs to be done to perfect this method.”


Dr David Clancy, Lecturer in the Department of Biomedical & Life Sciences, Lancaster University, said:

“It appears that the method described in Liang et al. is an improvement upon standard CRISPR/Cas9 techniques, however because the entire genome was not sequenced to look for off target changes, we cannot know how precise the technique was. Given the nature of biological complexity, a precise method which can be guaranteed not to induce off target mutations is some time away and precision plus efficiency may only be achievable once whole genome sequencing becomes very fast and very cheap.

“If we want to use techniques which carry a degree of error, and therefore danger to the offspring, like mitochondrial replacement therapy, we need to debate first the right to, and then the value of, genetic identity with one’s offspring. That is, the value we place, and therefore the lengths to which we will go, to enable people to have children who are genetically related to them, instead of avoiding inherited disease by using donor IVF. Wherever there is the chance of harm to the offspring, regulators and legislators need to be clear about this trade-off. It is a debate which has not yet been had, but this paper and others like it suggest that now might be the right time.”


Prof. Robin Lovell-Badge, Group Leader, The Francis Crick Institute, said:

“This new study by Liang et al. (the same group who published the very first paper in 2015 using genome editing techniques in human embryos) is interesting in several respects. They make use of a variation of the methods where CRISPR guide RNAs are used to effect a specific genetic change of a version of a gene (allele), but rather than relying on Cas9 to cut DNA, which usually leads to insertions or deletions, or with an added DNA template to make precise deletions, additions or exchanges, they use modified versions of Cas9 to allow so-called “base-editing”. This does not cut DNA, but relies on an enzyme that changes cytosine (C) to uridine (U) [by a process called deamination]. The uridine is then converted to a thymidine (T) by normal cellular processes. (Uridine is a base almost always found in RNA, not DNA.) This results in a C to T conversion, and because C pairs with G and T with A, in DNA, this can leads to a change of the CG base pair to a TA base pair.

“As the authors discuss, cases of beta-thalassemia can be due to mutations that have converted A to G in a particular location within the beta globin gene involved in regulation of its expression. The mutation significantly reduced the level of activity (expression) of the gene, meaning that there is insufficient globin made to carry oxygen in the blood. They set out to ask if using the base-editing methods, to convert C to T in the opposite strand of DNA would lead to a conversion of the mutant allele from G back to A.

“They first tested this in an established cell line, in which they had engineered a copy of the mutant allele, and showed that it did work to a reasonable extent. However, there was a second G close to the specific one they wanted to change, and this was also often changed to an A. They also tested the same methods in skin cells obtained directly from a beta-thalassemia patient. Again, conversion (often of both G’s) was not at very high frequencies, but at a rate that if it were applied in patients would give at least some benefit. Because they used fibroblasts that don’t express beta-globin, they could not assess whether they have restored expression to normal levels in those cells that have corrected the mutation and whether the conversion of the second G had an effect or not.

“They also wanted to test the methods to see if they would work in early human embryos, as a proof-of-principle towards eventually applying this to avoid inheritance of the mutant alleles and therefore beta-thalassemia in children. The ideal test of the methods would be to use homozygous mutant embryos to see if one or both mutant allele showed conversion to a normal allele. It would be difficult to obtain such embryos from donors, left over after IVF, because both parents would have to be carriers of the mutant allele (or beta-thalassemia patients). While homozygous mutant embryos may be obtained from couples undergoing PGD to avoid having children with beta-thalassemia, these will have developed too far to be of any use for the genome editing techniques, which need to be introduced into the unfertilised or fertilised egg. (PGD involves biopsy of 8- cell embryos or more usually blastocyst stages.) Obtaining sperm from carriers may be possible, but obtaining eggs also from carriers and then creating embryos for research would have been challenging practically and perhaps ethically.

“The authors therefore adopted a clever approach. This was to use somatic nuclear transfer (i.e. cloning methods). They used homozygous mutant fibroblasts from the patient, and transferred the nuclei from these into oocytes (unfertilised eggs) from which their own nuclear genetic material had been removed. [The oocytes had been donated for research by women undergoing fertility treatment where ICSI (intracytoplasmic sperm injection) was being used. They used oocytes that had not matured sufficiently for this, but which could then be matured in the lab.]

“The oocytes were injected with “base editing” components, this time using an improved version which was hoped would only target the mutation and not the adjacent G, just prior to the nuclear transfer. They found precise conversion of the mutant G to an A in a significant proportion of the cells of the resulting embryos; in some cells both copies of the allele had been corrected, in others just one copy. The second G had not been converted. This looks very promising, but all the embryos from which they were able to get information from more than one cell were clearly mosaic, with some cells still carrying both mutant alleles.

“I should stress that the authors are not talking about cloning genome edited people ! The cloning methods were used only to generate embryos that were kept in vitro for a few days to allow them to test the methods.

“It should also be noted that the authors are cautious and certainly do not claim that the methods should be applied – especially as they found that there were sometimes inappropriate edits, many cases where the editing had not occurred at all, and that mosaicism was common. Inappropriate editing, involving excision of a base rather than conversion, seems to have occurred in some cases, which would be unlikely to help at all in this example of a genetic disease. There is scope, however, for improving probably each step in the methods, for example using inhibitors to reduce the likelihood that excision would occur.

“It is a complex paper, with some interesting results that might indicate a route to avoiding certain genetic diseases in both somatic and germline treatments – although it is far too early to even consider applying the methods clinically.”


* ‘Correction of β-thalassemia mutant by base editor in human embryos’ by Liang et al. was published in Protein & Cell on Saturday 23rd September.


This story came from the BBC report found here.


Declared interests

Dr Dusko Ilic: No conflicts of interest.

Prof. Matthew Cobb: No conflicts of interest.

Dr Helen Claire O’Neill: No conflicts of interest.

Prof. Darren Griffin: No conflicts of interest.

Prof. Shirley Hodgson: No conflicts of interest.

Dr David Clancy: No conflicts of interest.

Prof. Robin Lovell-Badge: “Robin Lovell-Badge has no conflicts of interest.  He uses genome editing methods in his own research, but this does not involve human embryos.”

in this section

filter RoundUps by year

search by tag