Crossroads between transposons and gene regulation

Genomic cytosine methylation is a major mechanism to transcriptionally silence transposable elements across eukaryotes. Therefore, transposable elements are expected to evolve mechanisms to evade methylation to remain active in the host genomes. However, the discovery of several retrotransposons that have hijacked DNA methylation through the acquisition of host genes challenges this expectation. In one case, three distinct retrotransposon classes have acquired cytosine DNA methyltransferases, most spectacularly in the genomes of Symbiodinium dinoflagellates, where hundreds to thousands of copies of these retrotransposons are likely to have influenced the host’s exceptional epigenomic landscape. In another case, a group of arthropod retrotransposons have acquired a functionally active Methyl-CpG Binding Domain. This domain guides the retrotransposon integration into methylated CpG-rich regions, which correspond to repetitive regions of the genome, thus likely decreasing the potentially harmful effects of new insertions. These illustrate the convoluted arms race between transposable elements and their host genomes. Both cases exemplify the delicate balance between expression and silencing that retrotransposons need to achieve in order to co-exist with their host. These serve as the first examples of how transposons can domesticate host genes central to epigenome regulation to their own advantage.

Professor Wysocka’s team is interested in understanding how TEs can serve as a substrate for evolution of novel regulatory elements and functions in the primate lineage. During human pre-implantation development many TEs escape silencing and are transcribed at discreet stages, with most stage-specific regulation occurring at the class of LTR retrotransposons called endogenous retroviruses (ERVs), which are remnants of ancient retroviral infections. One such ERV class, called HERVK, retained coding potential and gives rise to the viral-like particles and proteins that are readily detected in human blastocysts and can influence innate immune response. The team discovered that HERVK LTR elements, called LTR5HS, function as ape-specific enhancers and regulate expression of nearly 300 human embryonic genes. A flip side to TEs serving as a playground for regulatory innovation is that mobile transposons are parasitic elements that pose an ongoing threat to the genome, and thus must be controlled by the host cell. The non-LTR retrotransposon LINE-1 (L1) is the only currently mobile autonomous TE in humans. In collaboration with the Bassik lab at Stanford, the team performed first genome-wide screen for regulators of L1 retrotransposition and identified ~150 human genes that either activate or repress L1 retrotransposition in human cells. Through this screen, the researchers discovered a novel mechanism dedicated to transcriptional silencing of evolutionarily young, full-length L1s immersed within transcriptionally permissive euchromatic environment, and showed that this silencing pathway also has a collateral effect on the host gene expression programs. Professor Wysocka will discuss these findings and talk about her team ongoing work.

Approximately half of our genome consists of repetitive DNA sequences, mainly contained in transposable elements (TEs). TEs are retrovirus-derived DNA elements which copy-pasted themselves throughout our genome during vertebrate evolution. TE invasions have been a major driver of genome evolution but it remains elusive to which extent TEs have influenced how genes in the human genome are regulated. In previous work (Jacobs et al, 2014; Nature) Dr Jacobs’s team showed that KRAB zinc finger genes (KRAB-ZNFs) in our genome are in a continuous battle against invasions of TEs, revealing how the human genome is actually in a war against itself. His lab currently investigates how the ‘evolutionary arms race’ between TEs and KRAB-ZNFs has re-shaped gene-regulatory networks involved in human brain development. The team find that TE invasions provided our genome with novel gene-regulatory elements, adding an extra level of complexity to how, where and when neuronal genes in our genome are switched-on or -off. In addition, the researchers find that evolutionary changes in KRAB-ZNFs, provide them not only with the capacity to recognise and repress TEs, but also results in them recognising gene promoters. This suggests that both TEs and KRAB-ZNFs have added new primate-specific layers of gene regulation, repeatedly innovating gene expression networks throughout primate evolution. Consistent with this concept, the team presents one of these KRAB-ZNFs which initially evolved to repress TEs, but now has become a modulator of genes central to brain development. The researchers’ work shows that the impact of TE invasions lasts long after the TE has lost its capacity to replicate, which ironically, is largely due to our own genome’s defense mechanism to protect our genome from uncontrolled spreading of TEs.

In mammals, the terminally differentiated sperm and oocyte fuse to create a totipotent zygote upon fertilisation. The mechanisms underlying the epigenetic reprogramming towards totipotency that follows fertilisation are not fully understood, and the molecular features of totipotent cells remain scarce. Embryonic cells remain totipotent only for a restricted time window. During this time, embryonic cells are characterised by an atypical chromatin structure and reactivation of specific families of retrotransposons. Recently, it was reported that totipotent-like cells arise in ES cell cultures in vitro. Like in the embryo, these cells are characterised by the expression of MERVL LTR retrotransposons. To address how the expression of these elements is regulated during the transition between totipotent and pluripotent states, Dr Torres-Padilla’s team first examined histone modifications and chromatin structure in early mouse embryos. Remarkably, the researchers have found that specific features of embryonic chromatin are also present in totipotent-like cells in vitro. Based on this analysis, they have begun to decipher key molecular regulators of repetitive elements in the embryo, and how they contribute to shaping the regulatory programme of the newly formed embryo. The team’s results have identified candidate proteins that regulate chromatin function and expression of these elements and show that they can induce totipotency. The researchers are currently examining the role of these molecules in sustaining totipotency in the embryo. Dr Torres-Padilla will present her team’s latest results that reveal a new role for chromatin integrity in promoting epigenetic reprogramming and sustaining molecular features of totipotent cells.

A substantial proportion of the genome of many species is derived from transposable elements (TEs). Moreover, through various self-copying mechanisms, TEs continue to proliferate in the genomes of most species. TEs have contributed numerous regulatory, transcript and protein innovations and have also been linked to disease. However, notwithstanding their demonstrated impact, many genomic studies still exclude them because their repetitive nature results in various analytical complexities. Fortunately, a growing array of methods and software tools are being developed to cater for them. Dr Bourque will present a summary of computational resources for TEs and highlight some of the challenges and remaining gaps to perform comprehensive genomic analyses that do not simply 'mask' repeats. As an example, Dr Bourque will also describe a tool called PopSV that was developed to detect copy number variants (CNVs), which are known to affect a large portion of the human genome and have been implicated in many diseases. Although whole-genome sequencing (WGS) can help identify CNVs, most analytical methods suffer from limited sensitivity and specificity, especially in regions of low mappability. Dr Bourque’s team demonstrate that the PopSV calls are stable across different types of repeat-rich regions and validate the accuracy of our predictions using orthogonal approaches. Applying PopSV to 640 human genomes, they find that low-mappability regions are approximately 5 times more likely to harbor germline CNVs, in stark contrast to the nearly uniform distribution observed for somatic CNVs in 95 cancer genomes.

Transposons form more than half the human genome, and several human diseases have been associated with aberrant activity of transposable elements (TEs) via a variety of mechanisms. Aberrant transposon activity has been shown to induce mutations, alter the regulation of adjacent genes, and produce toxic and/or immunogenic proteins. Despite the importance of TEs in human health and development, reads derived from these regions are frequently ignored in most sequencing data analysis protocols due to the difficulty in properly assigning TE-derived reads to the correct locus of origin. Compromises must often be made between certainty about the exact genomic locations of active TEs and sensitivity to the youngest and most potentially dangerous elements. To address this problem, the Gale-Hammell lab has been systematically creating computational approaches to probabilistically handle reads from repetitive regions in genomics datasets: the TEtoolkit. This has allowed for the identification of new factors contributing to TE silencing as well as new contexts in which TEs are considerably more active than expected. In particular, there is now mounting evidence that TEs contribute to neurodegenerative disease.

Transposable elements (TEs) are more mobile genetic units that have been found to play key regulatory roles in gene expression and impact eukaryotic evolution. Sequencing a new genome of any organism has always been accompanied with annotation of both genes and repeat elements like TEs. But, due to the mobile nature of TEs, there have been constant efforts in discovery of non-reference TEs by resequencing genomes especially with the advent of long read sequencing. However, there is a case to be made to move beyond TE discovery towards transcriptional annotation. It is common for genes to have annotated transcriptional start sites, stop sites and exon-intron boundaries whereas TEs are annotated as blocks of DNA without any transcriptional information. This talk will focus on the loss of information and bioinformatic artefacts that arise because of lack of TE transcriptional annotation. In his research Dr Panda has created a transcript-based functional annotation of Arabidopsis TEs using Nanopore full-length cDNA sequencing in a number of TE activated strains. He will show that characterising TE transcripts improves our investigation of TE RNAs by reducing the complex multi-mapping of short reads. This annotation also enables other genomic studies, for example, understanding how TE integration affects regional mRNA production.

Rice is the staple food crop for more than half of the world’s population. Many factors are now jeopardizing a sustainable production of this cereal, among which, population size increase, environmental changes and reduction of arable land caused by cities expansion. As a consequence, rice has been the focus of intense research over the last decades, which lead to the development of large genomic resources such as several high quality genome assemblies for cultivated rice and several of its wild relatives, genome sequences for 3,000 rice varieties and transcriptomic data for many developmental stages and various physiological stresses. Therefore, the genus Oryza (to which cultivated rice belongs) is now one of the best models to study genome evolution using full scale genomic analyses. We took advantage of these resources to address the question of the impact of transposable elements (TEs) on the structure and evolution of this plant genome. Here, we show that TEs contribute to genomic turn-over at a rate which is much higher than in animal kingdom. We then looked at the transpositional landscape of cultivated Asian rice, taking advantage of the availability of the 3,000 genomes dataset. This resource provides the opportunity to study the impact of TEs at the species level. We show that retrotransposons (the major TE type in plants) are very active in this crop and contribute to genome diversification in agro. Our results suggest that one could tentatively search for TE insertion polymorphisms that may have been adaptive in the recent history of rice cultivation.