The distracting genome

I'm always pleased to see the achievements of UK bioscience in the headlines, especially when the reporting is up to scratch. I was therefore delighted to see last week's front pages and airwaves filled with well-written, balanced articles about a significant development in UK embryo research. There was, however, something missing: a story. Rather than news of a scientific advance, the bulletins concerned a bureaucratic decision not to stop research using a standard genome-editing tool for a lawful purpose but to pass the ultimate decision to a research ethics committee (see BioNews 837).

The tool in question, CRISPR/Cas9, is of obvious utility to any of the 'principal purposes' listed by the Human Fertilisation & Embryology Act:

a) increasing knowledge about serious disease or other serious medical conditions;
b) developing treatments for serious disease or other serious medical conditions;
c) increasing knowledge about the causes of any congenital disease or congenital medical condition that does not fall within paragraph (a);
d) promoting advances in the treatment of infertility;
e) increasing knowledge about the causes of miscarriage;
f) developing more effective techniques of contraception;
g) developing methods for detecting the presence of gene, chromosome or mitochondrion abnormalities in embryos before implantation; or 
h) increasing knowledge about the development of embryos.

The decision to renew Dr Niakan's research licence should therefore never have been in doubt; she might as well have applied to use a new sort of test tube. But, of course, the HFEA was distracted by something beyond the law of research licences: the possibility that it might lead to the implantation of genome-edited human embryos. Although CRISPR/Cas9 has the potential to reduce the incidence of serious inherited disease, it would seem that some people would prefer babies to suffer than to bear the indignity of being born as a healthy someone else. It's an academic concern, not least because gene-edited embryos and gametes cannot be lawfully implanted anyway. However, as that position could change with surprisingly few amendments to the law, let's take it seriously.

The Human Fertilisation and Embryology Act 1990 was amended seven years' ago to provide a framework for mitochondrial donation under what are now the Mitochondrial Donation Regulations 2015, Article 7 of which permits two procedures – third party pronuclear transfer and maternal spindle transfer – for the purpose of preventing the transmission of serious mitochondrial disease. It would be straightforward to add a third procedure, mitochondrial DNA editing, to article 7. Having agreed that, perhaps other interventions should become legitimate. There's a significant, if under-appreciated, difference between mitochondrial DNA and mitochondrial genes. Only 37 mitochondrial genes are located in mitochondria: the remaining 1,500 or so are situated in the nucleus. Why not edit them to prevent the transmission of mitochondrial disease, too? And, if these, why not permit editing of other nuclear genes to prevent the transmission of other serious heritable diseases? There would be a brake: adopting the principles of the 2015 Regulations, no such intervention would be permissible unless the HFEA determined, in each specific case, (a) that an embryo created using the gametes in question would incur genetic abnormalities and (b) that there was a significant risk that a person with those abnormalities would have or would develop serious congenital disease. To slippery-slopers, however, even authorised interventions would violate the human germline and human genome. This, they claim, would contravene international law.

They're wrong on both counts. There is no international treaty on the human genome, and the UK is not a signatory to the 1997 UNESCO Declaration on the Human Genome or the Council of Europe's 'Oviedo' Convention on Human Rights and Biomedicine. The nearest the UK approaches to a prohibition is under the 1998 Biotechnology Directive, which excludes from patentability 'processes for modifying the germ line genetic identity of human beings' on the basis of a 'consensus within the Community that interventions in the germ line offends against ordre public and morality'. Putting aside the fact that the 'consensus' excludes the UK, we do at least have a good starting point for debate: a distinction between interventions that would change the 'germ line genetic identity of human beings' and those that would not. If something is patentable, it can hardly be illegal.

On this basis, splicing green fluorescent protein genes into implantable embryos would not be patentable because no known human gene codes for that protein. In the absence of a serious condition, implantation would not be permitted anyway. By contrast, human donor mitochondrial transfer, which is already permissible subject to HFEA approval, can't change the 'germ line genetic identity of human beings' one jot. Nor, crucially, would correction of mutant genes into wild type (normal) genes. The Biotechnology Directive refers to 'human beings' (plural) because it's concerned with the integrity of humanity's gene pool, not individual parent-to-child transmission. The UNESCO Declaration and Oviedo Convention also concern gene-pool integrity. It makes no sense to try to read them at the individual level. For example, no one has ever been born with 'the human genome', while the nearest thing to an inviolable uninterrupted person-to-person germ 'line' of inheritance is the collection of 37 maternally inherited mitochondrial genes that avoid the hurly burley of sexual reproduction – genes so highly conserved as to be among the least exclusively 'human' of all.

Strikingly, the Biotechnology Directive uses the expression 'genetic identity' instead of 'the human genome'. Presumably, 'genetic' refers to the two percent of the genome that comprises genes. We have less cause to be cavalier about 'junk' DNA today, but 'genetic' is at least a claim to exclude non-functional DNA. The word 'identity', too, is instructive: as identical proteins can be encoded by different DNA sequences, 'identity' presumably concerns expression and not arbitrary sequences.

Corrections to wild/expression type might change 'the germ line genetic identity' in the sense of reducing pathological gene frequencies, but there is no legal – let alone ethical – basis for such an objection. With this in mind, let's put issues of human enhancement to one side and acknowledge that it's unlikely for most practical purposes that the germline and human genome would be threatened by the clinical use of CRISPR/Cas9. For now, the important thing is to leave the researchers to get on with the essential work of basic science.