Assessing the impact of endogenous retroviruses on trophoblast gene regulation

Throughout evolution, retroviruses invaded virtually all mammals and became integrated within their genomes. Although these so-called endogenous retroviruses (ERVs) lost their ability to produce viral particles, for a while they could still self-replicate within genomes and be passed on to the next generation. Eventually, most became inactive and could not expand further. Currently there are no known active ERVs in humans and only a relatively small number remain active in mice. Nevertheless, ERVs make up ~8% of the mouse and human genomes.

It is thought that most ERVs serve no particular purpose. However, a small subset has been 'domesticated' by the species that host them to perform essential functions. Some have turned into genes, which produce proteins involved in, e.g., immunity and placental development. Others became important regulators of existing genes. That means that certain ERVs can determine whether a given gene is 'on' or 'off'. However, only few examples of this are known to date, and it remains unclear how important ERVs are for the regulation of gene activity.

One particular place where ERVs seem to be important is in the placenta. Placental shape and structure varies extensively between different mammals, and it is thought that these differences might come, at least in part, from the influence that ERVs have on gene activity. In this proposal we aim to test this hypothesis by comparing how ERVs affect gene activity in the placenta of mice, rats and humans. First, we will look for differences in the location of ERVs between these species and how it correlates with gene activity in the placenta. This will tell us which ERVs are likely to be important. Next, we will test for causal relationships between ERVs and gene activity by manipulating selected ERVs with molecular tools. We will take several approaches, first inactivating multiple ERVs simultaneously, then removing individual ERVs from the genome. We will test whether these experiments resulted in genes turning 'on' or 'off' and ultimately, any change in the behaviour of placental cells. This proposal will provide clues as to whether ERVs may have played a role in placental evolution, and potentially impact on reproductive success.

Endogenous retroviruses (ERVs) comprise around 8% of the human genome. Increasing evidence exists that ERVs can confer advantages to their host, either as co-opted genes or as gene regulatory elements. One of the most promising tissues where ERVs are likely to play a role is the placenta, where global DNA hypomethylation and its transient nature make trophoblast cells an ideal environment for ERV activity. Apart from the co-option of ERVs as fusogenic proteins, there are a number of examples where ERV-derived non-coding sequences are used as placenta-specific gene regulators. Placental structure is strikingly varied between different mammals, and it has been suggested that the large inter-species differences in ERV composition have contributed to the rapid evolution of the placenta by providing gene regulatory modules. However, evidence for this hypothesis remains scarce. Here we will take a cross-species comparative approach to test for roles of ERV-derived sequences in establishing gene regulatory networks in trophoblast. We will identify putative gene regulatory ERVs in trophoblast from mice and humans by profiling trophoblast stem cells (TSCs) as well as human primary cytotrophoblast. To gain a more detailed evolutionary view of the contribution of ERVs to the regulatory landscape of the mouse placenta we will also profile TSCs from several rodent species that diverge with respect to key ERV subfamilies. We will then perform extensive genetic and epigenetic editing experiments in TSCs to establish causal relationships between species-specific ERVs and gene expression. This proposal will reveal to what extent ERVs have driven essential gene regulatory innovation in trophoblast, providing a wider appreciation for the contribution of ERVs to placental evolution. As the impact of ERVs on organismal fitness become clearer, so will their potential involvement in human pathologies, such as those associated with placental dysfunction.

There is a current expanding interest in understanding the impact of ERVs across a variety of species and different cellular contexts. The genome-wide view provided by high-throughput sequencing and the power of genetic and epigenetic editing tools have led to a renewed appreciation of the potential roles that ERVs have in genome function, in both health and disease. Our work will constitute a deep functional assessment of the impact of ERVs on the evolution of the placenta, a vital extraembryonic tissue that supports fetal development. Correct establishment of the feto-maternal interface is essential not only for reproductive success, but also for long-term adult health.

Around 1 in 4 pregnancies end in miscarriage, and the fraction of these that is caused due to placental dysfunction is likely underestimated. As recently shown by Hemberger and colleagues, in a survey of mouse knock-out lines, almost 70% of utrauterine lethality was associated with placental dysmorphology, with early lethality almost always associated with placental malformation (Perez-Garcia et al. 2018 Nature). Notably, placental phenotypes were also frequently associated with defects in the development of the brain, heart and vasculature. Beyond mutations in coding regions of the genome, regulatory regions may also impact on placental phenotypes. The work proposed here will delineate the role of ERVs in placental gene regulation and as such, their potential to cause disease if aberrantly regulated or mutated.

In the cases where a poorly functioning placenta does not result in fetal demise, future health issues can ensue, as illustrated by the developmental origins hypothesis. Low placental perfusion, resulting in a reduction in nutrition and in utero growth restriction/small for gestational age infants are physiologically programmed to expect scarcity. As they age, these individuals are at an increased risk of metabolic disorders such as obesity, type II diabetes and heart disease. By revealing the mechanistic pathways that link aberrant genetic regulation to placental pathologies, our study will provide a greater understanding of how adaptation to conditions in utero may lead to adult disease.

The interplay between trophoblast cells and the maternal uterine decidua also plays a key role in reproductive success in later life. It was recently demonstrated that maternal ageing has a striking impact on the ability of the uterus to support pregnancy (Woods et al. 2017 Nat Comms). By evaluating the role of ERVs in the regulation of trophoblast gene networks and physiology, including regulation of cellular invasion into the decidua, we can contribute valuable knowledge on the genetic determinants of pregnancy outcomes throughout ageing.