TSL Statement on Gene Editing

The Sainsbury Laboratory, Norwich (TSL) advocates the view that

1. legislation should focus on the end-product and not the technology used,
2. regulation should cover any trait that could not be made with traditional technology (breeding, mutagenesis, TILLING, crossing, tissue culture, hybridisation, etc.), and
3. organisms modified by GE that do not contain foreign DNA should not be regulated as GMOs.

GE is the most precise genetic technology currently available and enables exact operator-specified DNA sequence changes, unlike other breeding methods. Modern genome sequencing methods allow the end-product to be fully characterised. These can robustly demonstrate the introduction of a desired trait at the genome level and the absence of any trace of the method used to introduce it. GE provides a much more rapid and predictable way to introduce a desired trait. The regulation should therefore focus on each trait on a case-by-case basis and approve those that are demonstrated to be beneficial.

Site-directed nucleases (SDN) enable the targeted breaking and repairing of DNA with or without a few base pairs of added DNA. Given the extensive genetic variation in natural and domesticated plant populations that are used by plant breeders, and the spontaneous mutation rate in all plants, the changes to be introduced by simple SDN1 and SDN2 versions of GE (small insertions or deletions or directed few-base changes) are minimal. Despite such changes involving only a few nucleotides, a targeted approach based on modern scientific insights enables plant breeders to provide varieties with valuable traits that confer improved performance. In particular, the work of plant breeders is not just about providing specific useful forms ("alleles") of genes, but about the combinations of alleles, and using recombination to efficiently recover such desirable combinations of alleles. GE methods enable breeders to combine useful alleles more quickly and more efficiently.

At The Sainsbury Laboratory, we are advancing knowledge about the very important wheat blast disease, which causes significant losses and recently has been found both in Bangladesh and Zambia; spread of the disease to neighbouring countries is almost certain to occur. As part of its virulence mechanism, proteins from the disease organism target so-called HMA proteins in the host. Using GE, we have mutated wheat genes encoding HMA proteins, resulting in wheat lines with reduced susceptibility to blast. It is not in the public interest for wheat lines with elevated blast resistance resulting from our use of GE methods to be regulated as GMOs. Although individual mutations that we introduced could have occurred "naturally", the use of GE has dramatically short-circuited the process, reducing the many years that it would have taken to find conventionally bred plants with the combination of mutations assembled to enable disease resistance.

Organisms produced using GE do not pose a greater risk of harm to human health or the environment than their traditionally bred counterparts

For plant breeding, use of GE (and also GM) poses a similar (and extremely low) risk of harm compared to conventional breeding varieties. There is no plausible and scientifically validated mechanism by which use of the methods per se could confer elevated risk of harm. The variation introduced by the use of these methods to create SDN1 and SDN2 events is tiny compared to pre-existing variation between crop varieties and breeding lines, and between cultivated lines and the wild relatives that are often used as sources of useful genetic variation for crop improvement. The methods currently employed in plant breeding can lead to the loss, gain or recombination of hundreds of genes at random – an outcome that is safe and accepted to be so.

Certain non-safety issues would need to be considered if organisms produced by GE were not regulated as GMO

1. If varieties emerge carrying gene editing events that can be used freely in the UK but are not approved for commerce in other markets (e.g. EU), this could create problems for exports. However, the EU has already approved many GM events for import, such as soybeans and maize for animal feed, so this should not be an insuperable barrier.
2. If edited varieties, and in particular, editing events are subject to patent protection rather than plant varieties rights protection, this could create problems for other breeders to breed from varieties carrying such events. Therefore, in the view of TSL, it is essential that gene edited events are NOT patented, and varieties incorporating them can be bred from by all plant breeders.

Mutations introduced by GE could also appear spontaneously through natural processes

"Natural" (or spontaneous) genetic mechanisms can create mutations every time a cell divides or is hit by UV light. The range and rate of resulting changes in nucleotide sequence are well defined and include deletions, transitions and transversions. Broadly, the bigger the change, the lower the probability of such a change occurring. Any change that results in transversions, transitions, or deletions could have occurred “naturally”. Only insertions of defined sequences of at least (say) 100 bp, either at a defined or random position, should be regarded as not likely to have been able to occur spontaneously.

The figures below illustrate natural mutation rates, based on measured Arabidopsis mutation rates of 1 mutation per 0.1Gb per generation. What they show is that in any wheat field, there will at least one seed that carries a mutation at any position in the genome. These mutation rates are occurring in wheat fields throughout the world, so almost any mutation that might have been introduced by gene editing is likely to be appearing regularly in fields throughout the world by natural processes. Thus, it makes no sense to impose excessively stringent regulation of edited events and any such regulation would be difficult to police given that "natural" mutations and editing-derived mutations will often be indistinguishable.

Consider 1 ha of wheat of cultivated wheat:

yield 10 t = 10⁴ kg = 10⁷ g
weight of wheat grain 50 mg = 20 grains per g
Arabidopsis: 1 mutation per 0.1 Gb haploid genome
wheat: 100 mutations per 10 Gb haploid genome = 100 mutations per grain
10⁷ g = 2 ´ 10⁸ grains = 2 ´ 10¹⁰ mutations

2 ´ 10¹⁰ per 2 ´ 10¹⁰ bp = 1 mutation per bp in every hectare of wheat.

Both GMO and non-GMO legislation need reform

No changes need to be made to existing non-GMO plant varieties rights legislation, except in that they should be extended to cover gene editing events. Overall, we strongly advocate that organisms modified by GE that do not contain foreign DNA should not be regulated as GMO.

We also believe that the present GMO regulations imported from the EU are too stringent and restricts many useful contributions to sustainable agriculture in the UK that could be made by plants improved using GM methods (usually via Agrobacterium).

We acknowledge that plants improved with the GM method can raise public concern. Therefore, if new DNA has been added to the gene repertoire of the crop, additional regulation is required to ensure public access to information about each GM event, the assessments carried out, and reasons for authorization. Some GM-specific regulation is justified to ensure that there has not been a breakdown in tracking procedures in the source laboratory, e.g. to make sure that any GM line does not carry an unintended gene such as a jellyfish green fluorescent protein. Ideally, regulation should solely consider the utility and safety of a new variety, without considering the method by which its qualities were introduced. However, we recognize that certain measures need to be put in place to inform the public and put safety concerns at rest. As each trait should be evaluated and regulated on its merits, case-by-case, the public should be informed about the basis and net benefits to address their concerns about each.

At TSL, we have used the GM method to move immune receptors against potato late blight from wild relatives of potato into the market-favoured variety Maris Piper. The resulting lines are essentially completely immune to all known races of late blight, have improved tuber properties, and yield indistinguishably from Maris Piper. In order for the UK to take advantage of the disease resistance of these new lines ("the blight resistant potatoes to feed the world"), we need reform of regulation governing crops improved by using GM as well as GE methods. We also need to communicate the merits of growing a blight-resistant GM potato in the UK with regards to how it benefits our biodiversity, the national economy, local farmers and consumers themselves.

Current GM regulations imported from the EU unhelpfully restrict the innovation required to address the huge challenge of sustainable increases in crop productivity while minimizing environmental collateral damage. We should therefore reconsider the GM regulations we imported from the EU. The regulations should ensure public health and environmental impact issues have been evaluated for each GM event. This needs to be conducted rationally, based on plausible scenarios (e.g. does a new potato variety carry elevated steroidal glycoalkaloids?) rather than addressing every conceivable "unknown unknown".

Existing regulations fail to balance the consequences of adopting an innovation (e.g. genetic control of crop diseases) with the consequences of NOT adopting it (continued spraying of agrichemicals). It would be helpful to factor that in. Furthermore, many agrichemicals such as neonicotinoids for aphid control or methyl bromide fumigation for nematode control are becoming more restricted in or removed from use. Thus, genetic resistance in our crop plants to aphid transmitted viruses and to nematodes is becoming of increased importance. If GM regulations were trait-based and considered ‘case-by-case’, the benefits of adopting a specific trait could be clearly justified against the consequences of continuing with increasingly vulnerable crops.

Existing regulations require extensive laboratory animal feeding studies for all crop GM events; these are uninformative, are questionable on animal welfare grounds and are inconsistent with a mandate to reduce unnecessary use of live animals in research.

The regulation of GM events should be based on scenarios for harm (to health or environment) that are based on realistic, well-defined and validated biological mechanisms, rather than fear of "unknown unknowns". The precautionary principle is based on the possibility that there might be "unknown bad scenarios that we don't even know we don't know about". This philosophy is no longer appropriate for the use of a method that has been extant for 38 years, has been used in crops and in the human and animal food chain for 26 years, and that has been used in thousands of labs worldwide for research purposes over the last 35 years. We now know an enormous amount about the use of this method.

In the future, any application of the precautionary principle, based on acknowledgement of incomplete knowledge, should be accompanied by an exit strategy. Such an exit strategy should be in the form of a post-cautionary principle that reviews regulations at defined intervals and assesses if they continue to be appropriate. We therefore welcome the current consultation and encourage regular reviews whatever the outcome.

DEFRA should carefully study the revised GM regulations that will apply in the US from 6th April 2021 (APHIS SECURE).

https://www.aphis.usda.gov/aphis/ourfocus/biotechnology/biotech-rule-revision#:~:text=The%20SECURE%20rule%2C%20which%20stands,they%20were%20established%20in%201987

In this revised regulation, the stringency of regulation of both gene editing and also GM events is substantially reduced. DEFRA should consider adopting many or all these revised policies, which were formulated after extensive consultation with the US plant science community and many other stakeholders.

This statement reflects TSL’s submission to the UK DEFRA consultation on the Regulation of Genetic Technologies, March 2021.