Monday 13 August 2018

EU law and genetic mutations: does a recent CJEU ruling show an irrational fear of mutation?

A recent CJEU judgment has made it more complicated to develop plants which have been cross-bred by mutagenic breeding. Two contributions to the blog (from Kathleen Garnett and Felix Beck) examine the judgment from different angles.  

Confédération Paysanne & Others, Case C-528/16: GMO cabbage or plain old cabbage?

Kathleen Garnett, Consultant and writer on EU government and law

In September 2016 Stefan Jansson, professor in Plant Cell and Molecular Biology at Umeå University in Sweden sat down to a meal of cabbage and pasta, which he shared with a good friend. Prof Jansson had grown the cabbage successfully in the back of his garden in Sweden using conventional cottage-garden husbandry. He claimed it was a historic first – no one else on planet earth (to the best of his knowledge) had ever eaten a CRISPR Cas-9 cabbage. He obtained the mutagenic cabbage seed from a scientist abroad (who preferred to stay anonymous).

Two years on and Prof Jansson along with many of his colleagues in the scientific community are hugely disappointed with the European Court of Justice’s ruling on mutagenic plants, which came out on 25 July 2018. Commenting on his cabbage in Nature Jansson noted “I took a photo yesterday, and I took another after the ruling. It’s still the same plant. Yesterday it wasn’t a GMO, and now it’s a GMO. I’m a bit curious what I have to do. Do I have to remove it?”

The CJEU’s ruling in the Confédération Paysanne & Others case has led to derision within the scientific community. Some have called the decision “catastrophic”, others “absurd". In this piece I set out the exact legal reasoning of the judgment. Since this case concerns a very specific form of plant breeding technique called “mutagenesis”, I briefly explain what is meant by mutagenesis in order to place the case in context.

Background on transgenesis and mutagenesis

The two most common methods for scientist to change the DNA structure of plants through artificial means are transgenesis and mutagenesis.

Transgenesis occurs when scientists transfer a gene from one species into the genome of another species. This happens through highly sophisticated, scientific techniques, such as the use of a gene gun or through the use of soil bacteria.

Cross-breeding between species does not occur in nature. A dog cannot mate with a cat and produce off-spring because they are from a different species. Similarly, in nature an apple cannot cross-breed with a grass because of the species barrier. Species which are closely related can produce off-spring but the off-spring are infertile – as is the case with the mule.

Through transgenesis, however, bio-engineers have been able to cross the species barrier by inserting the DNA from one species into another species to produce crops with enhanced traits that are capable of replicating. Transgenic techniques (although not defined in the Directive as such) are described in Annex I A, Directive 2001/18 (the EU law on GM foods) as:

-          recombinant nucleic acid techniques involving the formation of new combination of genetic material by the insertion of nucleic acid molecules produced by whatever means outside of an organism, into any virus, bacterial plasmid or other vector system and their incorporation into a host organism in which they do not naturally occur but in which they are capable of continued propagation;

-          direct introduction into an organism of heritable material prepared outside the organism including micro-injection, macro-injection and micro-encapsulation;

-          cell fusion (including protoplast fusion) or hybridisation techniques where live cells with new combinations of heritable genetic material are formed through the fusion of two or more cells by means of methods that do not occur naturally. (emphasis added in all cases).

In the US many maize, soya-beans, corn and cotton seeds are grown using this technology. Over 40% of US agricultural land cultivates transgenesis crops. In the EU there is only one commercial variety of transgenesis crop – the Mon 819 – which was approved in 1998 before the 2001 GMO Directive. Although cultivation of GMO crops in the EU is low to non-existent the EU does, nevertheless, import huge amounts of GMO soya beans for animal feed from over-seas.

By contrast, mutagenesis does not require the transfer or insertion of foreign DNA but it can alter the genetic make-up of a living organism. Mutation is generally defined as, “a process by which the genetic information of an organism is changed in a stable manner, resulting in a mutation.” More simply put, “having the ability to cause a permanent change in an organism’s genes”.

Mutations, as some of us may remember from our school biology classes, can occur spontaneously and randomly in nature every few generations or so. Mutations can also occur spontaneously as a result of background influences such as errors in DNA replication, environmental chemicals and radiation. Since the 1960s a number of scientists have been looking to induce rapid, unnatural mutation in plants in order to improve certain plant characteristics by, for example, making citrus fruit sweeter, with fewer seeds and thinner skins as with the KinnowLS.

I would like to stress at this stage that “mutant” plant is the scientific term for these particular novel plants. It is not an attempt to use alarmist terminology for the sake of a “click-bait” headline. It is the official term used for these novel plants by scientists themselves. Indeed, the international organisation responsible for cataloguing these novel plants (the FAO/IAEA) refers to its list of mutagenic plants as “The Mutant Variety Database” with applicants asked on the registration form to list their new “mutant” variety.

Conventional mutagenesis

In 2001 mutagenesis relied on two primary methods to induce rapid, forced, plant mutation: the use of mutagenic chemicals, or the application of ionizing radiation. For the purposes of this piece I shall refer to these two specific forms of technology as “conventional” mutagenesis.

When the GMO Directive was drafted the EU opted to exempt conventional plant mutagenesis from the Directive (2001/18/EC, Annex 1 B), presumably because it was a form of technology that has been “conventionally used” and has a “long safety record” (recital 17, Directive 2001/18/EC). As the Flemish Institute for Biotechnology states, conventional, in vito, mutagenesis creates thousands of mutant plants the vast majority of which are useless in that they either show undesirable product defects or are simply non-viable. Only a few new plants bred out of conventional mutagenesis show enhanced traits. It is the latter group of mutant plants which, in 2001 were exempted from the GMO Directive. For the past 17 years those working on classical mutagenesis have placed over 46 varieties of mutant herbicide resistant sunflower and six varieties of mutant herbicide resistant rape seeds on the EU catalogue of common plant varieties without having to undergo the stringent risk assessment procedures set out in the GMO Directive.

Novel mutagenesis

In the past decade, technological advances in mutagenesis have ensured that the technology no longer relies on either mutagenic chemicals or ionising radiation to affect artificial mutation in plants. A number of novel, directed and very precise technologies have emerged in the field of mutagenesis, which include:

-          site-directed nucleases (SDN) (including ZFN-1/2/3 and CRISPR systems);

-          oligonucleotide directed mutagenesis (ODM);

-          cisgenesis;

-          RNA-dependent DNA methylation (RdDM);

-          Grafting (non-GM scion on GM rootstock);

-          reverse breeding; and

-          agro-infiltration.

An amateur gardener’s shed may have a paint brush with which to brush the pollen from one plant onto the pistil of a sexually compatible plant, producing a hybrid that carries genes from both parents. In the case of novel, directed mutagenesis, however, this would be impossible. It requires highly sophisticated scientific technique to create a novel seed using these methods. No hobby, amateur gardener would have the sophisticated equipment needed to create a CRISPR Cas-9 cabbage seed. The eye alone is unable to distinguish the difference between a CRSIPR Cas-9 cabbage on the one hand and a natural, conventional cabbage on the other – only scientists using specialized DNA profiling would be able to identify the difference. Thus, although growing plants from mutagenic seeds can be done in a traditional, conventional manner (as Prof Jansson proves) the production of the seed is all but traditional or conventional.

Questions referred to the CJEU

In 2015 a small, not-for-profit organisation Confédération Paysanne together with eight other associations concerned with the protection of the environment and the dissemination of GMO information asked the then French Prime Minister to ban the cultivation and marketing of herbicide tolerant rape varieties created through the use of classical mutagenic techniques. Advocate General Bobek summarises the applicant’s concerns as follows:

“For the Applicants the use of herbicide resistant seed varieties obtained by mutagenesis carries a risk of significant harm to the environment and to human and animal health. It leads to an accumulation of carcinogenic molecules or endocrine disruptors in cultivated plants intended for human or animal consumption. The applicants refer, moreover, to the risks of unintentional effects, such as undesired or off-target mutations on other parts of the genome. They consider that this is the result of the techniques employed when modifications of the genome takes place in vitro and for the regeneration of plants from the cells thus modified.”

The French Prime Minister refused their request, so the applicants appealed to the French Conseil d’État. Feeling that this was a matter of EU interpretation the Conseil d’État referred the case to the CJEU asking, in summary, whether plants created as a result of novel, directed mutagenesis (i.e. those developed post 2001):

-          fall under the definition of a GMO as set out in Article 2, Directive 2001/18/EC, Article 2(2) Directive 2001/18 defines "genetically modified organism (GMO)" as “an organism, with the exception of human beings, in which the genetic material has been altered in a way that does not occur naturally by mating and/or natural recombination”;

-          are exempted from the impact assessment and traceability measures required of transgenic GMO’s in Directive 2001/18/EC; my review in this post focuses on these first two questions which have caused most controversy in the scientific community;

-          constitute GMO’s within the meaning of Article 4 of the Directive 2002/53/EC establishing a European common catalogue of agricultural plant species varieties;

-          constitute a harmonising measure prohibiting member states to all or some of the obligations laid in the GMO Directive – or do member states have some discretion to set the regime for organisms obtained by mutagenesis; and

-          fall under the precautionary principle guaranteed by Article 191 (2) [TFEU]?

In January of this year A.G. Bobek took a narrow interpretation of the GMO Directive. He agreed that novel, directed mutagenesis techniques fall under the definition of a GMO (Opinion at para 56) - but they are nevertheless exempted from the precautionary principle and the other stringent requirements set out in 2001 by virtue of the Annex I B exception (at para 56 and at para 81).

In Bobek’s opinion, the EU intended to exclude all forms of mutagenesis from the GMO Directive - past, present and future.  Bobek opined that in 2001 the EU understood that technology does not stand still and that its decision to exempt “mutagenesis” from the Directive’s obligations would apply to all future mutagenic technology regardless of novel techniques.

“The Applicants and several other interested parties have, to a great extent, relied on recital 17 to reach the conclusion that the EU legislature only intended to exempt safe mutagenesis techniques. I cannot agree. Neither the text, nor the historical context, nor the internal logic of the GMO Directive supports that proposition.” (at para 90 and at para 91).

The GMO Directive referred to transgenic GMOs only through Annex I A and as such it is only that sort of plant breeding to which the Directive applies. Had the intention in 2001 been to include all new plant breeding techniques in the Directive’s stringent obligations and requirements, the legislature would not have created the Annex I B exemption referring specifically to mutagenesis. In short, all mutagenic plants – both conventional and novel are exempt from the stringent requirements set out in the 2001 GMO Directive. As a result, mutant plants could be placed on the European catalogue of common agricultural plant varieties without the need for applying the GMO requirements set out in the Directive establishing the catalogue. In answer to the harmonisation question Bobek stated that the 2001 GMO Directive does not preclude member states from adopting measures governing mutagenesis, provided that, “in so doing they respect the overarching obligations arising from EU law.” (para 108 ff; sufice to say the CJEU agreed with this part of the Opinion).

The CJEU judgment diverged significantly from that of Bobek’s opinion. 

The CJEU ruled:

“Article 2(2) of Directive 2001/18 must be interpreted as meaning that organisms obtained by means of techniques/methods of mutagenesis constitute GMO’s within the meaning of that provision.” CJEU, Judgment, 25 July 2018, C-528/16, (para 54).

To recall, the 2001 Directive on the deliberate release into the environment of GMO’s defines a GMO as “…an organism, with the exception of human beings, in which the genetic material has been altered in a way that does not occur naturally by mating and/or natural  recombination.” Article 2 (2) Definitions 2001/18/EC.

Applying this definition to both transgenic and mutagenic techniques the CJEU rules. “…those techniques/methods alter the genetic material of an organism in a way that does not occur naturally, within the meaning of that provision. It follows that organisms obtained by means of techniques/methods of mutagenesis must be considered to be GMOs within the meaning of Article 2 (2) of Directive 2001/18”  (CJEU, para 29-30). (Emphasis added).

The CJEU further decided that although mutagenesis is not listed as a GMO in Annex I A this does not mean that novel, directed mutagenic techniques should not fall under the definition of GMO’s. The use of the word “inter-alia” in Annex I A suggests “the list of genetic modification techniques in that part is not exhaustive. Consequently, the list cannot be regarded as excluding genetic modification techniques other than those to which it specifically refers.” (CJEU, para 35).

“…only organisms obtained by means of techniques/methods of mutagenesis which have conventionally been used in a number of applications and have a long safety record are excluded from the scope of the directive.” (CJEU, para 54).

Here the CJEU turns to the Article 3(1) exemptions listed in Annex I B.

Firstly, the mutagenic exemption must be interpreted strictly. The Court here refers to its finding in Białowieża Forest (Commission v Poland) C-441/17, which itself builds upon a trail of precedent in the application of the Habitats Directive. Arguably the first in that line of authority, Case C-239/04 Commission v Portugal, at 35 simply observes the need for strict interpretation of exemptions from general protective regimes. However, it would not be far-fetched to say that the general principles of EU environmental law, in particular the principle of high level of environmental protection and the principles of prevention and precaution, play a role in that approach.

The need for strict interpretation having been established, the Court still required further guidance on how exactly that interpretation is to proceed. The wording of the provisions evidently serves as a first guidance principle. Here, the Directive is not of much help: at 43: the wording of Article 3(1) read in conjunction with the Annex does not on its own provide any conclusive guidance as to the types of techniques /methods that the EU legislature intended specifically to exclude from the scope of the Directive.

Consequently, the CJEU decides to examine the exemption contextually: ex multi: Pinckernelle, C-535/15. The context in which the exclusion is made is clarified in particular by recital 17 of the Directive, which states verbatim ‘this Directive should not apply to organisms obtained through certain techniques of genetic modification which have conventionally been used in a number of applications and have a long safety record.’ It is not at all surprising therefore that the Court does not extend the exemption to those techniques which do not so qualify: at 47: ‘the referring court is called upon to rule, in particular, on the techniques /methods of directed mutagenesis involving the use of genetic engineering which have appeared or have been mostly developed since Directive 2001/18 was adopted and in respect of which the risks for the environment or for human health have not thus far been established with certainty.

The Court further notes at para 48 that what these techniques purport to do must be taken at face value:

‘…the direct modification of the genetic material of an organism through mutagenesis makes it possible to obtain the same effects as the introduction of a foreign gene into that organism and, secondly, that the development of those new techniques/methods makes it possible to produce genetically modified varieties at a rate and in quantities quite unlike those resulting from the application of conventional methods of random mutagenesis.’ (emphasis added)

This is then where the precautionary principle firmly kicks in, again with reference to the recitals of the Directive: at 49:

…living organisms, whether released into the environment in large or small amounts for experimental purposes or as commercial products, may reproduce in the environment and cross national frontiers, thereby affecting other Member States. The effects of such releases on the environment may be irreversible. In the same vein, recital 5 of that directive states that the protection of human health and the environment requires that due attention be given to controlling risks from such releases.

At 50:

‘Furthermore, it has been emphasised, in recital 8 of that directive, that the precautionary principle was taken into account in the drafting of the directive and must also be taken into account in its implementation. Emphasis is also placed, in recital 55 of Directive 2001/18, on the need to follow closely the development and use of GMOs.’

The Court need not but repeats at 50 the perfectly logical chain of interpretation that leads to its conclusion: recital 17 explicitly justifies the exemption for mutagenesis by referring to the safety record of conventional mutagenesis. Article 4(1) emphasises the precautionary approach; such approach cannot support inclusion in the exception for new techniques/methods of mutagenesis which have appeared or have been mostly developed since Directive 2001/18 was adopted.

The impact of the Judgment

The ruling is significant for many reasons, primarily because the Directive is no longer restricted to the insertion of foreign DNA into the genome of a living species using the transgenesis technique. Any artificial modification of a plant using novel, directed mutagenesis must also apply the stringent obligations and requirements set out in Directive 2001/18.

Secondly, the terminology used in the ruling opens up the door to many unanswered legal questions. There is enough meat in this judgment to conclude that its reasoning extends not just to transgenic and mutagenic plants but to all forms of novel plant breeding techniques, which are likely to emerge in the foreseeable future.

Consider, for example, the repeated references in Confédération Paysannes to “naturally occurring”, “conventional use”, “at a rate and in quantities quite unlike those resulting from the application of conventional methods” and “long safety record”. The CJEU is at pains to point out that these are not unimportant words to be dismissed as irrelevant when determining the Annex I B exemption. Rather, the judges rule that these terminologies have deliberately and intentionally been woven into the very fabric of the 2001 Directive in order to distinguish them from novel, artificial techniques.

Along these lines, some argue, not unreasonably, that the law should apply to conventional mutagenesis itself (the very technique the Directive and this judgment does exempt) given that the use of mutagenic chemicals and ionising radiation is hardly a “naturally occurring” technique albeit in 2001 it was considered “conventional”. All of this begs the question how much Confédération Paysannes will apply to other novel plant breeding techniques which are sure to emerge in the coming years? If yes, is it even possible to define what is meant by “conventional use”?  Does conventional use refer to ten years, twenty years or five months? If the law understands what “naturally occurring” means is it possible to deduce a definition of “artificial”, “induced” and “forced” technique? All of these considerations could have wider implications on the commercialisation of the EU’s biotechnology policy in the months and years ahead.

Further, unlike AG Bobek, who took a very narrow interpretation of the Annex I B exemption, the CJEU took a far broader approach to the questions referred to it by the Conseil d’Etat. They did so by placing greater emphasis on the Directive’s over-arching objective namely to protect human health and the environment. By focusing on the potentially “irreversible” effect of the release of a mutant plant in the environment the CJEU pulls the law on mutagenic plants – potentially all new plant breeding techniques – firmly back into the orbit of the Directive’s original intention and purpose namely to protect human health and the environment from the unforeseen effects of GMO technology through application of the precautionary principle.


It should be stressed that the ruling does not stop researchers from continuing to research new plant breeding techniques. They are, however, required to do so in a regulated, controlled environment. Nor, does the ruling end all scientific endeavour in new plant breeding techniques. The risk assessment obligations set out in the Directive are not about gut-feeling. They ensure objective, impartial and transparent scientific analysis. “The environmental risk assessment should be carried out in a scientifically sound and transparent manner based on available scientific and technical data.” (Annex II Principles for the Environmental Risk Assessment B, 2001/18/EC). Prof. Jansson’s Plant Cell and Molecular Biology Department at Umeå University in Sweden will still be able to conduct research into plant mutagenesis – but probably not in a Professor’s back garden.

What the ruling does impact on, however, is the ease with which industry can commercialise novel, mutagenic plants in the EU. I would like to re-emphasise at this point that this ruling does not block the cultivation of these plants on European Union soil. In practice however, very few GMO varieties are grown in the EU given that industry seemingly finds the obligations too stringent to invest the time, effort and money in wanting to do so. Even in cases where industry has had varieties go through the whole process, some of them have been subject to national moratoriums in application of the ‘safeguard measures’ set out in Directive 2001/18/EC.

Taking the CRISP Cas-9 cabbage grown in Prof. Jansson’s back garden as an example of the challenges facing the commercialisation of mutagenic plants in the Europe: any CRISPR Cas-9 cabbage will now be subject to notification to the national competent authority, using a standard authorisation procedure. Environmental risk assessment must be carried out, any risks must be managed, and any further growing must be subject to regular inspections from national competent authorities. Keeping the seed supplier anonymous, for instance, clearly will no longer be possible. Two days after the ruling Bayer and BASF announced that they would pursue gene-editing of plants outside of the EU.

For those wishing to commercialise novel plant breeding techniques unchecked, this ruling is indeed a set-back. For consumers in the EU, already enjoying a wide variety of fresh produce, the immediate effects of this judgment seem less pressing.

To be or not to be… a regulated GMO: The CJEU’s verdict on genome editing

Felix Beck, Doctoral candidate, University of Freiburg/Germany

The recent CJEU judgment on the regulation of genome-edited crops has been debated heatedly throughout the European Union. Many scientists and plant-breeders condemned the verdict as a “backward step” and “hostile to progress“, while GMO sceptics hail it as upholding the precautionary principle and protecting consumer choice between GMO and non-GMO products. However, the legal reasoning of the Court has so far only received little attention. While the Court was rather tight-lipped on the biggest legal issues, the ruling still yields some interesting observations.

To recall the facts of the case: On referral by the French Conseil d'Ètat, the CJEU had to decide whether crops modified with so-called “genome editing” techniques are covered by the EU’s Directive 2001/18/EC on the deliberate release of Genetically Modified Organisms (GMOs). The notion “genome editing” refers to recently-developed techniques, like CRISPR, that allow for targeted modification of the DNA of virtually any organism. This allows to activate or knock-out specific genes, for example to render a crop plant more resistant to certain herbicides or pathogens. In contrast to conventional genetic engineering, genome editing is much more precise and can be used without necessarily inserting foreign DNA into the organism. Whether organisms resulting from these techniques are subject to the protracted GMO regulation in the EU or not has been controversial for several years.

First, the Court assumed with much ado that genome edited organisms constitute GMOs in terms of Art. 2(2) of Directive 2001/18/EC. It even saw no reason to comment on the controversial question whether the notion “altered in a way that does not occur naturally”, which is used to define what constitutes a GMO, relates to the technique of genetic modification or its result. This question was not expressly presented by the referring court, which allowed the CJEU to avoid taking a clear stand. However, the Court indicated that it prefers a strictly process-oriented interpretation, which means that any technique where the genome is modified in vitro would lead to GMOs in terms of Directive 2001/18/EC.

The second and more difficult issue concerned the question whether genome edited organisms are exempted from regulation as organisms resulting from “mutagenesis”. Conventional mutagenesis techniques rely on exposing the organism to certain chemicals or ionizing radiation, which increases the occurrence of genetic mutations. Afterwards, a plant breeder has to select individuals carrying the desired traits from a large number of treated organisms. Pursuant to Art. 3 and Annex I B para. 1 of Directive 2001/18/EC, organisms bred with “mutagenesis” are perceived to constitute GMOs, but are exempted from the directive's scope. The term mutagenesis, however, is not further defined or qualified under EU law, which posed the question whether this term is to be interpreted dynamically (incorporating new mutagenesis techniques) or whether it only refers to those mutagenesis techniques that were known in 2001 when the Directive was adopted.

For the Court, the decisive argument was recital 17, which provides that the Directive “should not apply to organisms obtained through certain techniques of genetic modification which have conventionally been used in a number of applications and have a long safety record”. In the view of the Court, targeted mutagenesis techniques simply have no such long safety record and thus do not fall under the mutagenesis exemption.

In my view, this is a legitimate way of reasoning. There is a difference between the generic meaning of mutagenesis (i.e. any method that induces genetic mutations) and the legal meaning of the same term: recital 17 of Directive 2001/18/EC clearly indicates that only those techniques that were known and conventionally used when the Directive was adopted in 2001 should be exempted from its scope, i.e. conventional mutagenesis techniques relying on chemicals or radiation. Otherwise, changes in scientific nomenclature would be able to modify the scope of regulatory regimes, which would raise questions for the democratic legitimacy of such an interpretation.

Importantly, and this appears to remain difficult to grasp for non-lawyers, the Directive does not leave room for the question whether targeted mutagenesis is equally safe or even safer than conventional mutagenesis; it simply has no comparably "long" safety record. Therefore, it was reasonable for the CJEU to reject the idea of adopting a dynamic interpretation of the term mutagenesis, as Advocate General Bobek suggested in his opinion.

Unfortunately, the CJEU judges did not confine themselves to legal arguments, but also relied on very general (and unsubstantiated) allegations that the risks associated with genome editing were similar to those of conventional GM techniques (which is disputed by the vast majority of scientists), and that genome editing would lead to the development of modified organisms at higher rates and quantities (which is true, but not a risk per se). These misguided statements may make it very hard for scientists to accept that the CJEU's judgment may be okay from a purely legal viewpoint.

After all, the CJEU never was the appropriate organ to assess the risks or non-risks associated with genome editing techniques or to decide on their future regulation. This is fundamentally a matter of policy and must be dealt with by the European Union legislator. Hence, the Court should not be blamed for choosing one out of two controversial options for interpreting the Directive, but it should rather be asked why the European Union legislator has remained inactive for so long. When the European Commission proposed the first Deliberate Release Directive in 1988 and its revision that was adopted in 2001, it promised to regularly update the Directive in order to “keep pace with scientific and technological progress”. It is time for the European legislator to live up to this promise.

Art credit: John Byrne and Terry Austin, Marvel Comics