On the Identifiability of Genome Editing in plants - Commentary of the ZKBS on Y. Bertheau (2019)

In spring 2019 the article „New Breeding Techniques: Detection and Identification of the Techniques and Derived Products” by Yves Bertheau was published in the book “Encyclopedia of Food Chemistry” [1].

In his article the author hypothesizes, that one can unequivocally identify whether mutations in the plant genome originate from natural mutagenesis processes or whether they were generated by Genome Editing. He further claims that it would just be a matter of establishing appropriate methods for this issue.

Firstly, Bertheau correctly notes that plant genomes are basically very stable and that mutations occurring spontaneously in somatic cells are very rarely passed on to the next generation. The use of New Breeding Techniques like CRISPR/Cas, TALEN, ZFN, ODM and others, referred to as Genome Editing, that generate targeted mutations would stress the cells to an exceptional extent and lead to a significant increase in general mutation frequency. Bertheau claims that the resulting genetic and epigenetic alterations are distinguishable from naturally occurring mutations so that in the future simple, fast and cost-effective/affordable routine techniques could easily be used to detect those mutations. According to Bertheau it would also be possible to conclude which New Breeding Technique was applied (identification). The author suggests combining different qualitative and quantitative methods, as for instance phenotypic characterization, investigation of gene expression patterns, various sequencing techniques, sequence comparisons and suchlike.

Bertheau suggests that unintentional and spontaneous mutations could be distinguished from intended alterations and claims that the applied New Breeding Technique could be identified as well, all by characterization of the organism, especially through phenotype characterization and different sequencing procedures.

This suggestion lacks a scientifically sound explanation and is unfeasible according to the state-of-the-art of science and technology.

Both, the natural spontaneously occurring mutations as well as the mutations generated by Genome Editing are based on the mechanism of the natural repair system of the plant cell that is used to repair double strand breaks, the so-called non-homologous end joining (NHEJ). This mechanism is activated after a double strand break and reseals the break, which is accompanied by the loss or insertion of base pairs ("indels"; mutations based on insertion or deletion). NHEJ per se is completely independent of the cause/trigger of the double strand break. The resulting mutation therefore cannot be used to identify its causal trigger.

Bertheau argues that the use of sequence specific nucleases would create traceable indels with specific patterns such as the frequency of appearance and specific sizes. However, because indels always arise from the NHEJ-mechanism, no statement regarding the causal agent of the double strand break is possible.

Furthermore, Bertheau argues that besides the desired mutation additional mutations as well as epigenetic changes occur due to the stress the New Breeding Techniques cause. These so-called “scars” would be characteristic for the respective New Breeding Technique.

Epigenetic changes are characterized by low stability and are therefore unsuitable as markers for routine identification techniques, especially for products with market authorization. Additional mutations that remain in the DNA besides the desired mutation after application of the New Breeding Techniques are primarily caused by in vitro-techniques per se. This is well known and happens mainly during single cell isolation, in vitro propagation, callus formation and plant regeneration. It is not possible to conclude for these well-known mutations, which are already applied in breeding (“somaclonal variation”) that they originate from New Breeding Techniques. The reason is that they arise likewise in the course of classic mutagenesis techniques, embryo rescue-techniques and tissue culture or even completely naturally, triggered by abiotic or biotic stress.

According to Bertheau artificial mutations generated by nucleases can be found in the proximity of distinct signature sequences such as the PAM sequence for CRISPR/Cas, because all site specific DNases would require a signature in addition to their recognition sequence in order to cut the target site. Hence, edited genomes would display/contain characteristic signatures close to the edited site that could be identified and would allow drawing a clear conclusion regarding the applied New Breeding Technique. This is incorrect. No such signatures are known/described for TALEN, ZFN or meganucleases. They also do not exist for ODM. Furthermore, as research progresses, it is by now possible to create a mutation using CRISPR/Cas at nearly every position in a genome, for instance by variation of the PAM sequence.

To ascertain whether a distinct plant harbors more mutations after applying a New Breeding Technique than would be expected naturally a precisely sequenced, eligible reference genome would be required for this plant/variety. However, there will not be a single qualified reference genome due to the massive variability of plant genomes. According to Bertheau the so-called “pan-genomes” that map existing variability of plant genomes in databases could provide remedy. However, in the opinion of the ZKBS these are inadequate to be used as reference genomes for identification due to their inherent fluctuation that would require a permanent update.

Furthermore, in the development of new plant varieties, various mutations are lost in the continuing cultivation by further out-breeding. The following generation not only receives variants of the other parent but moreover, new mutations might occur spontaneously in each generation. This camouflages an allegedly characteristic pattern of mutations generated by Genome Editing (techniques).

When it comes to surveillance of imported genome edited plants it is an unmanageable challenge that these plants are not regulated as genetic engineering in many important agricultural countries. Accordingly, no information regarding the production process as well as the introduced mutations has to be provided. Without this knowledge, it is impossible to determine the cause that accounts for a detected deviation from the reference genome, no matter whether the cause is natural or technical.

The option suggested by Bertheau to retroactively identify an edited base within a plant genome and the technique used to generate it is non-existent. The proposed methods are not based on current scientific knowledge and furthermore involve highly variable biological parameters (like epigenetic changes) that are no reliable base for identification.

In addition, the ZKBS refers to the report of the Joint Research Centre (JRC) of the European Commission and the European Network of GMO Laboratories (ENGL) which concludes that the hints from characterization of a plant that point towards genome editing as quoted by Bertheau are not sufficient to identify which technique was initially applied.

[1] Bertheau, Y. (2019). New breeding techniques: detection and identification of the techniques and derived products. In: Laurence Melton, Fereidoon Shahidi, Peter Varelis, Encyclopedia of food chemistry (p. 320-336). AMSTERDAM, NLD : ELSEVIER SCIENCE BV., DOI: 10.1016/B978-0-08-100596-5.21834-9.

published: September 2019