Gene Editing: A View Through the Prism of Inherited Metabolic Disorders

Abstract

Novel technological developments mean that gene editing – making deliberately targeted alterations in specific genes – is now a clinical reality. The inherited metabolic disorders, a group of clinically significant, monogenic disorders, provide a useful paradigm to explore some of the many ethical issues that arise from this technological capability. Fundamental questions about the significance of the genome, and of manipulating it by selection or editing, are reviewed, and a particular focus on the legislative process that has permitted the development of mitochondrial donation techniques is considered. Ultimately, decisions about what we should do with gene editing must be determined by reference to other non-genomic texts that determine what it is to be human – rather than simply to undertake gene editing because it can be done.

KEYWORDS:

• Gene editing
• inherited metabolic disorders
• mitochondrial disease
• mitochondrial donation
• germline
• germline gene editing
• embryo

The ‘technology of purposeful DNA modification’ (Rehmann-Sutter 2018), that is the ability deliberately to make targeted alterations in specific genes within an organism, has been developed to such a sophisticated level that gene editing is now a clinical therapeutic reality. Stem cells extracted from patients affected with severe paediatric neurodegenerative disorders such as metachromatic leukodystrophy have been edited ex vivo using lentiviral systems to insert a correctly functioning copy of the defective gene, and the cells returned to the patients as an autologous haematopoietic stem cell transplant, thereby preventing the progression of the disease (Biffi et al. 2013). More recently in vivo gene editing has been accomplished using zinc finger nuclease technologies (Sharma et al. 2015) to insert a functioning copy of the gene encoding specific lysosomal enzymes directly in to liver cells of patients with lysosomal storage disorders in an attempt to cure their disease, with the first patients treated widely reported in the press.^{1, 2} The development of RNA-guided gene-editing technologies such as CRISPR/Cas9³ or zinc finger nucleases means that it is now possible to edit single genes within an organism, including an embryo.

This special issue of The New Bioethics aims to take stock and review some of the complex issues that the technological possibilities generate. The articles included are drawn from various disciplines, ranging from philosophy, applied bioethics, to public policy development, and importantly from those at the forefront of the technological advances. This editorial reviews and introduces these diverse papers through the prism of the clinical field of inherited metabolic disorders.

Inherited metabolic disorders as a paradigm

Living organisms can be viewed as complex biological systems. Thousands of different chemical reactions occur every second in every cell of our bodies, for example generating energy that powers muscle contraction or neuronal synaptic firing, synthesizing new proteins to permit growth, or degrading or recycling waste-products. ‘Metabolism’ is the sum total of these chemical reactions. Virtually every chemical reaction is catalysed by a unique enzyme, a protein that recognizes specific substrates and facilitates the conversion to the product of the chemical reaction. The protein that forms a specific enzyme is itself encoded by a unique gene, whose nucleotide base sequence provides the ‘blueprint’ for the amino acid sequence of the protein. As Zwart comments (2018), ‘The genome is the primordial layer from where multiple circuits and complicated networks of molecular messages pervade living organisms’.

The inherited metabolic disorders arise when a mutation in a given gene results in a deficiency of the related enzyme. The majority of these conditions are inherited in an autosomal recessive manner, whereby both the maternally inherited and paternally inherited copy of the given gene must harbour a mutation for the disease to manifest. In the usual situation this means that both parents are heterozygous carriers of the condition, i.e. they each carry one mutated copy of the gene and one ‘healthy’ copy. Thus a deficiency of the enzyme phenylalanine hydroxylase caused by mutations in the PAH gene gives rise to the condition phenylketonuria (PKU), associated with pathologically high phenylalanine levels as this amino acid cannot be converted to tyrosine. Untreated, PKU results in significant progressive damage to the central nervous system, manifesting with microcephaly, developmental delay and significant cognitive impairment. PKU is eminently treatable with dietary phenylalanine restriction, and detection via newborn screening permits treatment to be commenced before damage occurs. Many of the metabolic disorders, however, carry a devastating prognosis, for example mucopolysaccharidosis type III (Sanfilippo syndrome) which is caused by deficiency of one of the several lysosomal enzymes that degrade complex macro-molecules (the glycosaminoglycans), with the result that these macro-molecules accumulate in different tissues of the body including the brain and cause a relentlessly progressive neurodegenerative disorder, with childhood onset dementia and significantly curtailed life-expectancy.

The inherited metabolic disorders thereby provide a relatively simple paradigm for diseases that could be amenable to gene-editing technologies: correction of a single gene should effectively remove the disorder. Somatic gene editing (such as the examples cited above) has the potential to treat an affected individual, with the presumption that earlier treatment before significant pathological damage has occurred will result in a better outcome. ‘Treating’ an early embryo with some form of germline gene editing would effectively prevent the disease from occurring at all, but in contrast to post-natal somatic gene editing would result in a permanent change to the human germline that would be passed to future generations.

Ethical dimensions

The ethical issues that arise in the clinical arena from these classical monogenic disorders are multifaceted, and often intensely painful for those affected. The autosomal recessive inheritance alluded to above means that when counselling parents with a child in whom a diagnosis of a metabolic disorder has been made, the ‘recurrence risk’ in future children has to be explained. Given that both parents are heterozygous carriers of the disorder, any and every future child they bear has a 25% risk of inheriting both ‘faulty’ genes and thereby affected by the disorder. The parents are then faced with difficult decisions about whether to consider having further children, or whether to explore options including pre-implantation genetic diagnosis (PGD, a form of in vitro fertilization (IVF) where only healthy embryos not affected by the disorder are implanted to the uterus), or to consider antenatal testing (to determine if the baby in utero is affected or not, and then to consider whether or not to terminate an affected pregnancy). Other parents may opt for post-natal testing of a baby once it has been born, and often we will provide a prospective neonatal management plan to instigate treatment while the diagnosis is clarified.

These ‘choices’ raise some fundamental questions. It is surely right to think that it would be better for an extant child not to have a devastating diagnosis such as Sanfilippo syndrome. Antenatal testing and PGD raise the corollary of whether it is better to not exist in the first place than to exist but have the disorder. This is a painful choice that parents are faced with. The prospect of germline (embryonic) gene-editing techniques will add to this complexity, but is editing any different to selecting? Several of the articles in this issue grapple with concepts related to these very real dilemmas, not least the ‘non-existence problem’ in considering our responsibilities to someone who does not yet exist (but who may or may not exist depending on my choices).

Selection and editing, and the status of the embryo

Ethically though, is there any difference between selecting embryos such that only those without the condition are born, and editing an affected embryo so that it does not have the disorder? Rehmann-Sutter argues that human germline gene editing does raise more ethical issues, because the manipulation of the human germline will introduce a permanent change that has effects on intergenerational relationships. He presents an analysis of the human germline, defining it as ‘a relationship that links the generations and has its lived reality in families’. He argues that germline gene editing is contentious primarily because the changes it instils are inherited by the next generation. These concerns encompass both issues of safety – the changes we make now may have unforeseen consequences that future generations have not consented to – and questions on the impact on relationships between generations (such as the perceived moral responsibility to ‘produce children with optimal genetic starting conditions’).

Shaw, on the other hand, argues in a different context that there is no meaningful moral distinction between modification and selection as means to achieving a given end. In considering the converse situation whereby some consider it acceptable to positively select for a disability such as deafness, he explores some fundamental questions about the status of the human embryo as an ‘object of moral consideration’. Are embryos analogous to ‘future persons’ or to ‘presently-existing individuals’, or ‘mere ‘symbols’ of human life’? Shaw suggests that they may be in a category of their own, ‘sui generis’. I would agree that how we view the status of the embryo is of utmost importance, as it will inform the real-world decisions we make around PGD, antenatal testing, as well as germline gene editing and embryonic manipulation technologies. If the embryo is a primitive person, certain ethical obligations should be extended towards it.

Craven et al. (2018), in reviewing the issues around ‘mitochondrial donation’ techniques, discuss the necessity of experimenting on human embryos to allow the development of these techniques for clinical application. They allude to the complex social, cultural and religious considerations fundamental to determining whether this is ethically acceptable without exploring these considerations in detail, but do review some practical concerns about the potential sources of these embryos including the steps taken to avoid ‘inducement’ to potential donors.

More than the sum of our genes?

In considering the inherited metabolic disorders, the prime biological importance of our genes is seen by the devastating impact of when ‘things go wrong’. A single point mutation in a single gene can result in a life-altering, and life-shortening disease.

Furthermore, studying the inherited metabolic disorders provides important insights into the normal physiological processes occurring in our bodies, and could be conceived of as naturally occurring ‘knockout models’. Indeed the remarkable complexity of metabolism and the underlying genomic system can invoke ‘an overwhelming sense of awe’, as Zwart (2018) states was the case for Francis Collins (sometime Director of the International Human Genome Sequencing Consortium) in decoding the human genome, a theme known to the biblical psalmist (‘I praise you because I am fearfully and wonderfully made; your works are wonderful, I know that full well.’ Ps 139:14).

Zwart discusses the great hope that surrounded the Human Genome Project (HGP), and the expectation that by ‘knowing’ this most fundamental layer of our biological systems, that humanity would finally ‘know itself’ and be able to control its future. However, he highlights the ‘narcissistic offence’ caused by the finding that in actuality, we humans are, at the genomic level, not significantly different from the roundworm, and have fewer protein-encoding genes that some species of flowering plant. The ‘HGP negates a genetic reductionist understanding of the genome as our blueprint.’ He concludes that ‘we are not only the product of our genes, but forged by culture as well’. The text of the genome is but one text or word (λόγος) that defines what it is to be human. And significantly, when considering if we should edit the genomic text, reference must be made to the other cultural and religious texts, that is to say that the social, cultural and religious voices must be heard. He advocates ‘zooming out somewhat from frontstage bioethical quandaries towards the more fundamental backdrop issues’.

Mitochondrial disorders

The pace of events, however, forces us back to these very frontstage bioethical quandaries. One prominent subclass of the inherited metabolic disorders is ‘mitochondrial disease’. The mitochondrial disorders are, like other inherited metabolic disorders, caused by mutations in one of many genes that encode for one of the numerous proteins required for normal mitochondrial function. The mitochondria have multiple biological functions, not least a primary role in ATP⁴ (‘energy’) production, and mitochondrial dysfunction is associated with a wide range of clinical phenotypic disease (Davison and Rahman 2017). Many mitochondrial disorders are autosomal recessive conditions, caused by bi-allelic mutations in a nuclear-encoded gene, but an additional level of complexity is added since the mitochondria themselves contain a small ring of mitochondrial DNA (mtDNA) that encompasses 37 genes that encode proteins and tRNAs⁵ required for mitochondrial function. Mutations or deletions in the mtDNA can also cause mitochondrial diseases such as Mitochondrial Encephalomyopathy, Lactic Acidosis and Stroke-like episodes (MELAS) syndrome.

Mitochondria are only inherited maternally (since only mitochondria from the ovum and not the sperm are included in the fertilized embryo), and therefore mitochondrial disorders due to mtDNA mutations are matrilineally inherited, and all offspring of a women who carries mtDNA mutations could be at risk of inheriting mitochondrial disease. ‘Mitochondrial donation’ or ‘nuclear genome transfer’ techniques have been developed that enable the generation of an embryo whose nuclear genetic material derives from the biological mother and father, but with mtDNA derived from a third person without a pathogenic mtDNA mutation (so-called ‘three parent babies’).

MtDNA inheritance patterns do challenge some aspects of the traditional concepts of parenthood. In one sense normal procreative processes are a form of genome editing or shuffling. Any child born will have nuclear gene combinations inherited from four different grandparents, while mtDNA inheritance can be followed down a single matrilineal germline. Similarly, the intertwined germlines of our ancestors are augmented by the combined social, cultural and other influences that go to make up ‘who I am’. The issues around the primacy of biological parenthood are an important consideration, yet parenthood is not restricted to the purely biological (the prime example of non-biological parenthood being adoption). The impact for all parties of introducing additional biological parents should be considered.

Governmental legislation in the United Kingdom has allowed for mitochondrial donation techniques to be implemented in order to prevent serious mtDNA-related disorders (Craven et al. 2018). Two articles in this issue scrutinize this legislative process and consider in detail many of the ethical issues that arise from mitochondrial donation/nuclear genome transfer techniques. Craven et al. from the Wellcome Centre for Mitochondrial Research in Newcastle upon Tyne where mitochondrial donation has been pioneered, provide a robust defence of the process and highlight the stringent governance structures in place around mitochondrial donation techniques, and argue that it should be used ‘cautiously and under strict regulation’, and highlight that the use of the technique is only legal ‘to avoid serious mtDNA disease’ (and not for other indications or applications such as treating infertility).

Cussins and Lowthorp, on the other hand, raise a number of concerns both about the fundamental ethical acceptability of mitochondrial donation/nuclear genome transfer techniques, and about the policy process underpinning the technology. They argue that in a sense, ‘it does not address a medical need’ since it does not treat an existing person (a non-existence argument). (That said, the suffering for a couple who have had multiple affected offspring must not be underestimated; it can be argued that this in itself is a medical need justifying treatment.) Questions about the safety of the technique are paramount, as both articles explore and concur that long-term safety surveillance is important, not least to monitor for any unexpected epigenetic effects arising from the cell manipulation.

The more fundamental concerns that Cussins and Lowthorp address relate to the precedent that the (currently narrowly focused) U.K. legislation may set internationally – that it is a ‘step towards designer babies’. The U.K. legislative process and framework may be a model followed for future germline editing legislation, and the suitability of this process being ‘fit for purpose’ in providing adequate safeguards is questioned.

Fixing and treating versus improving and enhancing

Finally, the articles in this issue raise a note of caution about the motives of some who would view germline gene editing as a means to support a eugenics agenda. We must be careful never to reach a point where, as Obasogie puts it (cited by Cussins and Lowthorp), it is considered ‘a sin of parents to have a child that carries the heavy burden of genetic disease’ – even for a severe inherited metabolic disorder.

The inherited metabolic disorders that I have used as the vantage point in this editorial to survey gene-editing technologies provide a relatively simple paradigm: a clear target for gene editing to be used as a therapeutic means. In that case, a single defective gene, with a demonstrably abnormal enzyme activity and subsequent severe disease, provides an obvious target that would benefit from improved treatment. To borrow (as Zwart advocates) from another non-genomic text, restorative medicine aims to help the ‘lame to walk, the blind to see’, as part of the mandate to subdue and master creation. Rightly used, gene-editing techniques may play a part in that.

However, caution is rightly sounded when the same technologies can be turned to ‘improving and enhancing’ – the fast to run even faster? It is surely correct that robust legislative safeguards must be implemented, but perhaps more fundamentally that we follow Zwart in ensuring that the genomic text is not the only one heeded in deciding the future.

Disclosure statement

No potential conflict of interest was reported by the author.

ORCID

James Davison http://orcid.org/0000-0002-9009-5538

Notes on contributor

James Davison is a consultant in paediatric metabolic medicine at Great Ormond Street Hospital NHS Foundation Trust, London, U.K.

Notes

1 https://investor.sangamo.com/press-releases/detail/381/sangamo-announces-treatment-of-first-patient-in-landmark;

2 https://www.theguardian.com/science/2017/nov/15/scientists-make-first-ever-attempt-at-gene-editing-inside-the-body

3 Clustered Regularly Interspaced Short Palindromic Repeats/CRISPR-associated protein 9.

4 Adenosine triphosphate.

5 Transfer ribonucleic acids.