My Lords, I thank the noble Baroness, Lady Bennett of Manor Castle, for raising the question of exogenous DNA, which I mentioned at Second Reading, as did she. I am sorry to say that, on this occasion, I disagree with the noble Baroness. I think her amendment goes too far, since, as she explained so clearly, the current techniques of gene editing require the use of exogenous DNA not only for the CRISPR-Cas9 construct but for the insertion mechanism—for instance, Agrobacterium—and for the antibiotic resistance filter used to check that the changes one intended to achieve have been achieved. Furthermore, at the end of the gene-editing process, there could be tiny fragments of this exogenous DNA left in the gene-edited organism. As I mentioned at Second Reading, although there are new techniques being developed that will obviate the need for gene editing with exogenous DNA, they are
not yet ready for mainstream use. Hence, in my view, Amendment 11, if accepted, would in effect kill off gene editing for the near future.
Should we be worried about exogenous DNA? The noble Baroness, Lady Bennett, clearly thinks we should. The Bill deals with the issue of exogenous DNA by stating in Clause 1(6) that any remaining exogenous DNA must not code for a protein if the precision-bred organism is to be considered as having been produced by a process equivalent to traditional breeding. Does that provide us with sufficient reassurance?
I am most grateful to Professor Wendy Harwood and her colleagues at the John Innes Centre, in Norwich, for their further advice. My initial thought in considering this problem was that the wording in the Bill could be tightened to say in Clause 1(6) that the exogenous DNA, if there is any left, should have no effect on the phenotype of the precision-bred organism; in other words, no effect on the appearance or biological properties of the edited organism. This would be a more stringent criterion than that requiring that the fragment of DNA could not code for a protein.
As the noble Lord, Lord Winston, so clearly explained to us in his excellent introduction, the principal way in which DNA is expressed in the phenotype is by genes coding for proteins that are the building blocks of life. However, there are also other ways in which nucleic acids can affect the phenotype; I am sorry if this is a bit technical. One example is RNA interference, a molecular process in which short strands of RNA can act to silence a gene. There are other examples of gene regulation by RNA strands that are not transcribed to produce a protein.
There are mechanisms not covered by Clause 1(6) by which exogenous DNA could affect the functioning of the precision-bred or gene-edited organism. Does this justify a change in the wording of the Bill? The view from the John Innes Centre is that it is unlikely that non-coding DNA could exert a phenotypic effect, although this is both theoretically and practically possible. It argues that these possibilities should be tested for in any gene-editing strategy before a product is developed. If this is the case, it would not be necessary to require it by regulation. Furthermore, the scientists at the John Innes Centre argue that the requirement of “no phenotypic effect” might lead to the conclusion that there has to be exhaustive testing for this in an unspecified range of environmental conditions.
I can see both sides of the argument. On the one hand, there is an argument for ensuring that any remaining exogenous DNA has no discernible effect on the phenotype of the precision-bred organism. That would be a more stringent criterion than Clause 1(6). On the other hand, one does not want disproportionate regulation that stifles innovation.
I do not expect the Minister to answer on the technical issues right now but could he—if not right now, before Report—put down in writing an explanation of why making the requirement in Clause 1(6) that any exogenous DNA should have no phenotypic effect would be disproportionate and, if it is disproportionate, whether other steps could be taken to manage the risk of non-coding effects on the phenotype of gene-edited organisms?