Gene editing creates a precise mess

September 2018

CRISPR-Cas9 gene-editing technology is causing the jitters. Hailed as "highly precise" and "virtually impossible to detect", CRISPR has become the GM technique which biotech and medical researchers are banking on [1]. The reason, it seems, for CRISPR's "highly precise" reputation is that it's designed (by humans) to latch onto a highly precise location in the genome.

So convinced have scientists been of the mechanistic nature of CRISPR-Cas9's seek-and-chop action, that checks on possible associated mutations have been limited to the immediate vicinity of the target site and to DNA sequences elsewhere with a known similarity to the target.

In fact, confidence is such that correction of faulty human genes using CRISPR has already moved to human trials. This may have been premature, because two studies have now pointed out that the human cells which allow their faulty genes to be sorted have other inherent faults which can lead to cancer [2].

The first major set-back happened in 2017 with a report that the successful use of CRISPR to restore the sight of blind laboratory mice was accompanied by thousands of mutations elsewhere in the genome [3]. This paper was later retracted due to doubts about the validity of the control mice*. However, two other papers published that year also indicate CRISPR-induced mutations in the wider genome and the authors of all three studies suggest the need for extended post-CRISPR DNA analysis [4].

Now another study has been published reporting wide-ranging CRISPR-induced mutations including small, large and huge (over 5000 nucleic acid units in length**) deletions, plus DNA insertions, inversions and exchanges. The large number of cells and DNA deletion events sampled in the experiment indicated a "vast" diversity of potential alterations in the genome.

Add to this that the true scale of the problem was almost certainly hidden because detection of the DNA sequences depends on having a known sequence for the test to latch on to. This means that some changes may have been rendered undetectable due to the scrambling present.

Most significantly, there was no repeatability in the outcome, and the observed disruption could certainly have pathogenic consequences.

One of the authors expressed concern that the alterations in DNA resulting from CRISPR-Cas9 editing "have been seriously underestimated before now".

London-based molecular biologist, Dr Michael Antoniou, pointed out that no matter how specific your CRISPR is designed to be, what happens after it has cut the DNA is down to holistic cell repair mechanisms.

Professor Paul Thomas, who led the widely-reported research published last year describing the successful correction of a faulty gene in experimental human embryos*** , said
"We're very good at cutting (DNA). We're still learning about what happens after you cut".
Other gene-editing tools such as TALENS, zinc fingers etc. [5] will likely produce the same co-lateral DNA disruption.


Prof. Thomas's paper sparked a scholarly set of 'Brief Communications' from other scientists suggesting a need for more comprehensive investigation of unsuspected DNA changes which gene editing might have induced in the wider genome. Thomas' team's responded by carrying out further analyses, acknowledging areas needing further research, and expressing the "hope" that the questions raised and their additional results would "serve as a useful platform for further discussions and studies". This is the way science should be done.

Compare and contrast this with what happens when a study suggests a need for more comprehensive testing for unsuspected outcomes in GMOs for food or feed. The history has been one of an immediate orchestrated barrage of fake news discrediting the scientists and disparaging their experimental technique: there's a resounding lack of encouragement to repeat or extend any study which might prove negative effects. If this makes you think the biotech industry and its henchmen have something they want to hide, it's probably because they have something they want to hide.

A lot of the questions which arose around Prof. Thomas' embryo research related to the difficulty of gene-editing an animal so that the whole body is altered; this will equally apply to attempts to gene-edit livestock. Such problems don't arise in plants because their individual cells can be gene-edited and regenerated into a whole plant. However, the need to check for potentially harmful DNA changes beyond the edited site, and the need to check for unpredicted toxins and allergens applies to all GM foods.

Don't be fooled by biotech industry propaganda: there may be such a thing as a precise, humanly-contrived, DNA change, but there's no such thing as a precise humanly-devised genomic outcome.


[1] CRISPR/Cas9 GENE EDITING - March 2016
[3] ERRORS IN CRISPR - November 2017

* Retraction doesn't mean the findings were necessarily wrong, but they may be, and the experiment certainly needs to be repeated with the appropriate rigour ** Nucleic acids are the 'NA' in DNA *** The faulty gene corrected was one known to lead to a heart condition which is the most common cause of sudden death in young athletes.

  • CRISPR causes greater genetic damage than previously thought, GM Watch 17.07.18
  • Michael Kosicki, et al., July 2018, Repair of double-strand breaks induced by CRISPR-Cas9 lead to large deletions and complex rearrangements, Nature Biotechnology
  • Liam Mannix, We can change human DNA. We're just not sure which bits,, 9.08.18
  • Hong Ma, et al., 2017, Correction of a pathogenic gene mutation in human embryos, Nature 548
  • Dieter Egli, et al., 2018, Inter-homologue repair in fertilized human eggs? Brief Communications, Nature 560
  • Fatwa Adikusuma, et al., 2018, Large deletions induced by cas9 cleavage, Brief Communications, Nature 560
  • Hong Ma, et al., 2018, Ma et al. reply, Brief Communications, Nature 560
Photo Creative Commons

No comments:

Post a comment

Thanks for your comment. All comments are moderated before they are published.