Genome Engineering, Chemical Exposure, and the Germline: An Ethical Synthesis

Abstract

Concerns about the human germline in the context of genome editing have been at the forefront of contemporary bioethics, with the clear recognition that the heritability of such intervention merits special moral consideration. In contrast, the question of moral responsibility for modifications of the germline genome resulting from anthropogenic environmental toxicants has received little attention. Yet, whether the impact of human technological activity on enduring shifts in human heredity occurs via purposeful genetic modification or non-directed changes that undermine genome stability, the result in both cases is irreversible genetic change in future generations. This article argues that the robust ethical reflection developed by the bioethics community to address human heritable genome editing can be used as a resource to address understudied questions of moral responsibility for anthropogenic insults to the germline, thereby outlining a future-oriented ethical framework for germline responsibility in a time of widespread concern about industrial chemicals and human futures.

“Some would-be architects of our future look toward a time when it will be possible to alter the human germ plasm by design. But we may easily be doing so now by inadvertence, for many chemicals, like radiation, bring about gene mutations. It is ironic to think that man might determine his own future by something so seemingly trivial as the choice of an insect spray.”¹

1. Introduction

Intense interest in and concern for the human germline in the context of intentional and directed genome editing is a core theme of contemporary bioethics. The emergence of CRISPR-Cas9 technology, and the announcement of the birth of children altered with it as embryos starkly highlighted what is at stake.² Genome editing in an early embryo modifies virtually all its cellular genome, including those of germ cell progenitors that will ultimately become gametes, constituting future offspring. Present not just in the child that the embryo develops into, but their offspring, and their offspring’s germ cell progenitors, and so on, such changes are carried forward to future generations and can disseminate through populations. As such, germline alterations are viewed as inherently different to modifications in cells of the soma, whose replication ends with the death of the individual they compose. To date, the germline has appeared as a line of intervention that should not be crossed, or, more recently, might be crossed only in limited ways and with very strong justification.³

Scientists, ethicists and publics have long debated the idea of total bans on intentional human germline modification, discussions structured by assumptions that this purposeful action lies in the future and could be prevented from taking place.⁴ Meanwhile, a less heralded realm of scientific work has been generating evidence that human-made chemical and radioactive technologies have likely already inadvertently altered uncounted germlines, with unknown long-term consequences. It is increasingly clear that germ cells can undergo interlinked genomic, epigenomic, and cytoplasmic changes ensuant to chemical and radioactive exposures, and some of these changes are stably transmitted to further generations. Perhaps anti-intuitively, seemingly inconsequential technologies that populate life in an industrialized global world induce heritable human germline modifications. Latent rather than spectacular, these toxic germ cell impacts are consequential for the health and fertility of future generations.

This article is concerned with a paradoxical disconnect between powerful moral concerns around the germline in genome editing debates, and the relative absence of germline considerations vis-a-vis other chemical technologies. Although the category of germ cell mutagenicity exists in international guidelines, chemical substances are rarely screened for toxicity using germ cells, and there exists no mandate to do so. Accordingly, not a single substance is currently positively categorized as a known human germ cell mutagen, even though many have been identified by proxy in rodent models.⁵ Rather, in the current paradigm of toxicity assessment for chemicals and radiation, compounds are tested in somatic cells, that is, non-reproductive cells, with the “general assumption … that evaluations based on somatic cell mutation assays protect the germline by default.”⁶ This assumption disregards the possibility that germ cells may be harmed at lower exposure levels than other cells, i.e., doses below legal thresholds at which no adverse effects have been observed. Subsuming one to the other also overlooks the vast disparity of impact between risks of alterations in somatic and germ cells: while the former generate a limited clonal lineage, the latter give rise to all cells in the individual descended from them, including their future germ cells, and therefore all their descendants.

Paradoxically, one reason that considerations of intentional germline editing have proceeded in relative isolation from discussions about the ethics of environmental toxicants’ impact on human heredity is that the latter are studied under the rubric of “epigenetics,” commonly characterized in opposition to genetics.⁷ “Epigenetic carcinogens” have been definitionally segregated from “genotoxic carcinogens” since the late 1970s as causing tumors in the absence of genetic damage.⁸ Likewise, epigenetic transgenerational inheritance is explicitly defined by the absence of genetic intervention as “germline transmission of epigenetic information and phenotypic change across generations in the absence of continued direct environmental exposure or genetic manipulation.”⁹ The apparent persistence of epigenetic effects across lifespans and generations has raised questions of social justice, with scholars coining the term epigenetic responsibility to argue that new forms of responsibility and considerations for transgenerational justice arise in light of this knowledge.¹⁰ Yet these considerations of responsibility for social and environmental factors impinging on the heritable epigenome take place in a notably separate literature than that bearing on transgenerational change mediated by genetic engineering.

Indeed, in considering transgenerational epigenetic inheritance, scholars for the most part contrast it against genetic inheritance, celebrating its disruptive possibilities for non-essentialist concepts of health and inheritance.¹¹ Accordingly, they have not considered the bioethical concept of the germline genome as a suitable framework to examine the normative implications of this novel form of inheritance. Quite on the contrary, Tim Lewens suggests that observing heritable changes to “non-genetic structures” should lead us to question the validity of the moral category of the germline interventions in the first place.¹² According to Lewens, interventions that “influence the physiology of numerous individuals,” causing enduring change in human heredity through non-genetic structures – interventions that “spread over space and time” in such a way that affected individuals cannot give consent – could encompass everything from urban planning to dietary advice.¹³ Are germ cells even the material site where changes might matter the most, as opposed to, say, breast milk as a cross-generational transmitter of consequential material change? According to Lewens, this “blurring of the germline” begs the very question of whether there is any “distinctive ethical significance” of germline interventions as a category.¹⁴

In this article, we argue that there is, indeed, a normative distinctiveness to the germline as a site of human intervention, as has been argued with respect to genome engineering. We contend that a framework of germline responsibility is a necessary step in the consideration of the future-oriented impact of chemical technologies on human heredity, whether or not those technologies were explicitly designed or intended to alter the human germline. The development of this framework requires that germline interventions, whether intended or not, be thought through a continuum that encompasses genetics and epigenetics rather than placing them in opposite camps. It is time to forgo the concept of the germline as an immutable entity that is transmitted from generation to generation except when altered by intentional engineering – developing instead a fuller understanding of the germ cell life cycle as a form of dynamic maintenance undergirding the stability of the material basis of continuity. Any human technologies that impinge on this continuity should be undertaken with special care. To devise a framework for germline responsibility, we must systematically embrace both the genome editing debate’s certainty that the germline deserves special consideration and protections, and the push from transgenerational epigenetics and other emerging biosciences to see DNA sequences and their integrity as ongoing processes embedded in cellular contexts. As such, germlines are susceptible to heritable alteration ensuant to a wide range of human technological activity, not just those transformations undertaken on purpose.

In outlining a synthetic framework of germline responsibility, we proceed as follows. First, we briefly review the biology of germ cells and the germ cell lifecycle, discussing the relationship between the material processes of germ cell development and the concept of the germline. Second, we draw on the literature of heritable genome editing to examine existing moral frameworks specific to the germline. Third, we analyze the evidence for germline-mediated transgenerational harm from synthetic chemicals with an emphasis on the continuity between genetic and epigenetic maintenance of genome stability over time. Finally, we show how thinking about heritable genome editing and environmental germline toxicants together offers direction to future-based ethics that apply to nonintentional interventions in the germline.

2. Germ Cells and the Germline

What is a germline?

Today, the germline is understood as the vehicle of heredity by which individuals of one generation give rise to the next generation. The term refers to a process: eggs and sperm fuse to form zygotes; zygotes differentiate to become embryos – embryos that already in the early days of life set aside the germ cells that will in the future adult organism generate eggs and sperm. Repeat. Following identification of the DNA double helix as the structural carrier of genetic information from one generation to the next in the 1950s, the human germline became synonymous with genetic inheritance, the means by which individuals inherit genetic traits from both parents and pass them on to their offspring.

Nonetheless, despite iconography that shows it as such, DNA does not exist naked in abstract space, handed on intact from generation to generation like a baton in an endless relay race. Indeed, imagine that this genome-baton must be painstakingly remade, incessantly, at a molecular level, in order to remain the same; repaired at every instant due to constant insults from metabolically produced free radicals or ionizing radiation; and carried across the vast (to a cell) expanse of the gastrulating embryo to find the location that will become an ovary or testis. Even passing on the baton is a long and complicated process involving many actors, as it must be divided exactingly in half to generate multiple copies of it, each with the same genes but different alleles. In other words, the apparent endlessness, continuity, immortality, or replication of the germline all paradoxically involve constant dynamic change to persist as the same thing over time, the very essence of heritability. Thus, the germline is better understood and treated as a process than a thing.¹⁵ A series of complex inductions and becomings give rise to the more familiar and reassuringly discrete cellular objects that we know as gametes: sperm and eggs.

The fertilized egg – the human zygote – through cell division becomes a blastocyst, containing an inner cell mass of pluripotent embryonic stem cells. As the blastocyst goes through gastrulation (about 15 days after fertilization, in humans), a subset of these pluripotent cells is induced by egg yolk sac signals to migrate across the dividing, growing embryo to the location of the future gonads. These are called primordial germ cells (PGCs). Primordial germ cells undergo further specification, then multiple rounds of mitosis (cell division), and eventually meiosis, a form of division unique to the process of making gametes. Meiosis involves genetic recombination – crossing-over – to generate new combinations of the parental genomes in each chromosome before a second stage that whittles down the cell’s genetic content to one copy of each chromosome rather than the usual two of each.

This whole developmental process is remarkably long. In developing human female embryos, oocytes in the ovary start meiosis, but then enter a phase of arrest that lasts until puberty. Many years later, at ovulation a few oocytes mature during each menstrual cycle, but meiosis is only fully completed after fertilization. In human males, primordial stem cells differentiate during embryonic development into spermatogonial stem cells. Located in the testes, these stem cells start producing spermatozoa at puberty. Spermatogenesis is then continuous during an individual’s life with each cycle of production lasting approximately three months.

The epigenome is central to germ cell specification and fate

Described this way, it sounds easy. Cells divide, they migrate, they get specified. Yet the actions that are usually cavalierly ascribed to DNA during reproduction and development – replication, crossing-over, gene expression, base excision and repair – depend on the epigenome, a set of molecules that collectively regulate gene expression, gene replication, and genetic repair. If the epigenome is altered, so is the cell’s capacity to divide, repair, or differentiate. These molecular modulations include (1) the modification of cytosines in the DNA sequence by methylation, changing its openness to transcription or repair; (2) processes that add small chemical groups to the histone proteins essential to the structure and configuration of chromatin, the complex of DNA and proteins that make up chromosomes; and (3) small RNA molecules that play a variety of roles in modulating gene expression. More recently, it has been proposed that lipids (fat molecules) present in the zygote can carry molecular non-DNA encoded information from gametes to offspring by patterning the distribution and mobilization of epigenetic changes during epigenome ‘reprogramming,’ in which most parental epigenetic marks are removed from the embryo genome and new ones are established.¹⁶

Maintenance and repair in germline continuity

At each of the points in the processes described above, there is possibility for error coupled with robust mechanisms evolved to detect and fix it. Cells of the germline, like other cells, suffer errors in DNA replication or chromosome segregation upon mitotic division, and additional slipups when they divide into gametes during meiosis. As appropriate to their reproductive role, germ cells have excellent systems of DNA surveillance and repair and respond to such errors with DNA repair mechanisms or apoptosis (an ordered form of self-destruction). Mature spermatozoa are the exception, not able to perform DNA repair due to high nuclear compaction, which is left to be completed by the oocyte after fertilization.

The massive epigenetic changes in germ cells and the processes of repair described above also represent a source of vulnerability that can lead to genome instability. The epigenome silences genome regions whose expression might be harmful to the organism, such as those encoding transposable elements. Such silencing is temporarily lifted as the epigenome resets. In addition, epigenetic marks need to be correctly reinstated when reprogramming ends. In other words, replication accuracy through cell division, or the suppression of the genome’s accumulated selfish mobile genetic elements is an intensive and incessant process of surveillance, maintenance, and repair. This germline process allows generations to proceed with relative genetic fidelity one from the other, a process that has a normative as well as a biological meaning.

3. The Germline as a Moral Barrier

Germline vs. soma

The understanding of the germline as the vehicle of heredity in which genetic changes impact all further generations has meant its distinction in moral consideration. The germline was seen as the primary site of intervention by twentieth-century eugenicists.¹⁷ After eugenics was discredited, the germline acquired an operational ethical meaning as a moral line or boundary not to be crossed by human engineering, in addition to its biological meaning as a cell lineage. A distinction between the germline and the soma was formalized in recombinant DNA technology policy in the United States.¹⁸ Genetic engineering of the soma was rendered acceptable by the understanding that, unlike heritable germline interventions, onward transmission by cell division would end with the death of the individual whose somatic genome was edited.

The strength of the germline as a moral barrier has notably eroded in recent years in the wake of CRISPR-Cas9 genome editing and the perception that “off-target” effects are a largely controllable safety issue.¹⁹ Nonetheless, something of the moral distinction between engineering of germline and soma remains essentially uncontroverted in assumptions of gravity of the former and acceptability of the latter. In terms of the argument we make here, what matters is that the heritability of germline editing interventions is still seen as the defining biological and moral difference distinguishing such interventions from actions on the soma. Such actions raise distinctive questions for human conduct because of their cross-generational impact. There is a consensus that human heritable genome editing (HHGE), even assuming that it will be found ethically permissible, calls for a particular kind of scientific and regulatory responsibility²⁰ and will warrant additional justifications and safeguards, by comparison to somatic genome editing. Key considerations highlighted in the literature fall into three main articulations of responsibility: for not harming future people, for the rights of individuals whose genome is impacts, and for social implications and inequality. We discuss these now in turn.

Responsibility for not harming future people

HHGE carries “the possibility of off-target alterations, as well as on-target events that have unintended consequences,” risking physical harm to individuals in the future.²¹ This literature highlights germline rather than somatic cell-associated risks exactly because modifications to germ cells manifest in every cell that descends from them and therefore the whole body, meaning modifications could have dramatic health impacts. The unknown stability of HHGE-engineered modifications after several generations and their effects in the context of changing environments also uniquely concerns the germline: a modification that was initially beneficial could become detrimental in a different environment. These risks of unintended off- and on-target effects are a key moral consideration weighing against HHGE according to the value of nonmaleficence, in other words, safety.²² While, according to a risk/benefit analysis, “[t]he mere fact that there are risks does not show that we should reject [HHGE] … for it is sometimes worth taking a risky course of action if the potential benefits are great enough,”²³ conducting HHGE requires that off-target effects remain minimal.

Responsibility for the rights of individuals whose genome is impacted

The consequences of germline interventions affect people in the future rather than people in the present. Frameworks centered on existing individuals whose autonomy and interests are taken as touchstones to assess the ethicality of a biomedical intervention are little suited to address future persons. By definition, future individuals cannot grant or refuse their consent to past editing that will persist in their genomes and lives.²⁴ While this conundrum is raised by other reproductive technologies, HHGE’s impact goes beyond immediate offspring to all subsequent generations. Certainly, parents regularly make therapeutic decisions on behalf of future children. Some argue that there may even be a moral obligation for intending parents to technologically ensure their child has the best possible health.²⁵ However, the distinction between therapy and enhancement on which justification of such interventions relies is far from clear-cut.²⁶ In the case of HHGE the criteria that could justify such a decision become particularly elusive. Long-term consequences of germline genome interventions are unpredictable, complicating efforts to calculate benefits and risks and undermining the ability of intended parents to provide informed consent.²⁷

Responsibility for social implications and inequality

Even if intended as an individual therapeutic tool, HHGE’s impacts are far-reaching in a way that makes them ethically questionable. As noted, any heritable modification has the potential to become widespread in the population, implicating the collective human genome. Moreover, HHGE would have far-reaching social effects. Unequal access to costly technology is likely to reinforce reproductive inequalities, while dominant values may be promoted by the technology at the expense of others. Enabling pre-selection in reproduction, HHGE could support eugenic goals and increased intolerance for human variation.

Proponents of HHGE address this objection by rehearsing distinctions between “bad” and “good eugenics,” first proposed in the context of debates over genome engineering in the 1970s.²⁸ Only when forced upon an individual by a government or third party, they argue, would eugenic practices be unethical; HHGE could be ethical if preserving individual procreative autonomy and therapeutic beneficence.²⁹ Critics underline the limitations of this line of argument: even if not directed by a contestable social ideology, HHGE affects communities and societies in other ways, as these technologies shape understanding of the acceptable range of variation for being human.³⁰ Differences that can be genetically corrected are more likely to appear as defects that should be corrected, leading to reduced social acceptance of and support for people with disabilities. Thus, the technology may substantially, albeit indirectly, harm people other than those whose genome were edited.

In sum, even with internal differences, a consensus may be identified in these discussions, that the issues raised by germline genome editing mark novel areas of specific moral significance that require careful consideration allied with regulation and oversight. It is all the more striking then that this reasoning has not been applied to human technological activity that causes non-directed germline genome modification, the aim of the next section of this paper. It has been difficult to think across these arenas exactly because of distinctions between germ and soma and between genetic and environmental processes. Such distinctions are actively challenged by the concept of the processual germline outlined above, and the material evidence of its capacity for irrevocable genetic alteration by environmental toxicants detailed below. The ethical principles developed around HHGE we just enumerated oblige us to pay special attention to present chemical technologies that present collective germline-mediated future harms, currently an arena of extremely active investigation in the biomedical sciences.

4. Reframing Germ Cells and the Germline: Repair, Stability, and the Capacity for Ongoing Integrity

From germ cell immutability to vulnerability

Ethical discussions of HHGE largely assume a view of the human genome as a stable or fixed entity unless altered by genetic tools invented by life scientists. Such assumptions are an understandable legacy of decades of theorization of the secluded and inviolate nature of the germline. For August Weismann and many who followed, the germline was embodied in specific material particles that gave body to heredity, constituting a continuous flow from generation to generation untouched by the life, experience and death of the somatic bodies it passed through.³¹ The resulting dogma of the “Weismann barrier” sees the germline as a set of instructions sufficient to make new organisms with each generation, transmitted in a form impervious to environmental influences.

However, as demonstrated in section 2 above, the germline is a process of constant maintenance, which shows the integrity of genetic information to be ongoing dynamic work. Recent research has begun to reframe the germline in terms of processes of DNA repair, genome stability, and epigenome integrity. Such processes underlie both a capacity for adaptation and flexibility in cells constituting the germline, and periods of vulnerability in which germline alterations can occur and be carried forward. Here we work systematically through ideas of how environmental toxicants interfere with the dynamic maintenance of the germline, thereby permanently altering its material continuity such that physiological or health impacts are seen in the descendants of exposed individuals – much as intentional germline modification would be.

A scientific review would enumerate the evidence and assess its strength, but as analysts of the ethical significance of the research we instead narrate these findings in terms of their reframing of the germline as a work of dynamic maintenance, a process whose vulnerabilities link environmental toxicants to transgenerational health. The significance of this reframing lies in its understanding of human chemical technologies as agents of genomic change that are carried forward in the germline, in contradistinction to older ideas of toxicants exclusively as poisons or carcinogens that only directly cause death or damage to present bodies.³² Resilience and adaptability underlie the fact that a gamete’s loss of integrity from chemical exposure does not always preclude its ability to fertilize, and thus the possibility that non-lethal alterations may become incorporated in ways that impact later generations. The mis-repaired or destabilized genome of the gamete can shift in its “ability to program a normal pattern of embryonic development,” yet development proceeds.³³ While such changes are not inevitably pathogenic, viable but altered germ cells can result in offspring that suffer health consequences or diminished capacities for reproduction.

Environmental exposures that cause a raised propensity for mutation via genome instability

Beginning with the mutagenicity of X-rays almost 90 years ago, the unique susceptibility of germ cells to toxicants has been highlighted for decades.³⁴ In rodents dozens of substances have been characterized as germ cell mutagens, and extensive evidence points to human germline-mediated heredity of effects of at least four types of common toxic exposures: ionizing radiation, chemotherapy, smoking, and air pollution.³⁵ Tobacco smoking drives chromosomal aberrations and DNA damage in human sperm, and individuals whose fathers smoked before conception show increased DNA damage compared to children of non-smoking fathers.³⁶ Occupational exposure to diverse workplace chemicals or chemotherapy also drives genetic changes in the sperm of exposed men, with effects sometimes lasting years after exposure despite relatively rapid turnover of spermatozoa in the body.³⁷

Identifying mutations in specific genes has proved difficult, but next-generation sequencing has opened out an expanded view of the genome, driving a shift in perception of toxic exposures as causing increased mutation propensity overall. For instance, a study of nine families exposed to Agent Orange (dioxin) during the Vietnam War demonstrated an increased rate of new mutations in the genomes of their children born long after the exposure.³⁸ While there hardly needs to be any further indictment of war as harmful, this finding nonetheless underscores the point that human violence towards other living contemporary humans via military technology is at the same time violence against future humans descended from combatants and civilians exposed to those technologies. What is at stake here for these descendants is an abrogation of the mechanisms that subdue mutation rates.

In the broader literature, as in the example of dioxin, causal reasoning around the interaction of chemicals and genomes has expanded considerably from the idea that a toxicant must directly cause a mutation in an important gene to be legible as a germline mutagen. The expansion of the number of repeated sequences in the genome during replication events, or interference with DNA-repair mechanisms can generate genome instability, defined as the increased tendency of the genome to acquire mutations.³⁹ Thus, a toxicant exposure can change the capacity for maintaining DNA sequence fidelity as the genome goes through the unending processes of replication, division, and specification – described above as constitutive of germline continuity.

These forces of destabilization and the propensity for mutation are evident in the example of the Chernobyl accident in 1986. Exposure to ionizing radiation in men caused increased genomic instability in their children.⁴⁰ Compared to controls born in the area before the accident, children whose fathers had been exposed to fallout had a higher incidence of mutations in tandem repeat sequences. Also called minisatellites, these short, repeated sequences of DNA are found primarily in the noncoding genome. Many other exposures, such as paternal tobacco smoking, also cause noncoding genome mutations in offspring. Initially, such studies were dismissed precisely because of the noncoding status of the affected DNA, thought to be biologically inconsequential.⁴¹ However, tandem repeats are now understood to contribute to cancer and other disease risks exactly because they confer genomic instability.

Thus, we see a shift in the literature away from direct mutagenic effects in single genes, towards an understanding of toxicity that encompasses damage to the genome architecture itself, and the cellular maintenance structures and resilience tools sustaining ongoing genomic integrity. This view profoundly undermines previous understandings of the germline genome as essentially stable unless altered by intentional targeted engineering.

Transgenerational epigenetic inheritance

As described in section two, the epigenome plays an active role in choreographing the dynamic series of changes to cell migration, specification, meiosis and development that constitute the germline and give rise to germ cells over the life span. Transgenerational epigenetics is defined as “germline-mediated inheritance of epigenetic information between generations in the absence of direct environmental influences, that leads to phenotypic variation.”⁴² It is hard to talk about transgenerational epigenetics without using the term “controversial,” as it has been challenging to pin down clear physical mechanisms for germline-mediated inheritance of epigenetic information distinct from genetic transmission.⁴³

However, to re-invoke the premise we started with, our concern lies here with the idea that germline integrity is an ongoing process embedded in cellular contexts. What matters is not the difference between genetic and epigenetic mechanisms but their entwinement in the process that is the germline. The stability of the material basis of continuity can be compromised when that process is impinged upon by environmental conditions. For example, high blood glucose has been shown to suppress the making of key enzymes in the oocyte; because these are epigenetic enzymes that perform methyl group removal, the normal removal and reprogramming of the paternal epigenome after fertilization is suppressed.⁴⁴ Again, we see a situation in which the exposure (high blood glucose/diabetes) does not explicitly break something or become a physical mark in the germ cell, rather it downregulates the capacity of the oocyte to participate in demethylation after fertilization. This shifts the epigenome in the next generation, altering gene expression in the pancreas such that progeny express less insulin and are themselves glucose intolerant. In this experimental setting, because the oocytes were removed by egg retrieval, fertilized in vitro with sperm from metabolically normal male mice, and then fostered in a healthy, non-diabetic surrogate female mouse, the oocyte is the only material link across the generations.

In humans, the endocrine disruptor diethylstilbestrol (DES) is perhaps the most well-studied example of the multigenerational impact of human pharmaceutical technology use; it is estimated that there are 50 million living descendants of women treated during pregnancy with this synthetic estrogen between the 1940s and the 1970s.⁴⁵ Although the great-grandchildren of DES-treated women – in scientific parlance the F3 generation – are still young, studies of the transgenerational impacts of DES have been done in model organisms that indicate the impacts of these drugs will reach far past the already-documented harms of direct exposure. Mice exposed to DES at levels equivalent to these original therapeutic applications bear progeny whose descendants followed out over sequential generations demonstrate changed fertility, anogenital distance, and puberty timing, even in F3 and F4 generations.⁴⁶ In other words, these wide-ranging impacts reach into generations far beyond those directly exposed to the intentionally administered technology, which, it should not be forgotten, was explicitly meant to control pregnancy by preventing miscarriage.

Of particular significance for our analysis is the figure of the exposed germ cell found in this work on DES and other endocrine disruptors. Since the discovery in the 1970s that DES and Thalidomide could cross the placenta, it has become clear that an exposed embryo’s somatic cells will carry forward effects into the fetus and child that is formed from them by cell division.⁴⁷ It is harder to think about the exposed embryo’s germ cells, whose genome and epigenome participate in every generation going forward; yet herein lies the distinctiveness of intergenerational germline responsibility and its link back to the large body of ethical consideration of germline engineering. Transgenerational epigenetic changes resulting from germ cell exposure are not separate from genetic impacts. If DES or high blood sugar affects the methylation status of DNA or chromatin, those changes also affect the likelihood of genomic rearrangements at that site; in turn, genetic variation in the sequences encoding epigenetic enzymes will mediate the impact of any toxicant exposures.⁴⁸

Indeed, this genetic-epigenetic continuity is visible in studies of how environmental toxicants impact every step of the germline process. It has been difficult to study the earliest stages of primordial germ cell (PGC) proliferation and migration to the developing gonads since they occur during embryonic development. Yet it is likely a uniquely vulnerable period in terms of toxicant exposure, given the number of decisive changes establishing body plan and cell fate occurring then. Recently these stages were modeled with mouse embryonic stem cells cued to differentiate as PGC-like cells in vitro. Exposed to environmentally-relevant concentrations of the plasticizer bisphenol-A (BPA), these PGC-like cells showed increased rates of proliferation with higher levels of DNA damage, and BPA-specific changes to gene expression indicating a remapping of the epigenome, showing that germ cell development is susceptible to toxic exposure in multiple ways.⁴⁹

5. Ethics and Environmentally Mediated Harm in the Germline

Unintended vs unanticipated

It is time to bring the different pieces together. We have argued thus far that the human germline is a dynamic ongoing biological process. Extensive moral and philosophical consideration of genetic engineering of the human germline indicates that it is marked out for special ethical consideration compared to genetic engineering of somatic cells because of its unique constitutive role for future humans. Technologies such as endocrine-disrupting pharmaceutical drugs and plasticizers have by contrast not been considered in terms of such responsibilities, either by ethicists or by regulators. Indeed, there is a general assumption in risk assessment that chemicals that do not show mutagenic effects in somatic cells do not need to be tested on germ cells, or if they are tested in germ cells, then the endpoints detected in somatic cell genotoxicity tests (point mutations or chromosome aberrations) are also the ones that are most relevant to germ cells, despite the clear relevance of impacts on DNA repair mechanisms or epigenetic patterning of development.⁵⁰ This is a technological presentism focused only on the urgency of diseases suffered by the currently living (soma) that overlooks the future lives embodied in the germline. By showing that intentional and unintentional germline modification exist on a continuum, we have suggested that reflections prompted by HHGE can be applied to a wider range of human technological projects. Anthropogenic toxic effects in the germline should, in parallel to genome engineering, be conceived as ways of engineering the germline, albeit involuntarily and cumulatively.

Yet, one might object that understanding these technologies as actions bearing on the germline does not automatically entail that they fall within the scope of morality and responsibility. Moral responsibility applies to actions for which an agent is not only causally connected but also that he or she chose with cognizance. Genome editing actions are unequivocally intentional with regard to the germline, leading to evaluation of the actions of involved scientists in terms of moral responsibility, as in the case of He Jiankui.⁵¹ By contrast, the goal of using industrial chemicals is not to modify the germline, and even less the germline of a particular individual, but to make agricultural or industrial products. Ensuing damage to human and animal germlines is rarely traceable to a single action or agent and may accumulate through repeated exposure. Therefore, unintentional germ cell effects ostensibly fall into the category of unintended consequences for which no one can be held responsible.

However, this objection conflates an often-overlooked distinction between unintended and unanticipated consequences of technologies.⁵² While intention commits one’s moral responsibility, absence of intention does not automatically release one from all responsibility. Rather, cognizance of consequences that one should reasonably have known may result from one’s action is sufficient to ground moral responsibility.⁵³ In other words, if the impossibility of anticipating an effect may relieve one of moral responsibility, the mere failure to do so does not. Indeed, in the debate on HHGE, unintended off- and on-target effects are considered to fall within the scope of the agent’s responsibility. Put differently, two forms of anticipation need to be distinguished. Certainly, we cannot predict all specific long-term effects of our technologies via the germline, for example whether an endocrine disrupting pesticide will cause a pathology in an individual whose grandparent was exposed. This involves many other factors, including additional stressors experienced by this individual in her own lifetime. Nonetheless, we can expect our chemical and radioactive technologies to impact the germline and cause health effects in future generations, not only in light of the scientific evidence but also because the biological activity and lack of degradability of industrial chemicals, far from being accidental, are very much designed into these technologies.⁵⁴ From such technological interventions ensue “duties to exercise due care.”⁵⁵

Thus, heritable genome editing and heritable, environmentally mediated alterations of the germline are not as dissimilar as they might seem at first blush: in both cases, while harm to the germline is not intended, we can anticipate that human activities create that risk; therefore, agents bear moral responsibility for the consequences of their actions. The principles that delineate an ethical framework for HHGE discussed above can help us envision analogous frameworks for germline responsibility in the use of chemical and radioactive technologies.

Responsibility for not harming future people

The failure to include in moral reasoning the unintended consequences of technologies on future humans is well illustrated by current uses of the insecticide dichlorodiphenyltrichloroethane (DDT) against malaria. DDT was banned in many countries in the 1970s for its reproductive toxicity, a fact paramount in Rachel Carson’s denunciation of its toxicity.⁵⁶ However, some countries continue to employ it as a low-cost mosquito control measure in malaria prevention. Such decisions seem the lesser of two evils: in at-risk regions, health risks from malaria are thought to outweigh those stemming from low-dose DDT exposure. Yet this calculus focuses on benefits for present human beings and fails to include DDT’s transgenerational toxicity.⁵⁷ Ethical reflection on HHGE clearly indicates that the risks that unintended and indirect consequences represent for future generations should be considered in assessing the moral permissibility of technology use.

Responsibility for the rights of individuals whose genome was impacted

Second, ethical reflection concerning HHGE has highlighted the moral difficulty posed by the inherent impossibility of securing consent of future individuals affected by germline interventions. The same issue of accountability occurs for individuals predisposed to disease by their ancestors’ exposures, even before their own life begins.⁵⁸ Such individuals are left with a chemical legacy, a fundamentally unidirectional relationship with no accountability. One might object that the need for accountability diminishes as time passes, with inherited harm from toxic exposure becoming presumably attenuated as generations are more removed from the initial event. This reasoning is at the core of the judicial concept of victim attenuation that justifies dismissal of liability.⁵⁹ However, toxicants and stressors may, through the germline mechanisms discussed above, manifest in the third or fourth generation even when they do not cause detectable harm to the exposed adult or fetus.⁶⁰ A low dose that is not toxic to a whole organism might cause lesions in germ cells that will affect the development of new organisms. Far from lessening with time, effects might expand in future generations, a harm for which future-oriented frameworks of accountability should be developed. Ethicists working in the context of HHGE have underlined the necessity of building institutional, legal, and financial infrastructures that ensure health follow-up and accountability beyond the time of technological intervention.⁶¹ A similar framework could address the misalignment between the long biological memory of latent germline effects and short individual and institutional temporalities.

Responsibility for social implications and inequality

As in considerations of HHGE, concern for the effects of chemicals in the 1970s emphasized the collective risk posed to the “shape of the future”⁶² or “the Future of Man” through the alteration of germlines.⁶³ Environmental reproductive justice scholars argue that environmental harm is unequally distributed along social and racial lines.⁶⁴ Accordingly, germline effects of toxic exposure are not just collective, but disproportionally impact underserved communities via environmental racism, and reinforce and reproduce inequalities into the future. Residents of Chemical Valley, site of a large petrochemical and chemical industrial complex on the St. Clair River in Canada, have experienced many reproductive harms, while fish and wildlife in this ecology also show many signs of endocrine disruptor burden.⁶⁵ The Aamjiwnaang First Nation who live there face pollutant exposure directly rooted in colonial history and shaped by current socioeconomic inequalities. As M. Murphy has remarked, “the sex ratio effects experienced by the Aamjiwnaang nation may be the latent response of exposures two generations ago. Or it could be the effect of continuous, multiple, accumulated, multigenerational exposures crossing a threshold that has not become an epistemologically legible measure.”⁶⁶ Through germline effects, injustice in the present reverberates through time.

Disability and indigenous studies scholars have warned that conceptualizing the effects of reproductive toxicity as unequivocal “harm” or “damage” is a research practice that leads to the reification of difference and the reduction of individuals and communities to these experiences, ultimately repeating violence through research.⁶⁷ Such damage-centered frameworks reveal implicit eugenic norms that govern assessments of interventions on the germline, yet the challenge remains of addressing transgenerational injustice and violence while acknowledging lives that differ from the norm. Above, we foregrounded the need to maintain capacities for repair as an important shift in perspective away from the enumeration of specific mutations. Following the lead of philosopher Kyle Whyte’s emphasis on the importance of Indigenous communities’ capacity for “collective continuance,” the perspective presented here highlights the responsibility to care for the deep and multilayered means of repair and adaptation that characterize reproduction in living systems as well as social resilience to environmental challenges.⁶⁸

6. Conclusion: A Framework of Responsibility for the Germline

We have argued, through a series of sequential steps, that ethical principles developed in considering human germline engineering should be extended more broadly to encompass radioactive and chemical technologies such as pesticides and plasticizers that cause genetic changes through promoting genome instability, increased mutation propensities, and decreased capacities for gene repair. While viewed in the past as a stably transmitted set of genetic instructions impinged upon only by intentional human genetic engineering technologies, we propose an updated concept of the germline and its constituent genome as a scene of constant maintenance whose ongoing fidelity through time depends on an interrelated suite of mechanisms incessantly active in genome and epigenome replication, genetic repair, chromosome segregation, cell differentiation, and cell division. Even mechanisms of genetic novelty, such as the crossing-over that occurs during meiotic gamete formation, are subject to a complex, robust and incessantly active mismatch repair system. Novelty occurs within tightly constrained parameters that ensure that chromosomes can pair and segregate properly such that the next stage in the germline can go forward.⁶⁹ Abrogation by human technologies of those processes of constancy in material continuity, whether intentional or non-intended, will manifest as changed life and health patterns in future generations within lineages, and disseminate between lineages through human populations via sexual reproduction.

On these grounds, we contend that it would be a conceptual error to separate epigenetic from genetic components of the maintenance or alteration of the germline across time and generations, whether one is addressing the biology or the ethics of the germline. While it has been argued that epigenetic facts as compared to genetic facts give rise to new forms of distinctly collective responsibility for environmental pollutants,⁷⁰ or that germline is “blurred” as a point of ethical consideration because epigenetic heritability is distributed through bodies and environments outside of genomic elements,⁷¹ these are only partial accounts of ethical implications of still-nascent epigenetic knowledge. This is not a choice between an epigenetic account of how life works and another genetic version of it. Skeptics, on the other hand, have declared that “epigenetics changes nothing,” and scientific evidence cannot “unilaterally necessitate changes in ethical theory or policy recommendations.”⁷² Neither of these positions captures what we have endeavored to demonstrate: that already existing ethical theory bearing on the human germline should be applied in cases of both HHGE and environmental germline toxicants. Our point is not that the scientific evidence has changed, but that the concept is evolving. Today’s germline concept is capacious enough to hold the genetic and epigenetic perspectives together as interacting elements in the maintenance of biological stability. Newer insights into the epigenetic control of DNA repair or the role of toxicants in genome instability do not blur, rather they sharpen the germline as an aspect of human life necessitating moral care and regulatory vigilance, a set of principles that have already been developed with respect to genome engineering. The germline was originally and has always been a historically and culturally situated concept; this concept is undergoing change in the wake of genome editing technology and embryo models and the epigenetics of DNA repair, as well as Agent Orange, DES, and widespread tobacco use, and it now incorporates a much wider set of processes and determinants.

We have argued here that these shifts in the germline concept have led to a double standard in moral considerations of the germline and social distribution of responsibility for it. While concerns for the germline are regarded as paramount when debating or regulating genetic engineering technologies, they rarely appear in considerations of chemical and radioactive risk. It is not that new scientific findings push us to new ethical positions, but rather the reverse: explicit consideration and an ethics of care regarding future persons in decisions about the social acceptance and regulation of technologies that irreversibly alter the human genome is a well-developed position with practical correlates that should be updated and more broadly applied. Harm to living people from carcinogenic and endocrine disrupting pollutants is evident; the germline constitutes the material ramification into the future of the harm caused by these alterations. While being careful not to reify notions of purity or integrity, calling for the right of individuals and populations to safeguard their means of resilience and repair in the face of anthropogenic chemicals that undermine those resources seems a logical next step in the work of bioethics to answer fundamental questions of how we should live, and what kind of care for the future might be implicit in our ethical and regulatory stance toward germlines embodied in people here and now.