Findings in human embryos are relevant to plants and farm animals. Report: Claire Robinson

CRISPR-Cas9 genome editing can lead to unintended mutations (DNA damage) at the targeted section of DNA in early human embryos, researchers have revealed. This highlights the need for further research into the effects of CRISPR-Cas9 genome editing, especially when used to edit human DNA in laboratory research.*

The findings also have relevance to gene-edited plants and farm animals. But advocates for the deregulation of plant and livestock gene editing remain silent on the potential consequences, which as usual are only being highlighted by researchers in the clinical field. In the case of gene-edited plants, these consequences could include the production of novel toxins or allergens or raised levels of existing toxins or allergens.

In the new study published in the journal Proceedings of the National Academy of Science (USA), scientists retrospectively analysed data from previous research in which they had gene-edited human embryos during the first few days of development.

The team found that while the majority of CRISPR/Cas9-induced mutations were small insertions or deletions, in approximately 16% of samples there were large unintended mutations that would have been missed by conventional screening methods to assess DNA changes.

Limitations of CRISPR

Professor Kathy Niakan, group leader of the Human Embryo and Stem Cell Laboratory at the Francis Crick Institute and Professor of Reproductive Physiology at the University of Cambridge, and senior author of the study, commented, “It is important to understand these events, how they arise and their frequency, so we can appreciate the current limitations of the technology and inform strategies to improve it in the future to minimise these mutations.”

Gregorio Alanis-Lobato, lead author and a former postdoctoral training fellow in the Human Embryo and Stem Cell Laboratory at the Crick, said: “Conventional tests used to check the accuracy of CRISPR-Cas9 can miss the types of unintended on-target mutations we identified in this study. There’s still so much for us to learn about the effects of CRISPR-Cas9 technology and while this valuable tool is refined, we need to thoroughly examine all changes.”  

The researchers have developed an open-source computational pipeline to identify whether CRISPR/Cas9 has caused unintended on-target mutations, based on different types of sequencing data.

Chromosomal mayhem

The newly published study is one of three papers that were previously posted on the pre-print website bioRxiv in 2020, describing major unintended DNA damage (mutations) at the intended gene editing site in human embryos using the CRISPR/Cas9 tool. This led a Nature News article to label the outcomes as “chromosomal mayhem”. GMWatch reported on the three studies and the Nature News commentary in July 2020.

Relevance to plants

The mechanisms of gene editing and the subsequent DNA repair process by the cell that constitutes the "edit" are the same in plants as in humans and other animals. So there is no reason to expect that gene-edited plants would be free from such genetic errors or that plant gene editing is more "precise" than human/animal gene editing. Indeed, many studies have documented such errors in plants, as well as other organisms.

However, as with human gene editing, the screening methods that are generally used to look for genetic errors in edited plants are inadequate, meaning that many such errors will have been missed. Most studies on gene-edited plants and livestock animals use short-range PCR and short-read genetic sequencing to look for errors. But what is needed is long-range PCR and long-read genetic sequencing, which could pick up a far wider range of errors. The use of inadequate screening methods has contributed to the trend whereby plant gene editors make false claims of precision and predictability for the technology.

In a scientific review, Kawall and colleagues confirmed that the “vast majority” of studies on gene-edited plants used inadequate and biased detection methods to screen for genetic errors, meaning that they will miss many such errors. Among studies on gene-edited animals, none included a thorough analysis of genetic errors.

The computational pipeline developed by the researchers on the new study could be used to analyse sequencing data from gene-edited plants and livestock animals.

Types of genetic error are the same between animals and plants, but consequences may differ

While the types of genetic error arising from gene editing will be the same in animals (including humans) and plants, the consequences of the errors can differ markedly between species and kingdoms. For example, a gene-edited animal won't be toxic, since if it were, it would poison itself. But alterations in the function of multiple genes that can inadvertently or intentionally take place through gene editing in plants could lead to unintended and unpredictable changes in the plant’s biochemistry, including the production of toxins or allergens. And because this quality won't be obvious to the naked eye, it's unlikely to be spotted by developers unless testing is forced by regulation.

Deformities can affect both plants and animals, but because many such effects are obvious, they can be more easily be screened out from the breeding program. However, the deformities and premature deaths that can arise during research and development of gene-edited animals raise serious ethical and welfare concerns.

Can genetic errors be bred out of gene-edited plants?

Plant gene editors often say that the genetic errors produced by gene editing are bred out during the development of gene-edited crops, since there is no point in trying to sell a crop that is not fit for purpose.

However, when screening gene-edited plants, plant gene editors only look for the success or otherwise of the intended trait and whether the plant grows and performs acceptably well. They do not carry out in-depth compositional analyses known as “omics” molecular profiling to look for unintended toxins and allergens or changed levels of the same. Neither do they look for changed composition or interactions with the environment that might affect wildlife. If anyone wishes to argue that these tests are done, then they must explain why they are not published in the peer-reviewed literature.

It is possible to breed out some of the unintended genetic errors by backcrossing with normal plants – but not all of the errors will be eliminated. And it’s not easy to know the consequences of the remaining genetic errors, without further investigation such as through "omics". Also, you cannot breed out genetic errors that are linked to the gene of interest that you have changed, because if you breed out the unintended errors, you also lose the intended genetic change.

With vegetatively propagated crops, such as potato, banana, and fruit trees, you cannot breed out genetic errors, so any genetic errors that inevitably arise at the different stages of the gene editing process (plant tissue culture, plant cell genetic transformation, gene editing) will persist into the final marketed product.

In conclusion, the findings in human embryos are relevant to plant and farm animal gene editing. Researchers need to face up to this fact, employ adequate screening methods and analytical tools (including the one featured in the new publication) to look for genetic errors in gene-edited organisms. And then they need to do further research to investigate the possible consequences. In the case of plants, that means “omics” analyses and long-term animal feeding studies.

What is also clear is that the full effects of gene editing methods are only now starting to be understood and that claims that gene editing is so precise that its use in animals and plants can be safely deregulated are not supported by a rapidly growing body of research.


* In the UK, gene editing of human embryos is regulated and is currently only allowed for research purposes. Research is restricted to the first 14 days of development and embryos are not allowed to be implanted into a womb. Countries outside the UK have also adopted the 14-day rule. However, it is being challenged by scientists who want to extend this limit.

Source of comment on the new study in human embryos: Francis Crick Institute
https://www.crick.ac.uk/news/2021-04-09_researchers-call-for-greater-awareness-of-unintended-consequences-of-crispr-gene-editing
---

The new study:

Alanis-Lobato, G., et al: Frequent loss-of-heterozygosity in CRISPR-Cas9-edited early human embryos. PNAS, April 2021. DOI: 10.1073/pnas.2004832117